Frankie Front 8/20/03 7:44 PM Page I

Home , Aellopos titan, Andira inermis, Camino (web browser), Chira Island, Curatella, Enyo ocypete

frankie front 8/20/03 7:44 PM Page i

BIODIVERSITY CONSERVATION in COSTA RICA frankie front 8/20/03 7:44 PM Page ii frankie front 8/20/03 7:44 PM Page iii

BIODIVERSITY CONSERVATION in COSTA RICA Learning the Lessons in a Seasonal Dry Forest

Edited by Gordon W. Frankie, Alfonso Mata, and S. Bradleigh Vinson

UNIVERSITY OF CALIFORNIA PRESS Berkeley Los Angeles London frankie front 8/20/03 7:44 PM Page iv

University of California Press Berkeley and Los Angeles, California

University of California Press, Ltd. London, England

Library of Congress Cataloging-in-Publication Data

Biodiversity conservation in Costa Rica : learning the lessons in a seasonal dry forest / edited by Gordon W. Frankie, Alfonso Mata, and S. Bradleigh Vinson. p. cm. Includes bibliographical references. ISBN 0-520-22309-8 (cloth : alk. paper).—ISBN 0-520-24103-7 (pbk. : alk. paper). 1. Biological diversity conservation—Costa Rica. 2. Forest ecology— Costa Rica. I. Frankie, G. W. II. Mata, Alfonso. III. Vinson, S. Bradleigh, 1938– QH77.C8 B56 2004 333.95′16′097286—dc21 2003000593

Manufactured in the United States of America 13 12 11 10 09 08 07 06 05 04 10987654 321

The paper used in this publication meets the minimum requirements of ANSI/NISO Z39.48-1992 (R 1997) ∞ frankie front 8/20/03 7:44 PM Page v

CONTENTS

Preface / vii 4. IMPACT OF GLOBAL CHANGES ON THE REPRODUCTIVE BIOLOGY OF TREES IN 1. INTRODUCTION / 1 TROPICAL DRY FORESTS / 38 Alfonso Mata and Jaime Echeverría Kamaljit S. Bawa

Part I. Biodiversity and 5. TROPICAL DRY-FOREST MAMMALS OF Ecological Studies PALO VERDE: ECOLOGY AND CONSERVATION IN A CHANGING LANDSCAPE / 48 Section A. Costa Rican Dry Forest Kathryn E. Stoner and Robert M. Timm

2. FLOWERING PHENOLOGY AND POLLINATION SYSTEMS DIVERSITY IN THE SEASONAL 6. THE CONSERVATION VALUES OF DRY FOREST / 17 BEES AND ANTS IN THE COSTA RICAN DRY FOREST / 67 Gordon W. Frankie, William A. Haber, S. Bradleigh Vinson, Kamaljit S. Bawa, S. Bradleigh Vinson, Sean T. O’Keefe, and Peter S. Ronchi, and Nelson Zamora Gordon W. Frankie

3. BREEDING STRUCTURE OF NEOTROPICAL 7. ECOLOGY OF DRY-FOREST WILDLAND DRY-FOREST TREE SPECIES IN INSECTS IN THE AREA DE FRAGMENTED LANDSCAPES / 30 CONSERVACIÓN GUANACASTE / 80 James L. Hamrick and Victoria J. Apsit Daniel H. Janzen frankie front 8/20/03 7:44 PM Page vi

Section B. Biotic Relationships with Other Part II. Transferring Biodiversity Costa Rican Forests Knowledge into Action: The Record 8. DIVERSITY, MIGRATION, AND CONSERVATION OF BUTTERFLIES IN NORTHERN COSTA RICA / 99 17. BIODIVERSITY INVENTORIES IN COSTA RICA AND William A. Haber and Robert D. Stevenson THEIR APPLICATION TO CONSERVATION / 229 Paul Hanson 9. WATERSHED ECOLOGY AND CONSERVATION: HYDROLOGICAL RESOURCES IN THE 18. CONFLICT RESOLUTION: RECOGNIZING NORTHWEST OF COSTA RICA / 115 AND MANAGING DISCORD IN Alfonso Mata RESOURCE PROTECTION / 237 Gregory A. Giusti 10. WHERE THE DRY FOREST FEEDS THE SEA: THE GULF OF NICOYA ESTUARY / 126 19. CONSERVATION AND ENVIRONMENTAL José A. Vargas and Alfonso Mata EDUCATION IN RURAL NORTHWESTERN COSTA RICA: LEARNING THE LESSONS OF A NONGOVERNMENTAL ORGANIZATION / 247 11. MANGROVE FORESTS UNDER DRY SEASONAL CLIMATES IN COSTA RICA / 136 Gordon W. Frankie and Jorge A. Jiménez S. Bradleigh Vinson

Section C. Biotic Relationships with Other 20. THE MEDIA AND BIODIVERSITY Geographical Areas CONSERVATION / 257

12. GEOGRAPHICAL DISTRIBUTION, ECOLOGY, Gilda Aburto AND CONSERVATION STATUS OF COSTA RICAN DRY-FOREST AVIFAUNA / 147 21. THREATS TO THE CONSERVATION OF TROPICAL DRY FOREST IN COSTA RICA / 266 Gilbert Barrantes and Julio E. Sánchez Mauricio Quesada and Kathryn E. Stoner 13. AN ULTRASONICALLY SILENT NIGHT: THE TROPICAL DRY FOREST WITHOUT BATS / 160 22. ENVIRONMENTAL LAW OF COSTA RICA: DEVELOPMENT AND ENFORCEMENT / 281 Richard K. LaVal Roxana Salazar 14. BIODIVERSITY AND CONSERVATION OF MESOAMERICAN DRY-FOREST 23. DISPUTE OVER THE PROTECTION OF THE HERPETOFAUNA / 177 ENVIRONMENT IN COSTA RICA / 289 Mahmood Sasa and Federico Bolaños Julio Alberto Bustos

15. PARQUE MARINO LAS BAULAS: CONSERVATION 24. THE POLICY CONTEXT FOR CONSERVATION IN LESSONS FROM A NEW NATIONAL PARK AND COSTA RICA: MODEL OR MUDDLE? / 299 FROM 45 YEARS OF CONSERVATION OF SEA TURTLES IN COSTA RICA / 194 Katrina Brandon

James R. Spotila and Frank V. Paladino 25. CONCLUSION AND RECOMMENDATIONS / 311

16. PROSPECTS FOR CIRCA SITUM TREE Gordon W. Frankie, Alfonso Mata, and CONSERVATION IN MESOAMERICAN Katrina Brandon DRY-FOREST AGRO-ECOSYSTEMS / 210 David H. Boshier, James E. Gordon, List of Contributors / 325 and Adrian J. Barrance Index / 327 frankie front 8/20/03 7:44 PM Page vii

PREFACE

he idea of producing this book resulted lished in Costa Rica, including the dry-forest area, T from several realizations for the editors in there had been no comprehensive assessments 1997. The first was an awareness that a large of whether these designated areas had been ef- body of biological research from major regions fective in protecting biodiversity. of Costa Rica was available in the literature. In We decided to focus on the seasonal dry for- particular, extensive research had been carried est in the northwestern region of the country out on the lowland Atlantic wet forest, the mid- because it lacked a review comparable to that of dle elevation cloud forest (1,400–1,800 m), and the lowland wet and cloud forests of Costa Rica. the lowland Pacific dry forest over a period of Furthermore, this type of forest was rapidly dis- about 30 years. Further, the wet forest had just appearing, in large part because it was so easily received a comprehensive biological review by converted to agriculture. Daniel Janzen estimated McDade et al. (1994), and a biological review of that only 2 percent of the original Middle Amer- the cloud forest was in progress (see later in this ican tropical dry forest remained. preface). Second, from the 1980s onward, many Building on extensive biological knowledge biologists and nonbiologists made enormous and modern trends for conserving biodiversity, human and financial investments to protect bio- we determined that the book should address diversity in all parts of the country, using mod- three main questions: What do we know about ern conservation approaches and methods. De- the biodiversity and status of the most promi- spite these efforts, little attention had been paid nent groups of plants and animals in the dry to assessing effectiveness of their work. Finally, forest? What have we learned biologically, so- in the more than 30 years that had passed since cioeconomically, and politically about conserv- the first national parks and reserves were estab- ing these specific groups? What do we need to

vii frankie front 8/20/03 7:44 PM Page viii

consider and do in the future to ensure im- In the current volume we chose to focus on proved conservation and protection of all bio- the lowland seasonal dry-forest region of Costa diversity in the dry forest, as well as other major Rica, which includes several related life zones, regions of the country? according to Holdridge et al. (1971). The book Several major publications have influenced represents the first effort to treat comprehen- our thinking in developing the conceptual goals sively the findings from a wide variety of plant for the book. One of the first workers to investi- and animal biologists investigating a highly gate Costa Rican life zones was Leslie R. Hold- seasonal tropical environment. We were also in- ridge of the Tropical Science Center in Costa terested in whether modern principles of con- Rica. Over a period of several years, he and his servation biology had been put into practice to colleagues intensively studied many plant for- study and conserve dry-forest biodiversity. Thus, mations and their environmental determinants. we asked the biological contributors to assess This work established an ecological foundation the status of the particular taxonomic groups that eventually led to a lengthy treatise on the life they studied and, where applicable, to generalize zone system (Holdridge et al. 1971). Daniel H. for the entire country. In several cases, this re- Janzen (1983) and numerous invited colleagues quest took some authors into other Costa Rican prepared an overview of the natural history of life zones and beyond into adjacent countries the country. His classic 1983 volume and its later of Middle America. Further, because biodiversity Spanish translation (1991) also provided exten- conservation by definition encompasses more sive lists of the plant and animal diversity found than just biology, we also invited several con- in several selected life zones countrywide. tributors to present socioeconomic, policy, legal, More recently, in the 1990s, large scientific and political perspectives on biodiversity conser- publications focused on compiling ecological vation to provide the important social contexts. papers from major life zones that have received Finally, all authors were asked to offer their per- extensive study. For example, McDade et al. sonal recommendations on future directions, (1994) presented a large series of edited papers policies, and actions to better conserve and pro- on the ecology and natural history of a lowland tect biodiversity. These recommendations form Atlantic wet forest, La Selva. This is the site the final synthesis chapter. where researchers associated with the Organi- The book begins with an introductory chap- zation for Tropical Studies first began to study ter, and the chapters that follow are divided into the biota intensively, in 1968; research at La Selva two parts. The final chapter presents conclu- has continued since that time. Nadkarni and sions and recommendations. Chapter 1 (“In- Wheelwright (2000) published a dozen edited troduction”) is concerned with the physical, bio- papers on the natural history and ecology of the logical, human, and conservation environment Monteverde Cloud Forest and its several life of the dry forest. To exemplify these features, zones. In addition, their book offers a limited the chapter focuses on two prominent regions, view of conservation issues. the Tempisque Valley and Peninsula of Nicoya A publication by Bullock et al. (1995) on the (see maps). In addition to the terrestrial envi- seasonal dry tropical forests of the world also ronments, these major areas are ecologically influenced the development of the current vol- connected with major riparian corridors from ume. Most papers in the Bullock volume are other ecosystems, coastal areas, and the marine concerned with botanical information, with only environment. limited coverage of the fauna. Further, there is Part I consists of biological and ecological surprisingly little mention of conserving bio- studies of biodiversity in the dry forest and adja- diversity in any of the examined forests (Frankie cent ecosystems. It is divided into three sub- 1997). parts, each based on different biogeographical

viii PREFACE frankie front 8/20/03 7:44 PM Page ix

considerations of biodiversity. The first of these of experiences and projects aimed at applying consists of three chapters (2–4) on flowering accumulated biological knowledge to solve prob- phenology, breeding systems, pollination, plant- lems in conserving and protecting biodiversity breeding structure in fragmented landscapes, and natural resources in general. In chapter 17 and possible global changes in tropical plants Costa Rica’s rich inventories of biodiversity are and their pollinators. Chapters 5–7 in the same described and actual and potential uses explored. subpart are concerned with mammals, bees, and Chapter 18 is the only contribution that does not ants as bioindicators of environmental health deal directly with a Costa Rican experience. The and ecological/evolutionary adaptations of in- author was invited to present a model that has sects to the seasonal dry forest. wide potential use for bringing diverse stake- Chapters in the next subpart explore the eco- holders together to collaborate on projects of logical relationships between dry-forest organ- mutual interest and benefit. The next two chap- isms and other ecosystems in Costa Rica. Chap- ters are concerned with transferring bioconser- ter 8 presents the diversity of butterflies that live vation knowledge to diverse audiences. In chap- in dry forests and migrate regularly to other ter 19 the transfer is from biologists and other habitats. Chapter 9 deals with hydrological re- professionals to elementary school children, high sources in the dry forest and watersheds that schoolers, and selected adult audiences. The connect the dry forest to upland ecosystems. importance of transferring information between This is followed logically by a critical examina- biologists and the media and vice versa is em- tion in chapter 10 of the flow of sweet water into phasized in chapter 20. the marine gulf of Nicoya estuary, with all its The next four chapters form a loose unit of biological and socioeconomic implications. Fi- contributions dealing with threats to conser- nally, in chapter 11, mangrove forests that form vation (chapter 21); the environmental laws that a transitional habitat between the dry forest and are designed to protect the environment and pe- the marine environment are examined in terms nalize offenders (chapters 22 and 23); and the of their ecology, uses to humans, and potential policy context for conserving all biodiversity in for conservation. Costa Rica (chapter 24). These four chapters are The third subpart examines selected dry- considered to be extremely important for under- forest organisms over broad geographical areas, standing the basic challenges that currently face which include the Costa Rican dry forest. Chap- biodiversity conservation and those likely to ters 12–14 deal respectively with birds, bats, and confront biodiversity protection in the future. the herpetofauna. Each author develops a case In the final chapter (25), a synthesis is pro- for the necessity of evaluating the diversity— vided of 12 of the major lessons that emerge from over a wide geographical area of Middle America the 24 contributions. Some of these lessons are —of the taxa he or she has studied. In chap- repeatedly mentioned or explained in more deter 15, sea turtle diversity, conservation problems, tail in several of the chapters. Some represent and projections for the future are offered for all special cases, which the editors considered to be of Costa Rica, Middle America, and beyond to important, and still others could only be inferred open marine environments where these animals from several authors. Minor lessons can be found migrate. The final chapter (16) in this subpart at the conclusion of most chapters. deals with an ongoing bio-socioeconomic study to conserve valuable tree species in dry-forest ACKNOWLEDGMENTS agro-ecosystems of Middle America. Part II explores transferring and applying We are grateful to the following colleagues for biodiversity and conservation knowledge. The their time, interest, and helpful comments dur- eight chapters in this part examine a wide range ing the early stages of developing the book and

PREFACE ix frankie front 8/20/03 7:44 PM Page x

for their reviews of many of the chapters: Mario REFERENCES Boza, Harry Greene, Bill Haber, Peter Kevan, Bullock, S. H., H. A. Mooney, and E. Medina, eds. Felipe Noguera, Sean O’Keefe, Paul Opler, Jerry 1995. Seasonally dry tropical forests. New York: Powell, Mauricio Quesada, Peter Ronchi, Mary Cambridge University Press. 450 pp. Schindler, Stan Schneider, Kathryn Stoner, Rob- Frankie, G. W. 1997. Endangered havens for diver- bin Thorp, and Jorge Vega. Several others who sity. Book review of S. H. Bullock et al., eds. Seasonally Dry Tropical Forests (1995). BioScience reviewed early manuscript drafts are cited indi- 47:322–24. vidually at the end of each chapter. Collectively, Holdridge, L. R., W. C. Grenke, W. H. Hatheway, the thoughtful and critical comments of re- T. Liang, and J. A. Tosi Jr. 1971. Forest environ- viewers helped greatly to improve the quality ments in tropical life zones: A pilot study. Oxford: and clarity of the book. We also thank Pablo Mata Pergamon Press. 400+ pp. for his excellent translations. Marilyn Tomkins, Janzen, D. H., ed. 1983. Costa Rican natural history. Chicago: University of Chicago Press. 816 pp. Genesis Humphrey, Margaret Przybylski, Megan ———. 1991. Historia natural de Costa Rica. Chi- Konar, and Mary Schindler greatly assisted the cago: University of Chicago Press. 822 pp. editors by keeping communications constantly Spanish translation of Janzen’s 1983 volume. ﬂowing among all contributors. Special thanks McDade, L. A., K. S. Bawa, H. A. Hespenheide, are due to Mary Schindler, who assisted in edit- and G. S. Hartshorn, eds. 1994. La Selva, ecology and natural history of a Neotropical rain forest. ing several manuscripts prior to their ﬁnal sub- Chicago: University of Chicago Press. 486 pp. mission to the University of California Press. Nadkarni, N., and N. Wheelwright, eds. 2000. Monteverde, ecology and conservation of a tropical GORDON W. FRANKIE cloud forest. New York: Oxford University Press. AND ALFONSO MATA 573 pp.

x PREFACE frankie chap 01 8/20/03 7:46 PM Page 1

chapter 1

Introduction

Alfonso Mata and Jaime Echeverría

he chorotega region in northwestern square kilometer. Of this population, 43 percent TCosta Rica is one of the most important is located in urban centers (~60 inhabitants per areas of this republic; it covers primarily the square kilometer), whereas 57 percent lives in Tempisque River Basin (TRB), Nicoya Peninsula, rural areas and tends to move toward cities such and other nearby lands (see map 1.1). The coun- as Liberia, Cañas, and Nicoya (see map 1.2). try’s only seasonal dry forest is located here. En- There has been a slow migration of rural and joying a climate of contrasts, varied geological urban residents toward other parts of the coun- formations, very attractive natural scenic areas, try, mainly owing to lack of employment and the and a rich cultural heritage, the Chorotega re- mechanized monocultures that require less hu- gion is perhaps the second most important eco- man labor. These activities involve seasonal crops nomic region in the country, after the Central (melons, sugarcane) and are primarily carried Valley (Mata and Blanco 1994), where the capi- out by a large contingent of nonresident Nica- tal city of San José is located. Politically, this area raguans. The tourism industry in this area is one constitutes the province of Guanacaste, with of the most important of the country. approximately 275,000 inhabitants distributed The entire region has been notably altered and in 11 counties and an average density of 26 in- transformed, undergoing substantial changes habitants per square kilometer. In addition, three in land use and ownership, as well as in the qual- counties of Puntarenas Province occupy the tip ity and quantity of its natural resources. The of the Nicoya Peninsula. The TRB is made up impact is a cause of concern for the region’s of nine counties of Guanacaste Province. The inhabitants, public institutions, and nongovern- approximate population in this area is 157,000, mental organizations, which are making efforts with an average density of 30.6 inhabitants per to prevent further damage and repair that which

1 frankie chap 01 8/20/03 7:46 PM Page 2

MAP 1.1. Principal river basins of Costa Rica.

has already been done. This introductory chap- sants, or basins, with similar areas (map 1.1). The ter presents the region’s main physical, biologi- Paciﬁc Basin covers a territory of 26,585 km2 cal, and general environmental characteristics, as (Herrera 1985), modeled by a network of rivers well as some anthropogenic environmental ef- whose ﬂow and eroding behavior are deter- fects, topics discussed in subsequent chapters of mined by marked climatic seasons. This basin, this volume. particularly the northwestern part of Guanacaste, is characterized by a dry seasonal climate, represented by the tropical dry-forest life zone GEOGRAPHY (sensu Holdridge 1967; see the section “Life The massive mountain range system that runs Zones” later in this chapter and map 1.2) with its along the length of Costa Rica, with a northwest– transitions and the tropical wet forest of atmos- southeast orientation, creates two principal ver- pheric association, which is characteristic of the

2 MATA AND ECHEVERRÍA frankie chap 01 8/20/03 7:46 PM Page 3

MAP 1.2. Life zones, river matrix, and important geographical points of northwestern Costa Rica.

Nicoya Peninsula. On the other side, the Carib- Along with the continental northwest, the pen- bean Basin has a more homogeneous and much insula encloses the Gulf of Nicoya, one of the more humid climate for most of the year. This most important ecosystems in the country (chap- large basin is divided into two main watersheds; ter 10). The coastal hills and mountains deter- in one, the rivers empty into Lake Nicaragua and mine the watershed of the northwestern coastal the San Juan River, which in turn reach the strip, which begins at the border with Nicaragua Caribbean Sea, and in the other, the rivers empty and ends at the southern tip of the peninsula directly into the Caribbean. (map 1.1). The Guanacaste Mountain Range (Cordillera The vast region around the Gulf of Nicoya, Guanacaste) separates the watersheds with a with this common drainage, is known as the few rivers ﬂowing to Lake Nicaragua and the San Gulf of Nicoya Basin (GNB), covering approxi- Juan River from those of the Gulf of Nicoya mately 12,000 km2. The basin’s land and estu- (Mata and Blanco 1994). The Nicoya Peninsula arine ﬂuvial components are hydrologically is located in the southern sector of Guanacaste. connected with a common receptor: the Gulf of

INTRODUCTION 3 frankie chap 01 8/20/03 7:46 PM Page 4

MAP 1.3. Parks, reserves, and conservation areas of Costa Rica.

Nicoya (see maps 1.1–1.3). The surface area of zones that occur in Costa Rica. It also features the GNB constitutes half of the country’s entire the country’s only tropical dry-forest zone. The Paciﬁc Basin (map 1.1) and represents nearly 25 TRB, which includes the Bebedero River, is sit- percent of the entire country (Mata and Blanco uated here. It is the most extensive hydrological 1994: 47). The GNB has a variety of land and subregion of Guanacaste (5,460 km2, which aquatic ecosystems comprising nearly all the life represents 54% of the province and approxi-

4 MATA AND ECHEVERRÍA frankie chap 01 8/20/03 7:46 PM Page 5

mately 10% of the total national area). The Mountain Range (map 1.1). The Nicoya complex largest extant territory of dry forest in the Pacific- consists of the oldest rocks in Costa Rica, formed side ecoregion of Mesoamerica (World Wildlife around 74 million years ago. In addition, the re- Fund/World Bank 1995), which is protected gion has sedimentary formations such as Rivas, (Boza 1999), is in this subregion. Sabana Grande, Brito, and Barra Honda. Forma- tions of intrusive rocks and fluvial, colluvial, and GEOPHYSICAL ASPECTS coastal deposits are common, as well as swampy areas. Three main formation processes con- CLIMATE tributed to the relief in the TRB. The first is Precipitation in the northwestern region of the volcanic, which occurred in the northern and country shows a marked seasonal variation, with northeastern sectors of the area. The second, dean average of 1,800 mm annually; the TRB has nudation, occurred in different sectors but more 1,746 mm of rain per year. About 95 percent of toward the southwestern sector. The main allu- the rainfall occurs from May to November, and vial sedimentation forms, the result of the third the dry period extends from December to April. process, are located in the flat and lower parts There are recurrent periods of prolonged rains of the TRB (Bergoeing et al. 1983). that induce extensive flooding in the lowlands of Sulfur, clays, alluviums, limestone, and di- the TRB (Asch et al. 2000). The southern region atomite are among the geological resources of of Nicoya Peninsula has the most abundant an- Guanacaste. There have been claims of illegal nual precipitation, between 1,800 and 2,300 mm, exploitation of alluvial material in public-access and the least abundant is in the central and riverbeds, and there are authorized quarries for northwestern zones of the TRB, with around road maintenance and construction. Geothermal 1,400 mm per year. The monthly mean temper- energy is generated on the volcanic mountain ature lies between 24.6°C and 30°C. The high- range slopes. Seismic activity is lively throughout est temperatures occur in April and the lowest the province, with local faults. The subduction between September and December. Strong trade process of the Coco’s and Caribbean tectonic winds are predominant during the dry season, plates affects the whole country, and conse- with speeds between 10 and 30 km per hour and quently the probability of a strong energy release gusts of roughly 60 km per hour. Winds from is high in the entire Chorotega region. the Pacific are predominant during the rainy season, bringing humidity with them, with speeds SOILS between 3 and 8 km per hour, making the Nicoya Soils vary from volcanic types in the upper parts Peninsula the most humid sector of the entire of the mountain range to flooding alluvial in region. the lower part of the TRB. The region has valleys of varying sizes, with high-fertility soils (e.g., GEOLOGY AND GEOMORPHOLOGY Tempisque, Curime, Nacaome, Nosara) and poor The northwestern region of Costa Rica was soils (ignimbrite deposits of Liberia-Cañas). Some formed through various processes. Notable are soils are light in texture, such as the volcanic the volcanic and sedimentary rocks from the types, and others are heavy, such as soils with a Mesozoic, which crop up extensively in the penin- high clay content in the floodplains. Among the sula (Castillo 1984), as well as the Bagaces and soil orders in the TRB are alfisols (13%), entisols Liberia formations and the sedimentary rocks of (26%), inceptisols (38%), mollisols, and vertisols the Quaternary (Castillo 1983) in the TRB. The (TSC 1999). Nicoya complex, the Aguacate Group, and the Quaternary volcanic formations are included in LAND-USE CAPACITY AND CONFLICTS this geological evolution, as well as recent vol- The soils of Guanacaste are shallow or stony, or canic structures that make up the Guanacaste both, and the long dry period and strong winds

INTRODUCTION 5 frankie chap 01 8/20/03 7:46 PM Page 6

further limit their productiveness (Echeverría et the strong and sustained rainfalls of September al. 1998). Many sectors with small slopes are and October, particularly under the influence of classified as lands for watershed protection, for the tropical storms in the Caribbean Basin. example, north of Bagaces and Liberia and the hills of the Nicoya Peninsula. During the first HYDROGEOLOGY half of the past century, deforestation was exten- Groundwater is used in the entire region for sive in the entire region, but particularly in the human as well as agricultural, industrial, and lowlands during the period 1940–65, with the cattle needs. The Chorotega region has 62 per- opening of the Pan American Highway. After cent of the country’s rural aqueducts, which are the cattle-ranching decline, at the end of the supplied by wells (Echeverría et al. 1998). The century, and regardless of natural regeneration principal sources are the volcanic aquifer of on many hills, the effects of tree cutting are still the Bagaces formation, which supplies water to noticeable in the inner parts of the peninsula’s the people of Liberia, Bagaces, and Cañas, and coastal mountain range. the colluvial-alluvial aquifer alongside the right- Official data indicate that the Chorotega area hand region of the Tempisque River, which has some of the highest land-use imbalances in supplies water to several towns and smaller pop- the country. In this region 38 percent of the ulation centers and may be the most important land—almost 600,000 ha—is overused, only aquifer in the entire basin. Colluvial deposits in 16 percent is used sustainably, and the rest is Nicoya Peninsula’s valleys can provide effective underused. In the case of the TRB, which is the flows of up to 65 liters per second (e.g., Nosara flattest land in the region, 30 percent of the area Valley and smaller ones). Their waters are gen- is overused, 40 percent used sustainably, and erally potable and suitable for agriculture. 30 percent underused. Overuse brings about de- Several of these aquifers could supply coastal terioration of soils and other resources, as is the tourist projects between Culebra and Brasilito case when cattle are kept in lands with forest Bays, transporting water across the TRB’s west- capacity. An example of underuse in the basin is ern divider and therefore diminishing risks of raising livestock in lands with the potential for saline intrusion in those limited coastal aquifers. agricultural activity. There have been documented cases of saliniza- A considerable part of the Guanacaste flood- tion from overexploitation of the northwestern plains is used for agricultural purposes and for coastal strip, specifically in Flamingo, El Coco, some urban and semiurban areas (such as Fila- and Tamarindo and particularly in the area of delfia and Ortega); these zones are extremely Puntarenas (TSC 1983: 95), among other cases. vulnerable because they include wetlands and plains with recurrent flooding (see the section HYDROLOGY “Geohazards and Disasters”). There are no ordi- The rivers of northwestern Costa Rica can be nances for the safe construction and location of divided into three geographical sectors: the Lake urban developments in these areas. Nicaragua Basin, to the north; the northwestern The main erosive process in the region is coastal strip; and the GNB. In the first sector, hydrological, but eolian erosion is also a prob- four rivers originate in the volcanic massifs of lem. Because of the inadequate use of soils, the Cerro El Hacha and Orosi Volcano, the north- sediment load in rivers is directly related to the ern extreme of the Guanacaste Mountain Range erosion. The most abundant production of sed- (map 1.2). These are the Sapoa, Sabalos, Mena, iments coincides with the rainy season (May to and Haciendas, and they empty into Lake Nica- November), particularly in the hills of the Nicoya ragua. The Haciendas is the boundary between Peninsula, which become microbasins produc- the provinces of Alajuela and Guanacaste. The ing large quantities of water during short periods. largest watercourse of Guanacaste is the Temp- Landslides and fast floods are frequent during isque River, born in the southern flanks of the

6 MATA AND ECHEVERRÍA frankie chap 01 8/20/03 7:46 PM Page 7

Orosi volcanic massif. Together with the Bebe- primarily by detouring rivers and creeks dero River it forms the largest watershed in Costa through the Arenal Tempisque Irrigation Sys- Rica. The western sector of the Nicoya Peninsula tem. When completed, the system will service has low coastal sierras (50 to 250 m above sea about 45,000 ha, with a water volume of 45 m3 level) where creeks and a few rivers originate, per second. This project, the largest hydrological their microbasins emptying directly into the system in the country, consists of surface water Pacific Ocean along the entire northwestern for multiple purposes collected from the Arenal coastal strip. Most of these small streams remain River watershed (Atlantic Basin), which passes dried up during the summer but rapidly grow and through a tunnel to feed a cascade of three flood with the strong and persistent precipita- power stations on the Pacific Basin side. These tion of the rainy season. All of them are born in power stations are part of the Arenal-Corobici- the driest life zones of the country (see map 1.2), Sandillal Hydroelectric Project developed by the such as the tropical dry forest, as well as in the Costarrican Institute of Electricity. The presence tropical moist and premontane wet forests. of water of excellent quality in the dry-zone area The northwestern coastal strip is narrow has stimulated the invasion of the floodplains compared with the TRB, which empties into the and many wetlands for agricultural purposes, gulf, and it widens heading south in the penin- without sufficient environmental studies. The sula as its mountains become higher. For ex- entire area is subject to recurrent natural floods. ample, the Cerros La Carbonera and Cerro Vista Stream corridors are being altered, and levees, al Mar (983 m above sea level) and Zaragoza channels, and ditches have been constructed later average between 200 and 500 m above sea without a master plan. These and other activities level all the way to the extreme of the peninsula. are making the area highly vulnerable to envi- Although most of the rivers are nearly dry dur- ronmental damage (see the section “Deforesta- ing the summer, the Nosara and Ario Rivers, tion of Basins, Wetlands, and Stream Corridors” the two longest and most affluent of the entire and chapter 9). Balance between ecological al- northwestern coastal strip, keep a perceptible teration and economic development of the en- base flow, with a more noticeable decrease in tire region, although it seems to be positive, has April and May. Two of the most important wet- not yet been studied. lands of the coastal strip are the outstanding Tamarindo National Wildlife Refuge (man- GEOHAZARDS AND DISASTERS groves and estuary), annexed to Las Baulas Ma- The main geohazards of the region are hydro- rine National Park and the Ostional National meteorological; they increase during times of Wildlife Refuge, at the mouth of the Nosara River strong rainfall and storms, which in some in- (map 1.3). stances are indirect effects of hurricanes origi- Water is used in many ways in the region: nating far in the Caribbean but which always public supply, domestic use, agricultural irriga- have strong repercussions. As a result of the tion, hydroelectricity generation, tourism, and vulnerability caused by uncontrolled human ac- industrial activities. Water services to cities and tivity (deficient urban planning, urban invasion large towns are provided by the national system of lands susceptible to the effects of these strong of aqueducts. There are rural aqueduct councils recurring phenomena), there are damages from in smaller population centers, but some are floods, avalanches, and landslides, as well as from continuously mismanaged, and the dispersed hydrological erosion. They occur in the flood- population obtains its water supply from artesian plains of rivers and on steep slopes, and they wells. Groundwater resources largely supply affect crops, bridge infrastructure, roads, and industrial and agricultural activities. urban areas, particularly in the wetland sectors Irrigation water, used for the extensive sug- of the Tempisque, Cañas, and Las Palmas Rivers. arcane, rice, and melon industries, is obtained However, these lands have been overtaken by

INTRODUCTION 7 frankie chap 01 8/20/03 7:46 PM Page 8

agricultural activities, settlements, neighbor- human activities, undergone by the original nat- hoods, and even cities (case in point: the city ural vegetation. It includes categories such as of Filadelfia and areas nearby). Although the intervened primary forests, secondary growths National Emergency Commission acts efficiently and their initial successional stages, and agri- during disasters, there is no ordinance for ap- cultural and other matrixes in the landscape. Ad- propriate and obligatory construction of houses vantages of this system include the prediction of (e.g., perched on piles) or for development of the type of natural vegetation that should exist urban centers on more elevated lands. Levee in a deforested area, starting from local physical construction continues along the Tempisque environmental factors; the prediction of carbon River without environmental studies or official offsets by terrestrial biota in a determined life restrictions. zone (Tosi 1997); and the analysis of possible There is a potential seismic threat through- changes in type of vegetation from climatic out the entire Pacific region of the country. Vol- change. canic hazards are present only in the sector of The tropical dry-forest zones, which are the Guanacaste Mountain Range, where the Rin- strongly seasonal, and the related transitions to con de la Vieja and Arenal are the only volcanoes moist forest constitute around 30 percent of currently having eruptions and fumaroles activ- the Chorotega region. The largest remaining ity; other volcanoes show less activity. dry forest in the Pacific-side ecoregion of Meso- america (World Wildlife Fund/World Bank 1995), BIOLOGICAL ASPECTS which is protected (Boza 1999), lies in this territory. The tropical moist forest predominates, LIFE ZONES particularly in the Nicoya Peninsula (map 1.2). The great bioclimatic diversity of the Chorotega The greatest biodiversity occurs in the Gua- region (map 1.2) is divided into seven life zones nacaste and Tilarán Mountain Ranges, where with various transitions. The method of classi- several life zones are found as narrow eleva- fication of plant formations, or World Life Zone tional bands surrounding the volcanic moun- System (Holdridge 1967), consists of three main tain chain. levels of detail. The first is the main life zone category, which considers climatic variables of PROTECTION AND annual mean precipitation and annual mean BIODIVERSITY CONSERVATION biotemperature and the potential evapotranspi- The northwestern region of Costa Rica has three ration defined by the first two variables and a conservation areas: Tempisque (dry), Guanacaste constant empirical value. The biotemperature is (dry), and Arenal (midelevation/cloud forest) an adjustment of the annual mean temperature, (map 1.3). The three manage a total of 35 wild in the 0°C–30°C range, at which there is sup- areas that have varying protection categories, posed to be a better development of life. Each life such as national parks and wildlife refuges. One zone has a definite range for each variable, and of the most outstanding examples of protection the plotting of these factors on a graph forms a is the Area de Conservación Guanacaste (ACG), diagram of hexagons (Hartshorn 1983). The covering 120,000 terrestrial and 43,000 marine plant association, or ecosystem, is the second ha (Janzen 2000; chapter 7). The second dry- level. It considers local environmental factors, forest conservation area, Area de Conservación such as dry periods, soils, geological relief, and Tempisque (ACT), consists of about 35,000 ha prevailing winds; therefore, associations such as of terrestrial and wetland habitat and is still be- hydric, wet and dry edaphic, cold and warm ating defined and administered. mospheric, and combinations of them are found. An essential part of the conservation process The third level is the successional stage, refer- is maintenance of the biodiversity that is threat- ring to the degree of intervention, mainly due to ened by human activities. In the dry forest it is

8 MATA AND ECHEVERRÍA frankie chap 01 8/20/03 7:46 PM Page 9

still necessary to increase the number of pro- decline in faunal loss. Even so, poaching still ex- tected areas and to connect them with appropri- ists, whether with temporary permits or illegally ate corridors, so that all existing ecosystems by unscrupulous people, and even fire is some- and biomes of the region can be encompassed. times used to flush wildlife. Furthermore, polit- Conservation organizations have developed ical pressure to develop tourist facilities and outstanding protected areas (e.g., Monteverde hotels in national refuges is common, and there Cloud Forest Preserve) and have been making is illegal fishing during prohibited seasons. The efforts to protect other important zones already National System of Conservation Areas (map 1.3) targeted for development projects; a few are has recently become an instrument for imple- nearly established, and several corridors are mentation and interinstitutional coordination, under consideration. with a geographical decentralization, and watches over integrated forestry management, wildlife, WILDLIFE and areas for biodiversity protection. However, In general, the Chorotega region still has a rich the system is not completely efficient, as it lacks flora and fauna. The different life zones of the appropriate finances and staffing. area have an ample diversity of plants and animals. In the past 25 years, important sectors of WETLANDS the forest habitat in the Nicoya Peninsula have These productive ecosystems are abundant naturally recovered (secondary growth) after the in Guanacaste, particularly around the Gulf of recession of cattle ranching (TSC/CIEDES/CI Nicoya. Wetlands alone represent 20 percent of 1997); many species of mammals and birds have the area of the TRB, that is, 1,025 km2, not in- had noticeable increases in their populations cluding the riparian area or uplands. These eco- (A. Mata, pers. obs.). This is an important devel- systems present conditions of great ecological, opment because Guanacaste is one region of economic, and social value, and the inhabitants Costa Rica where wildlife has suffered the great- of the region have exploited them throughout est negative impact, resulting from loss of natu- history. In the case of the middle TRB area, the ral habitat, overexploitation through extensive continuous invasion of wetlands by agriculture cattle ranching and agriculture (chapter 21), and has resulted in a large negative impact. There hunting/poaching, particularly during the mid- have also been cases of agricultural, domestic, dle years of the past century. Furthermore, the and industrial contamination of wetlands, as hunting of sea turtles (chapter 15) and dolphins, well as overexploitation of their resources and as well as overfishing in the Gulf of Nicoya (chap- the alteration of natural drainages. ter 10), has prompted legal actions for environ- The most extensive wetlands are those be- mental protection. tween 3 and 30 m above sea level, located in the There are almost 30 endangered bird spe- floodplains of the rivers in the TRB, the so-called cies in the region (chapter 12), 12 mammal palustrine wetlands. Those influenced by tidal species (chapter 5), and several hardwood tree estuarine waters are located near the Tempisque species having commercial value. On the other River mouth (near Palo Verde; map 1.2) and sur- hand, about 20 animal species—among them rounding the Gulf of Nicoya (Echeverría et al. birds, rodents, and mammals—are considered 1998). Small lakes or lagoons cover a lesser area pests because of alleged damages done to agri- (e.g., Cañas River); the rest of wetlands are the cultural crops and domestic animals. banks of the stream corridors, which extend up The creation of protection areas, such as the to the headwaters. The natural vegetation in- ACT and ACG, and wildlife refuges or private cludes herbaceous gramineae and cyperaceae preserves (e.g., Monteverde Cloud Forest) has shrubs, floating and submerged aquatic vegeta- brought about a change in the consciousness of tion, rooted vascular plants, mangroves, and the population, resulting in a slow but effective dry-forest trees in naturally drained areas. The

INTRODUCTION 9 frankie chap 01 8/20/03 7:46 PM Page 10

wildlife supported by different wetland types is 189); the old custom of slash-and-burn land clear- of great environmental significance to inhabi- ance is still a problem during the dry season. tants of the area. The government protects a few of these areas, such as the wetlands in the ACG WATER EXTRACTION BY INDUSTRY, and Palo Verde in the ACT. IRRIGATION, AND HUMAN POPULATIONS Although smaller in size, other important There are several notable water consumers in wetlands have diverse wildlife, such as those lo- the dry forest. Once they use this resource, those cated on the northwestern coastal strip (map 1.1). consumers become producers of contaminated On the southern coast of the Nicoya Peninsula water, in varying degrees of output. Among them some remain in protected areas such as in Ta- are sugar industries, rice growers, coffee mills, marindo and at the mouth of the Nosara River fruit-packaging plants (melons), the Arenal- (map 1.3); they are in national wildlife refuges. Tempisque Irrigation System, aquaculture, cities, and towns (Echeverría et al. 1998). ENVIRONMENTAL IMPACTS The sugar industry taps water from the Cañas, Liberia, and Tempisque Rivers, with volumes DEFORESTATION OF BASINS, WETLANDS, varying from 5 to 20 m3 per second or more for AND STREAM CORRIDORS industrial and irrigation purposes. There have The disappearance of extensive forest areas been shortage problems during the dry season and fragmentation caused by human actions has and when the sugar harvest is ending, and it has strongly affected the ecosystem of the entire re- even been necessary to obtain water from the ir- gion, especially lowland areas subjected to agri- rigation system to satisfy the industrial demands, culture and extensive livestock use. Damage to to the extent that in some years the Tempisque riparian systems, which had traditionally been River can be easily crossed on foot by the end protected, is visible in the majority of rivers in of the dry season. Taxes for exploitation rights Guanacaste—as it is nationwide. The fluvial- are quite low, and, according to municipal au- riparian continuum is essential to maintain bio- thorities, there is an evident lack of controls for diversity of the ecosystem (chapter 9). However, the amount of water pumped. each bridge, road, or crop established along a river zone results in various degrees of frag- MODIFICATION OF THE NATURAL mentation, which could be prevented by pro- COURSE OF THE TEMPISQUE RIVER tecting stream corridors and by environmentally Near those same industries, fluvial morphology sensitive construction, as well as through en- has been altered by channeling and dam con- forcement of current laws (chapters 22 and 23). struction for water deviation and containment, Cattle grazing in wetlands and stream corridors, as well as by intervention of riparian forests. If which is common, is even more damaging to sugarcane expansion in these fragile areas is not understory vegetation. Slum settlements, as well regulated, there will probably be more interven- as expanding urbanization from main cities, tion with channels, levees, and fluvial detours, have reached the natural floodplains and banks which would bring about certain alteration of of rivers and their wetlands, resulting in unde- nearly all wetlands in this region. Water con- sirable environmental effects and increasing the sumption during the dry season drastically vulnerability of those same settlements. changes the volume of the river, which has al- Although outstanding advances have been ready been tapped upstream. There is no legis- made in fire prevention, thousands of hectares lation that regulates levee or channel construc- in the region were affected by fires from 1997 to tion, and some residents of the area who have 1998, in part as a result of the effects of the El experienced floodings feel that this infrastruc- Niño phenomenon (Estado de la nación 1999: ture, built without specific environmental plan-

10 MATA AND ECHEVERRÍA frankie chap 01 8/20/03 7:46 PM Page 11

ning, is contributing to the prolonged contain- The public zone consists of the first 50-m strip ment of floodwater in the middle TRB. of land starting from the high-tide water level and all surface uncovered during low tide and in- POLLUTION: WATER AND SOLID WASTES cludes any extension covered by estuaries, man- The entire dry-forest region, but particularly the groves, coastal lagoons, and floodplains. The TRB, is subject to various types of pollution; interior band of 150 m of continental or insular water pollution is the most relevant, in the form land constitutes the restricted zone. According of sewage, wastewater from industry and urban to the law, both contiguous bands should be un- centers, and water from agriculture activities. der protection and careful management with re- The initial capacity of urban sewage treatment gard to urban, tourist, industrial, or agricultural plants was reached some time ago, and now they developments. However, there have been serious cannot handle wastes produced by new urban de- transgressions. Several disasters have occurred velopments (Echeverría et al. 1998; chapter 9). since approval of the law, particularly as a result Point-source pollution is generally caused by of ignorance, lassitude, or connivance involving the discharge of urban wastewater, tourist de- the same municipalities in charge of surveil- velopments, industry (sugar mills), mining, pig lance of resources. Even worse, some of these farms, tilapia aquaculture, and garbage dump- disasters were the result of inaction on the part sites, among others. Aside from routine govern- of the Ministry of Environment and Energy ment controls of bacteriological quality of water (MINAE) and the Tourism Ministry, which were from aqueducts and sediment monitoring of perhaps afraid of hampering economic devel- some rivers by the Costarrican Institute of Elec- opment. This lack of action may contribute, tricity, there is almost no surveillance of surface- paradoxically, to the destruction of the natural water quality in the entire region. attraction that is the driving force for tourist de- The legal frame for pollution control in the velopment and a reason for the country’s over- country is sufficient. However, the lack of regular seas environmental prestige. The problem is monitoring, failure to reinforce laws, lack of ap- of special relevance in the Chorotega region: of plication of new technologies, and limited capac- 484 official tourism concessions and permis- ity of governmental offices in charge work against sions registered in 1999 for the country (Estado improved control of emissions. Despite this sit- de la nación 1999: 191), 388 were for Guanacaste uation, several agreements have been reached be- Province. tween the government and the sugar industry (as well as coffee mills and pig farms) regarding the ASSESSMENT OF BIODIVERSITY control of these industries’ liquid and solid wastes CONSERVATION BY SPECIALISTS (chapter 9); a positive change has been noticed. Nonpoint sources of pollution include the use of Superimposed on today’s complex dry-forest chemicals in agricultural areas, such as in the ex- environment are the numerous biologists who tensive rice paddies. Almost no studies have been have studied the extant flora and fauna, princi- done on this subject. Along with the pollution pally in Costa Rica. Working with them, directly from the Central Valley Basin, the contamination or indirectly, are the specialists who have dealt of the TRB has repercussions on the Gulf of with socioeconomic, political, and legal aspects Nicoya that are yet to be estimated (chapter 10). of conserving this biota. In the chapters that follow, major players present their individual and ALTERATION OF THE interrelated stories on what has been learned TERRESTRIAL-MARITIME ZONE about biodiversity and its conservation in the dry Certain laws protect this coastal strip, which is forest and beyond and what must be done in the made up of a public area and a restricted area. future to better protect this biodiversity. The

INTRODUCTION 11 frankie chap 01 8/20/03 7:46 PM Page 12

authors provide the first overall assessment of Herrera, W. 1985. Clima de Costa Rica. In Veg- how well one Central American country has ex- etación y clima de Costa Rica, ed. L. D. Gomez, amined, valued, and conserved its biological 2:15–19. San José: Editorial Universidad Estatal a Distancia. heritage. Holdridge, L. R. 1967. Life zone ecology. San José: Tropical Science Center. 206 pp. Janzen, D. H., ed. 1983. Costa Rican natural history. ACKNOWLEDGMENTS Chicago: University of Chicago Press. 816 pp. ———. 1991. Historia natural de Costa Rica. San We thank Gordon Frankie and Brad Vinson for José: Editorial de la Universidad de Costa Rica. their valuable commentaries and suggestions 822 pp. and Vladimir Jimenez at the Tropical Science ———. 2000. Costa Rica’s Area de Conservación Center for the preparation of maps. Guanacaste: A long march to survival through non-damaging biodevelopment. Biodiversity 1: 7–20. REFERENCES Mata, A., and O. Blanco. 1994. La cuenca del Golfo de Nicoya: Reto al desarrollo sostenible. San José: Asch, C., G. Oconitrillo, and J. L. Rojas. 2000. De- Editorial de la Universidad de Costa Rica. limitación cartográfica y otras consideraciones so- 235 pp. bre las áreas afectadas por las inundaciones de la Tosi, J. A. 1997. An ecological model for the prediction cuenca baja del Río Tempisque, Guanacaste. San of carbon offsets by terrestrial biota. Occasional José: National Geographical Institute of Costa Papers No. 17. San José: Tropical Science Cen- Rica. 24 pp. ter. 34 pp. Bergoeing, J., et al. 1983. Geomorfología del Pacífico TSC (Tropical Science Center). 1983. Costa Rica: Norte de Costa Rica. San Pedro, Costa Rica: Country environmental profile, ed. G. Hartshorn. Oficina de Publicaciones Universidad de Costa San José: Editorial Trejos Hnos. 124 pp. Rica. 110 pp. ———. 1999. Map of Soils of Costa Rica (scale Boza, M. 1999. Biodiversity conservation in Meso- 1:200.000). San José: Geographical Informa- america. In Managed ecosystems: The Mesoamer- tion Center of the Tropical Science Center. ican experience, ed. L. Hupton Hatch and M. E. TSC/CIEDES/CI (Tropical Science Center/Center Swisher, 51–60. New York: Oxford University for Sustainable Development Research/Conser- Press. vation International and Fondo Nacional de Castillo, R. M. 1983. Geology of Costa Rica. In Costa Financiamiento Forestal). 1997. Estudio de Co- Rican natural history, ed. D. H. Janzen, 44–62. bertura Forestal Actual (1996/97) y Cambio de Chicago: University of Chicago Press. Cobertura para el período entre 1986/97 y 1996/ ———. 1984. Geología de Costa Rica: Una sinopsis. 97 para Costa Rica. San José: Tropical Science San José, Costa Rica: Editorial de la Universidad Center, Center for Sustainable Development de Costa Rica. 187 pp. Research/University of Costa Rica and Conser- Echeverría, J., A. Echeverría, and A. Mata. 1998. vation International/FONAFIFO. 20 pp. Plan de Acción para la Cuenca del Río Tempisque. Vargas, G. 1999. The geography of dryland plant Antecedentes del Estudio y Resumen Ejecutivo. San formations in Central America. In Managed eco- José: Tropical Science Center/Association for systems: The Mesoamerican experience, ed. L. Hup- the Tempisque River Basin Management. 39 pp. ton Hatch and M. E. Swisher, 88–97. New York: Estado de la nación en desarrollo humano sostenible. Oxford University Press. 1999. 5° informe. Proyecto Estado de la Nación. World Wildlife Fund/World Bank. 1995. Una eval- Costa Rica: Editorama S.A. 338 pp. uación del estado de la conservación de las ecorre- Hartshorn, G. 1983. Plants. In Costa Rican natural giones terrestres de America Latina y el Caribe. history, ed. D. H. Janzen, 118–57. Chicago: Uni- Map. Washington, D.C.: World Wildlife Fund/ versity of Chicago Press. World Bank.

12 MATA AND ECHEVERRÍA frankie chap 02 8/20/03 7:48 PM Page 13

PART ONE

Biodiversity and Ecological Studies frankie chap 02 8/20/03 7:48 PM Page 14 frankie chap 02 8/20/03 7:48 PM Page 15

Section A. Costa Rican Dry Forest frankie chap 02 8/20/03 7:48 PM Page 16 frankie chap 02 8/20/03 7:48 PM Page 17

chapter 2

Flowering Phenology and Pollination Systems Diversity in the Seasonal Dry Forest

Gordon W. Frankie, William A. Haber, S. Bradleigh Vinson, Kamaljit S. Bawa, Peter S. Ronchi, and Nelson Zamora

hen comparing Neotropical life zones, cal patterns and pollination systems in the dry Wone of the first generalizations to emerge forest. Second, we compare the diversity with is that seasonal dry forests have lower species that of other tropical forests. Third, we explore diversity than wetter or more aseasonal life the significance of this diversity and how it zones (Janzen 1983; Bullock et al. 1995). This might be affected by environmental disturbances pattern is easily recognized. The species-level caused by humans. count, however, is only one aspect of a much larger picture of biodiversity. Noss and Cooper- PHENOLOGY STUDIES rider (1994: 5) provide a useful definition of biodiversity that is relevant to this discussion: EARLY PHENOLOGY STUDIES, 1969–1973 “Biodiversity is the variety of life and its pro- Community-level periodicity patterns of leaf fall, cesses. It includes the variety of living organ- leaf flushing, flowering, and fruiting were deter- isms, the genetic differences among them, the mined for Costa Rican dry forest trees and shrubs communities and ecosystems in which they oc- from 1969 to 1970 and from 1971 to 1973. Pat- cur, and the ecological and evolutionary processes terns were found to be closely associated with that keep them functioning, yet ever changing the highly seasonal dry and wet periods (Frankie and adapting.” et al. 1974; Opler et al. 1975, 1976, 1980). The In this chapter we focus on selected processes major seasons are the long dry period from early of plant reproduction in the highly seasonal dry November to early May and the wet period from forest of Costa Rica and their evolutionary im- late May to early November, with a brief, but plications. The chapter has three goals. First, we variable, dry spell from July to August (Frankie document the diversity of flowering phenologi- et al. 1974). The forestwide patterns, as well as

17 frankie chap 02 8/20/03 7:48 PM Page 18

individual species patterns, have remained largely mination in mid-2003. Expected flowerings were the same every year since 1969 (G. Frankie based on cumulative experiences for each species pers. obs.). by one of us (GF) from 1969 to 1995. Expected flowering was expressed as the most RECENT PHENOLOGICAL likely month(s) of flowering for a given species. STUDIES, 1996 TO PRESENT We also used three supplementary sources to Early survey work provided a partial examination determine expected flowering. These were field of dry-forest phenology, focusing on only tree and notes from one of us (GF) that had been col- shrub life forms. Goals of the early work were lected yearly from 1971 to 1995; published ac- limited to marking and monitoring a restricted counts of selected flowering patterns (Janzen number of plant species in replication in se- 1967; Frankie et al. 1974, 1983; Heithaus et al. lected habitats. The approach was labor-intensive 1975; Opler et al. 1975, 1980; Frankie and Haber and tedious, but it provided the first forest over- 1983; Haber and Frankie 1989); and herbarium view of flowering patterns of the most common records from the Institute of Biodiversity (INBio) tree and shrub species. These findings also in- in Heredia (see chapter 17). Because we were dicated that flowering patterns of the other plant assessing abundance and timing of floral re- life forms needed to be factored into the forest- sources for pollinators at the population and wide picture. community levels, population and community In mid-1996 we began another flowering phe- constituted the levels of analysis for collecting nology study at a forest site in close proximity to phenological records in this study. (For discus- the sites of the earlier studies. The goal of this sion of analyses in phenology, see Newstrom et al. new study was to survey flowering patterns of 1994a,b.) Therefore, data were not collected on as many native angiosperm species of all forms individually tagged plants because this level of as possible in a lowland dry-forest area (100 m analysis was not pertinent to our question. More- elevation) that contained a variety of habitat types. over, it would have been cost-prohibitive to at- The new site (Bagaces) was an area measuring tempt to collect for a long-term study of an entire 10 × 10 km with an approximate center at Ha- community. By restricting the study to popula- cienda Monteverde, just to the northwest of the tion and community levels, we were able to sur- small town of Bagaces (see map in Frankie et al. vey thousands of plants for more than six years. 2002: 329; maps in chapter 1). It consisted of The composition of plant life forms observed mostly wooded savanna with two riparian forest in the Bagaces study area is presented in table 2.1. corridors and several creeks (both with perennial Herbs were the most common plant life form water), limited dry deciduous forest, and scat- (36%), followed closely by the trees. The percent- tered oak forest (70%). Some of the nonriparian ages of shrubs (16.0%) and climbers (16.2%) habitats had experienced variable damage from (lianas and vines combined) were each about wildfires during the past 20 years. The site also half that of the herbs. Epiphytes formed only a had regenerating second-growth forest (10%), small percentage of plant life forms, and parasitic cattle pasture (12%), cropland (6%), and the town plants (mistletoes) were represented by only two of Bagaces (2%). With the exception of one no- species. A terrestrial cactus rounded out the list table grass species, exotic plants were rare in the with only one species. The total number of flower- wildland habitats (Frankie et al. 1997). ing plants observed to date was 487 species. Based The survey was conducted by periodically on the current rate of discovery, we expect the visiting 14 different subsites within the Bagaces total to reach about 550 species. This projection study area. Each was usually visited at least is also based on the number of flowering plant once a month, and several were visited more species, about 1,000, that are known from the en- frequently. Field observations were made on all tire lowland dry-forest zone of Costa Rica (Janzen flowering plants from mid-1997 to project ter- and Liesner 1980; D. H. Janzen pers. comm.).

18 FRANKIE ET AL. frankie chap 02 8/20/03 7:48 PM Page 19

TABLE 2.1 Composition of Plant Life Forms Recorded at the Bagaces Study Area

plant life form number of species percent of total

Herbsa 175 36.0 Trees 144 29.6 Shrubs 78 16.0 Vines 56 11.5 Lianas 23 4.7 Epiphytesb 8 1.6 Parasites 2 0.4 Other (terrestrial cactus) 1 0.2 Total 487

aAbout 25 percent of herbs are woody based; the rest are herbaceous. bEpiphytes are characteristically rare in the dry forest.

FIGURE 2.1. Annual ﬂowering phenology of tree and nontree angiosperm species in the Bagaces study site. n = 140 tree species; n = 343 nontree species.

compared with 113 species in the 1969–70 study AREAWIDE FLOWERING PHENOLOGY (Frankie et al. 1974). The addition of 27 tree Flowering phenologies for tree and nontree species in this study changed the forestwide tree species in our study area are presented in ﬁg- picture slightly by increasing the large number ure 2.1. A total of 140 tree species were plotted of February and March blooming species. The

FLOWERING PHENOLOGY AND POLLINATION SYSTEMS 19 frankie chap 02 8/20/03 7:48 PM Page 20

small second peak in flowering recorded in 1974 moths (family Sphingidae) that pollinated them. at the onset of the wet season did not appear Frankie and Haber (1983) and Frankie et al. with the new compilation (see also Janzen 1967). (1983) characterized the large- and small-bee There was a wide diversity of flowering times (or pollination systems (see also Heithaus 1979). In patterns) among tree species, but most could be each of the studies cited in this paragraph, one categorized as strictly dry-season or wet-season type of pollination system was described. No at- bloomers. Within these patterns, in both dry- tempt has been made to investigate the various and wet-season flowering species, duration of pollination systems as an interactive community flowering ranged from brief through intermedi- of plants and pollinators in a given area. ate to extended. Most species had annual pat- In 1996, four of us (GF, SV, PR, and NZ) ini- terns, with one flowering period per year, but a tiated a study to examine in detail the relation- few species had subannual patterns, with flower- ships of Africanized honeybees and native bees ing twice a year in two separate seasons. A few foraging on the flora at the Bagaces study site species had a supra-annual pattern in which (Frankie et al. 2000, 2002). Although the study they skipped one or more years of flowering. Di- was focused on bees, which number about 250 verse patterns of flowering behavior have also species in this area (Frankie et al. 1983), the sur- been observed in wet-forest trees of La Selva in vey also provided an opportunity to reexamine the Atlantic lowlands (Frankie et al. 1974; Opler most other pollination systems in the area. Gris- et al. 1976, 1980; Newstrom et al. 1994a,b). wold et al. (2000) calculated that the entire Costa In every month, substantial flowering was Rican bee fauna consists of at least 98 genera observed at the community level consisting and 785 species. of herbs, shrubs, climbers, epiphytes, parasitic plants, and a terrestrial cactus (n = 343 species). POLLINATION SYSTEMS OF THE DRY FOREST In the lowest month, April, at least 42 nontree In this section we provide an overview of the pol- species were in flower (fig. 2.1). After April, the lination systems in our study area and the most numbers of species in flower increased contin- prominent characteristics of both the plants uously each month until a peak of 160 species and their pollinators. We used three methods for was reached in October, which is characteristi- categorizing a plant according to its pollination cally the wettest month of the year in the dry for- system. The most important was through re- est. Flowering in the nontrees declined sharply peated field observations of each plant in several after November. A slight rise in flowering was different habitats. Secondarily, we used the clas- recorded in February, which corresponds to the sic floral and pollinator syndromes for guidance period when a high number of climber species (Proctor et al. 1996). Finally, we called on pub- bloom. lished accounts and our general field experiences with pollinators and pollination that began in 1970 in this dry forest. SURVEY OF POLLINATION SYSTEMS We found four prominent pollination systems Considerable community pollination work was in both tree and nontree life forms. These were conducted between 1970 and 1980 in the low- large-bee, small-bee/generalist, moth, and bat land dry forest of Costa Rica. Heithaus et al. (1974, pollination types (table 2.2). The large-bee sys- 1975) analyzed foraging patterns and resource tem occurs frequently in the trees (26.4%) and utilization in bats and the plant species they pol- about half as frequently (12.5%) in the nontrees. linate. He and his colleagues also studied the Large-bee flowers are usually large, brightly col- fruits consumed and seeds dispersed by the bats. ored, and bilaterally symmetrical, and they com- Haber and Frankie (1989) and Haber (1984) de- monly bloom during the long dry season (Frankie veloped detailed information on floral character- et al. 1983). They are mostly adapted for visita- istics of dry-forest plants in relation to the hawk- tion and pollination by bees measuring 12 mm

20 FRANKIE ET AL. frankie chap 02 8/20/03 7:48 PM Page 21

TABLE 2.2 Pollination Systems of Dry-Forest Plants in the Bagaces Study Area

trees nontree plants all plants

percent percent percent pollinator group species of total species of total species of total

Large beea 38 26.4 40 12.5 78 16.8 Small bee/generalista,b 39 27.1 206 64.2 245 52.7 Mothc 25 17.4 9 2.8 34 7.3 Bat 9 6.2 2 0.6 11 2.4 Hummingbird 0 0.0 10 3.1 10 2.1 Wasp 1 0.7 0 0.0 1 0.2 Beetle 4 2.8 1 0.3 5 1.1 Butterﬂy 0 0.0 4 1.2 4 0.9 General insect 14 9.7 5 1.6 19 4.1 Unassigned insectd 8 5.5 44 13.7 52 11.2 Fig waspe 2 1.4 0 0.0 2 0.4 Wind 4 2.8 0 0.0 4 0.9 Total 144 321 465

aSee Frankie et al. (1983) for description of large-bee and small-bee systems. bSmall bees predominate in this category. cIncludes hawkmoths and perching moths. dLimited opportunity for study. eFicus species are pollinated by specialized chalcid wasps in the family Agaonidae.

or more in length, however, small bees such as (table 2.2). This system accounted for slightly megachilids also visited many of these species more than half of all pollination types. Small- and probably account for some of the pollination bee/generalist flowers are usually small, often (Frankie et al. 1976). Large bees included mostly cream colored, and radially symmetrical and anthophorids in the genera Centris (the most typically occur in inflorescences with numerous abundant genus, with 12 species, in our study flowers (Frankie et al. 1983). The blooming pe- area), Epicharis, Mesoplia, and Mesocheira. Sev- riods of species within this system are spread eral species of Xylocopa and Euglossini (orchid throughout the year, and flowers are visited by a bees) were occasional visitors to large-bee flowers wide variety of small bees (less than 12 mm in (Frankie et al. 1983). With the exception of Eu- length), social and solitary wasps, and several fly glossini, most large bees confine their breeding families. Some are also visited by certain groups and nesting to the long dry season (Sage 1968; of beetles. Common bee taxa included several Frankie et al. 1983; S. Vinson and G. Frankie genera of anthophorids, megachilids, halictids, unpubl. data). and stingless bees (meliponines). Small bees are The small-bee/generalist system was more the most important pollinators in this system, than twice as common in the nontree life form although wasps and flies may also account for (64.2%) as compared with the trees (27.1%) some of the pollination.

FLOWERING PHENOLOGY AND POLLINATION SYSTEMS 21 frankie chap 02 8/20/03 7:48 PM Page 22

The moth pollination type is common in the tial examination of bat floral resources and their trees (17.4%) and noticeably less frequent in interaction with seven flower-visiting bat species the nontrees (2.8%) (table 2.2). This system con- they studied. sists of flowers that are morphologically vari- Other distinct pollination types in this forest able, mostly white or cream-colored and varying site, which are all rare, are also presented in from narrow tubed to broad, open cup-shaped, table 2.2. The general insect type represents plant or brush flowers. Pollinators consist of two basic species that have a wide variety of visitors with types of moths, hawkmoths (about 70 species no clear group as the primary visitor or pollina- in the Sphingidae), which hover while feeding, tor. With more study, we expect that many of and the smaller perching moths of several fam- these species might be placed in the small-bee/ ilies (Baker and Baker 1983; Haber and Frankie generalist system. There has been insufficient 1989; W. A. Haber unpubl. data). Flowering times time and opportunity to study those in the cate- of moth plants are distributed throughout the gory “unassigned insect,” but the flowers show year with a prominent peak in the wet season morphological adaptation for insect visitors. Fi- (Haber and Frankie 1982, 1989; Haber 1983, nally, plants adapted for wind pollination are rare 1984). Wet-season concentration of moth flowers in this forest. corresponds to the period when adult moths Despite the relative ease of placing most are most abundant and also coincides with the plant species in a particular pollination system enormous flush of new leaves available to moth (table 2.2), there were many pollinator cross- larvae shortly after the onset of the wet season overs between systems. Examples include small (Frankie et al. 1974; Janzen 1988; Haber and bees (and wasps) commonly visiting large-bee Frankie 1989). flowers and, on occasion, large bees visiting An interesting group of trees (e.g., Albizia small-bee/generalist plants (Frankie et al. 1976, caribaea, Enterolobium cyclocarpum, Lysiloma 1983, 1997). Hawkmoths commonly visited bat auritum [Mimosaceae], and Guarea glabra [Meli- flowers, although the reverse was rare (Heithaus aceae]) with flowers adapted for perching moths et al. 1975; Haber and Frankie 1989). There bloom during mid- to late dry season well before were also several cases of large numbers of bee the first rains. These trees are apparently polli- species visiting nocturnal bat and moth flowers nated by moths that are active as adults in the in the morning for residual resources. Plants dry season, whose caterpillars take advantage of such as Crescentia alata (Bignoniaceae), Bomba- several tree species that flush new leaves before copsis quinata (Bombacaceae), Hymenaea cour- the rainy season begins; some species eat the baril (Caesalpiniaceae), Enterolobium cyclocarpum, flowers. Inga vera, Lysiloma auritum, Pithecellobium lance- The bat pollination system was found in the olatum, Samanea saman (all Mimosaceae), and trees (6.2%) and to a much lesser extent in Manilkara zapota (Sapotaceae) often attract high the nontrees (0.6%). Overall, it accounted for numbers of Africanized honeybees or native only 2.4 percent of the systems in the study area bees (or both) for relatively short periods (about (table 2.2). As with the moth flowers, bat flowers 30 min.) during and just after sunrise. Some bee are morphologically variable (Heithaus et al. groups are also attracted to wind-pollinated 1975). Compared with moth flowers, they are plants such as Quercus oleoides (Fagaceae) and a larger and supply their pollinators with more few grass species for their pollen. The flowers of nectar per flower (and pollen in some cases). Trigonia rugosa (Trigoniaceae), a common liana, They are also structurally strong and often pro- attract hawkmoths just prior to sunrise; then vide landing bases for bats. In contrast to moth large bees arrive for brief foraging at sunrise flowers, flowering times of bat flowers tend to be only, followed for the next few hours by a mix of spread out evenly over the year and are broadly small bees, wasps, flies, butterflies, and beetles. overlapping. Heithaus et al. (1975) provide a par- Diverse visitor/pollinator assemblages at flowers

22 FRANKIE ET AL. frankie chap 02 8/20/03 7:48 PM Page 23

of particular pollination systems have been flowering bee trees that are so abundant late in known for some time (Baker and Hurd 1968). the dry season. They have not, however, received enough eco- The small-bee/generalist system is the least logical and evolutionary study (Johnson and charismatic of all the other systems (table 2.2). Steiner 2000). Yet, by its abundance and pervasiveness, the system plays a vital ecological role in the commu- COMMUNITY STRUCTURE OF nity of dry-forest pollinators and visitors. It is DRY-FOREST POLLINATION SYSTEMS found commonly in trees and nontrees, and it is The recent and intensive phenological and pol- not restricted to any particular season. Further, lination studies provide important insights on a wide variety of small bees are probably the most how most flowering resources and their pollina- important pollinators of the system, but other tors (fig. 2.1 and table 2.2) are structured in di- groups, such as flies, wasps, and beetles, may verse pollination systems within the dry-forest also contribute to pollination. At the community community. Overall combined flowering times level, the frequent occurrence of this flower type (fig. 2.1) indicated that substantial floral re- in every month ensures that a wide variety of sources are available every month, with a dry- flower visitors will always have floral foods. season peak for trees and wet-season peak for nontree species. When large-bee, small-bee/ OTHER COMMUNITY POLLINATION STUDIES generalist, and moth systems (table 2.2) are Large community studies on tropical pollination superimposed on the annual flowering pattern, systems are rare. In this section we present re- three immediately new patterns emerge. First, sults of investigations on pollination systems of large-bee flowers mostly bloom in the dry sea- the Monteverde Cloud Forest Reserve and the low- son. Nesting times of most large bees are also land wet forest of La Selva (both of Costa Rica). largely restricted to this season (Sage 1968; These are followed by a review of recent pollina- Frankie et al. 1983; S. Vinson and G. Frankie in tion findings from an Asian Dipterocarp forest. prep.). A few Centris species and other large bees Pollination Systems of the Monteverde Cloud Forest. are occasionally observed in low numbers in the Pollination systems of the dry forest (table 2.2) wet season, but their consistently low diversity were compared with those from the nearby mid- and abundance are considered of minor impor- elevation cloud forest at Monteverde (see maps tance for pollination. in chapters 1 and 8). In this examination we were The second pattern is that most moth-adapted interested in similarities and differences in the plant species flower in the wet season. This is systems and also in the diversity and composi- also the period when a peak in caterpillars feed- tion of pollinators in this cooler, wetter environ- ing on new foliage leads to production of new ment. Monteverde was also chosen for compar- wet-season adult moths for several months (Jan- isons because it is the only other location in the zen 1988; Haber and Frankie 1989). The rela- country where extensive community pollination tionships of new foliage to larval feeding and the work on trees and nontrees has been done (Haber production of new moths is obvious and well 2000a,b; Murray et al. 2000). based on fundamental biological needs of these The cloud forest study site consisted of about lepidopterans. What is not yet clear is the rela- 40 km2 ranging from 1,200 to 1,860 m in eleva- tionship between moths and some moth-adapted tion. Three habitat types occur there: (a) Pacific flowers that bloom at the end of the dry season, slope wet forest, (b) Pacific slope, or leeward, when leaflessness reaches its extreme in the dry cloud forest, and (c) Atlantic slope, or wind- forest. Caterpillars of these moths may special- ward, cloud forest. The almost exclusively wind- ize on those plant species that flush new leaves pollinated families Cyperaceae and Poaceae were before the rainy season begins. In addition, cater- omitted from the species count, as well as the pillars of some Lepidoptera eat flowers of mass- highly specialized, often nectarless, and poorly

FLOWERING PHENOLOGY AND POLLINATION SYSTEMS 23 frankie chap 02 8/20/03 7:48 PM Page 24

TABLE 2.3 Pollination Systems of Cloud-Forest Plants in the Tilarán Mountain Range near Monteverde

trees nontree plants all plants

percent percent percent pollinator group species of total species of total species of total

Large bee 28 7.4 65 9.0 93 8.5 Small bee/generalist 146 38.5 248 34.4 394 35.9 Motha 54 14.2 14 1.9 68 6.2 Bat 2 0.5 32 4.4 34 3.1 Hummingbird 11 2.9 95 13.2 106 9.6 Beetle 6 1.6 20 2.8 26 2.4 Butterﬂy 8 2.1 13 1.8 21 1.9 Wasp 10 2.6 8 1.1 18 1.6 Fly 0 0.0 2 0.3 2 0.2 General insect 49 12.9 0 0.0 49 4.5 Unassigned insectb 30 7.9 145 20.1 175 15.9 Fig waspc 6 1.6 0 0.0 6 0.5 Arboreal mammal 1 0.3 0 0.0 1 0.1 Wind 26 6.9 31 4.3 57 5.2 Unknownb,d 2 0.5 48 6.7 50 4.5 Total 379 721 1,100

Note: The study included only those plants occurring at elevations above 1,200 m. Species of Cyperaceae, Orchidaceae, and Poaceae were omitted. aIncludes hawkmoths and perching moths. bLimited opportunity for study. cFicus species are pollinated by specialized chalcid wasps in the family Agaonidae. dIncludes 48 species of Piperaceae in which the distribution of insect versus wind pollination is unclear.

studied Orchidaceae, leaving a list of 1,100 plant eral frequencies in both areas. The small-bee/ species in the study area. Results presented here generalist type is the most frequently observed are based on more than 20 years of inventory, system at each site; however, it is more common phenology, and pollination work by one of us in the dry forest (52.7% versus 35.9%). (WH) and other researchers at Monteverde Distinct differences between the two forest (Haber 2000a,b; Murray et al. 2000). types can be observed in the large-bee, bird, but- Many similarities in pollination systems be- terﬂy, and wind pollination systems (tables 2.2 tween the two sites can be observed in tables 2.2 and 2.3). The large-bee type occurs much less and 2.3. With rare exceptions (i.e., arboreal mam- commonly at Monteverde (8.5%). In the dry for- mals), both sites have the same general pollina- est this system is twice as abundant (16.8%). tion types. Three of these, moth, bat, and beetle This difference may be due to the large-bee fauna systems, occur at approximately the same gen- of the cloud forest, consisting of 31 known species

24 FRANKIE ET AL. frankie chap 02 8/20/03 7:48 PM Page 25

(W. A. Haber unpubl. data), which is much less beetles, and wasps, in addition to small bees; diverse and abundant than in the dry forest (2) those with very small flowers with unspecial- (Frankie et al. 1976, 1983; Heithaus 1979). ized morphology presumably pollinated by small An example of dry-forest bee diversity and flies and wasps, with very low representation by abundance was demonstrated in 1972 through bees (W. A. Haber unpubl. data). Finally, wind intensive sampling conducted on the large bee- pollination occurs only rarely in both forests; flower tree, Andira inermis (Fabaceae) at the however, it is more common in cloud-forest southern limit of the town of Liberia (see maps plants (5.2% versus 0.9%). in chapter 1 and Frankie et al. 2002: 329). At that Pollination Systems of La Selva. Kress and Beach time high numbers of Centris bees, consisting (1994) provide an overview of the pollination sys- of several species, were sampled from several tems of a relatively high proportion of understory trees and shown to make intertree movements at and overstory plants in this lowland wet-forest rates that accounted for cross-pollination of this site (excluding epiphytes). They conclude that self-incompatible species (Bawa 1974; Frankie et bees may pollinate as many as 60 percent of the al. 1976). Further, it was estimated that at peak canopy (n = 51 species) and subcanopy (n = 74 foraging periods there could be up to 50,000 species) plants and almost 40 percent in the un- bees on a single large flowering crown of A. in- derstory plants (n = 151 species). Other overall ermis, and most of these belonged to the genus common pollination systems were, respectively, Centris. A comparable legume tree at Monteverde hummingbird, beetle, small diverse insect, and might attract a maximum of 100–200 bees moth. As in the dry forest, Kress and Beach state (W. A. Haber unpubl. data). With regard to bee that plants in the “small diverse insect” category flowers, it is also noteworthy that the small-bee/ may ultimately turn out to be pollinated by bees. generalist system is more common in trees of Pollination Systems of a Malaysian Forest. Mo- the cloud forest as compared with the dry forest mose et al. (1998) conducted a plant-pollinator (38.5% versus 27.1%). This system occurs almost survey in a lowland dipterocarp forest in Sarawak, twice as frequently, however, in the nontree Malaysia, where they monitored 576 individu- plants of the dry forest (64.2% versus 34.4%). ally marked plants (including trees, lianas, and The hummingbird pollination system is ex- epiphytes) from 1992 to 1996 to determine flow- tremely conspicuous in the cloud forest with its ering periodicity, breeding systems, floral char- colorful yellow, orange, red, or blue flowers and acteristics, and pollinator types. They recognized abundant hummers. It accounts for 9.6 percent 12 pollination systems in 270 plant species (of of the Monteverde species, with epiphytes or 999 known species from the general study area). shrubs the most representative plants. In sharp Slightly more than 50 percent of all pollination contrast, the bird system is rare in the dry forest systems were bees. Further, they recognized five (2.1%). The difference between the two forests different bee systems, which they described in is also evident in the high species diversity and some detail. The beetle system was the second abundance of hummers in the cloud forest; 20 most common (20%), followed by two general cloud-forest species versus 11 dry-forest species insect systems (17%). The only other notable sys- (see chapter 12; see also Feinsinger et al. 1986; tem was the bird system (9.5%). Together, these Murray et al. 1987, 2000). four major systems made up 95 percent of the The general insect system occurred at about known pollination systems in this forest. the same frequency in both forest types (4.1% in dry forest versus 4.5% in cloud forest). This pol- POLLINATION IN THE CHANGING lination type has received more study in cloud DRY FOREST OF COSTA RICA forests and consists of two groups of plants: (1) those whose flowers are regularly visited by Given that dry-forest plant communities show a wide variety of insects including butterflies, definite reproductive structure from a pollination

FLOWERING PHENOLOGY AND POLLINATION SYSTEMS 25 frankie chap 02 8/20/03 7:48 PM Page 26

perspective, one must wonder how this structure tropical and temperate regions as well (Frankie will endure present and future human distur- et al. 2000, 2002). These topics are briefly ex- bances (see also Johnson and Steiner 2000 and plored in this section. references therein). In this regard, Hamrick and Apsit (chapter 3) consider the effects of for- MORE BASIC PLANT INVENTORY WORK est fragmentation on pollen and gene flow in the Plant inventory surveys in the Neotropics are ex- Costa Rican dry forest and conclude that, at pres- tremely important in order to determine the full ent, fragmentation is not significantly affecting range of species present in given regions, eco- outcrossing. But what happens in the future as systems, or habitats. Dry forests of Mesoamerica this forest becomes even more developed (chap- especially need plant inventory work because of ters 1 and 16)? Bawa (chapter 4) considers the the many distinct habitats (Frankie 1997) that effects of local and global changes on pollinators make up this forest type. In addition to species and their ability to continue providing pollina- lists, these surveys help to (1) determine frequen- tion services to plants (see also Janzen 1974). cies of species occurrence, (2) define geograph- Our past and ongoing monitoring studies of ical and habitat limits of species, (3) construct large bees in the Bagaces study area and nearby phenological patterns, and (4) identify which Liberia indicated that these important pollinators habitats are in urgent need of attention. are progressively declining (Frankie et al. 1976, 1997, unpubl. data). This is understandable in INFORMATION ON GENERALIZED the vicinity of Liberia, which is the fastest-growing VERSUS SPECIALIZED POLLINATION town in the region. What is not clear is why large- Common occurrence of the small-bee/general- bee numbers also seem to be declining in the ist and general insect systems in both the dry Bagaces study area, which still has much intact forest and cloud forest raises many questions habitat. Our recent assessments suggest that about the effectiveness of each visitor type in (1) loss of habitat due to agricultural development pollination (Johnson and Steiner 2000). This scattered throughout the region, (2) increased system is also common in the La Selva site of the and intensified wildfires over the past 20 years Atlantic lowland forest (see maps in chapter 1) (Frankie et al. 1997), and (3) reduction of pre- (K. Bawa and G. Frankie pers. obs.). More spe- ferred floral hosts (Frankie et al. 2002) all seem cifically, we need to know the comparative ca- to be negatively and cumulatively affecting these pacity of each visitor type for carrying pollen and bees (G. Frankie and S. Vinson unpubl. data). which visitors are moving between plants to promote outcrossing. We also need to know more about periods of stigmatic receptivity in relation FUTURE STUDIES AND APPLICATIONS to visitor activity. FOR CONSERVATION WORK

Despite the threats that development poses (chap- MONITORING AND RESTORING POLLINATORS ters 21–23) to the dry-forest plants, their pollina- There is little doubt that at least some pollinators, and other ﬂoral visitors, there is still much tor types, such as large bees, are declining in the to be learned about the pollination ecology of dry forest (Frankie et al. 1997) as habitat loss lowland dry-forest plants, especially in areas that through human development continues (see have largely intact habitats. We urge that these chapters 1, 4, 7, 12, 13, and 21). In large areas that studies be initiated as soon as possible while valu- still contain substantially protected natural habi- able biological information can still be gathered tat, long-term monitoring programs need to be and effectively applied to addressing conserva- developed for the important pollinator groups, tion issues. In addition to the obvious value of especially large and small bees, bats, and moths. this information, there is the recognition that At the same time, basic biological and ecological lessons learned here have application in other information must be obtained for use in restor-

26 FRANKIE ET AL. frankie chap 02 8/20/03 7:48 PM Page 27

ing populations of pollinators where downward actively enhancing other pollinator resources trends are obvious (Vinson et al. 1993; chap- (e.g., nesting material for bees) has great appeal ter 6). It is important to stress that this kind of to private landowners and some public land stew- work must start now, while appropriate habitat ards in contrast to just “setting habitat aside for and pollinator populations are still extant (see wildlife.” also discussion on this topic by Janzen 1974). In all three cases, the pollinator and pollination biologist must be an active member of the APPLICATION OF BIOLOGISTS’ KNOWLEDGE outreach effort. Further, the biologist will need TO CONSERVE AND RESTORE POLLINATORS to continue the outreach for an extended period, Pollinator and pollination biologists are just be- and this could mean employing follow-up pro- ginning to recognize and document the decline fessional monitors and perhaps training volun- of pollinators worldwide (Allen-Wardell et al. teers from local nongovernmental organizations 1998). They are also the professionals who know to monitor, as well. The extended project period best how to monitor declines and sound the alarm would provide the principal investigator of the about impending losses to plants and eventually project with many opportunities for transferring to humans, with their dependence on plants. environmental knowledge to types of various But how does a classically trained biologist be- audiences. come involved in taking action on decline issues? There is no clear answer to this question, but it is certain that to do nothing will probably lead ACKNOWLEDGMENTS to pollinator population levels that are unable to We thank the California Agriculture Experi- provide vital plant reproductive services (Janzen ment Station and the Texas Agricultural Exper- 1974; Daily 1997, esp. chap. 8). iment Station for their support of this research. There are at least three courses of action that Several grants from the National Science Foun- pollinator biologists could pursue to assist de- dation and the National Geographic Society sup- clining pollinator populations. First, they can ported the early periods of this work. The David collaborate with other biologists/land stewards A. Stewart family kindly allowed us to use their who are also concerned about decline of specific property for our research. They also generously organisms (e.g., birds, mammals) and habitat. provided us with logistical help and were agree- Second, biologists could also work in a variety of able to setting aside sections of their land for ways toward conserving specific areas known to long-term monitoring of bees and plants. Linda naturally harbor healthy populations of polli- Newstrom and Robbin Thorp kindly read an nators, preferably several types. Biologists are early draft of this chapter. aware, through years of field experience, which areas have good diversity and abundance of bees and moths, for example. As in the first case, this REFERENCES kind of project will require that pollinator biologists collaborate with land stewards to ensure Allen-Wardell, G., et al. 1998. The potential consequences of pollinator declines on the conserva- that high-quality habitats will be given special tion of biodiversity and stability of food crop attention. yields. Conservation Biology 12:8–17. In areas where much is known about the re- Baker, H. G., and I. Baker. 1983. Floral nectar sugar quirements of pollinators and where decline is constituents in relation to pollinator type. In strongly suspected, a third possible course of Handbook of experimental pollination biology, ed. C. E. Jones and R. J. Little, 117–141. New York: action is restoration. In this case the goal of Van Nostrand-Reinhold. restoration should be to add known floral and Baker, H. G., and P. D. Hurd Jr. 1968. Intrafloral other resources preferred by pollinators (see ecology. Annual Review of Entomology 13:385– chapter 6). Hands-on work such as planting or 414.

FLOWERING PHENOLOGY AND POLLINATION SYSTEMS 27 frankie chap 02 8/20/03 7:48 PM Page 28

Bawa, K. S. 1974. Breeding systems of tree species Griswold, T., F. D. Parker, and P. E. Hanson. 2000. of a lowland tropical community. Evolution 28: An inventory of the bees of Costa Rica: The 85–92. myth of the depauperate tropics. In Proceedings Bullock, S. H., H. A. Mooney, and E. Medina, eds. of the Sixth International Conference on Apicul- 1995. Seasonally dry tropical forests. Cambridge: ture in Tropical Climates, Costa Rica, 152–16. San Cambridge University Press. 450 pp. José: International Bee Research Association. Daily, G. C., ed. 1997. Nature’s services, societal de- 226 pp. pendence on natural ecosystems. Covelo, Calif.: Haber, W. A. 1983. Hylocereus costaricensis. In Island Press. Costa Rican natural history, ed. D. H. Janzen, Feinsinger, P., K. G. Murray, S. Kinsman, and W. H. 252–53. Chicago: University of Chicago Press. Busby. 1986. Floral neighborhood and pollina- ———. 1984. Pollination by deceit in a mass- tion success in four hummingbird-pollinated flowering tropical tree, Plumeria rubra L. (Apo- cloud forest plant species. Ecology 67:449–64. cynaceae). Biotropica 16:269–75. Frankie, G. W. 1997. Endangered havens for diver- ———. 2000a. Plants and vegetation. In Monte- sity. BioScience 47:322–24. verde: Ecology and conservation of a tropical cloud Frankie, G. W., and W. A. Haber. 1983. Why bees forest, ed. N. Nadkarni and N. Wheelwright, move among mass flowering neotropical trees. 34–94. New York: Oxford University Press. In Handbook of experimental pollination biology, ———. 2000b. Vascular plants of Monteverde. In ed. C. E. Jones and R. J. Little, 361–72. New York: Monteverde: Ecology and conservation of a tropical Van Nostrand-Reinhold. cloud forest, ed. N. Nadkarni and N. Wheel- Frankie, G. W., H. G. Baker, and P. A. Opler. 1974. wright, 457–518. New York: Oxford University Comparative phenological studies of trees in Press. tropical wet and dry forests in the lowlands of Haber, W. A., and G. W. Frankie. 1982. Pollination Costa Rica. Journal of Ecology 62:881–919. ecology of Luehea (Tiliaceae) in Costa Rican Frankie, G. W., P. A. Opler, and K. S. Bawa. 1976. deciduous forest. Ecology 63:1740–50. Foraging behavior of solitary bees: Implica- ———. 1989. A tropical hawkmoth community: tions for outcrossing of a Neotropical forest tree Costa Rican dry forest Sphingidae. Biotropica species. Journal of Ecology 64:1049–57. 21:151–72. Frankie, G. W., W. A. Haber, P. A. Opler, and K. S. Heithaus, E. R. 1979. Flower-feeding specialization Bawa. 1983. Characteristics and organization of in wild bee and wasp communities in seasonal the large bee pollination system in the Costa Neotropical habitats. Oecologia 42:179–94. Rican dry forest. In Handbook of experimental Heithaus, E. R., P. A. Opler, and H. G. Baker. 1974. pollination biology, ed. C. E. Jones and R. J. Little, Bat activity and pollination of Bauhina pauletia: 411–47. New York: Van Nostrand-Reinhold. Plant-pollinator coevolution. Ecology 55:412–19. Frankie, G. W., S. B. Vinson, M. A. Rizzardi, T. L. Heithaus, E. R., T. H. Fleming, and P. A. Opler. Griswold, S. O’Keefe, and R. R. Snelling. 1997. 1975. Foraging patterns and resource utiliza- Diversity and abundance of bees visiting a mass tion in seven species of bats in a seasonal trop- flowering tree in disturbed seasonal dry forest, ical forest. Ecology 56:841–54. Costa Rica. Journal of the Kansas Entomological Janzen, D. H. 1967. Synchronization of sexual re- Society 70:281–96. production of trees within the dry season in Cen- Frankie, G. W., S. B. Vinson, S. O’Keefe, and M. A. tral America. Evolution 21:620–37. Rizzardi. 2000. Assessing impacts of African- ———. 1974. The deflowering of Central America. ized honey bees in Costa Rica and California: Natural History 83:48–53. A progress report. In Proceedings of the Sixth ———. 1983. Natural history of Costa Rica. Chi- International Conference on Apiculture in Tropi- cago: University of Chicago Press. 816 pp. cal Climates, Costa Rica, 157–61. San José: Inter- ———. 1988. Characterization of a Costa Rican dry national Bee Research Association. 226 pp. forest caterpillar fauna. Biotropica 20:120–35. Frankie, G. W., S. B. Vinson, R. W. Thorp, M. A. Janzen, D. H., and R. Liesner. 1980. Annotated Rizzardi, N. Zamora, and P. S. Ronchi. 2002. check list of plants of lowland Guanacaste Coexistence of Africanized honey bees and Province, Costa Rica, exclusive of grasses and native bees in the Costa Rican seasonal dry for- non-vascular cryptograms. Brenesia 18:15–90. est. In Proceedings of the Second International Con- Johnson, S. D., and K. E. Steiner. 2000. General- ference on Africanized Honey Bees and Bee Mites, ization versus specialization in plant pollina- ed. E. H. Erickson, R. E. Page, and A. A. Hanna, tion systems. Trends in Ecology and Evolution 15: 327–39. Medina, Ohio: A. I. Root Co. 140–43.

28 FRANKIE ET AL. frankie chap 02 8/20/03 7:48 PM Page 29

Kress, W. J., and J. H. Beach. 1994. Flowering plant ural history of a Neotropical rain forest, ed. L. A. reproductive systems. In La Selva, ecology and McDade et al., 142–60. Chicago: University of natural history of a Neotropical rain forest, ed. Chicago Press. L. A. McDade et al., 161–82. Chicago: University Noss, R. F., and A. Y. Cooperrider. 1994. Saving na- of Chicago Press. ture’s legacy: Protecting and restoring biodiversity. Momose, K., T. Yumoto, T. Nagamitsu, M. Kato, H. Covelo, Calif.: Island Press. 416 pp. Nagamasu, S. Sakai, R. D. Harrison, T. Itioka, Opler, P. A., H. G. Baker, and G. W. Frankie. 1975. A. A. Hamid, and T. Inoue. 1998. Pollination bi- Reproductive biology of some Costa Rican Cor- ology in a lowland Dipterocarp forest in Sara- dia species (Boraginaceae). Biotropica 7:234–47. wak, Malaysia. I. Characteristics of the plant- Opler, P. A., G. W. Frankie, and H. G. Baker. 1976. pollinator community in a lowland Dipterocarp Rainfall as a factor in the release, timing, and forest. American Journal of Botany 85:1477–1501. synchronization of anthesis by tropical trees Murray, K. G., P. Feinsinger, W. H. Busby, Y. B. Lin- and shrubs. Journal of Biogeography 3:231–36. hart, J. H. Beach, and S. Kinsman. 1987. Evalu- Opler, P. A., G. W. Frankie, and H. G. Baker. 1980. ation of character displacement among plants Comparative phenological studies of shrubs in in two tropical pollination guilds. Ecology 68: tropical wet and dry forests in the lowlands of 1283–93. Costa Rica. Journal of Ecology 68:167–18. Murray, K. G., S. Kinsman, and J. L. Bronstein. Proctor, M., P. Yeo, and A. Lack. 1996. The natural 2000. Plant-animal interactions. In Monte- history of pollination. Portland, Oreg.: Timber verde: Ecology and conservation of a tropical cloud Press. 479 pp. forest, ed. N. Nadkarni and N. Wheelwright, Sage, R. D. 1968. Observations on feeding, nesting, 245–302. New York: Oxford University Press. and territorial behavior of carpenter bees genus Newstrom, L. E., G. W. Frankie, and H. G. Baker. Xylocopa in Costa Rica. Annals of the Entomolog- 1994a. A new classification for plant phenology ical Society of America 61:884–89. based on flowering patterns in lowland tropical Vinson, S. B., G. W. Frankie, and J. F. Barthell. rain forest trees at La Selva, Costa Rica. Biotrop- 1993. Threats to the diversity of solitary bees in ica 26:141–59. a Neotropical dry forest in Central America. In Newstrom, L. E., G. W. Frankie, H. G. Baker, and Hymenoptera and biodiversity, ed. J. LaSalle and R. K. Colwell. 1994b. Diversity of long-term I. D. Gauld, 53–81. Wallingford, England: CAB flowering patterns. In La Selva, ecology and nat- International.

FLOWERING PHENOLOGY AND POLLINATION SYSTEMS 29 frankie chap 03 8/20/03 7:49 PM Page 30

chapter 3

Breeding Structure of Neotropical Dry-Forest Tree Species in Fragmented Landscapes

James L. Hamrick and Victoria J. Apsit

andscapes that once featured continu- high levels of allozyme genetic diversity and that L ously distributed, seasonal dry tropical forests the majority (86.5%) of the genetic diversity is are now characterized in much of Central Amer- found within rather than among populations ica by a matrix of pastures and agricultural lands (Hamrick 1994). Indirect estimates of gene flow punctuated by occasional patches of remnant among populations (Nm = the number of mi- forest, secondary forests, and narrow riparian grants per generation) of tropical tree species forest corridors. Fragmentation of these once indicate that historical levels of gene flow have continuous forests could adversely affect several been high enough to counteract the effects of aspects of the biology of tropical dry-forest tree genetic drift (i.e., Nm > 1.0). Contemporary meas- species (Harris 1984; Bierregaard et al. 1992). In ures of gene flow made in relatively undisturbed particular, changes in pollinator densities and continuous forests indicate that pollen flow rates behavior may disrupt or highly modify normal above 25 percent often occur over distances of breeding patterns in remnant populations (e.g., several hundred meters (Hamrick and Muraw- Frankie et al. 1997). Such changes in breeding ski 1990; Chase et al. 1996; Stacy et al. 1996). patterns can, in turn, modify levels and distri- The question, then, is, Does forest fragmen- bution of genetic diversity throughout local tation change pollinator behavior so that pollen landscapes (Nason and Hamrick 1997). movement is reduced to the extent that over Most tropical forest tree species are predom- several generations genetic diversity is lost via inantly outcrossing (Bawa 1974; Opler and Bawa genetic drift and rates of inbreeding increase 1978; Loveless 1992; Nason and Hamrick 1997). within the remaining fragments? In the follow- Population genetic studies have demonstrated ing sections we first examine theoretical expec- that, on average, tropical tree species have quite tations of the effects of fragmentation on the

30 frankie chap 03 8/20/03 7:49 PM Page 31

genetic composition of tropical dry-forest tree deforested areas would be lost, decreasing over- populations. We then review case studies of the all genetic diversity. breeding structure of several Neotropical dry- forest tree species located in fragmented land- SUBSEQUENT EFFECTS scapes to determine if gene flow is sufficient to With little gene flow via pollen or seeds between increase the low effective population sizes that fragments (i.e., Nm < 1) and with low effective often characterize fragmented tropical tree pop- population sizes within fragments, in subse- ulations. quent generations genetic drift would decrease genetic diversity within fragments, and genetic THEORETICAL EFFECTS OF differentiation among fragments would increase. FOREST FRAGMENTATION Inbreeding would also increase because of either increased self-fertilization or matings be- IMMEDIATE EFFECTS tween related individuals. However, most of The genetic composition of forest tree popula- the original genetic variation will be maintained tions immediately after fragmentation depends within the region; the genetic variation is only on several factors: (1) number, size, and distri- distributed differently. If individual fragment bution of fragments; (2) original density of the populations are destroyed or become extinct af- species; and (3) original distribution of genetic ter drift has restructured genetic diversity, over- variation. If numerous fragments are left after all genetic diversity will be reduced. disturbance and/or if fragments are large and In contrast, with gene flow between frag- close together, most of the genetic diversity in mented populations, the effects of genetic drift the original populations will be preserved. If, as and inbreeding will be reduced—that is, genetic is more likely, there are relatively few, small, diversity will be maintained within each frag- widely dispersed fragments, some low-frequency ment, and genetic differentiation among frag- alleles may be lost, but most genetic diversity ments will be relatively low. Thus, as a result of present in the original population will remain gene flow, subsequent loss of fragments will not (Young et al. 1996; Nason and Hamrick 1997). appreciably reduce the region’s genetic diversity. The density and dispersion of trees in the origi- In many highly disturbed landscapes, iso- nal forest will determine the number of indi- lated, freestanding trees are located in pastures, viduals in fragments and fragment-to-fragment agricultural fields, and fence rows. Such lone in- variation in population numbers. Finally, if ge- dividuals are reservoirs of genetic diversity and netic variation was distributed at random within may also play an important role in local breed- the original population (i.e., low genetic struc- ing populations (Levin 1995: chap. 16). Pollina- ture), each of the fragments should retain a rel- tors or seed dispersal agents that might not atively high proportion of the overall genetic di- move between fragments because of distance or versity and should be genetically similar to one their inability to perceive the fragment might another. On the other hand, if genetic variation move from a fragment to an isolated individual was patchily distributed originally (i.e., high ge- and then to a second fragment. Such movements netic structure), fragment populations will each would increase gene flow among fragments, in- contain less of the overall genetic diversity, and crease genetic diversity within fragments, and genetic differentiation among fragments will lower or maintain genetic differentiation among be higher. Overall genetic diversity for the total fragments. landscape will be maintained, however. An ex- Gene flow among fragments is, then, a key ception would occur if deforestation was not ran- element determining the effect of forest frag- dom with remnant forests occupying particular mentation on the genetic composition of rem- habitats (e.g., steep hillsides, ridgetops, riparian nant populations. Changes in pollinator or seed areas). Alleles that preferentially occur in the disperser behavior due to changes in the physical

BREEDING STRUCTURE OF TREE SPECIES 31 frankie chap 03 8/20/03 7:49 PM Page 32

configuration of biological landscapes could sites (separated by 100 and 200 m from other therefore have direct effects on the amount S. mombin), total rates of pollen immigration and spatial distribution of genetic variation. The averaged 45 percent (table 3.1). Average rates of < genetically effective size (Ne, which is generally pollen flow into small island populations ( 9 less than the number of reproductive adults) of individuals) isolated by distances as great as fragmented populations and the genetic related- 1,000 m were nearly twice those observed for ness of the adults within fragments will also the larger BCI populations. Fruit set and seed affect genetic changes due to genetic drift and germination were reduced in the small island inbreeding. populations relative to the larger BCI populations, indicating either that fruit production was pollen-limited or that higher rates of inbreeding BREEDING STRUCTURE OF NEOTROPICAL and subsequent abortion occurred. These re- DRY-FOREST TREE SPECIES sults indicate that the small insect pollinators of Recently molecular genetic markers have been S. mombin can move large distances across in- used to estimate rates and distances of pollen hospitable habitat (i.e., water). Although S. mom- movement in fragmented tropical landscapes. bin occurs at higher densities in the tropical dry From these analyses, breeding patterns and forests of Palo Verde National Park in Guana- pollen movement distances within study sites caste Province, Costa Rica, levels of gene flow can be described, and estimates can be made were generally consistent with the BCI results; of rates of pollen and seed dispersal into the study sites in continuous forests experienced sites (i.e., gene flow). In the sections that follow less gene flow than sites isolated by several hun- we discuss the results of several case studies of dred meters (J. Hamrick unpubl. data). the breeding structure of Neotropical dry-forest tree species. The studies involve species that oc- ENTEROLOBIUM CYCLOCARPUM (JACQ.) cur in the tropical dry forests of Costa Rica, al- GRISEB. (FABACEAE; MIMOSOIDAE; though in some instances the studies were con- COMMON NAME, GUANACASTE) ducted elsewhere within the range of the species. E. cyclocarpum is a large dominant tree of the Thus, it is possible that some differences might tropical dry forests of Central America and occur between locations owing to differences in northern South America. It is pollinated by the quantity and quality of pollinators or popu- moths, hawk moths, beetles, and other small lation densities of the trees. nocturnal insects and possibly by diurnal insects such as bees (Janzen 1982; Rocha and Aguilar SPONDIAS MOMBIN L. (ANACARDIACEAE; 2001a; K. Bawa pers. comm.; G. Frankie pers. COMMON NAME, JOBO) comm.). Mimosoid legumes such as E. cyclo- S. mombin, a canopy tree, is pollinated by small carpum have a pollination system that ensures diverse insects (including small bees, beetles, that seeds within individual fruits are usually flies, wasps, moths, and butterflies; Bawa et al. full-sibs. With mimosoid legumes, a single ball 1985). In the more mesic parts of its range it is of pollen (a polyad) originating from a single an early successional species, but in the seasonal pollen parent is deposited on the stigma. Poly- dry forests of Costa Rica it is a major compo- ads contain more pollen grains than the maxi- nent, along with its congener S. radlkoferi, of pri- mum number of ovules in the ovary (Kendrick mary forests (Janzen 1983). Nason and Hamrick and Knox 1989). As a result, the diploid geno- (1997) studied pollen flow into two semi-isolated type of the pollen parent can be inferred from S. mombin populations located in the continuous the genetic composition of the full-sib progeny forests of Barro Colorado Island (BCI), Panama, array. as well as into several small islands located near Apsit (1998) used standard paternity analysis BCI in Lake Gatun. In the two continuous forest procedures to describe the breeding structure of

32 HAMRICK AND APSIT frankie chap 03 8/20/03 7:49 PM Page 33

TABLE 3.1 Estimates of Pollen Flow Rates and Minimum Distances of Pollen Immigration for Tropical Dry-Forest Tree Species

rate of species location pollen ﬂow (%) distance(s) reference

spondias mombin Continuous forest BCI, Panama 45 100–200 Nason and Hamrick 1997 Islands 60–100 100–1,000

enterolobium cyclocarpum Paternity analyses Guanacaste, CR 80–100 120–350 Apsit et al. in review Full-sib analysis Guanacaste, CR 70–74 250 Apsit et al. 2001 Coto Brus, CR 50 500 Hamrick and Aldrich unpubl. data Cordia alliodora Turrialba, CR 3 280 Boshier et al. 1995 Cedrela odorata Coto Brus, CR 100 300 James et al. 1998

Note: BCI = Barro Colorado Island; CR = Costa Rica.

six fragmented populations located in two areas full-sib analyses over a three-year period to study of Guanacaste Province. The six populations each the breeding structure of a cluster of four E. cy- consisted of 3–12 adults. She showed that the clocarpum adults and four isolated trees located dominant pollen parents varied among mater- in a small area of dry forest in southern Costa nal trees within the same fragment and that there Rica. In all three years the same tree within the was statistically significant heterogeneity in the cluster was the dominant pollen parent (approx- genetic composition of the pollen received by imately 30%) of the other three trees. The four various maternal parents. Comparisons between isolated trees tended to receive pollen from a years also indicated significant heterogeneity in wider array of pollen parents than trees in the the genetic composition of pollen received by fragment. Approximately 50 percent of the effec- individuals. Estimates of pollen flow rates were tive pollen immigrated into the four-tree cluster high (80–100%) over distances ranging from from beyond 500 m. Finally, a more extensive 120 m to 350 m (table 3.1). The paternity analy- study of the breeding structure of E. cyclocarpum ses also indicated that isolated pasture trees do- populations located in habitats ranging from nated pollen to trees within forest fragments. primary forest to highly disturbed agricultural Apsit et al. (2001) also used full-sib prog- sites in the vicinity of Palo Verde National Park eny from individual fruits to analyze further (J. Hamrick unpubl. data) supported these ear- the breeding structure of one forest fragment lier studies and further indicated that the number (5 adults) over a three-year period. Estimates of pollen donors was largest for single trees lo- of the number of pollen parents per maternal cated in highly disturbed habitats. These results tree ranged from 27 to 83 trees across 5 mater- are consistent with those of Koptur (1984), who nal trees for the 1994 fruit crop and from 15 to demonstrated for seven Inga species that pollen 60 trees for the 1995 crop. An average of 70 often traveled as much as 500 m. to 74 percent of the pollen parents came from out- To summarize, it appears that most E. cyclo- side the study area (minimum distance 250 m). carpum trees in fragments receive pollen from Hamrick and Aldrich (unpubl. data) also used a diverse array of pollen donors and that more

BREEDING STRUCTURE OF TREE SPECIES 33 frankie chap 03 8/20/03 7:49 PM Page 34

than 50 percent of the pollen immigrates from compatible and inbred (Corner 1954; Baker 1959; beyond 500 m. Furthermore, trees isolated in Federov 1966) but are consistent with observa- pastures or agricultural fields play an important tions (e.g., Janzen 1971; Stiles 1975; Frankie et al. role in the overall breeding dynamics of these 1976) that tropical forest pollinators move large populations. distances between conspecific individuals. The high species diversity of tropical forests often CORDIA ALLIODORA (BORGAINACEAE; produces situations in which conspecifics are COMMON NAME, LAUREL) separated by a few hundred meters. As a result, C. alliodora, a large hermaphroditic, moth-polli- their pollinators have evolved the ability to locate nated tree (Bawa et al. 1985), is widespread on and fly between trees separated by several hun- the Atlantic and Pacific coasts of Central Amer- dred meters. Furthermore, in contrast to the ica. Boshier et al. (1995) estimated gene flow in findings of Ghazoul et al. (1998) on Shorea sia- a dense, nearly monospecific population located mensis from Thailand, forest fragmentation does on the Atlantic slope near Pavones, Turrialba, not appear to prevent intertree movement by the Costa Rica. First, they documented pollen move- pollinators of Neotropical dry-forest tree species. ment with a rare allele that occurred in a single Pollinator visitation to single isolated trees indi- tree within the study site. They also used standard cates that pollinators fly long distances over dis- paternity analyses. Both procedures indicated turbed habitats to reach flowering individuals. that most of the pollen was dispersed within In some cases pollinator movement may have 75 m of its source but that 3 percent of the pollen increased, perhaps in response to lower tree movements were beyond 280 m (table 3.1). densities (Murawski and Hamrick 1991; Stacy et al. 1996). The more limited pollen movement CEDRELA ODORATA (MELIACEAE; documented by Boshier et al. (1995) for a dense COMMON NAME, SPANISH CEDAR) population of Cordia alliodora is consistent with C. odorata is a functionally monoecious canopy such conclusions. tree that is locally abundant in the dry forests of These genetic-marker-based studies do not, Costa Rica. Its small white flowers are thought however, provide information on whether polli- to be moth-pollinated (Bawa 1985). James et al. nator visitation rates and fruit set are quanti- (1998) conducted a study of the mating system tatively lower for isolated trees. The available of C. ordorata in the highly fragmented dry forests evidence indicates that pollinator visitation rates of Coto Brus in southern Costa Rica. C. ordorata may be, on average, lower for trees in disturbed is predominantly outcrossing even in very small habitats (e.g., Andira inermis; Frankie et al. 1997) fragmented populations. One tree isolated in a and that probably, as a result, fruit set is lower pasture more than 300 m from the nearest for these trees (e.g., S. siamensis; Ghazoul et al. conspecific successfully set outcrossed fruit. 1998; E. cyclocarpum; Rocha and Aguilar 2001b). Evidently, its small-moth pollinators disperse pollen over several hundred meters of disturbed CONSERVATION IMPLICATIONS habitat. Estimates of gene flow rates into fragmented populations of tropical dry-forest tree species are CONCLUSIONS CONCERNING quite high. It is therefore tempting to generalize POLLEN MOVEMENT from these results that forest fragmentation In each of the studies just described, pollen was has had or will have few adverse effects on the often dispersed over a minimum of several hun- maintenance of genetic diversity in these dis- dred meters. These results are in stark contrast turbed landscapes. Such generalizations are to the predictions of early tropical biologists unwarranted, however, because the reproductive who felt that most tropical tree species were self- biology of other species may not be consistent

34 HAMRICK AND APSIT frankie chap 03 8/20/03 7:49 PM Page 35

with these results. A case in point is a study by forest tree populations should maintain the ge- Aldrich and Hamrick (1998) of Symphonia glob- netic structure seen in continuous, relatively un- ulifera populations located in a fragmented pre- disturbed forests. However, a pollen flow rate of montane wet forest near San Vito, Coto Brus, in 50 percent represents a gene flow rate of 25 per- southern Costa Rica. In this area S. globulifera is cent (i.e., 50% of the gametes come from local predominantly hummingbird-pollinated and bat- maternal trees, 25% of the gametes are locally dispersed. Using microsatellite loci to recon- produced male gametes, and 25% are immigrant struct genealogies of seedlings established in a male gametes). In a forest fragment with five

1-ha fragment, Aldrich and Hamrick (1998) adults and an effective population size (Ne) of demonstrated that more than 50 percent of the four (owing to uneven male and female repro- gametes within these seedlings originated from ductive success) that experiences a 50 percent

two trees located in an adjacent pasture. Similar rate of pollen flow, Nem would equal 1.0 (i.e., analyses of seedlings and saplings in nearby un- 4 × 0.25). At gene flow-drift equilibrium, 20 per- disturbed forests indicated that parentage was cent of the total genetic diversity would reside more evenly spread among adults in these sites. among populations. Thus, as a result of frag- Evidently, the release from competition of the mentation, genetic differentiation among popu- pasture trees led to increased flower production. lations might increase fourfold assuming that The larger floral resource apparently led to the gene flow rates and effective population sizes formation of territories by hummingbirds that are constant. In addition, low-frequency alleles typically move in a trap-line fashion among sev- may be lost, and inbreeding may increase. Es- eral adults in undisturbed forests. Bats concen- timates of Nm for fragmented populations of trate seeds of pasture trees below feeding roosts E. cyclocarpum are consistent with this example in the nearby fragment. Thus, for certain tropi- (Apsit 1998). Thus, even the high rates of ob- cal tree species, forest fragmentation can change served pollen flow may not be adequate to pre- pollinator behavior and adversely affect the ge- vent changes in the genetic composition of small, netic composition of subsequent generations. fragmented populations of dry-forest trees. Our generally inadequate knowledge of the nat- Isolated individuals appear to play a signifi- ural history and reproductive biology of tropical cant role in the overall breeding populations dry-forest tree species will therefore make gen- of tropical trees. Not only are such individuals eralizations concerning the effects of forest reservoirs of genetic diversity, but also they con- fragmentation on genetic diversity a chancy tribute pollen and perhaps even seeds to nearby proposition. fragments, increasing effective sizes of these At a scale of several kilometers, outcrossing populations. Often when management decisions tropical trees have approximately 95 percent of are being made for a region, the primary focus their total genetic diversity within their popula- is to preserve large, undisturbed habitats. Isolated tions (Hamrick and Loveless 1989). This result trees and small fragments are often regarded as is consistent with direct estimates of high rates unimportant and are destroyed. If the studies of gene movement in undisturbed, continuously presented earlier are generally applicable, the distributed forests (table 3.1). Assuming that loss of these lone trees could have significant these populations were in gene flow-drift equi- negative effects on breeding populations. Simi- librium prior to fragmentation, the number of larly, destruction or extinction of fragmented migrants per generation (Nm) would be approx- populations reduces overall size of a breeding imately 4.75, well above the level (Nm > 1) at population. Not only will genetic diversity be lost which genetic drift would be the predominate to the metapopulation of fragments, but the ac- factor influencing genetic structure. Superficially tive role of isolated individuals or fragment pop- it would appear that the high rates of pollen flow ulations in the local breeding populations of the observed among fragmented Neotropical dry- species will also be lost. It is possible that loss of

BREEDING STRUCTURE OF TREE SPECIES 35 frankie chap 03 8/20/03 7:49 PM Page 36

a few key individuals or of whole fragments may does not provide the information needed to un- produce a domino effect that further decreases derstand fully and predict the long-term evolu- effective population sizes of remaining frag- tionary and ecological responses of tropical dry- ments, increasing the inﬂuence of genetic drift forest trees to the human-modiﬁed landscapes and inbreeding on their genetic composition. they now occupy.

RESEARCH PRIORITIES REFERENCES The use of genetic markers to describe patterns Aldrich, P. R., and J. L. Hamrick. 1998. Reproduc- of mating within populations of Neotropical tree tive dominance of pasture trees in a fragmented tropical forest mosaic. Science 281:103–5. species produces quantitative estimates of selfing Apsit, V. J. 1998. Fragmentation and pollen move- and outcrossing rates in viable seeds, of the ge- ment in a Costa Rican dry forest tree species. netic relatedness of individuals, and of the rates Ph.D. diss., University of Georgia. 122 pp. and patterns of gene movement. Availability of Apsit, V. J., J. L. Hamrick, and J. D. Nason. 2001. such genetic markers allows unprecedented in- Breeding neighborhood size of a Costa Rican sights into many aspects of the reproductive dry forest tree species. Journal of Heredity 92: 415–20. biology of tropical tree species. However, these Baker, H. G. 1959. Reproductive methods as fac- analyses provide little or no information on other tors in speciation in flowering plants. Genetics aspects of the reproductive biology of these and Twentieth Century Darwinism: Cold Spring species. For example, genetic-marker-based ap- Harbor Symposium on Quantative Biology (24th), proaches do not provide information on the pro- 177–99. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory. portion of flowers receiving pollen, the propor- Bawa, K. S. 1974. Breeding systems of tree species tion of flowers that produce mature fruits or in a lowland tropical community. Evolution 28: viable seeds, or the relative viability of seeds or 85–92. seedlings produced by trees experiencing differ- Bawa, K. S., S. H. Bullock, D. R. Perry, R. E. Coville, ent habitat conditions. Nor do genetic-marker- and M. H. Grayum. 1985. Reproductive biology based studies provide insights into the effects of of tropical lowland rain forest trees. II. Pollina- tion systems. American Journal of Botany 72: changes in the quantity or quality (or both) of 346–56. pollinators. Genetic markers allow us to quan- Bierregaard, R. O. J., T. E. Lovejoy, V. Kopos, A. A. tify the final results of a number of processes dos Santos, and R. W. Hutchings. 1992. The that influence the breeding structure of popula- biological dynamics of tropical rainforest frag- tions, but they do not provide insights into how ments: A prospective comparison of fragments and continuous forest. BioScience 42:859–66. these results came about. Boshier, D. H., M. R. Chase, and K. S. Bawa. 1995. To understand more completely the effects Population genetics of Cordia alliodora (Boragi- of habitat disturbance and population fragmen- naceae), a neotropical tree. 3. Gene flow, neigh- tation on the reproductive biology of species, borhood and population structure. American studies that thoroughly integrate observations Journal of Botany 82:484–90. of pollinator visitation and pollen deposition Chase, M. R., C. Moller, R. Kessell, and K. S. Bawa. 1996. Distant gene flow in tropical trees. Na- with experiments and observations of fruit and ture 383:398–99. seed set are needed. These analyses should be Corner, E. J. H. 1954. The evolution of tropical coupled with quantitative estimates of the pop- forests. In Evolution as a process, ed. J. Huxley, ulation genetic parameters (e.g., gene flow) that A. C. Hardy, and E. B. Ford, 34–46. London: result. Such integrated studies should be per- Allen and Unwin. Federov, A. A. 1966. The structure of the tropical formed on species with varying densities, distri- rain forest and speciation in humid tropics. Jour- butions, pollinators, and the like so that robust nal of Ecology 54:1–11. comparisons and generalizations can be gener- Frankie, G. W., P. A. Opler, and K. S. Bawa. 1976. ated. The largely piecemeal approach of the past Foraging behavior of solitary bees: Implications

36 HAMRICK AND APSIT frankie chap 03 8/20/03 7:49 PM Page 37

for outcrossing of a neotropical forest tree In Advances in legume biology, ed. C. H. Stirton species. Journal of Ecology 64:1049–57. and J. L. Zarucci, 1–31. Monographs in System- Frankie, G. W., S. B. Vinson, M. A. Rizzardi, T. L. atic Botany 29. St. Louis: Missouri Botanical Griswold, S. O’Keefe, and R. R. Snelling. 1997. Garden. Diversity and abundance of bees visiting a mass Koptur, S. 1984. Outcrossing and pollinator limi- flowering tree species in disturbed seasonal dry tation of fruit set: Breeding systems of neotrop- forest, Costa Rica. Journal of the Kansas Ento- ical Inga trees (Fabaceae: Mimosoideae). Evolu- mological Society 70:281–96. tion 38:1130–43. Ghazoul, J., K. A. Liston, and T. J. B. Boyle. 1998. Levin, D. A. 1995. Plant outliers: An ecogenetic Disturbance induced density-dependent seed set perspective. American Naturalist 145:109–18. in Shorea siamensis (Dipterocarpaceae), a tropi- Loveless, M. D. 1992. Isozyme variation in tropical forest tree. Journal of Ecology 86:462–73. cal trees: Patterns of genetic organization. New Hamrick, J. L. 1994. Genetic diversity and conser- Forests 6:67–94. vation in tropical forests. In Proceedings: Inter- Murawski, D. A., and J. L. Hamrick. 1991. The ef- national Symposium on Genetic Conservation and fect of density of flowering individuals on the Production of Tropical Forest Tree Seed: 14–16 June mating systems of nine tropical tree species. 1993, Chiang Mai, Thailand, ed. R. M. Drysdale, Heredity 67:167–74. S. E. T. John, and A. C. Yopa, 1–9. Muak-Lek, Nason, J. D., and J. L. Hamrick. 1997. Reproductive Saraburi, Thailand: ASEAN-Canada Forest Tree and genetic consequences of forest fragmen- Centre. tation: Two case studies of neotropical canopy Hamrick, J. L., and M. D. Loveless. 1989. The trees. Journal of Heredity 88:264–76. genetic structure of tropical tree populations: Opler, P. A., and K. S. Bawa. 1978. Sex ratios in Association with reproductive biology. In Plant tropical forest trees. Evolution 32:812–21. evolutionary biology, ed. J. H. Bock and Y. B. Lin- Rocha, O., and G. Aguilar. 2001a. Variation in the hart, 131–46. Boulder, Colo.: Westview Press. breeding behavior of the dry forest tree, Enter- Hamrick, J. L., and D. A. Murawski. 1990. The olobium cyclocarpum (guanacaste) in Costa Rica. breeding structure of tropical tree populations. American Journal of Botany 88:1600–1606. Plant Species Biology 5:157–65. ———. 2001b. Reproductive biology of the dry for- Harris, L. D. 1984. The fragmented forest: Island bio- est tree Enterolobium cyclocarpum Jacq. (Gua- geography theory and the preservation of biotic nacaste) in Costa Rica: A comparison between diversity. Chicago: University of Chicago Press. trees left in pastures and trees in continuous 211 pp. forest. American Journal of Botany 88:1607–14. James, T., S. Vege, P. Aldrich, and J. L. Hamrick. Stacy, E. A., J. L. Hamrick, J. D. Nason, S. P. Hub- 1998. Mating systems of three tropical dry for- bell, R. B. Foster, and R. Condit. 1996. Pollen est tree species. Biotropica 30:587–94. dispersal in low-density populations of three Janzen, D. H. 1971. Euglossine bees as long- neotropical tree species. American Naturalist distance pollinators of tropical plants. Science 148:275–98. 171:203–5. Stiles, G. F. 1975. Ecology, flowering phenology, ———. 1982. Variation in average seed size and and hummingbird pollination of some Costa fruit seediness in a fruit crop of a Guanacaste Rican Heliconia species. Ecology 56:285–301. tree (Leguminosae: Enterolobium cyclocarpum). Young, A., T. Boyle, and T. Brown. 1996. The pop- American Journal of Botany 69:1169–78. ulation genetic consequences of habitat frag- ———. 1983. Costa Rican natural history. Chicago: mentation for plants. Trends in Ecology and University of Chicago Press. 816 pp. Evolution 11:413–18. Kendrick, J., and R. B. Knox. 1989. Pollen-pistil interactions in Leguminosae (Mimosoideae).

BREEDING STRUCTURE OF TREE SPECIES 37 frankie chap 04 8/20/03 7:50 PM Page 38

chapter 4

Impact of Global Changes on the Reproductive Biology of Trees in Tropical Dry Forests

Kamaljit S. Bawa

he most spectacular feature of tropical flowers mature and after fruits initiate develop- Tdry forests of Guanacaste is perhaps the mass ment, many insects protect reproductive parts of flowering of many tree species during the dry the plants (Bentley 1977). Following seed and fruit season. As the dry season begins toward the end development, a wide variety of animals disperse of December, a number of species start to bloom, seeds (Howe and Smallwood 1982). Immature displaying flowers of various shapes, sizes, and and mature seeds and fruits are subject to pre- colors in the leafless canopy until the end of dation by a diverse array of invertebrates and April (Janzen 1967; Frankie et al. 1974). The sea- vertebrates, and such interactions in turn have sonal progression of flowering of various species profound effects on the structure and diversity is also mirrored in the daily rhythms, as flowers of tropical forests (Janzen 1970; Connell 1971). of various species start opening at dawn, prog- The complex plant and animal interactions ress through the morning, and then are replaced are being affected by global environmental by nocturnally flowering species at dusk. Since change, resulting principally from deforestation almost all tree species are animal-pollinated, the and forest fragmentation, climate change, and activity of various pollinators—insects, birds, and invasive species. In particular, mutualistic inter- mammals—is tied to the phenology of flowers actions between plants and animals, so vital to of their host plants, providing a kaleidoscope of the reproduction and survival of both groups of plant-pollinator interactions resulting from mil- organisms, are likely to be disrupted with severe lions of years of coevolution (Frankie et al. 1983; consequences on the composition of tropical chapter 2). Flower-pollinator interactions are only biota. Here, I focus on the effects of deforesta- one of many interactions involved in the re- tion and forest fragmentation on reproductive productive processes of tropical plants. Before processes of forest trees in tropical dry forests of

38 frankie chap 04 8/20/03 7:50 PM Page 39

Costa Rica. I address the following questions: abundance of native species (Mooney and Hobbs (1) What are the effects of habitat loss and habitat 2000). Finally, global climate change is likely to fragmentation on plant-pollinator interactions? alter forest structure and composition with un- (2) What are the consequences of changes in known consequences for the biodiversity and plant-pollinator interactions caused by anthro- function of dry forests (Bawa and Dayanandan pogenic pressures on the reproductive output 1998). and mating systems of plants? (3) What are the cumulative effects of changes in reproductive REPRODUCTIVE PROCESSES processes on fitness of plant populations and eventually on species composition of tropical PHENOLOGY dry forests? I start with a consideration of vari- Flowering phenology of trees in tropical dry ous threats to tropical forests, then briefly review forests is strongly seasonal and, as expected, the reproductive processes, and finally discuss correlates with the phenology of their insect pol- how these processes might be affected by various linators. A vast majority of dry-forest tree species forces that threaten tropical forests, primarily are pollinated by bees, and the seasonal flowering habitat changes. of many bee-pollinated tree species in the dry months is now well documented in dry forests in different parts of the world (Janzen 1967; THREATS Bullock and Solís Magallanes 1980; Frankie Tropical dry forests may be among the most et al. 1983; Sivaraj and Krishnamurthy 1989). threatened of all terrestrial ecosystems. These Flowering of moth-pollinated species is also sea- ecosystems are more suited for agriculture and sonal. The number of moth-pollinated species human settlements than wet evergreen forests increases steadily from April, which marks the because the pronounced dry season can restrain beginning of the wet season; it peaks in the wet the growth of insect pests and weeds (Janzen season and then declines (Haber and Frankie 1973). As a result, dry forests throughout the 1989; chapter 2). world have disappeared at a rapid rate. In the Strongly seasonal flowering and associated Western Hemisphere only 0.09 percent of phenologies of pollinators are products of many the tropical dry forests that existed when the years of evolution and are influenced by a vari- Spaniards colonized the continent are now un- ety of biotic and abiotic factors. Abiotic mecha- der some form of protection, and only 2 percent nisms mediating flowering are expected to be of the area is sufficiently intact to be of interest diverse (Janzen 1967; Frankie et al. 1974; Bor- to conservationists (Janzen 1988). chert 1983; Bullock 1995). Nevertheless, water Contemporary threats to tropical dry forests stress in the dry season seems to play an impor- have multiple origins. Deforestation and forest tant role in triggering flowering (Borchert 1983, fragmentation, caused by a host of social and 1998). Thus climate change and habitat alter- economic factors (Brown and Pearce 1994; ations that modify rainfall patterns and soil Bawa and Dayanandan 1997 and the references moisture, as discussed later in this chapter, are therein) are the main factors. Apart from land- expected to initiate a cascade of effects on flow- use changes in the form of deforestation and ering, pollinators, and mating patterns, thereby habitat fragmentation, forests are being de- affecting the fitness of both plant and pollinator graded by extraction of timber and nontimber populations. forest products (Chambers et al. 1989; Vasquez and Gentry 1989; Nepstad et al. 1992; Clay MATING SYSTEMS 1997). Anthropogenic effects also include the in- Most dry-forest trees are obligately outcrossed. troduction and spread of exotic invasive species, Bawa (1974) showed that in Guanacaste a vast which can further influence the distribution and majority of tree species with hermaphroditic

IMPACT OF GLOBAL CHANGES 39 frankie chap 04 8/20/03 7:50 PM Page 40

flowers (78%) are self-incompatible. Dioecious demonstrated that certain bee species can move species characterized by separate male and fe- within hours between trees as far away as 0.8 km. male plants account for 22 percent of all tree Similarly, bats, the principal pollinators of many species (Bawa and Opler 1975). Genetic analyses tree species in the dry forest, are known to for- of mating systems have also revealed high levels age over distances of several kilometers in a of outcrossing (Boshier et al. 1995a; James et single night (Heithaus et al. 1975). Africanized al. 1998; chapters 3 and 16). Foraging ranges of honeybees have been shown to transport pollen pollinators (see later in this chapter) and esti- more than 20 km among trees in Amazonia mation of gene flow suggest that extensive gene (Dick 2001), though generally their foraging exchange occurs over large distances among ranges are within a 2 km radius. Studies utiliz- canopy trees (Boshier et al. 1995a; chapters 3 and ing genetic markers show that long-distance for- 16). Thus habitat alterations can have a marked aging does result in long-distance gene flow by effect on outcrossing rates and genetic diversity pollen (Boshier et al. 1995b; Chase et al. 1996; in trees. Nason et al. 1998). Genetic analyses of populations also reveal frequent migrations among pop- POLLINATORS ulations via pollen and seeds (chapter 3). Dry-forest tree species display a tremendous Apart from extensive foraging ranges of many diversity of pollination mechanisms. Pollinators tropical pollinators, several pollen vectors mi- range from small wasps, less than 1 mm in grate among habitats that may be spread over length, to large bats with a wing span of more vast areas. For example, in Guanacaste, moths than 1 m in the Old World tropics. Many tree that pollinate many tree species in the rainy sea- species in the dry forests of Guanacaste are pol- son migrate to adjacent wet evergreen forests linated by medium-sized to large bees (Frankie in the dry season (Janzen 1987; chapter 8). There et al. 1983) and many others by moths (Haber is extensive migration of moths and butterflies and Frankie 1989) and bats (Heithaus et al. 1975) across the continental divide at the end of the (see also chapter 2). In most species, pollination dry season. Foraging across diverse habitats can mechanisms are specialized in the sense that also occur on a daily basis. Many moths captured plants are pollinated by a restricted set of vectors, in Monteverde, in montane forests, have been belonging to the same genus or same family found to carry pollen from trees in the dry forests (but see Johnson and Steiner 2000). Our in- to surrounding lowlands (chapter 8). ability to distinguish actual pollinators from a wide array of floral visitors to a particular plant EFFECTS OF DEFORESTATION species has been a major impediment to deter- AND FOREST FRAGMENTATION mining the extent of specialization. The almost total dependence on animal pollen vectors makes In considering the effects of various forces that tree species vulnerable to changes affecting the threaten tropical dry forests, it is important to habitats and populations not only of the trees emphasize a few points at the outset. First, there but also of their pollinators, which require a is a great deal of uncertainty about effects such wide array of natural habitats for persistence of as global climate change. Models of climate their populations. change make predictions of changes in green- house gasses, temperature, rainfall, and other FORAGING RANGES OF POLLINATORS parameters at spatial scales much larger than There are extensive data on foraging ranges of scales typically used for ecological studies. Sec- pollinators of dry-forest species. Janzen (1971) ond, for threats such as invasive species, there was among the first to suggest that large bees, the are relatively few data on the effects of exotics on major pollinators of most tree species, forage over the reproductive biology of other species. Third, long distances. Frankie et al. (1976) empirically effects of many threats are similar. For example,

40 BAWA frankie chap 04 8/20/03 7:50 PM Page 41

deforestation, fragmentation, degradation, and the trends are not consistent. Wright (1999) ob- invasive species all are expected to reduce the served high levels of fruit production in El Niño amount of total habitat for certain species, re- years—1969, 1992, 1997—in a wet forest in duce population sizes of many species, increase Panama, but the fruit production declined in the prospect of drift or inbreeding (or both), and 1970, 1993, and 1998, the years following the El alter species composition. Fourth, several fac- Niño years. Wright attributes the high produc- tors are likely to interact synergistically with one tion of fruit to higher amounts of light available another, further compounding deleterious effects during the prolonged drought in the El Niño induced by one or more factors. The discussion years. The drought itself is assumed to trigger that follows pertains largely to effects of defor- flowering. Lower levels of fruit production in estation and forest fragmentation. 1970, 1993, and 1998, according to Wright, may be due to exhaustion of reserves during the pre- REPRODUCTIVE OUTPUT ceding years of high fruit production. Effects of Reproductive output may be reduced as a result fluctuations in fruit size on population dynam- of several factors associated with the loss and ics of plants have not been investigated in these degradation of dry tropical forests. Reduction forests. could occur owing to effects on flowering, abun- Pollinator Populations. Shifts in timing and du- dance of pollinators, and mating systems. De- ration of flowering could have marked effects on creased output may set in motion changes in pollinator populations. In 1998, following the El recruitment patterns, eventually influencing Niño year of 1997, the intensity of flowering was species composition. reduced in trees in Guanacaste. Reduction in Flowering. Global climate change may influence food resources influenced the abundance of bees flowering phenology of tropical plant species almost throughout the dry season (G. Frankie (Bawa and Dayanandan 1998; Borchert 1998; pers. comm.). Corlett and LaFrankie 1998). As mentioned Deforestation and forest fragmentation are earlier, most tree species in dry forests flower also likely to result in the decline of pollinator during the dry season. One predicted effect of populations, simply owing to loss of food re- climate change in Central America is the exten- sources and nesting habitats. The decline in sion of the dry season (Hulme and Viner 1998). pollinator populations may, however, be dis- Data also indicate that the amount of rainfall in proportionate to the loss of forest habitat for three Guanacaste has steadily decreased over the past reasons. First, many pollinators require habitat 70 years (Borchert 1998: fig. 4). Such a reduc- heterogeneity over hundreds of square kilome- tion may indirectly lead to an increase in the ters for persistence. Thus, for example, if polli- intensity of the dry season, through a decrease nators utilize a range of forest types, population in soil moisture, which is an important deter- numbers may decrease with the destruction of minant of flowering time and duration (Borchert one or more forest types, even though a given 1983, 1998). Changes in duration, intensity, and forest type may not suffer any attrition. Second, the onset of the dry season are likely to affect the many factors may act synergistically. Fragmen- phenology of species flowering in the dry months tation, for instance, may increase the probability as well as those flowering in the interphase of of fire (Malcolm 1998), which in turn may re- dry and wet seasons. Alterations may be mani- duce floral resources without affecting the total fested as changes in the timing, intensity, and forest area. Fires can also affect nesting sites for duration of flowering. bees (Frankie et al. 1997). Third, factors respon- Long-term phenological data are required to sible for deforestation and forest degradation assess effects of climate change. Such data for may have a ripple effect on processes involved in other tropical forests reveal significant alteration plant-pollinator interactions. Dryness due to frag- in the reproductive behavior of plants; however, mentation, increased incidence of fire, or climate

IMPACT OF GLOBAL CHANGES 41 frankie chap 04 8/20/03 7:50 PM Page 42

change may adversely affect nesting sites, lead- cal area. As mentioned earlier, nectarivorous bats ing to a reduction in pollinator populations, and hawkmoths in Guanacaste may forage over which in turn may eventually lead to repro- several kilometers in a single night and require ductive failure of host plants and, in the case of a diverse array of habitats to sustain themselves. annual plants, fewer floral resources for the Little is known, however, about changes in the next generation of pollinators. Negative feedback abundance of these pollen vectors. loops, once established, can severely reduce num- Decline in the abundance of pollinators bers of pollinators. should be reflected in changes in seed and fruit Decline in the abundance of pollinators has set of plant species. Although there are no tem- been suspected in ecosystems throughout the poral data on alterations in reproductive output, world (Nabhan 1996; Allen-Wardell et al. 1998). many authors have documented reductions in For dry forests of Guanacaste, Frankie and his seed and fruit set in fragmented habitats (Kearns associates have studied changes in bee popula- et al. 1998; Paton 2000). In the dry deciduous tions over many years (Frankie et al. 1997; Vin- forests of Guanacaste, Rocha and Aguilar (2001) son et al. 1997). Frankie et al. (1997) compared have compared seed and fruit sets of Entero- the number of bee species and their abundance lobium cyclocarpum, and trees in pastures have on the same trees in 1972 and 1996. Over the lower seed and fruit sets than those in contigu- 24-year period, the number of bee species cap- ous forests. tured declined from 70 to 28, and the average Apart from quantity, the quality of the seed number of individuals sampled per tree de- can also be affected. In a lowland tropical wet creased from 824 ± 210 to 92 ± 5, or about forest in the Sarapiqui region of Costa Rica, seed 90 percent (chapter 6). germination in Pithecellobium elegans in de- Deforestation, habitat fragmentation, in- graded forests and pastures was lower than in creased incidence of fire, and introduction of ex- contiguous forests (K. Bawa unpubl. data). Al- otic species (Frankie et al. 1997) have apparently though comparable data from dry forests of Gua- led to a marked reduction in populations of a wide nacaste are lacking, Rocha and Aguilar (2001) variety of native bees over the 24-year period. found that seedlings of Enterolobium cyclocarpum The major factor in deforestation has been con- in pastures have less vigor than those from trees version of forestland to pasture for raising cattle in contiguous forests. (Frankie et al. 1997). Negative effects of this Mating Systems. Effects on seed germination land-use change on native biota have been com- may be a manifestation of changes in mating pounded by the introduction of the exotic jaragua systems, or the alle effect. Decrease in pollinator grass, Hyparrhenia rufa, which has spread over abundance by itself may have only limited effect large areas. As forest fires burn trees and shrubs, on seed viability. Most tropical tree species are the grass invades resulting open areas, increas- self-incompatible, or dioecious. Self-incompati- ing fuel loads for future fires. Thus, deforesta- bility, however, is not absolute. There is a certain tion, the exotic grass, and fire interact to reduce proportion of individuals or seeds on the same floral resources. Moreover, fires burn deadwood individuals that result from selfing (Bawa 1974). and important habitat for cavity and ground- Although decline in pollinator populations may nesting bees (Frankie et al. 1997; chapter 6). increase the proportion of seeds that result from Except for bees, populations of pollinators selfing and these selfed seeds may have lower ger- have not been monitored in tropical forests. Pol- mination rates than outcrossed seeds, low viabil- linators with large foraging ranges are expected ity is likely to result from increased inbreeding. to be the most susceptible to decline in the qual- Inbreeding in forests affected by deforesta- ity and quantity of habits because such pollen tion, fragmentation, degradation, and invasive vectors depend for their survival on a range of species may increase in two ways. First, popula- heterogeneous habitats over a large geographi- tion size may be considerably reduced in frag-

42 BAWA frankie chap 04 8/20/03 7:50 PM Page 43

mented habitats, and the incidence of mating than pollen dispersal. Most seeds fall in the among neighbors or relatives may increase. Sec- vicinity of the parent tree (Howe and Smallwood ond, in degraded forests, the density of individ- 1982), and many tree species have no known uals may decrease. Reduction in density may mechanisms of dispersal (Van der Pijl 1982). lead to alterations in foraging behavior of polli- Further, a large proportion of tropical trees have nators. For example, Franceschinelli and Bawa wind-dispersed seeds with no mechanism of (2000) found that when density is low, hum- long-distance dispersal, except during storms mingbirds move less frequently among plants and hurricanes. Second, the seeds of many than when the density is high. As a result, the species are dispersed by nonflying mammals progeny arrays show a higher rate of inbreeding that may not readily move across inhospitable than plants in high-density patches. habitats separating fragmented forests. Even in Thus, change in phenology or pollinator contiguous forests, individuals over as large an abundances can have a cascade of effects leading area as 10 ha may all be derived from a single to fewer and more inbred seeds than those in maternal parent (Hamilton 1999). Thus, forest contiguous forests. The smaller quantities of fragmentation is likely to severely curtail gene seeds that are highly inbred may have a low via- flow via seed in many species. bility and lead to a decrease in regeneration. Deforestation and forest fragmentation may Poor regeneration may further decrease the also reduce genetic diversity of populations in density of adult plants, resulting in ever in- relatively large contiguous reserves. Although creasing losses in reproductive output and more population sizes of tropical trees are not known, inbreeding. Interestingly, although there is a it is known that gene flow via pollen is exten- clear relationship between changes in phenol- sive. Trees in fragments surrounding contigu- ogy and plant fitness, no study has documented ous forests may be a part of the same population the direct link between changes in reproductive and contribute to the maintenance of genetic di- parameters arising from human disturbance of versity within contiguous forests. With increas- one type and fitness as manifested by a decrease ing deforestation, fragments may disappear over in regeneration or the density of individuals. time, reducing overall population size. Extraction, too, can alter genetic diversity of GENETIC DIVERSITY populations. First, extraction can lower popula- Changes in pollinator abundance and mating tion density with a concomitant increase in in- systems of plants are also expected to influence breeding as described earlier. Second, extraction genetic diversity of tree populations. Fragmented can lead to dysgenic selection of harvested pop- populations, considerably reduced in size, may ulations if extracted individuals are more vigor- lose genetic variation as a result of inbreeding ous and the degree of vigor is positively corre- and genetic drift. The rate and extent of loss may lated with heterozygosity. be determined by gene flow via pollen and seed among populations. Gene flow via pollen in CHANGES IN SPECIES COMPOSITION tropical trees is extensive. In contiguous forests, pollen vectors can transport pollen among con- The deleterious effects of deforestation and for- specific trees over long distances. As a result, in- est fragmentation should lead to a decline in the terbreeding individuals in a population can cover fitness of populations and eventually to the local vast areas of tens of square kilometers, if not extinction of tree species. Is there any evidence more. that species have become locally extinct in trop- The impact of deforestation and other forces ical dry forests as a result of contemporary an- threatening tropical forests may be greater on thropogenic pressures? There are two lines of gene flow via seed than on movement of genes evidence that suggest such extinctions may be via pollen. First, seed dispersal is more limited common.

IMPACT OF GLOBAL CHANGES 43 frankie chap 04 8/20/03 7:50 PM Page 44

First, dioecious species may be more prone species have the potential to influence various to extinction as a result of fragmentation than aspects of ecosystem structure and function. hermaphroditic species for two reasons. In small The ongoing changes call for two types of fragments, sex ratios may become heavily biased responses. First, policy makers need to take in favor of one sex. Imbalances in sex ratio could appropriate action to stem the pace of global reduce effective population size and ultimately change resulting from climate change, habitat fitness. In addition, a greater proportion of dioe- loss, and introduction of invasive species. Vin- cious species have fleshy fruits and are animal- son et al. (chapter 6) further elaborate on this re- dispersed rather than hermaphroditic species sponse. Second, the scientific community needs (Bawa 1980). Specialized vertebrate frugivores to undertake appropriate research to reduce un- may be absent in small fragments. Indeed, Gille- certainty about effects of changes, accumulate spie (1999) has found that the proportion of adequate evidence for predicted changes, and dioecious species in a number of Central Amer- present clear options to policy makers for miti- ican forests is positively correlated with fragment gation measures (chapter 24). size. Only long-term phenological records can in- Second, Ganeshaiah et al. (1998) compared dicate the extent to which climate change has or the proportion of species whose seeds are dis- will affect phenological patterns of plants. Yet persed by animals and wind in a dry deciduous there are few, if any, places in the world where forest in two locations, one closer to human set- well-designed studies that quantify flowering tlements and the other farther away. They found and fruiting on a long-term basis are under way. that forests closer to human settlements have a Ironically, many authors associated with this greater proportion of wind-dispersed tree species volume were involved in long-term phenological than forests farther away from settlements. The studies in the 1960s, but the studies were dis- implication is that the absence of animal dis- continued because of lack of funding and the per- persal agents in areas inhabited by humans ception by funding agencies at the time about leads to domination of the flora having an un- the value of long-term phenological studies. usually high representation of wind-dispersed Now, 25 years later, many funding agencies species. and others believe that such studies are critical Both examples demonstrate that changes in for assessment of possible effects of climate forest composition may be subtle and, in gen- change. eral, may lead to depauperization of biological Long-term phenological studies must be ac- diversity. companied by monitoring of pollinators. Except for medium-sized to large solitary bees in the Neotropics (Frankie et al. 1997), there are no RECOMMENDATIONS data on long-term trends in the abundance of Climate change, fragmentation and degradation tropical pollinators. Several questions need to of habitats, and invasive species are likely to be addressed: Are populations of pollinators have profound consequences on biotic interac- declining? Is the reduction in number dis- tions involved in the reproduction of trees. I proportionate to the reduction in the habitat? have considered only consequences for plant- Do various classes of pollinators differ in their pollinator interactions and the concomitant ef- sensitivity to changes resulting from habitat loss, fects on mating patterns and genetic diversity of alterations in climate, and the introduction of trees. Undoubtedly there are consequences for alien species? many other biological processes and interactions. Other issues relate to possible reduction in The cumulative and synergistic effects of global pollinator abundance on the fitness of plants. changes in climate, habitats, and the composi- Here we have virtually no data. Roubik (2000) tion of biota due to the introduction of alien cites studies that show reduction in seed and

44 BAWA frankie chap 04 8/20/03 7:50 PM Page 45

fruit set in plants in nonforested habitats result- Bentley, B. 1977. The protection function of ants ing possibly from displacement of pollinators by visiting extrafloral nectarines of Bxca orellan Africanized honey bees. However, we also need L. (Biscaceae). Journal of Ecology 65:27–38. Borchert, R. 1983. Phenology and control of flow- to document trends in seed and fruit production ering in tropical trees. Biotropica 15:81–89. in plants in natural habitats to assess how natu- ———. 1998. Responses of tropical trees to rain- ral communities are being affected by an overall fall seasonality and its long-term changes. Cli- decline in pollinators. Furthermore, alterations mate Change 39:381–93. in mating patterns of plants and the attendant Boshier, D. H., M. R. Chase, and K. S. Bawa. 1995a. Population genetics of Cordia allio- effects on genetic diversity of plant populations dora (Boraginaceae), a neotropical tree. 2. Mat- due to decline in pollinators remain unexplored. ing system. American Journal of Botany 82: Clearly an integrated approach is required. 476–83. First, we need to frame questions regarding the Boshier, D. H., M. R. Chase, and K. S. Bawa. 1995b. relative and synergistic impact of habitat loss, Population genetics of Cordia alliodora (Boragi- climate change, and the introduction of alien naceae), a neotropical tree. 3. Gene flow, neighborhood, and population substructure. American species, both plants and animals, on pollinators. Journal of Botany 82:484–90. Second, we need to frame specific hypotheses to Brown, K., and D. W. Pearce, eds. 1994. The causes examine effects of abundance on plant-pollinator of tropical deforestation: The economic and statis- interactions and plant fitness and how changes tical analysis of factors giving rise to the loss of the in interactions and fitness may in turn influence tropical forests. London: University College Press. 338 pp. population and community structure. Long-term Bullock, S. H. 1995. Plant reproduction in neotrop- studies at specific sites on specific groups of pol- ical dry forests. In Seasonally dry tropical forests, linators and plants pollinated by these vectors ed. S. H. Bullock, H. A. Mooney, and E. Med- are required. The resolution of scientific issues ina, 277–303. Cambridge: Cambridge Univer- will be critical to the eventual success of mitiga- sity Press. tion measures that might be designed to counter Bullock, S. H., and J. A. Solís Magallanes. 1980. Phenology of canopy trees of tropical deciduous losses in populations of pollinators and to pre- forests in Mexico. Biotropica 15:292–94. serve the integrity of plant pollinator interactions. Chambers, R. C., N. C. Saxena, and T. Shah. 1989. To the hands of the poor: Water and trees. Boulder, Colo.: Westview Press. 273 pp. REFERENCES Chase, M. R., C. Moller, R. Kesseli, and K. Bawa. 1996. Distant gene flow in tropical trees. Nature Allen-Wardell, G., et al. 1998. The potential conse- 383:398–99. quences of pollinator decline on the conserva- Clay, J. 1997. The impact of palm harvest in the tion of biodiversity and stability of food crop Amazon estuary. In Harvesting wild species: Im- yields. Conservation Biology 12:8–17. plications for biodiversity conservation, ed. C. H. Bawa, K. S. 1974. Breeding systems of tree species Freese, 283–314. Baltimore: John Hopkins Uni- of a lowland tropical community. Evolution 28: versity Press. 85–92. Connell, J. H. 1971. On the role of natural enemies ———. 1980. The evolution of dioecy in flowering in preventing competitive exclusion in some plants. Annual Review of Ecology and Systematics marine mammals and in rain forest trees. In Dy- 11:15–40. namics of populations, ed. P. J. Boer and G. Grad- Bawa, K. S., and S. Dayanandan. 1997. Causes well, 298–312. Proceedings of the Advanced of tropical deforestation and institutional con- Study Institute on Dynamics of Numbers in straints to conservation. In Tropical rain forest: Populations, Oosterbeck, 1970. Wageningen, A wider perspective, ed. B. Goldmith, 175–98. Netherlands: Center for Agricultural Publish- London: Chapman and Hall. ing and Documentation. ———. 1998. Climate change and tropical forest Corlett, R. T., and J. V. LaFrankie Jr. 1998. Potential genetic resources. Climate Change 23:449–66. impacts of climate change in tropical Asian Bawa, K. S., and P. A. Opler. 1975. Dioecism in forests through an influence on phenology. Cli- tropical forest trees. Evolution 29:167–79. mate Change 39:439–53.

IMPACT OF GLOBAL CHANGES 45 frankie chap 04 8/20/03 7:50 PM Page 46

Curran, L. M., I. Caniago, G. D. Paoli, D. Astianti, Howe, H. F., and J. Smallwood. 1982. Ecology of M. Kusneti, M. Leighton, C. E. Nirarita, and seed dispersal. Annual Review of Ecology and Sys- H. Haeruman. 1999. Impact of El Niño and tematics 13:201–28. logging on canopy tree recruitment in Borneo. Hulme, M., and D. Viner. 1998. A climate change Science 286:2184–88. scenario for the tropics. Climate Change 39: Dick, C. 2001. Genetic rescue of remnant tropical 145–70. trees by an alien pollinator. Proceedings of the James, T., S. Vege, P. Aldrich, and J. L. Hamrick. Royal Society of London B. 268:2391–97. 1998. Mating systems of three tropical dry for- Franceschinelli, E. V., and K. S. Bawa. 2000. The est species. Biotropica 30:587–94. effect of ecological factors on the mating sys- Janzen, D. H. 1967. Synchronization of sexual re- tem of a South American shrub species (He- production of trees within the dry season in licters brevispira). Heredity 84:116–223. Central America. Evolution 21:620–37. Frankie, G. W., P. A. Opler, and H. G. Baker. 1974. ———. 1970. Herbivores and the number of tree Comparative phenological studies of trees in species in tropical forests. American Naturalist wet and dry forests in the lowlands of Costa 104:501–28. Rica. Journal of Ecology 62:881–919. ———. 1971. Euglossine bees as long-distance pol- Frankie, G. W., P. A. Opler, and K. S. Bawa. 1976. linators of tropical plants. Science 171:203–5. Foraging behavior of solitary bees: Implica- ———. 1973. Tropical agroecosystems. Science tions for outcrossing of a neotropical forest tree 182:1212–19. species. Journal of Ecology 64:1049–57. ———. 1987. How moths pass the dry season in Frankie, G. W., W. A. Haber, P. A. Opler, and K. S. a Costa Rican dry forest. Insect Science and Its Bawa. 1983. Characteristics and organization of Application 8:498–500. the large bee pollination system in the Costa ———. 1988. Tropical dry forests, the most en- Rican dry forest. In Handbook of experimental dangered major tropical ecosystem. In Biodiver- pollination biology, ed. C. E. Jones and R. J. Little, sity, ed. E. O. Wilson, 130–37. Washington, D.C.: 411–47. New York: Van Nostrand-Reinhold. National Academy Press. Frankie, G. W., S. B. Vinson, M. A. Rizzardi, T. L. Johnson, D., and J. E. Steiner. 2000. Generaliza- Griswold, S. O’Keefe, and R. R. Snelling. 1997. tion versus specialization in plant pollination Diversity and abundance of bees visiting a mass systems. Tree 15:140–43. flowering tree species in disturbed seasonal dry Kearns, C., D. Inouye, and N. Waser. 1998. Endan- forest, Costa Rica. Journal of the Kansas Ento- gered mutualisms: The conservation of plant- mological Society 70:281–96. pollinator interactions. Annual Review of Ecology Ganeshaiah, K. N., R. Uma Shaanker, K. S. Murali, and Systematics 29:83–112. Uma Shankar, and K. S. Bawa. 1998. Extraction Malcolm, J. R. 1998. A model of conductive heat of nontimber forest products in the forests of flow in forest edges and fragmented landscapes. Biligiri Rangan Hills, India. 5. Influence of dis- Climate Change 39:487–502. persal mode on species response to anthro- Mooney, H. A., and R. J. Hobbs, eds. 2000. Inva- pogenic pressures. Economic Botany 52:316–19. sive species in a changing world. Washington, D.C.: Gillespie, M. B. 1999. Plant life history character- Island Press. 457 pp. istics and rarity in tropical dry forest fragments Nabhan, G. P. 1996. Pollinator redbook. Vol. 1: Global of Central America. Journal of Tropical Ecology list of threatened vertebrate wildlife species serving 15:637–49. as pollinators for crops and wild plants. Tucson: Haber, W. A. 1984. Pollination by deceit in a mass- Arizona-Sonora Desert Museum and Forgotten flowering tropical tree Plumerica rubra L. (Apo- Pollinators Campaign Monographs. cynaceae). Biotropica 16:269–75. Nason, J. D., E. Allen Herre, and J. L. Hamrick. Haber, W. A., and G. W. Frankie. 1989. A tropical 1998. The breeding structure of a tropical key- hawkmoth community: Costa Rican dry forest stone plant resource. Nature 391:685–87. Sphingidae. Biotropica 21:155–72. Nepstad, D. C., F. Brown, L. Luz, A. Alechandra, Hamilton, M. B. 1999. Deforestation affects the and V. Viana. 1992. Biotic impoverishment of genetic structure of the surviving forest frag- Amazonian forests by rubber tapers, loggers, ments. Nature 401:129–30. and cattle ranchers. In Non timber products Heithaus, E. R., T. H. Fleming, and P. A. Opler. from tropical forests: Evaluation of a conservation 1975. Foraging patterns and resource utiliza- and development strategy, ed. D. C. Nepstad and tion in seven species of bats in a seasonal trop- S. Schwartzman, 1–14. New York: New York ical forest. Ecology 56:841–54. Botanical Garden.

46 BAWA frankie chap 04 8/20/03 7:50 PM Page 47

Nepstad, D. C., A. Verissimo, A. Alencar, C. Nobre, Roubik, D. W. 2000. Pollination system stability E. Lima, P. Lefebvre, P. Schlesinger, C. Potter, in tropical America. Conservation Biology 14: P. Moutinho, E. Mendoza, M. Cochrane, and 1235–36. V. Brooks. 1999. Large-scale impoverishment Sivaraj, N., and K. V. Krishnamurthy. 1989. Flow- of Amazonian forests by logging and fire. Nature ering phenology in the vegetation of Sher- 398:505–8. varoys, South India. Vegetation 79:85–88. Opler, P. A., G. W. Frankie, and H. G. Baker. 1980. Van der Pijl, L. 1982. Principles of dispersal in higher Comparative phenological studies of treelet and plants. Berlin: Springer-Verlag. shrub species in tropical wet and dry forests in Vasquez, R., and A. H. Gentry. 1989. Use and mis- the lowlands of Costa Rica. Journal of Ecology use of forest-harvested fruits in the Iquitos 68:167–88. area. Conservation Biology 3:350–61. Paton, D. C. 2000. Disruption of bird pollination Vinson, S. B., H. J. Williams, G. W. Frankie, and systems in southern Australia. Conservation Bi- G. Shrum. 1997. Floral lipid chemistry of Byr- ology 14:1232–34. sonima crassifolia (Malpigheaceae) and a use of Rocha, O. J., and G. Aguilar. 2001. Reproductive floral lipids by Centris bees (Hymenoptera: Ap- biology of the dry forest tree Enterolobium cyclo- idae). Biotropica 29:76–83. carpum (Guanacaste) in Costa Rica: A compar- Wright, S. J. 1999. El Niño events influence plant ison between trees left in pastures and trees in reproduction in a wet tropical forest. CTFS News continuous forests. American Journal of Botany (Center for Tropical Science, Smithsonian Insti- 88:1607–14. tution, Washington, D.C.) (Summer 1999): 10.

IMPACT OF GLOBAL CHANGES 47 frankie chap 05 8/20/03 7:51 PM Page 48

chapter 5

Tropical Dry-Forest Mammals of Palo Verde

ECOLOGY AND CONSERVATION IN A CHANGING LANDSCAPE

Kathryn E. Stoner and Robert M. Timm

esoamerica contains some of the son from December through May, an annual pre- Mworld’s most diverse forests. It has at least cipitation of approximately 1,500 mm, and an 20 major life zones, based on variations of tem- average annual temperature higher than 24°C perature and precipitation that can be broadly (Maldonado et al. 1995). These dry forests are summarized in five tropical forest types—dry largely deciduous today and encompass hetero- forest, wet forest, montane forest, coniferous geneous habitats varying in species composi- forest, and mangrove swamp (Holdridge et al. tion, abundance, rainfall, and soils. These char- 1971). When the Spaniards arrived in the New acteristics contribute to creating a harsh and World, there were perhaps 550,000 km2 of dry heterogeneous, yet seasonally resource-rich, forest on the Pacific side of lowland tropical environment for the native mammals. Mesoamerica. This dry forest occupied as much Mesoamerica has a diverse mammal fauna or more of the Mesoamerican lowlands as did that includes elements from both North and wet forests. Unfortunately, no habitat type in South America as well as endemic species. More Mesoamerica has been more influenced by hu- than 275 species in 28 families are recognized mans than the tropical dry forest; today less than from the region, at least 17.8 percent of which 1 percent remains intact, with less than 0.01 per- are endemic to Mesoamerica. The mammals of cent under protection. the tropical dry forest are among the most poorly In Costa Rica tropical dry forests occur known of any of the bioclimatic life zones. Mam- throughout the Pacific lowlands of Guanacaste mals that inhabit tropical dry-forest areas must Province and adjacent Puntarenas Province from be capable of dealing with high temperatures sea level to about 500 m. Costa Rica’s dry forest (to 40–41°C), very low precipitation in the dry is characterized by a five- to six-month dry sea- season, and large fluctuations in the availability

48 frankie chap 05 8/20/03 7:51 PM Page 49

of food resources over time. Most mammals of lowed by 11 species of rodents, 7 species of mar- the dry forest can be characterized as resident supials, 6 species in the weasel family, 5 species of generalists that shift their diets to utilize season- cats, 3 species in the raccoon family, 3 species ally available food resources, as resident special- of primates, 3 species of artiodactyls, 2 species of ists that forage on insects, seeds, or fruit and canids, 2 species of xenarthrans (edentates), 1 rab- nectar, or as migrants that occupy dry forests bit (Sylvilagus floridanus), and 1 tapir (Tapirus only seasonally and migrate to different habitats bairdii). Species reaching the southern boundary during periods of low food availability in search of their distribution in Costa Rica’s dry forest of available food sources. include opossum (Didelphis virginiana), gray sac- As in all tropical ecosystems, a wide variety winged bat (Balantiopteryx plicata), gray short- of mammals contribute to the maintenance of tailed bat (Carollia subrufa), long-tongued bat dry tropical forests through their role in seed (Glossophaga leachii), Salvin’s spiny pocket mouse dispersal and pollination (Heithaus et al. 1975; (Liomys salvini), slender harvest mouse (Reithro- Chapman 1989; Helversen 1993; see chapter 13). dontomys gracilis), harvest mouse (R. paradoxus), Bats visit, and presumably pollinate, at least and hooded skunk (Mephitis macoura). Raccoons 14 species of flowers in the tropical dry forest of (Procyon lotor), coyotes (Canis latrans), Mexican Palo Verde, and 29 species of fruits are con- porcupines (Coendou mexicanus), and Under- sumed by bats, which in turn disperse their seeds wood’s long-tongued bats (Hylonycteris under- (K. Stoner and R. Timm unpubl. data). In Guana- woodi) are found as far south as southwestern caste there are two peak periods of flowering Panama. No species reaches the northern limit activity: one during the long dry season and the of its distribution in Costa Rica’s dry forest. other during the middle of the rainy season Dry forests are believed to be less diverse (Frankie et al. 1974). Primates, many rodents, than wet forests because of the harsh seasonal and several generalist carnivores also are impor- environment. However, in Costa Rica the docu- tant seed dispersers in Costa Rica’s tropical dry mented mammal fauna of the Pacific lowlands forests. Thus the preservation of wildlife and its consists of 114 species, which is only slightly habitats are interrelated challenges, and conser- lower than that of lowland rain forest of La Selva vation efforts in Guanacaste’s tropical dry forests in northeastern Costa Rica, with 123 species need to consider both of these issues together. (Timm 1994; R. Timm unpubl. data). The dry forest has 66 species of bats and La Selva 67 species. The pattern of lower diversity in tropi- DIVERSITY, DISTRIBUTION, cal dry-forest habitats has been observed in AND ENDEMISM Mexico for other groups of vertebrates (Ceballos At least 207 species of mammals, including 110 1995) as well as for flora (Gentry 1995). species of bats, have been documented within Most of the resident mammal species of Costa Rica’s borders, and more will undoubtedly the dry forest are generalists that have a broad be discovered (Timm 1994; Rodríguez and Chin- diet allowing them to survive changes in food chilla 1996; Timm and LaVal 1998; Timm et al. abundance. Some of these generalists include 1999). The majority of mammals found in Costa mantled howler monkeys (Alouatta palliata), Rica’s tropical dry forest are distributed north- white-faced capuchins (Cebus capucinus), white- ward through Mesoamerica (sometimes as far tailed deer (Odocoileus virginianus), coyotes, white- as western Mexico), and many occur southward nosed coatis (Nasua narica), raccoons, opossums, into South America. Of the approximately 114 and some frugivorous bats. Although howler species of mammals originally present in Gua- monkeys are mainly folivorous and are selective nacaste’s tropical dry forest, perhaps 110 are still of the leaves they ingest, they can consume more found in this habitat. Bats are by far the most than 60 different species of plants within any diverse group, with more than 66 species, fol- one area (Glander 1978), allowing them sufficient

MAMMALS OF PALO VERDE 49 frankie chap 05 8/20/03 7:51 PM Page 50

flexibility to find food throughout the year. Dur- bat; these species are found at mid- and high ing the driest months, when more than 80 per- elevations (Timm 1994). All species that his- cent of the trees have lost their leaves, howlers torically were found in the dry forest of Costa Rica may still find edible leaves in some of the ever- also were found in the dry forest of adjacent Nica- green species such as Cecropia peltata and the ragua or farther north; thus Costa Rica has no wild cashew Anacardium excelsum. White-faced truly endemic dry-forest mammals. However, one capuchins also fare well in the dry forest even species of harvest mouse (R. paradoxus) is a dry- when fruit is scarce at the beginning of the rainy forest endemic restricted to the Pacific lowlands season because they then shift their diet to con- of Costa Rica and Nicaragua. In contrast to Costa sume mostly insects (Chapman 1988), which Rica, tropical dry forests in western Mexico con- are especially abundant at this time (Janzen and tain as many as 26 endemic mammal species Wilson 1983). Capuchins also consume other (Ceballos 1995). sources of protein, such as birds’ eggs, young birds, and baby coatis. Although white-tailed deer EXTIRPATED SPECIES populations in dry forests experience periodic crashes during particularly harsh years, deer are It is likely that more than 114 mammal species browsers and consume many species of both historically were present in the dry forest of Gua- herbaceous and woody plants and are thus well nacaste. Although most of these can still be found adapted to cope with the dry tropical forest envi- in some areas of this dry forest, several have ronment (Vaughan and Rodríguez 1991). Coy- been extirpated from the Palo Verde region and otes consume significant amounts of insects, throughout much of Costa Rica’s Pacific low- fruits, and grasses in tropical dry forests, and lands during the past several decades. These in- their diet varies seasonally in Costa Rica (Vaughan clude the water opossum (Chironectes minimus), and Rodríguez 1986), as it does throughout giant anteater (Myrmecophaga tridactyla), Hoff- their range. Coatis, raccoons, and opossums are mann’s two-toed sloth (Choloepus hoffmanni), the generalists and consume a wide variety of inver- brown-throated three-toed sloth (Bradypus varie- tebrates (especially insects), fruits, seeds, and gatus), grison (Galictis vittata), southern river ot- smaller vertebrates such as frogs and snakes. ter (Lutra longicaudis), white-lipped peccary (Tay- Some frugivorous bats, such as Seba’s short- assu pecari), and Baird’s tapir (Tapirus bairdii). tailed fruit bat (Carollia perspicillata), consume Giant anteaters were once found throughout more insects during periods in which fruit is Costa Rica from the Pacific and Caribbean coasts not available (Fleming 1988; but see the section to nearly the highest elevations (Timm et al. “Migratory Species” later in this chapter). 1989). Populations of giant anteaters have been Several dry-forest specialists with specific severely reduced throughout their range in the diets are present throughout the year. These past several decades as a result of overhunting include various species of insectivorous bats and habitat destruction. Giant anteaters must be (funnel-eared bats [Natalus stramineus], leaf- considered extremely rare in Costa Rica and in chinned bats [Pteronotus spp.]), the smaller ro- danger of extinction. dents, and several carnivores. Seasonal migrants Water opossums and southern river otters, that are more abundant during certain periods and perhaps grisons, require fresh running of the year when specific food sources are pres- streams. River otters are still present in Lomas ent include spider monkeys (Ateles geoffroyi) and Barbudal, Hacienda Monteverde, on the Coro- some nectarivorous (Glossophaga spp.) and fru- bicí River, and at Parque Nacional Guanacaste. givorous bats (Stoner 2001). We have historical reports of otters in the Cañas Costa Rica’s mammal fauna includes sev- River near its confluence with the Tempisque eral species that are endemic to the country, in- River, and we suspect that they were found cluding at least six rodents, two shrews, and one throughout the Tempisque River Basin. Al-

50 STONER AND TIMM frankie chap 05 8/20/03 7:51 PM Page 51

though we have no specific records of water opos- most agricultural areas the crops often run up sums and grisons in Palo Verde, we strongly to the rivers’ edge. suspect that they were present historically, as With such a diverse group of species having they were widely distributed in both the Pacific disappeared from Palo Verde, it is informative to and Caribbean lowlands from sea level to mid- ask what traits these taxa share that may account elevations. It is likely that stream contamination for their extirpation or susceptibility. The species and erosion caused by both the sugarcane and that are extirpated are all either highly prized rice industries in this region contributed sig- game species that have been eliminated by over- nificantly to the disappearance of these aquatic hunting (white-lipped peccaries and tapirs) or and semiaquatic animals from much of the specialists that either feed on specific foods or Pacific lowlands. have very specific habitat requirements. In gen- There have been no observations of white- eral, species that first disappear from a region lipped peccaries or tapirs reported in the Palo following alterations are those that have a large Verde region for several decades. David Stewart body size, low initial population density, large informed us that during the 1950s tapirs were territory size, or narrow habitat tolerance. found only as far west as the vicinity of the Pan American Highway, and neither species was at ENDANGERED SPECIES Palo Verde. Both tapirs and white-lipped peccaries are still found in the tropical dry forest of Costa Rica recognizes 13 species of mammals as Parque Nacional Guanacaste, but they are best endangered species in the country—the mantled considered extirpated from the vast majority of howler monkey, spider monkey, squirrel mon- the Guanacaste lowlands. key (Saimiri oerstedii), giant anteater, jaguar (Pan- Although we do not have the historical doc- thera onca), puma (Puma concolor), ocelot (Leop- umentation of how widely distributed two-toed ardus pardalis), margay (Leopardus wiedii), oncilla sloths and three-toed sloths were in the tropical (Leopardus tigrinus), jaguarundi, West Indian dry forest, we suspect that they were much more manatee (Trichechus manatus), white-lipped pec- widely distributed when there were larger ex- cary, and tapir (MINAE 1999). An additional 14 panses of mature forest. Mature stands of tropi- species of mammals are recognized as being cal dry forest would provide a variety of tree found in reduced populations. Of these, 7 are species that would be seasonally available for the found within the dry forest of Costa Rica—the folivorous sloths as well as cool, shady habitat. great false vampire bat (Vampyrum spectrum), Because sloths have a low metabolic rate, main- white-faced capuchin, Hoffmann’s two-toed tain a low body temperature, and are imperfect sloth, Deppe’s squirrel (Sciurus deppei), Under- homeotherms (McNab 1985), they may be phys- wood’s pocket gopher (Orthogeomys underwoodi), iologically less able to survive in the harsh, hot- grison, and southern river otter. Additionally, ter, drier habitats created by opening up mature several species that are especially sensitive to stands of tropical dry forest. It is likely that habitat destruction and forest fragmentation sloths in the dry forest originally occupied ri- include spider monkeys, felids, and predaceous parian habitats that provided both evergreen trees bats of the family Phyllostomidae, subfamily for them to forage on throughout the year and Phyllostominae. Spider monkeys require a large shade for thermoregulation. Forested areas sur- home range because of their dietary preference rounding riparian habitats throughout the Gua- for ripe fruits (Chapman 1988, 1989). Habitat nacaste lowlands have been largely destroyed fragmentation not only necessitates that they owing to the development of agriculture. Costa must travel farther to find the required ripe fruit Rican law prohibits the destruction of habitat resources but also sometimes eliminates ar- within 15 m of rivers (Law No. 7575, Article boreal passages that allow access to resources. No. 33); however, this is not enforced, and in Felids also require a large home range based on

MAMMALS OF PALO VERDE 51 frankie chap 05 8/20/03 7:51 PM Page 52

their carnivorous diet. The need to pass through The frugivorous Seba’s short-tailed fruit bat and open areas in a fragmented landscape makes the Jamaican fruit-eating bat (Artibeus jamaicen- them more vulnerable to illegal hunting. Previ- sis) account for more than 50 percent of the bats ous studies on the predaceous phyllostomine captured at this site. An additional 12 species bats have shown that they are sensitive to habi- account for 40 percent of captures, whereas the tat destruction, and some species are rarely found majority of species (33) account for less than in disturbed environments (Timm 1994; Schulze 10 percent of total captures, and most of these et al. 2000). Our data on bat populations in the were captured on fewer than 5 nights of the 56 dry forest suggest that a number of species are nights sampled. Excluding the aerial insectivores, rare in this habitat and should be considered as of the 10 species captured only once, there were endangered (table 5.1). 4 nectarivores, 3 carnivores, 2 gleaning insectivores, and 1 frugivore. Although mist-net data do not accurately sample the abundance of aerial MIGRATORY SPECIES insectivores, they provide useful data for com- Seasonal migrations along elevational gradients paring relative abundance of most other bat have been well documented for several species species within a community. Our mist-netting of tropical birds and butterflies; however, it only efforts at Palo Verde suggest that several of the recently has been suggested that migrations nectarivorous and predaceous bats are found at may occur in Neotropical bats (Timm and LaVal very low densities, likely a result of disturbance 2000; Stoner 2001, 2002). The abundance of within Palo Verde (i.e., fire and cattle), lack of several species of bats at Palo Verde changes mature trees, habitat fragmentation surround- significantly over seasons, suggesting that they ing Palo Verde, and lack of forested corridors shift habitats seasonally or migrate into and out within the region. of the region (Stoner 2001). Patterns of capture for some other species Bats were mist-netted at Palo Verde approxi- of bats also suggest that abundances vary sea- mately once every three weeks from January sonally. The pygmy fruit-eating bat (Artibeus 1994 through July 1997, for a total of 56 nights. phaeotis) and Thomas’ fruit-eating bat (Artibeus A total of 1,245 individuals representing 47 watsoni) were never captured in March, April, species were captured at one site, the Guayacán May, July, or August. The wrinkled-faced bat (Cen- waterhole (10°21′ N, 85°20′ W) (table 5.1). All turio senex) was captured in December (1995), netting at the waterhole was with ground-level January (1996), and February (1996, 1997) but mist nets, which are excellent for sampling most rarely in August (one individual was captured in frugivores, nectarivores, and some insectivores August 1995). The insectivorous orange-throated but underestimate higher-flying insectivores. Be- bat (Micronycteris brachyotis) was caught from cause Palo Verde has a marked dry season, bats December through April (and two individuals in are concentrated around waterholes, and sam- July), and the tiny big-eared bat (Micronycteris pling there produces a higher number of taxa minuta) was caught from January through April and greater numbers of individuals than would (and one was captured in June). Although sam- be expected to be captured at any one site in the ple sizes for these species were not sufficient to surrounding forest. Including our data from compare statistically, their abundance over sea- netting at Palo Verde, which began in the 1970s sons and the fact that the same pattern was ob- and continues to the present, we have captured served over several years suggest that most of or observed an additional 12 species in the area, these species are likely moving in and out of the for a total of 59 species. area on a seasonal basis. These patterns of abun- Significant differences occur in abundances dance and absence are not one of source/sink (estimated as bats/m2FD net × hour) of the most because individually marked animals return af- common frugivorous bats over different seasons. ter months of absence (Stoner 2001).

52 STONER AND TIMM frankie chap 05 8/20/03 7:51 PM Page 53

TABLE 5.1 Abundance of 47 Species of Bats Captured at the Guayacán Water Hole in Palo Verde from January 1994 through July 1997

total percent number percent number of total of nights of nights species caught bats caught caught caught

Carollia perspicillata 472 37.9 47 84 Artibeus jamaicensis 174 13.9 40 71 Sturnira lilium 111 8.9 33 59 Artibeus lituratus 65 5.2 27 48 Rhogeessa tumida 49 3.9 17 30 Pteronotus davyi 47 3.8 17 30 Glossophaga soricina 43 3.4 17 30 Artibeus phaeotis 38 3.0 21 38 Centurio senex 38 3.0 11 20 Micronycteris brachyotis 37 3.0 15 27 Artibeus watsoni 25 2.0 14 25 Pteronotus parnellii 17 1.4 12 21 Natalus stramineus 16 1.3 9 16 Desmodus rotundus 13 1.0 8 14 Pteronotus gymnonotus 11 0.9 9 16 Trachops cirrhosus 10 0.8 8 14 Micronycteris minuta 9 0.7 8 14 Uroderma bilobatum 7 0.6 2 4 Carollia brevicauda 6 0.5 4 7 Noctilio leporinus 6 0.5 4 7 Platyrrhinus helleri 5 0.4 2 4 Micronycteris microtis 5 0.4 5 9 Carollia castanea 4 0.3 4 7 Lasiurus blossevillii 4 0.3 NA NA Vampyressa nymphaea 4 0.3 4 7 Micronycteris nicefori 3 0.2 3 5 Micronycteris schmidtorum 3 0.2 3 5 Chrotopterus auritus 2 0.2 2 4 Diphylla ecaudata 2 0.2 2 4 Tonatia brasiliensis 2 0.2 2 4 Glossophaga commissarisi 1 0.08 1 2 frankie chap 05 8/20/03 7:51 PM Page 54

TABLE 5.1 (continued)

total percent number percent number of total of nights of nights species caught bats caught caught caught

Glossophaga leachii 1 0.08 1 2 Pteronotus personatus 1 0.08 1 2 Noctilio albiventris 1 0.08 1 2 Macrophyllum macrophyllum 1 0.08 1 2 Phyllostomus discolor 1 0.08 1 2 Vampyrum spectrum 1 0.08 1 2 Hylonycteris underwoodi 1 0.08 1 2 Carollia subrufa 1 0.08 1 2 Lichonycteris obscura 1 0.08 1 2 Micronycteris hirsuta 1 0.08 1 2 Cyttarops alecto 1 0.08 NA NA Rhynchonycteris naso 1 0.08 NA NA Saccopteryx bilineata 1 0.08 NA NA Myotis elegans 1 0.08 NA NA Myotis riparius 1 0.08 NA NA Molossus molossus 1 0.08 NA NA Balantiopteryx plicataa Saccopteryx lepturaa Diclidurus albusa Chiroderma villosumb Peropteryx kapplerib Peropteryx macrotisb Myotis albescensb Myotis nigricansb Eumops auripendulusc Molossus ater c Molossus pretiosusc Molossus sinaloaec Micronycteris sylvestrisd Mimon cozumelaed Mimon crenulatumd Tonatia bidensd frankie chap 05 8/20/03 7:51 PM Page 55

TABLE 5.1 (continued)

total percent number percent number of total of nights of nights species caught bats caught caught caught

Tonatia silvicolad Uroderma magnirostrumd Vampyrodes caracciolid Eumops underwoodid Total 1,245

Note: All data presented in this table are based on 56 nights of netting. Species listed as NA (not applicable) refer to aerial insectivores whose abundance is poorly estimated with mist-netting techniques and thus would be underestimated by our sampling method. Twelve additional species that are listed were observed or captured within the area, and eight additional species listed are expected to occur at Palo Verde (see individual notes). aBats observed roosting within the area. bBats captured within the dry forest in the Palo Verde region. cSpecies detected by their echolocation signals (E. Kalko pers. comm.). dSpecies that are expected to be found in Palo Verde.

At Palo Verde, Peter’s tent-making bat (Uro- migratory. Spider monkeys are present in Palo derma bilobatum) also exhibits a seasonal pat- Verde only during the rainy season and appear tern in abundance and reproduction (Timm to migrate out of the area during the dry sea- and Lewis 1991 and subsequent observations by son. Fruit availability is strongly seasonal here Timm through January 2003). During June and (Frankie et al. 1974), and the migration of spider July (mid–rainy season) adult males and females monkeys is probably related to the lack of abun- as well as juvenile bats are present in a breeding dant fruit resources during the dry season. Large colony. All adult females captured during the herds of white-lipped peccaries originally were mid–rainy season were either pregnant or lac- found in the dry forests of Guanacaste and also tating. During January and February (dry season), were migratory (Janzen 1986). We suspect that only a couple or in most cases no bats were ob- the peccary herds would have fed heavily on the served. The few bats that were captured during seasonally abundant acorns (Quercus oleoides) the dry season were males with testes only mod- and palm fruits such as Acrocomia vinifera, At- erately developed. talea butyracea (= Scheelea rostrata), Bactris major, Although we have not yet identified where and B. minor. the bats are moving when they are not in the area, we suspect two areas. The first possibility FACTORS AFFECTING MAMMALS is that during periods of reduced resource avail- OF THE TROPICAL DRY FOREST ability bats change habitats to riparian areas within the lowland forest of Guanacaste, pos- HUNTING sibly near the Bebedero River, Piedras River, or Hunting and deforestation were the first impor- Tempisque River. Another possibility is that bats tant influences on the recent distribution and in the lowland tropical dry forest migrate eleva- density of mammals of the tropical dry forest, as tionally to higher areas with more abundant re- they were in other areas within the Neotropics. sources during certain seasons. All large and some small mammals in tropical In addition to bats, some arboreal and ter- dry forests have been subjected to extreme hunt- restrial mammals of the tropical dry forest are ing pressures. The Chorotega, the dominant

MAMMALS OF PALO VERDE 55 frankie chap 05 8/20/03 7:51 PM Page 56

pre-Columbian human inhabitants of the Gua- fragmented-forested habitats, and extensive ar- nacaste lowlands, hunted a wide variety of eas in various stages of succession. Hence, large mammals (Quesada López-Calleja 1980). The areas of land are only marginally inhabitable by most common mammals hunted today as a most native species. Opening up forest in the source of protein in the lowlands of Guanacaste harsh, highly seasonal environment of this area include white-tailed deer, pacas (Agouti paca, has had additional consequences, including a known in Costa Rica as tepezcuintle), collared subsequent increase in dryness, higher temper- peccaries (Pecari tajacu), armadillos (Dasypus atures, reduced availability of appropriate forest novemcinctus), and opossums. Pacas are highly habitat for both food resources and living space prized but infrequently obtained because they for mammals, and erosion of topsoil, which lim- occur in such low numbers. Hunters with whom its nutrients. The few fragments of dry forest that we have spoken are opportunistic, taking any remain in Costa Rica have all been influenced by of the preferred game animals when they are the surrounding habitat alterations. available. Much of the recent conversion is due to agri- Poaching continues to be one of the most cultural development, especially sugarcane and serious problems that threaten Parque Nacional rice production. Sugarcane fields, not present in Palo Verde, and this is largely attributed to the the Guanacaste area until the 1960s, accounted lack of guards to protect this approximately for approximately 16 percent of the area in the 20,000 ha park (Vaughan et al. 1995). White- basin by 1992–93, and rice and other agricul- tailed deer and collared peccaries are the most tural crops accounted for an additional 13.9 per- common animals poached in the park. Most il- cent. The lower Tempisque Basin is one of the legal poaching occurs within protected areas areas that has been designated by the Ministerio such as Palo Verde because these areas provide de Agricultura y Ganaderia and the Servicio Na- the best wildlife habitat and consequently the cional de Aguas Subterraneas Riego y Avena- highest densities of game species. In 1997 in miento as a rice production area, and many the Area de Conservación Tempisque, of the pastures and small forest fragments are being hunters arrested for poaching, approximately converted to rice fields. Because of the extensive three were released for every one that was actu- rice fields between Palo Verde and Lomas Bar- ally convicted (MINAE 1998). Poaching likely budal, the movement of most terrestrial mam- will continue in national parks and other pro- mals between these two important reserves is no tected areas until more park personnel are avail- longer possible. able to monitor these areas and stronger laws are enacted and enforced to protect wildlife. CATTLE, JARAGUA, FIRE, CATTAILS, AND LAND MANAGEMENT DEFORESTATION AND CONVERSION It is our intent here to review briefly the history OF TROPICAL DRY FOREST of cattle use and the introduction of African The single factor that has most strongly in- grasses in the region and to assess their impact fluenced the current distribution and abun- on native mammals. In recent years, the use of dance of mammals in the dry forest is land cattle as a habitat management tool in Palo conversion—the loss of tropical dry forest. Con- Verde has been the source of considerable conversion of tropical dry forest in Costa Rica has troversy (McCoy 1994; McCoy and Rodríguez resulted from various activities, including cattle 1994; Stern et al. 2002; see chapter 21). Cattle ranching, the timber industry, agricultural de- ranching has been important in Guanacaste for velopment, and the tourism industry (see chap- more than three centuries, and all of what is now ter 21). Regardless of the reasons for deforesta- Parque Nacional Palo Verde was a working cat- tion, the result has been the creation of vast tle ranch until the creation of the wildlife refuge expanses of open pastures, agricultural fields, in 1977. Cattle were slowly removed from the

56 STONER AND TIMM frankie chap 05 8/20/03 7:51 PM Page 57

refuge during the period 1978–81 (when it was are not well adapted for large-scale production designated a national park); however, some cat- in open pastures (Sáenz-Maroto 1955; León S. et tle were reintroduced into Palo Verde in 1987 as al. 1982). The jaragua, which came to Central part of an active management plan to control America via Brazil, was introduced into the cattails (Typha dominguensis—locally called enea), Palo Verde area from Puntarenas in the 1920s

which had expanded dramatically and quickly in (Sáenz-Maroto 1955; Parsons 1972). As in all C4 the lagoon, effectively eliminating much of the grasses, the nutrient levels of growing stems open water needed by waterfowl. In 1991 more and leaves during the rainy season are high, and cattle were introduced into Palo Verde as part the young grasses provide good forage for cattle of a management plan to control fires within (McCammon-Feldman 1980); however, the the park. During the late 1990s, some 1,500 to mature plants during the dry season have little 6,000 head of cattle were present in the park, nutritional value (Daubenmire 1972) and are with some 600 head in the Palo Verde lagoon not consumed by cattle. This species is native to and the rest in forested areas and pastures. the plains of Africa, grows to a height of 2 m or One of the earliest ranches in what is now more, and is highly adapted to fires. Throughout Costa Rica and adjacent Nicaragua was estab- the Neotropics, regular burning was, and con- lished at Santa Rosa in the late 1500s (Janzen tinues to be, a management strategy to remove 1986). By 1800, large ranches were present mature, nondigestible woody stems during the throughout Guanacaste (Boucher et al. 1983). dry season, promoting growth during the suc- Cattle were raised in Guanacaste primarily for ceeding rainy season. This practice has been the the hides, which were exported to Europe for use source of numerous uncontrolled fires since its as leather; to a lesser extent lard and dried beef introduction into the area. The fuel provided by also were exported. The finqueros (ranch hands) this non-native grass allows the fires to burn hot- hunted native mammals, especially peccaries, ter and to be more destructive than they are by for food and killed jaguars, pumas, and coyotes the burning of native vegetation.

because they were a threat to livestock. C4 grasses such as jaragua grow best in warm, The rapid deforestation of Costa Rica’s dry dry conditions, and they mature much more

forest during the 1950s, 1960s, and 1970s, con- quickly than C3 grasses. Tall C4 grasses often verting mature forest into pastureland, was en- become so dense that other plant species are couraged by low-interest loans from the national unable to compete. Annual burning of dry above- banks, as well as by support from the Agency for ground vegetation releases nitrogen, enriching

International Development and the World Bank the soil for more dominant C4 grasses, enhanc- (Parsons 1983). During this rapid expansion of ing their growth and allowing them to outcom- the cattle industry, ranchers experimented with pete other plant species (Collins et al. 1998). Fires several breeds, progressing from the Spanish also warm the soil, which favors the growth of

criollo to various beef breeds, including Here- C4 grasses. In temperate prairies of the mid- fords, Angus, and Charolais. Today, zebus and western United States, annually burned water- Brahman are found throughout the lower eleva- sheds had the lowest plant species richness, as

tions of Costa Rica, as well as throughout the burning increases the dominance of C4 grasses Neotropics, because they are well adapted to the and reduces plant species diversity (Collins et al. hot climate and extremes of rainy and dry sea- 1998). sons and are highly resistant to bites from ticks, In tropical rain forests of the Amazon Basin, flies, and other arthropods. fires have been found to create a positive feed- Jaragua, or African star grass (Hyparrhenia back system whereby periodic burning causes rufa), and other African savanna grasses were an increase in fuel loading, fire intensity, and introduced into Costa Rica in the 1920s as cattle fire susceptibility (Cochrane et al. 1999). The forage because the native Costa Rican grasses first fires characteristically kill only the smaller

MAMMALS OF PALO VERDE 57 frankie chap 05 8/20/03 7:51 PM Page 58

trees, especially those with thin bark. Because of Other direct effects of cattle include compe- fuel buildup, second fires are faster moving and tition with native mammalian herbivores (Janzen more intense, often killing larger, thicker-barked and Wilson 1983; Robinson and Bolen 1984) and trees. The long-term effect of recurrent fires in selective grazing on many native plant species. Brazil is to create an open canopy savanna or Cattle directly compete with deer for forage, and scrub habitat. In the Mesoamerican dry forest, inadequate habitat may be available for native several tree species with smooth, photosynthetic wildlife because of intensive cattle grazing. A bark (especially species such as Bursera simaruba) number of previous studies in other habitats are often killed by the first fires. Although some demonstrate that competition with cattle may tree species that have bark of medium thickness, increase annual deer mortality by as much as such as Guazuma ulmifolia, Crescentia alata, Byr- 40 percent (Robinson and Bolen 1984). Conklin sonima crassifolia, and Curatella americana, may (1987), in a study of several species of plants that survive initial burns, continued exposure to sur- are potential browse for cattle at La Pacifica, face fires from adjacent pastures eventually pen- found that cattle readily consume a large num- etrates and eliminates the forest (B. Williamson ber of native dry-forest trees and shrubs, eating pers. comm.). leaves, stems, fruits, and seeds of a wide num- Forest fragments have persisted only where ber of species. Saplings of native species provide they have been protected by humans or by nat- excellent protein, nutrients, and crude fiber for ural firebreaks such as roads and limestone out- cattle, which are able to digest the leaf tannins in crops. Remnant trees, such as cenizero (Samanea native saplings. Unlike jaragua, many browse saman) and guanacaste (Enterolobium cyclocar- species retain their nutrient levels fairly con- pum), can be found in pastures, but usually only stantly throughout both the rainy and dry sea- as older, isolated individuals. Recruits into pas- sons. In contrast, grasses, especially jaragua, tures are limited to a few species, namely, Byr- show significant reduction in digestibility and sonima crassifolia, Crescentia alata, Curatella amer- nutrient levels during the dry season. Native dry- icana, and Guazuma ulmifolia. These four species forest trees and shrubs are both more palatable are widespread as tree islands throughout the and more nutritious than jaragua for grazing by pastures of Mesoamerica and provide some of domestic livestock, and free-ranging cattle shift the few fruits and roost sites available for wild- their foraging to native species during the dry life in this anthropogenic landscape (Hartshorn season when given the opportunity. Hartshorn 1983). (1983: 131) reported that he was “unable to find The development of large-scale cattle ranch- seedlings or small saplings of Brosimum alicas- ing in Guanacaste in the early 1900s, combined trum” in the inventory plots at Palo Verde after with the introduction of jaragua, not only reduced cattle had grazed in the plots. No native mam- the available forested habitats for mammals but mals of Costa Rica will feed on the stems or also stimulated a series of other changes that leaves of mature jaragua, and it is unlikely that dramatically affected the remaining habitat and even native grazers such as cotton rats (Sigmo- fauna. Previous studies on cattle document that don hispidus) and eastern cottontail rabbits feed livestock alter ecosystem processes by reducing on anything but the youngest shoots. the cover of herbaceous plants and litter, dis- Our survey efforts in jaragua for small mam- turbing and compacting soils, reducing water mals have identified only one species of native infiltration rates, and increasing soil erosion mammal that can occupy this grass, cotton rats. (Belsky and Blumenthal 1997). Furthermore, Daubenmire (1972: 37), in a year-long study on they suggest that forests subjected to grazing the ecological consequences of converting dry pressure are less resilient to natural disturbances forest to pasturelands near Cañas, noted the lack such as fire and diseases. of mammals (only two rabbits seen) and other

58 STONER AND TIMM frankie chap 05 8/20/03 7:51 PM Page 59

animals in Hyparrhenia, stating that “[al]though and the stems, and leaves are primarily dried rodents and insects were present, they were very cellulose. few in species and numbers and their use of the Cattle as a management tool for controlling vegetation [Hyparrhenia] was negligible.” Cattle cattails in the Palo Verde marsh have proved also destroy bee nests, especially those of the ineffective (figs. 5.1–5.3). Although cattle will large, ground-nesting anthophorid bees that are consume some young, actively growing cattail the major pollinators of much of the dry forest shoots, they do not consume the mature leaves (Frankie et al. 1997). We suspect that cattle also (D. A. Stewart pers. comm.), and they prefer will negatively affect the nests of small mam- other species when given the opportunity to feed mals, as many nests along the periphery of the on them. The explosion of cattails in the marsh marsh are at the surface (Oryzomys and Sig- began shortly after the transfer of land from a modon nest at the surface or in aboveground cattle ranch to a national park. At this time, ce- vegetation). The combination of even moderate ment gates or weirs that were used in the dry grazing by cattle and regular burning has an ex- season to maintain water in the lagoon after tremely negative impact on rodent populations. high tide were abandoned, thus contributing to The presence of single-species stands of the flourishing of cattails in the lagoon. Within jaragua and of artificially created, open savan- five years of the change, cattails had eliminated nalike habitat within the tropical dry forest of most of the open areas of the marsh, which were Guanacaste has reduced or eliminated the pop- critical for waterfowl. Cattle in the marsh pri- ulations of most native dry-forest mammals from marily feed on floating water hyacinths (Eich- those habitats. However, a few species have hornia crassipes and E. heterosperma), which are increased in numbers and distribution with the locally called lirio de agua or choreja, but they will creation of these grasslands. Cotton rats are eat a wide variety of species, especially Pistia much more widespread and abundant in Costa stratiotes (locally called lechuga). Cattle will eat Rica today than they were in the past. Cotton rats young cattails, but they prefer other species. are an open-grassland species, and they would When cattle are allowed to graze in upland ar- have been rare in Guanacaste before the removal eas, they have a significant and detrimental im- of the forests and the introduction of grasses. pact on the composition and structure of native Other mammals that probably are more com- forest trees and shrubs. Overgrazing by cattle, mon now than before the creation of the open the presence of vast areas of jaragua, and un- savannalike habitats include opossums, vampire controlled fires continue to be among the most bats, cottontail rabbits, armadillos, coyotes, and serious threats to the native flora and fauna in gray foxes (Urocyon cinereoargenteus). the dry forest of Costa Rica. In recent years, we Controlled cattle grazing potentially may be have observed that in areas where cattle were an effective management tool in reducing the grazing, most of the smaller saplings had been biomass of jaragua when cattle are grazed on either consumed or trampled and that there the growing grass during the rainy season and were very few rodents on the forest floor in heav- are then removed. However, the presence of ily grazed areas. The use of cattle within Parque cattle in tropical dry forest will change the struc- Nacional Palo Verde is having a significant neg- ture and composition of that habitat (Stern et al. ative impact on the regeneration of the forest 2002). Cattle cannot be sustained on jaragua and on the abundance of native mammals. throughout the year and must be allowed to forage on other species. Cattle lose weight on a diet PESTICIDE CONTAMINATION consisting solely of jaragua during the dry sea- Pesticide contamination is a problem worldwide, son (D. A. Stewart pers. comm.) because most and in recent years, with the increase in agricul- of the plants’ nutrients are stored in the roots tural development in Guanacaste, it has become

MAMMALS OF PALO VERDE 59 frankie chap 05 2003-09-11 12:12 Page 60

FIGURE 5.1. Laguna Palo Verde in February 1970. The marsh included diverse characteristic aquatic vegetation and open areas between patches of Parkinsonia aculeata trees, water hyacinth, water lilies, and bull- rush. Large areas of open, shallow water were present. No cattails were visible because they occurred only in small patches along the airstrip (shown in the lower part of photo) near a freshwater spring draining into the marsh. Photograph by Gordon Frankie.

FIGURE 5.2. Laguna Palo Verde in February 1970. A close-up view of the marsh. Photograph by Gordon Frankie. frankie chap 05 2003-09-11 12:12 Page 61

FIGURE 5.3. Laguna Palo Verde in February 2001. Since the mid-1980s, the diversity of both plants and waterfowl has declined dramatically, and the marsh has become choked with cattails (Typha dominguensis, Typhaceae) and two aquatic perennial grasses, Hymenachne amplexicaulis and Paspalidium geminatum (Araceae). Photograph by Robert Timm.

a serious threat to the fauna in this region. Or- for example, by lowering sperm count. Die-offs ganophosphates and carbamates, the most com- of both birds and mammals occur even when mon insecticides in use today throughout the pesticides are applied responsibly because many world, are known as cholinesterase-inhibiting animals consume the pesticide, either directly pesticides because they kill by interfering with or indirectly. Pesticide residues in the stomachs the enzyme vital for nerve transmission. Organo- of poisoned mammals and birds are known to phosphates and carbamates work well against a kill predators and scavengers. wide range of insect pests and are often less Cropdusters in Guanacaste (as well as else- expensive than many alternatives, which adds to where in Costa Rica) have customarily dumped their popularity. In addition to affecting insects, into rivers unused pesticides remaining from many are acutely (immediately) toxic to most aerial spraying, and this practice continues. A vertebrates and other invertebrates. Because number of toxic pesticides have been isolated they break down quickly in soil and water, they from the Tempisque River, including aldrin, often need to be applied to crops more than once chlordane, DDT, heptachlor, and lindane (Mata during the growing season. Organochlorine in- and Blanco 1994). These ﬁve insecticides are secticides, such as DDT, also are very effective all organochlorine compounds and are known in killing a broad range of insects; however, they to cause an increase in cancer incidence in hu- are slow to break down, remain toxic for a con- mans and to persist in the environment for siderable length of time, and accumulate in body many years. Aldrin, chlordane, and heptachlor fat. Additionally, the long-lived organic pesticides are cyclodienes that are similar to, but more act as endocrine disrupters, mimicking naturally toxic than, DDT. Chlorinated hydrocarbon pesti- occurring androgens (estrogens), and may be cides are notorious for their severe effects on detrimental to mammalian reproductive cycles— nontarget organisms, whereas target species,

MAMMALS OF PALO VERDE 61 frankie chap 05 8/20/03 7:51 PM Page 62

particularly many species of insects, develop re- have recognized three categories of species sistance (Laws 1993). Thirty pesticides have been whose interactions in ecosystems are important identified as commonly used in rice fields in for providing information about the quality of the area adjacent to Palo Verde (Robinson 1993). the habitat: (1) keystone species: species whose Some of these pesticides are extremely toxic to disappearance results in the disappearance of wildlife (and humans), and their use is illegal several other species; (2) indicator species: in Costa Rica, but they are still available on the species whose population changes are thought black market and are commonly used (Hilje to indicate the effects of management activities; 1988). High levels of pesticide residuals, includ- and (3) mobile link species: species who are im- ing organic chlorides and their metabolites, have portant links to more than one food chain, plant- been found in eggshells of herons that nest on animal association, or ecosystem (Soulé 1989). Isla de Pájaros in Parque Nacional Palo Verde Research efforts on mammals in tropical dry for- (Hidalgo 1986). The effects of agrochemicals on est should concentrate on species that fall into tropical mammals have yet to be studied. these categories with the goal of conserving both fauna and flora of dry forest regions. PREDICTIONS AND RECOMMENDATIONS TRAINING AND EDUCATION RESEARCH PRIORITIES FOR PROGRAMS FOR PARK PERSONNEL DRY-FOREST MAMMALS The implementation of long-term programs for Because little is known about the abundance the monitoring and conservation of dry-forest and distribution of most species of mammals in mammals in Costa Rica requires trained profes- the tropical dry forest, monitoring populations sionals working in the national parks. The lim- in its many distinct habitats should be a top re- ited budget that currently supports protected search priority. In order to identify critical areas areas in Costa Rica, however, is not sufficient for protection and adopt the best strategies for to employ specialized professionals, and most successful conservation, it is imperative to col- park personnel have no formal training in biol- lect basic information about population densities ogy. Creative proposals to increase the budget of and changes in densities over time within this national parks should be evaluated, including life zone. This information will help to identify imposing a tourist tax on hotels with the funds vulnerable species as such and to concentrate re- going directly to protected areas (see chapter 21). search and conservation efforts. Furthermore, The goal should be to have at least one special- monitoring the abundance of mammals within ized, professionally trained biologist working these various habitats throughout the year will within a protected area to initiate monitoring help to identify potential migratory patterns; and conservation programs and train park per- this information, in turn, will help determine sonnel in fieldwork. which areas to protect as biological corridors connecting tropical dry forest to other habitats. ECONOMIC ALTERNATIVES AND Finally, coordinated efforts should be established SUSTAINABLE DEVELOPMENT between Latin American countries that still have Although the consensus among conservation tropical dry forest in order to accumulate infor- biologists is to protect as much land as possible mation on the distribution and abundance of as quickly as possible, the human factor of pop- tropical dry forest mammal species over their ulation expansion and economic needs cannot entire range. be easily ignored. Unless conservation programs In addition to documenting the distribution take into account the needs of human societies, and abundance of dry-forest mammals, it is im- it is unlikely that the goals of long-term conser- portant to identify ecologically significant species vation will be achieved. Sound economic alter- for management and monitoring. Researchers natives need to be provided to rural Costa Ricans

62 STONER AND TIMM frankie chap 05 8/20/03 7:51 PM Page 63

in areas of tropical dry forest if we are to con- same environmental benefits for mammalian serve this fragile ecosystem and the mammal communities as do old-growth forests. Com- fauna that it supports. Alternative means of bined with projects that foster regeneration of generating economic benefits beyond the tradi- degraded areas, continued efforts need to focus tional use of the land for timber, agricultural on conserving old-growth forests, as many na- crops, and cattle grazing should be evaluated and tive mammals may be found only in this habitat. encouraged by the Costa Rican government. One Since dry forests in Costa Rica are largely possibility is to provide the highest economic in- restricted to protected areas, the habitats imme- centives for reforestation programs that provide diately surrounding these areas should be eval- total protection. Under the current forestry pol- uated before any type of development is carried icy, programs in which land is devoted to refor- out. Buffer areas surrounding national parks estation with plantations and selective logging and other protected areas theoretically exist in programs receive larger economic incentives Costa Rica, but little has been done to evaluate than programs providing total protection (see or restrict activities in these areas. For example, chapter 21). the agricultural development of rice fields bor- It is of utmost importance to encourage re- dering Palo Verde and Lomas Barbudal destroyed search on the sustainable development of eco- the biological corridor that connected them. The systems and to evaluate potential alternative potential effect that agricultural development on means of land and water exploitation that will the borders of the national parks has on native have a minimum impact on ecosystems. In par- mammal fauna needs to be evaluated more fully. ticular, in the dry-forest region of Guanacaste, studies evaluating the effects of forest fragmen- CONCLUSIONS tation, large-scale irrigation projects, and rice cultivation on local mammals should be under- No habitat type in Costa Rica (and throughout taken. Before sustainable development pro- Mesoamerica) has been more affected by hu- grams are implemented within or surrounding mans than the tropical dry forest. Open pasture- protected areas, feasibility studies should be con- land and grazed forested tracts are common ducted in order to estimate not only the eco- habitats within this region today. In recent years nomic benefit of the new activity but also the rice, sugarcane, and hay fields have replaced nat- ecological cost to the area. For example, the man- ural habitats throughout the Guanacaste region. agement plan implemented in Palo Verde using Palo Verde has lost at least eight species of na- cows within the park was part of a sustainable tive mammals to date, and it is likely that more development plan; unfortunately this activity was mammals in this habitat will become extinct implemented without first evaluating the eco- if efforts are not made to reduce the effects of nomic benefits or the ecological costs. The humans on this ecosystem. Some species of economic benefits are at best minimal and the mammals are still abundant at Palo Verde, and ecological costs high (Mozo 1995; see chapter 21). a few species (those that live in the savanna) are undoubtedly more abundant today than they PROTECTION, REGENERATION, were prior to settlement; however, the long-term AND BUFFER AREAS effects that the fragmented landscape will have If the native mammals of Costa Rica’s tropical on populations of native mammals are still un- dry forest are to be conserved, additional efforts known. Because there is no simple solution to must be made to protect the forest that remains guarantee the successful conservation of any and allow natural succession to regenerate particular species or group of species, we sug- mature stands of dry forest. Although increased gest a combined effort that includes (1) research; forest regeneration is a positive development, (2) education and training; (3) economic alter- young, regenerating forests do not provide the natives and evaluation of alternative uses of land

MAMMALS OF PALO VERDE 63 frankie chap 05 8/20/03 7:51 PM Page 64

and water; and (4) protection and regeneration versity by grazing and mowing in native tall- of habitats. We believe that such a multifaceted grass prairie. Science 280:745–47. approach will be the most successful way to pro- Conklin, N. L. 1987. The potential nutritional value to cattle of some tropical browse species from tect dry-forest mammals and their habitat. Guanacaste, Costa Rica. Ph.D. diss., Cornell University, Ithaca, N.Y. 329 pp. ACKNOWLEDGMENTS Daubenmire, R. 1972. Some ecologic consequences of converting forest to savanna in northwestern We thank the Ministerio del Ambiente y Energía Costa Rica. Tropical Ecology 13:31–51. and Sistema Nacional de Areas de Conservación, Fleming, T. H. 1988. The short-tailed fruit bat: A especially Javier Guevara Sequeira, for giving us study in plant-animal interactions. Chicago: Uni- versity of Chicago Press. 365 pp. the opportunity to work at Parque Nacional Palo Frankie, G. W., H. G. Baker, and P. A. Opler. 1974. Verde. Gordon Frankie, Eric Fuchs, Doug Gill, Comparative phenological studies of trees in Richard LaVal, Melissa Panger, Mauricio Que- tropical wet and dry forest in the lowlands of sada, and Bruce Williamson unselfishly shared Costa Rica. Journal of Ecology 62:881–919. with us original observations they have made. Frankie, G. W., S. B. Vinson, M. A. Rizzardi, T. L. Griswold, S. O’Keefe, and R. R. Snelling. 1997. David A. Stewart generously provided us with a Diversity and abundance of bees visiting a mass history of the Stewart Ranch, details of cattle flowering tree species in disturbed seasonal dry ranching, and historical information on wildlife forest, Costa Rica. Journal of the Kansas Ento- in the area. Gordon Frankie, Doug Gill, Marion mological Society 70:281–96. Klaus, Richard LaVal, Deedra McClearn, Mauri- Gentry, A. H. 1995. Diversity and floristic compo- cio Quesada, and Heather York offered editorial sition of neotropical dry forests. In Seasonally dry tropical forests, ed. S. H. Bullock, H. A. Mooney, comments on earlier drafts of this chapter, sig- and E. Medina, 146–94. Cambridge: Cambridge nificantly improving it. University Press. Glander, K. E. 1978. Howling monkey feeding be- REFERENCES havior and plant secondary compounds: A study of strategies. In The ecology of arboreal folivores, Belsky, A. J., and D. M. Blumenthal. 1997. Effects ed. G. G. Montgomery, 561–74. Washington, of livestock grazing on stand dynamics and D.C.: Smithsonian Institution Press. soils in upland forests of the Interior West. Con- Hartshorn, G. S. 1983. Plants: Introduction. In servation Biology 11:315–27. Costa Rican natural history, ed. D. H. Janzen, Boucher, D. H., M. Hanson, S. Risch, and J. H. 118–57. Chicago: University of Chicago Press. Vandermeer. 1983. Agriculture: Introduction. Heithaus, E. R., T. H. Fleming, and P. A. Opler. In Costa Rican natural history, ed. D. H. Janzen, 1975. Foraging patterns and resource utiliza- 66–73. Chicago: University of Chicago Press. tion in seven species of bats in a seasonal trop- Ceballos, G. 1995. Vertebrate diversity, ecology, ical forest. Ecology 56:841–54. and conservation in neotropical dry forests. In Helversen, O. von. 1993. Adaptations of flowers to Seasonally dry tropical forests, ed. S. H. Bullock, the pollination by glossophagine bats. In Animal- H. A. Mooney, and E. Medina, 195–220. Cam- plant interactions in tropical environments, ed. bridge: Cambridge University Press. W. Barthlott, C. M. Naumann, K. Schmidt- Chapman, C. A. 1988. Patterns of foraging and Loske, and K.-L. Schuchmann, 41–59. Bonn, range use by three species of Neotropical pri- Germany: Zoologisches Forschungsinstitut und mates. Primates 29:177–94. Museum Alexander Koenig. ———. 1989. Primate seed dispersal: The fate of Hidalgo, C. C. 1986. Determinación de residuos dispersed seeds. Biotropica 21:148–54. de plaguicidas organoclorados en huevos de Cochrane, M. A., A. Alencar, M. D. Schulze, C. M. ocho especies de aves acuáticas, colectados du- Souza Jr., D. C. Nepstad, P. Lefebure, and E. A. rante 1983–1984 en la Isla Pájaros, Guanacaste, Davidson. 1999. Positive feedbacks in the fire Costa Rica. Tesis de Grado, Escuela de Biología, dynamics of closed canopy tropical forests. Sci- Universidad de Costa Rica, San José. ence 284:1832–35. Hilje, L. 1988. El uso de los plaguicidas en Costa Collins, S. L., A. K. Knapp, J. M. Briggs, J. M. Blair, Rica. San José: Editorial Universidad Estatal a and E. M. Steinauer. 1998. Modulation of di- Distancia.

64 STONER AND TIMM frankie chap 05 8/20/03 7:51 PM Page 65

Holdridge, L. R., W. C. Grenke, W. H. Hatheway, ———. 1999. Lista oficial de la República de Costa T. Liang, and J. A. Tosi Jr. 1971. Forest environ- Rica: Lista de especies de fauna silvestre con ments in tropical life zones: A pilot study. Oxford: poblaciones reducidas y en peligro de extinción Pergamon Press. 747 pp. para Costa Rica. Decreto 26435-MINAE. In Lista Janzen, D. H. 1986. Guanacaste National Park: Trop- de fauna de importancia para la conservación en ical ecological and cultural restoration. San José: Centroamérica y México: Listas rojas, listas oficiales Editorial Universidad Estatal a Distancia. 103 pp. y especies en apéndices CITES, ed. V. Solís Rivera, Janzen, D. H., and D. E. Wilson. 1983. Mammals: A. Jiménez Elizondo, O. Brenes, and L. Vil- Introduction. In Costa Rican natural history, ed. nitzky Strusberg, 127–40. Sistema de Integra- D. H. Janzen, 426–42. Chicago: University of ción Centroamericana. Direccion Ambiental, Chicago Press. con el apoyo técnico de UICN–ORMA y WWF Laws, E. A. 1993. Aquatic pollution: An introductory Centroamérica. San José: WWF, UICN, SICA. text. 2d ed. New York: John Wiley and Sons. Mozo, E. T. 1995. Pastoreo con ganado vacuno, una 611 pp. alternativa del ACT para prevención de incendios León S., J., C. Barboza V., and J. Aguilar F. 1982. forestales, recuperación de humedales y restaru- Desarrollo tecnológico en la ganadería de carne. ración del bosque tropical seco. Bagaces, Costa San José: Consejo Nacional de Investigaciones Rica: Convenio MIRENEM–Opción Colombia, Científicas y Tecnológicas. Universidad Sergio Arboleda. 56 pp. Maldonado U., T., J. Bravo, G. Castro S., Q. Ji- Parsons, J. J. 1972. Spread of African pasture ménez M., O. Saborío, and L. Paniagua C. 1995. grasses to the American tropics. Journal of Evaluación ecológica rápida región del Tempisque Range Management 25:12–17. Guanacaste, Costa Rica. San José: Fundación ———. 1983. Beef cattle (ganado). In Costa Rican Neotrópica, Centro de Estudios Ambientales y natural history, ed. D. H. Janzen, 77–79. Chi- Políticas. 104 pp. + 8 appendixes. cago: University of Chicago Press. Mata, A., and O. Blanco. 1994. La cuenca del Golfo Pepper, I. L., C. P. Gerba, and M. L. Brusseau, eds. de Nicoya: Un reto al desarrollo sostenible. San 1996. Pollution science. New York: Academic José: Editorial de la Universidad de Costa Rica. Press. 397 pp. 235 pp. Quesada López-Calleja, R. 1980. Costa Rica: La fron- McCammon-Feldman, B. 1980. A critical analysis tera sur de Mesoamérica. 2d ed. Madrid: Instituto of tropical savanna forage consumption and Costarricense de Turismo. 288 pp. utilization by goats. Ph.D. diss., University of Robinson, T. H. 1993. Fate and transport of agri- Illinois, Urbana. cultural contaminants from rice paddies: Im- McCoy, M. B. 1994. Seasonal, freshwater marshes pact sampling strategies and the potential en- in the tropics: A case in which cattle grazing is vironmental degradation to dry tropical coastal not detrimental. In Principles of conservation bi- wetlands—Guanacaste, Costa Rica. Master’s ology, ed. G. K. Meffe and C. R. Carroll, 352–53. thesis, University of California, Santa Barbara. Sunderland, Mass.: Sinauer Associates. 47 pp. McCoy, M. B., and J. M. Rodríguez. 1994. Cat- Robinson, W. L., and E. G. Bolen. 1984. Wildlife tail (Typha dominguensis) eradication methods ecology and management. New York: Macmillan. in the restoration of a tropical, seasonal, fresh- 478 pp. water marsh. In Global wetlands: Old World Rodríguez, F. J., and F. A. Chinchilla. 1996. Lista and New, ed. W. J. Mitsch, 469–82. New York: de mamíferos de Costa Rica. Revista de Biología Elsevier. Tropical 44:877–90. McNab, B. K. 1985. Energetics, population biology, Sáenz-Maroto, A. 1955. Los forrajes de Costa Rica. and distribution of xenarthrans, living and ex- San José: Editorial Universitaria, Universidad tinct. In The evolution and ecology of armadillos, de Costa Rica. sloths, and vermilinguas, ed. G. G. Montgomery, Schulze, M. D., N. E. Seavy, and D. F. Whitacre. 219–32. Washington, D.C.: Smithsonian Insti- 2000. A comparison of the phyllostomid bat tution Press. assemblages in undisturbed Neotropical forest MINAE (Ministerio de Ambiente y Energía). 1998. and in forest fragments of a slash-and-burn Un recuento de los logros del Área de Conservación farming mosaic in Petén, Guatemala. Biotropica Tempisque, 1997. San José: Ministerio del Am- 32:174–84. biente y Energía, Sistema Nacional de Áreas Soulé, M. E. 1989. Conservation biology in the de Conservación, Área de Conservación Temp- twenty-first century: Summary and outlook. isque. 30 pp. In Conservation for the twenty-first century, ed.

MAMMALS OF PALO VERDE 65 frankie chap 05 8/20/03 7:51 PM Page 66

D. Western and M. Pearl, 297–303. New York: Nadkarni and N. T. Wheelwright, 223–44, 553– Oxford University Press. 60. New York: Oxford University Press. Stern, M., M. Quesada, and K. E. Stoner. 2002. Timm, R. M., and S. E. Lewis. 1991. Tent construc- Changes in composition and structure of a tion and use by Uroderma bilobatum in coconut tropical dry forest following intermittent cattle palms (Cocos nucifera) in Costa Rica. Bulletin grazing. Revista de Biología Tropical 50:1021–34. of the American Museum of Natural History 206: Stoner, K. E. 2001. Differential habitat use and 251–60. reproductive patterns of frugivorous bats in Timm, R. M., D. E. Wilson, B. L. Clauson, R. K. tropical dry forest of northwestern Costa Rica. LaVal, and C. S. Vaughan. 1989. Mammals Canadian Journal of Zoology 79:1626–33. of the La Selva–Braulio Carrillo complex, Costa ———. 2002. Murciélagos nectarívoros y frugí- Rica. North American Fauna 75:1–162. voros del bosque caducifolio de la Reserva de la Timm, R. M., R. K. LaVal, and B. Rodríguez- Biosfera Chamela-Cuixmala. In Historia natu- Herrera. 1999. Clave de campo para los mur- ral del bosque caducifolia de Chamela, ed. F. A. ciélagos de Costa Rica. Brenesia (Museo Nacional Noguera, M. Quesada, J. Vega, and A. Garcia- de Costa Rica) 52:1–32. Aldrete, 379–95. Mexico City: Instituto de Bi- Vaughan, C., and M. Rodríguez. 1986. Compara- ología, Universidad Nacional Autónoma de ción de los hábitos alimentarios del coyote (Ca- México. nis latrans) en dos localidades en Costa Rica. Timm, R. M. 1994. The mammal fauna. In La Selva: Vida Silvestre Neotropical 1:6–11. Ecology and natural history of a Neotropical rain ———. 1991. White-tailed deer management in forest, ed. L. A. McDade, K. S. Bawa, H. A. Hes- Costa Rica. In Neotropical wildlife use and con- penheide, and G. S. Hartshorn, 229–37, 394–98. servation, ed. J. G. Robinson and K. H. Redford, Chicago: University of Chicago Press. 288–99. Chicago: University of Chicago Press. Timm, R. M., and R. K. LaVal. 1998. A ﬁeld key Vaughan, C., M. McCoy, J. Fallas, H. Cháves, to the bats of Costa Rica. Occasional Publication G. Barboza, G. Wong, M. Carbonell, J. Rau, and Series, Center of Latin American Studies, Univer- M. Carranza. 1995. Plan de manejo y desarrollo sity of Kansas 22:1–30. del Parque Nacional Palo Verde y Reserva Biológica ———. 2000. Mammals. In Monteverde: Ecology Lomas Barbudal. Heredia, Costa Rica: Universi- and conservation of a tropical cloud forest, ed. N. M. dad Nacional. 206 pp.

66 STONER AND TIMM frankie chap 06 8/20/03 7:52 PM Page 67

chapter 6

The Conservation Values of Bees and Ants in the Costa Rican Dry Forest

S. Bradleigh Vinson, Sean T. O’Keefe, and Gordon W. Frankie

ver the past 30 years of solitary-bee towns of Liberia and Cañas (see maps in chap- Ostudies in the Costa Rican dry forest we ter 1). The second factor is the introduction of have observed a steady decline in native solitary- jaragua grass, Hyparrhenia rufa, from Africa. bee populations. In 1972 bee diversity was sur- The third factor, often linked to the second, is veyed from a small population of the fabaceous human-caused fire. Jaragua grows 2–3 m higher tree Andira inermis, at a site just south of the than the short native grasses, becomes highly town limits of Liberia (see maps in chapter 1). At flammable during the dry season, and burns that time about 70 species were collected (Frankie very hot. Jaragua is also fire-adapted and readily et al. 1976). In 1989 we casually sampled bees resurges after a fire. It is fanned by dry-season again from several A. inermis trees in the vicin- trade winds that allow fire to burn well inside the ity of Liberia and found only 37 species (Vinson dry-forest edge, which opens the forest to fur- et al. 1993). Another sampling at the original ther jaragua invasion. Besides slowly converting Liberia site in 1996, during a year of robust forest to grassland, fires sometimes invade deep A. inermis flowering, revealed that diversity had into the forest, resulting in loss of deadwood that declined from a high of 70 species in 1972 to is used by nesting anthophorid bees including only 28 species. Further, overall numbers of bees some Centris (Frankie et al. 1988) and several Xy- had declined by slightly more than 90 percent locopa species. Deadwood-nesting species were (Frankie et al. 1997). rare in the 1996 captures (Frankie et al. 1997). What brought on this decline? Our data Fire also leads to other problems for bees. Im- suggest four major factors. The first is habitat portant resource trees such as three Tabebuia destruction, caused primarily by agricultural species, Dalbergia retusa, Caesalpinia eriostachys, development and urbanization, between the and Cochlospermum vitifolium are resistant to

67 frankie chap 06 8/20/03 7:52 PM Page 68

“cool” native-grass fires, but not the hot-burning tems (Christensen et al. 1996), especially aquatic jaragua that results in tree loss following one ecosystems (Loeb and Spacie 1994). or two fires (Frankie et al. 1997). Fire also re- One major dilemma of this approach is deter- moves ground cover, increasing vulnerability of mining which groups of organisms are the best ground-nesting bees to predation (S. Vinson indicators. Pearson (1994) offered seven criteria pers. obs.) and increasing solar radiation, which that should be met by indicator groups: (1) the alters microclimate and solitary-bee nesting group is taxonomically well known, (2) their success (Frankie et al. 1988). The fourth factor biology and general life history are understood, is introduction of ornamental flowering plants (3) their populations can be surveyed and ma- and reforestation using exotic tree species that nipulated, (4) some species of the group occupy do not supply proper bee requisites. Intensive a breadth of habitats with a broad geographical studies on the relationship of Africanized honey- range, (5) some populations should be special- bees to native bees from 1996 to 2001 in this ized, (6) observed patterns in the indicator group area revealed little evidence that this exotic bee are reflected in other related and unrelated taxa, is having an impact on the native bee fauna and (7) the group is of potential economic im- (Frankie et al. 2002). portance. These criteria suggest the use of species It is clear that over the past 30 years, as the from different trophic levels, but Williams and dry forest declined, solitary bees also declined. Gaston (1994) advocated using only higher tax- Despite the establishment of reserves and na- onomic groups, Oliver and Beattie (1996) sug- tional parks, factors such as fire, exotic invaders, gested using selected morphospecies, and Lam- and human encroachment threaten the Costa beck (1997) suggested using an umbrella species, Rican dry forest (Frankie et al. 1997; chapters 7, a single species that may indicate areas for 9, 12, 13, 21–24). Further, habitat fragmentation conservation. Birds and mammals are com- and isolation continues (chapter 3). To protect monly used for monitoring because they are remaining forests and their diversity we must well known, often easier to identify, have larger first identify threatened areas. This requires long- habitat requirements, and are not so numerous term monitoring of habitats with consistently as to overwhelm a study. Further, their numbers applied, quantitative methods that can detect and frequency of occurrence are influenced by changes in organism populations and their habi- environmental disturbance. Sea turtles, for ex- tats and at the same time recognize new prob- ample, can be used as “flagship” organisms be- lems. Developing methods and procedures for cause they generate sympathy from financial monitoring the health of protected areas becomes donors (see chapter 15). In addition, many bird critical. Here we propose using both bees and assemblages have a loyal human following, the ants as bioindicators. bird watchers. Although birds and mammals are useful in monitoring large landscapes, they are not use- MONITORING ECOSYSTEM HEALTH ful for monitoring small unique habitat frag- Although there are many approaches to moni- ments embedded within larger conserved areas toring and assessing ecosystem health, the most or smaller isolated fragments remaining on practical and widely implemented includes the private lands (both referred to in this chapter as use of indicator organisms. Originally, this “habitat fragments”). This is because different approach was used to assess the effects of pol- groups of organisms operate at different scales lutants on selected organisms—for example, in order to live and reproduce and thus differ in lichens to monitor for air pollution (Richardson the ways they can be used as indicators. Scale is 1992). Indicator organisms are also used for extremely important, and organisms that de- broader ecological monitoring, such as measur- pend on diversity and derive their requisite needs ing effects of management practices on ecosys- within the scale in question must be selected. For

68 VINSON ET AL. frankie chap 06 8/20/03 7:52 PM Page 69

such habitat fragments, specific insect groups genera, and Costa Rica alone has at least 785 (Hammond 1994), such as aquatic insects (Loeb species and 98 genera (Griswold et al. 2000). and Spacie 1994) and ants (Agosti et al. 2000), The bee fauna in Costa Rica appears to be rich have been frequently used. But should we be on a per-area basis, as there are more than concerned about habitat fragments? Habitat frag- 129 species/104 sq km in Costa Rica, compared ments may provide resources, act as stepping- with 8 in Mexico and 4.8 in the United States. stones to other areas, or serve as refuges or In other words, in Costa Rica there are 16 times sinks. They may harbor unique organisms, have as many bee species per area as in Mexico and strong populations of species that are less com- 27 times as many as in the United States (Han- mon or threatened in other areas, or contain son and Gauld 1995). The major bee groups different species or genotypes (chapter 16). In are the Halictinae (195 spp.), Anthophorinae addition, habitat fragments may contain the (143 spp.), Megachilinae (96 spp.), Euglos- only remnants of original habitat, which may sinae (58 spp.), Meliponinae (48 spp.), and Xy- include representative species on which restora- lopinae (37 spp.) (Hanson and Gauld 1995). In tion projects could draw. Unfortunately, habitat the dry forest, Meliponines along with Centris, fragments and investment in their preservation Epicharis, Xylocopa, and introduced honeybees are largely ignored, either because they are con- are common. The Tempisque dry-forest region sidered safe within larger protected areas or be- of Costa Rica alone has more than 250 bee cause they are considered too small or of low species (chapter 2). Michener (2000) provides value. modern keys to identify bee genera and sub- We suggest that Hymenoptera are an ideal genera, but there is no single source for identi- indicator group for habitat fragments and focus fication to species, and many smaller species on bee and ant taxa that access different habitat await description. attributes. Bees are dependent on a number of Worldwide there are about 300 genera of specific requisite plant species and nesting con- ants (Formicidae) with some 8,800 species (Höll- ditions distributed in several microhabitats. Be- dobler and Wilson 1990). Ants of Costa Rica are cause bees depend on several microhabitats, they better known than those of any other New World can provide information on more than the health region south of the United States. Longino of the location in which they are recorded. In and Hanson (1995) recorded 81 genera and 620 contrast, ants are more sessile and dependent species in Costa Rica, and many more await de- on local microclimate and availability of local re- scription, particularly the smaller, cryptic forms. sources. In this chapter we discuss the diversity Most ant genera are known, and many are found and ecological relationships of these two taxa in the dry forest, but genera new to Costa Rica and argue why they may be good bioindicators. are still being found (O’Keefe and Agosti 1998). We then discuss how they may be monitored and Bolton (1994) has provided a thorough key to how the data may be interpreted and used to ef- identify ants to genus. Identification to species fect change. is more problematic because there is no synthetic treatment of all species (MacKay and Vin- son 1989). DIVERSITY OF BEES AND ANTS

In any monitoring effort it is important to iden- ECOLOGICAL RELATIONSHIPS OF BEES tify the species encountered. Hanson and Gauld AND ANTS AND THEIR REQUISITE NEEDS (1995) provide an excellent review of the taxonomy and biology of Costa Rican Hymenoptera. BEES Bees are classiﬁed as a single family, the Apidae Bees and ants have numerous ecological func- (Michener 2000). Worldwide, there are 25,000 tions in the Costa Rican dry forest, and under- to 30,000 species representing more than 400 standing these speciﬁc interactions is necessary

CONSERVATION VALUES OF BEES AND ANTS 69 frankie chap 06 8/20/03 7:52 PM Page 70

if these taxa are to be used in monitoring. Bees some species (Centris flavifrons) nest in the require a number of resources and conditions, ground and prefer soil, others (C. flavofasciata) referred to as their biological and ecological req- prefer sand, and still others (C. bicornuta) nest uisites. Many bee species are specialists in their in preexisting wood cavities (Coville et al. 1983 requisite needs, requiring particular resources in Vinson et al. 1993). Bees construct nests con- and conditions to survive. As some requisite sisting of a series of cells using secretions and needs are fulfilled, select plants are pollinated. resources from the surrounding environment, Plants have also evolved specializations that in- such as sand, wood borings (sawdust), spider fluence the species of bees that visit. Pollen- webs, plant leaves, petals, waxes, saps, and resins producing specialists include trees such as (Eickwort et al. 1981). Further, most solitary bees Cochlospermum vitifolium, numerous Cassia and seal the nest entrance with a plug often com- Senna species, Curatella americana, and Byrson- posed of other materials that are found in the ima crassifolia (Vinson et al. 1993). Some plants, local environment (Roubik 1989). Nests must such as C. vitifolium and the shrub Cassia biflora, be located in appropriate habitats. For example, contain their small, dry pollen in tubular an- we found that some bee species produce fewer thers, which bees collect with a behavior called cells per nest in recently burned over or rocky “buzz-collecting” (Thorp 1979). Bees of several soil or fail to complete development when their genera perform buzz pollination, such as those nests were located in exposed, hot areas (see of Xylocopa, Centris, Epicharis, Gaesischia, Exoma- Frankie et al. 1988, and Vinson and Frankie 1988 lopsis, Melipona, and Bombus (Buchmann 1985); in Vinson et al. 1993). the introduced honeybee or stingless bees of the tribe Trigonini cannot. ANTS A primary energy source for all adult bees is Ants are even more dependent than bees on nectar, which also serves as an important provi- specific habitats. The two major ant guilds are sioned resource. Important nectar plants in- those associated with plants and those that clude Tabebuia species, Gliricidia sepium, Cae- are predators. Ant-plant associations include ant- salpinia eriostachys, Andira inermis, Myrospermum plant mutualisms, seed dispersal, and herbivory. frutescens, Pterocarpus michelianus, and Dalbergia Two well-studied ant-plant mutualisms include retusa (Vinson et al. 1993). relationships between Azteca ants and Cecropia Centris species are an important pollinator plants and those between Pseudomyrmex ants group in the Costa Rican dry forest (Frankie et and Acacia plants, each involving several species al. 1976). In addition to pollen and nectar, they of ants and plants (Huxley and Cutler 1991). need floral oils as provisioned larval food or for Seed-dispersing ants, ecologically important nest construction (or both) (Vinson et al. 1997). in many arid regions, include only Solenopsis The major oil resource plants are the tree Byr- and Pheidole in Costa Rica (Longino and Hanson sonima crassifolia (Malpighiaceae) and, to a lesser 1995). Ants belonging to the genera Atta and extent, several malpighiaceous vines (Vinson et Acromyrmex (known as leaf-cutters and grass- al. 1993). Resource accessibility also influences cutters) are dominant herbivores in the Neotrop- a bee’s impact on plant pollination as well as its ics, consuming far more vegetation than any ability to fulfill its requisites. Small bees are other animal group (Stradling 1991). more effective in servicing smaller flowers that Predatory ants in Costa Rica can be divided they can manipulate (Gottsberger 1986), but into four subguilds: roaming army ants, large large bees can gain access to large flowers and predatory ponerines, cryptic soil-litter ants, and can move pollen over long distances (Frankie et generalist predators. Army ants have a major al. 1976; Michener and Brooks 1984). impact on tropical ecosystems (Gotwald 1995). In addition to nutrition, bees require specific Eciton burchelli form fronts 3–15 m wide, feeding substrates for nest construction. For example, mainly on other arthropods, including other

70 VINSON ET AL. frankie chap 06 8/20/03 7:52 PM Page 71

ants (Franks 1982a). Other species of army ants span. Species diversity, ranges of sociality, num- raid in columns and are insect prey specialists. bers of individuals, diverse requisite needs, and Ponerines, which have a powerful sting, are gen- numerous ecological roles are factors that make eralist predators that form small colonies but this order ideal for evaluating forest fragments. represent a significant portion of the dry-forest Scale is important (Lawton et al. 1998), and ant fauna (Longino and Hanson 1995). The bees and ants are sensitive to, but differ in rarely seen soil-litter ants are mostly generalist their responses to, the scale of environmental predators, feeding on whatever they can catch, disturbances, such as wildfires, invasion by ex- with a few specialist predator species. This fauna otic organisms, or the loss of critical requisite relies on a deep and established litter layer, in resources. which as many as 30–35 genera can be encoun- There is much to justify the use of bees and tered. The last group consists of the generalist ants as indicators of ecosystem health. Bees predators and scavengers such as Crematogaster can move between several resources present in and Solenopsis. Ants of this last group feed on one or more habitats and thus may be depend- almost any insect they can subdue but will also ent on several different nearby habitat frag- feed on carrion, seeds, select plant parts, or ments (Banaszak 1996). Bees have not been used honeydew from homopterans (Vinson 1997). extensively to monitor, although this situation Further details of predatory ants can be found is changing as more information is gathered in Hölldobler and Wilson (1990). (Kearns and Inouye 1997; Frankie et al. 1998, Ants influence their ecosystem in many 2002; Kevan 2001). In contrast, because native ways, such as modifying habitat near their nests ant colonies persist over time, have limited through translocation of large amounts of or- ranges, and require an extensive but local forag- ganic material into deeper soil layers. As a result, ing area, they are less susceptible to fragmen- soil pH is altered, soil water-holding capacity is tation than vertebrates or even bees. Thus they increased, and salts are increased—all of which are dependent on resources that are in specific are important to plant root health (Pe˛tal 1978). habitats in which they reside (Andersen 1990). Ants are a major soil-turnover group that rival or Their disappearance is an early sign of local surpass termites and earthworms (Wilson 1990). stresses in an ecosystem. However, several ant Ants are also an important food resource for species are good invaders of disturbed habitats, many vertebrates (Hölldobler and Wilson 1990), and their presence is a warning of detrimental particularly birds. Willis and Oniki (1978) listed changes occurring in the habitat. Some ant gen- 50 bird species that depend on and consume a era and species are widespread and thus provide great number of arthropods fleeing army ant comparative data. Other ant genera have high raids (Franks 1982b). species diversity with many specialist species. Finally, ants occupy higher trophic levels, are easily sampled, are relatively easy to identify to BEES AND ANTS ARE GOOD genus, and are responsive to changing environ- CANDIDATES AS BIOINDICATORS mental conditions (Majer 1983). Great species OF ENVIRONMENTAL CHANGE diversity, filling numerous ecological roles, and Members of the order Hymenoptera are abun- sensitivity to disturbance make ants an impor- dant in most ecosystems and are involved in tant ecological indicator group (Alonso 2000). many interactions with other taxa (LaSalle and Gauld 1993a,b). Hymenoptera fit the indicator DOCUMENTING CHANGES IN group criteria discussed by Pearson (1994), and, ORGANISMS USED FOR MONITORING compared with vertebrates, they rapidly respond in numbers to environmental change because New (1999) reviewed aspects of conducting in- of their high reproductive capacity and short life vertebrate surveys for conservation that included

CONSERVATION VALUES OF BEES AND ANTS 71 frankie chap 06 8/20/03 7:52 PM Page 72

ants and wasps but not bees. Through the use of parameters, ant diversity can be examined by trap nests (Strickler et al. 1996; Frankie et al. ecological roles if the ant genera are first organ- 1998) or sweep netting (Gess and Gess 1993; ized into functional groups to reduce complex- Kevan et al. 1997), populations of bees can now ity (Greenslade 1985; Andersen 1986). Ander- be surveyed in specific habitats (Kevan 2001). In sen (1991) described seven Australian ant-fauna addition, a relatively noninvasive monitoring functional groups based on their relative domi- technique, involving floral visitation counts, has nance within the ant community, resource needs, been developed to assess relative floral attrac- foraging preferences, and environmental pref- tion, diversity, and abundance of bees in an area erences. He included opportunists, which are (Frankie et al. 2002). characteristic of disturbed habitats and share In contrast, ants have been extensively sam- several attributes (i.e., generalized habits, omniv- pled by numerous methods, such as pitfall or orous diets, flexible foraging times, poor com- bait traps, leaf litter sampling, beating or canopy petitors, easily displaced from food resources, fogging, and observation (see review by Bestel- most abundant at sites where dominant ants are meyer et al. 2000). Many of these methods sam- least abundant), because their increase is an ple different ecological groups with little species indicator of habitat disturbance. The occurrence overlap. For example, bait traps placed on vege- of highly invasive (tramp) ants (Passera 1994) is tation sample canopy-foraging species, whereas of concern, and Saks and Carroll (1980) showed the same trap placed in holes in the ground col- that tramp ants could become abundant in lect only subterranean-foraging species. disturbed areas while remaining uncommon in nearby undisturbed forests. Exotic tramp ant species have been found to produce major USING BEES AND ANTS changes in abundance and diversity of inver- AS BIOINDICATORS tebrate species in other regions (Lubin 1984; Bees have been rarely used as a monitoring tool, Zenner-Polania 1994). Tennant (1994) found, although honeybees (Bromenshenk et al. 1985) however, that the tramp Wasmannia auropunc- and solitary bees (Kevan et al. 1997) have been tata was not a pest in its native Costa Rica. Kas- used to detect and monitor for pollutants. One pari and Majer (2000) give a more thorough reason that they have rarely been used is the discussion of using ants to monitor environ- lack of simple nondestructive monitoring meth- mental changes. ods. Another is the high variation in diversity (Williams et al. 2001) and population numbers CHANGES THAT ARE IMPORTANT (Frankie et al. 1998; Roubik 2001) of the native TO BEES AND ANTS bee fauna from one place and time to another. The development of floral visitation counts Bees require a consistent supply of diverse, crit- (Frankie et al. 2002) and grouping of bees into ical resources that are available in an intact dry functional groups, as suggested in table 6.1, forest. Changes in one or two plant resource should provide new opportunities for the use of species that provide either nesting or food ma- bees in biological monitoring. terials can result in a void that can be detri- Ant monitoring has been used for fire man- mental. As forest fragments become smaller and agement, pesticide contamination, habitat dis- more isolated, the loss of critical resources in- turbance, and reserve design (Andersen 1990). creases. For example, fewer nest sites can result Majer (1983) listed seven parameters for using in increased nest site competition and usurpa- ants as indicators: species richness, species den- tion (Vinson and Frankie 2000) and an increase sity, Shannon’s diversity index, Eveness index, in mortality due to disease, parasites, and pred- Mountford’s similarity index, indicator species, ator activity. As the population declines, abun- and indicator groups. In addition to Majer’s dant resources, such as nectar, may be under-

72 VINSON ET AL. frankie chap 06 8/20/03 7:52 PM Page 73

TABLE 6.1 Functional Groups of the More Common Bees in Costa Rica Based on Foraging and Nest Habits

functional groups (subfamily) genus

Large to medium flower pollinators—deadwood nesters (Xylocopinae) Xylocopa, (Apinae) Centris Large to medium flower pollinators—ground nesters (Apinae) Centris, Epicharis Oil flower pollinators (Apinae) Centris, Paratetrapedia, Tetrapedia Moderate to small flower specialist pollinators (Panurginae), (Apinae) Ancyloscelis, Melitoma, Peponapis, Xenoglossa, (Megachilinae) Lithurgus Moderate to small flower generalists—cavity nesters (Apinae) Melipona Moderate to small flower generalists—ground nesters (Colletinae), Colletes Moderate to small flower generalists—leaf collectors (Megachilinae) Osmia, Anthidium, Megachile Moderate to small flower generalists—resin collectors (Megachilinae), (Apinae) Melipona Moderate to large flower specialists—fragrance collectors (Apinae) Nocturnal foragers (Halictinae) Megalopta Dawn foragers (Apinae) Peponapis, Xenoglossa Cleptoparasitic (Megachilinae) Coelioxys, Dolichostelis, (Nomadinae) Nomada, (Apinae) Mesoplia Buzz pollen collectors (Diphaglossinae) Ptiloglossa, (Halictinae) Megalopta, (Apinae) Centris, Epicharis;Gaesischia, Exomalopsis, Thygater, (Xylocopinae) Xylocopa Social generalist pollinators (Apinae) Apis, Melipona

utilized, causing nectar-producing plant species synchronizations. For example, the premature to receive limited pollination (Frankie et al. emergence of an important pollinator or pre- 1997). If a resource within a fragment is lost, mature flowering of a resource could have serious even temporarily, causing the extinction of the reproductive consequences for both (see chap- consumer, consumers from other habitat frag- ter 4). Another serious problem for bees is the ments may be unable to locate or reach to recol- introduction of nonpreferred floral resource onize the fragment. plants, planting of exotic flowers, and reforesta- As forests are reduced or altered (or both), tion with non-native trees (Frankie et al. 1997, challenges to their health increase. Local envi- 1998). For ants these same plants may provide ronmental changes can also alter timing of new prey or opportunities for invasive species. important biological events, disrupting seasonal As a result, there is a change in the balance of

CONSERVATION VALUES OF BEES AND ANTS 73 frankie chap 06 8/20/03 7:52 PM Page 74

biological interactions that can release some used bee functional groups as a survey or mon- species from their controls or increase the sus- itoring tool. As Stebbins (1979) noted, as habi- ceptibility of an area to invasion. tats are fragmented, remnant plant populations become isolated, resulting in “ecological traps.” Resident pollinators may be unable to locate CONSIDERATIONS IN required resources at the right time and thus may MONITORING CHANGES IN suffer declines (Jennersten 1988; Cane 2001). BEE AND ANT POPULATIONS But such problems are not specific to habitat Although recommendations regarding conser- fragments. Janzen et al. (1982) reported that vation of large areas and corridors (see chap- different habitats even in a conserved region ters 3, 4, 9, 21, and 23) are essential, preserving differed dramatically in bee diversity. Some diversity of habitat fragments is also important bees, such as euglossines, travel great distances, to maintain the region’s overall diversity (see whereas others, such as Centris flavofasciata, are chapter 16). To use Hymenoptera as a conser- more local (Vinson et al. 1987). Further, a par- vation tool, monitoring techniques need to be ticular bee species may nest in one habitat type implemented and parameters for evaluation de- and travel to another for nest construction re- veloped. Parameters that target specific focus or sources and then to a third or fourth for provi- functional groups (e.g., army ants or oil-collecting sions. Thus, changes in bee species composition bees) that include the bee or ant fauna (i.e., num- should be revealing, considered from a func- ber of genera or functional groups present) must tional group standpoint. be considered. Bees or ants can then be surveyed With regard to establishing threshold levels and monitored in an undisturbed area to estab- for ants or bees in Costa Rica, there are no data lish baseline data. These data are essential as a for ants and only some for bees (Frankie et al. comparison with disturbed areas or other areas 2002). However, some preliminary suggestions of concern. Parameters already known to be use- can be offered. To establish upper threshold ful include the presence or absence of certain fo- levels—levels at which the fauna can be consid- cus or functional groups, such as army or tramp ered “healthy”—climax-growth habitats and spe- ant species or buzz-pollen or nectar-collecting bee cialized embedded habitats in areas already pro- species, and a survey of total numbers of genera tected should be surveyed. The establishment present. of lower threshold levels is more problematic. For example, the Costa Rican dry-forest ant Future research lies in establishing quantitative fauna could be organized—in a manner similar measures of functional groups, determining to Andersen’s (1991) organization—to include lower threshold levels of target groups, and im- Atta, Pseudomyrmex, Azteca, Eciton, and Labidus plementing warning systems for habitats in peril. as dominates; Solenopsis, Camponotus, Gnampto- One method for measuring ecosystems’ genys, and Pheidole as subordinates; and Cremato- health would be to create a scale by establishing gaster, Leptothorax, and Monomorium as the an upper threshold level for the number of ant generalist predators. The cryptic leaf-litter fauna genera or species present in a healthy habitat include ponerine and myrmecine genera; op- and comparing it, and the species composition, portunists include Solenopsis, Wasmannia, and with the same information from habitats with Pheidole; and Pachycondyla and Ectatomma would various levels of disturbance. The upper thresh- round out the list as the large solitary foragers. old of genera present in a pristine area may be Although 67 percent of the world’s flowering 50–60 genera, whereas in very disturbed areas, plants depend on insects for pollination (Tepe- at least 5–8 genera largely composed of tramp dino 1979) with bees predominating (see chap- ants should still be found. With ant functional ter 2), to our knowledge no one has developed or groups as a parameter, intact, healthy habitat

74 VINSON ET AL. frankie chap 06 8/20/03 7:52 PM Page 75

patches should have all six functional ant groups lems and then attempting restoration to some present, whereas highly disturbed areas may have desired level. For bees, this may involve planting only opportunistic species. missing preferred plant species or providing nest Bees could be dealt with similarly, but at the sites (see chapter 2). Ant restoration may require species level. For example, a decrease in the selective removal of tramp ant species, restoring number of wood-nesting Centris and Xylocopa certain associated plants, or providing other species would suggest a loss of deadwood nest requisite resources. resources, whereas an overall decrease in Centris but not other bee species would suggest a loss RECOMMENDATIONS of oil resources. In the case of bees, another factor to consider is resource utilization in terms Biologists involved in habitat or wildlife manage- of flower visitation counts (Frankie et al. 2002). ment, field ecology, or pest management should A decline in Centris at both nectar and oil re- become involved in monitoring both native and source trees in which the nectar is not com- exotic insects, including bees and ants, using pletely removed but the oil resource is would mostly nondestructive methods. Monitoring suggest that the oil resource was limiting. As a should be scheduled for several distinct seasons, result, restoring more oil-producing B. crassifolia over a broad area, and ideally at several key times trees to an area would be of more benefit to Cen- of day. tris than planting nectar-producing trees. The most essential component to any con- In summary, upper and lower threshold lev- servation program is education. This requires els can be determined by surveying healthy, un- development of educational materials that em- affected habitats, habitats of various sizes, and phasize importance and value of insects such habitats with various levels of disturbance. Once as bees and ants. In many cases it will be up to thresholds are established, habitat patches can biologists to develop outreach materials for dif- be monitored, and these results can be com- ferent audiences such as children, farmers, pared to assess levels of impact or disturbance ranchers, and rural communities. Information or success of restoration efforts. Historical col- is needed regarding the importance of bee di- lections from undisturbed habitats are useful versity for efficient pollination of plants in con- for long-term comparisons (Frankie et al. 1997). served areas and many crops and the importance of native ants in managing pest outbreaks in natural and managed systems. Also needed is IDENTIFICATION AND RESTORATION an increased awareness of invader species and OF SPECIFIC HABITATS AND ways in which they gain access to an area. Most CRITICAL RESOURCES landowners and stewards could also use this First, various habitats should be characterized information to learn about the value of setting according to the bee and ant species present, aside land for conservation and for being vigi- including associated plants, animals, and envi- lant against invader organisms. ronmental requirements. Second, bee and ant Attitudes toward natural resource use and service or functional groups should be defined exploitation must be changed, and governmental for each habitat type (table 6.1). Third, once bee policy should play a role in implementing some and ant species and their various associations of the outlined recommendations. The govern- are known, a monitoring program can be initi- ment could provide incentives to landholders to ated. This program should include abundance, restore marginal or nonproductive agricultural distribution, species’ composition, and occur- land. Government policy could encourage and rence of their requisite needs. Finally, the value support biologists who interpret data collected of monitoring is in identifying habitats with prob- in monitoring programs in public lands and in

CONSERVATION VALUES OF BEES AND ANTS 75 frankie chap 06 8/20/03 7:52 PM Page 76

other conserved areas. Advisory groups should Bestelmeyer, B. T., D. Agosti, L. E. Alonso, C. R. F. be established in selected regions that consist of Brandão, W. L. Brown, J. H. C. Delabie, and several biologists (botanists, zoologists, and con- R. Silvestre. 2000. Field techniques for the study of ground-dwelling ants: An overview, servation-oriented specialists) and people from description, and evaluation. In Ants: Standard the agricultural community, tourist industry, mu- methods for measuring and monitoring biodiver- nicipal government, and educational commu- sity, ed. D. Agosti, J. D. Majer, L. E. Alonso, nity who can provide advice and suggestions to and T. R. Schultz, 122–44. Washington, D.C.: higher-level governmental agencies (see chap- Smithsonian Institution Press. Bolton, B. 1994. Identification guide to the ant gen- ters 18 and 19). era of the world. Cambridge, Mass.: Harvard Uni- versity Press. 224 pp. ACKNOWLEDGMENTS Bromenshenk, J. J., S. R. Carlson, J. C. Simpson, and M. J. Thomas. 1985. Pollution monitoring The authors thank Ronald Weeks, Peter Kevan, of Puget Sound with honey bees. Science 227: and Stan Schneider for manuscript suggestions. 632–34. Buchmann, S. L. 1985. Bees use vibration to aid pollen collection from non-poricidal flowers. Jour- REFERENCES nal of the Kansas Entomological Society 58:517–25. Buchmann, S. L., and G. P. Nabhan. 1996. The Agosti, D., J. D. Majer, L. E. Alonso, and T. R. forgotten pollinators. Covelo, Calif.: Island Press. Schultz, eds. 2000. Ants: Standard methods 292 pp. for measuring and monitoring biodiversity. Wash- Cane, J. H. 2001. Habitat fragmentation and native ington, D.C.: Smithsonian Institution Press. bees: A premature verdict? Conservation Ecology 299 pp. 5(1):3. [On-line] URL: http://www.consecol.org/ Alonso, L. E. 2000. Ants as indicators of diversity. vol5/iss1/art3 In Ants: Standard methods for measuring and Cherrett, J. M. 1989. Leaf-cutting ants. In Eco- monitoring biodiversity, ed. D. Agosti, J. D. Majer, systems of the world. 14B. Tropical rain forest L. E. Alonso, and T. R. Schultz, 80–88. Wash- ecosystems, ed. H. Lieth and M. J. A. Werger, ington, D.C.: Smithsonian Institution Press. 473–88. Biogeographical and Ecological Studies. Andersen, A. N. 1986. Patterns of ant community Amsterdam: Elsevier. organization in mesic southeastern Australia. Christensen, N. L., A. M. Bartuska, J. H. Brown, Australian Journal of Ecology 11:87–97. S. Carpenter, C. D’Antonio, R. Francis, J. F. ———. 1990. The use of ant communities to eval- Franklin, J. A. MacMahon, R. F. Noss, D. J. Par- uate change in Australian terrestrial ecosystems: son, C. H. Petersen, M. G. Turner, and R. G. A review and a recipe. Proceedings of the Ecolog- Woodmansee. 1996. The report of the Ecolog- ical Society of Australia 16:347–57. ical Society of America Committee on the sci- ———. 1991. Parallels between ants and plants: entific basis for ecosystem management. Eco- Implications for community ecology. In Ant- logical Applications 6:665–91. plant interactions, ed. C. R. Huxley and D. F. Eickwort, G. C., R. W. Matthews, and J. Carpenter. Cutler, 539–58. Oxford: Oxford University Press. 1981. Observations on the nesting behavior of ———. 1997. Using ants as bioindicators: Multi- Megachile rubi and M. texana with a discussion scale issues in ant community ecology. Conser- of the significance of soil nesting in the evolu- vation Ecology 1(1):8. [On-line] URL: http://www. tion of Megachilid bees. Journal of the Kansas consecol.org/vol1/iss1/art8 Entomological Society 54:557–70. Banaszak, J. 1996. Ecological basis of conserva- Frankie, G. W., and S. B. Vinson. 2001. Diversity, tion of wild bees. In The conservation of bees, ed. structure, and variations in pollination systems A. Mattheson, S. L. Buchmann, C. O’Toole, of the seasonal dry forest of Costa Rica. In Trop- P. Westrich, and I. H. Williams, 55–62. Linnean ical ecosystems: Structure, diversity and human Society Symposium Series 18. London: Acade- welfare. Proceedings of the International Con- mic Press. ference on Tropical Ecosystems, ed. K. N. Gane- Bestelmeyer, B. T., and Wiens. 1996. The effects of shaiah, R. Uma Shaanker, and K. S. Bawa, land use on the structure of ground-foraging 202–3. New Delhi: Oxford-IBH. ant communities in the Argentine Chaco. Eco- Frankie, G. W., P. A. Opler, and K. S. Bawa. 1976. logical Applications 6:1225–40. Foraging behavior of solitary bees: Implications

76 VINSON ET AL. frankie chap 06 8/20/03 7:52 PM Page 77

for outcrossing of a neotropical forest tree Greenslade, P. J. M. 1985. Some effects of season species. Journal of Ecology 64:1049–57. and geographic aspect on ants (Hymenoptera: Frankie, G. W., S. B. Vinson, L. Newstrom, and Formicidae) in the Mt. Lofty Ranges, South J. F. Barthell. 1988. Nest site and habitat pref- Australia. Transactions of the Royal Society of erences of Centris bees in the Costa Rican dry South Australia 109:17–23. forest. Biotropica 20:301–10. Griswold, T., F. D. Parker, and P. E. Hanson. 2000. Frankie, G. W., S. B. Vinson, M. A. Rizzardi, T. L. An inventory of the bees of Costa Rica: The Griswold, S. O’Keefe, and R. R. Snelling. 1997. myth of the depauperate tropics. In Proceedings Diversity and abundance of bees visiting a mass of the Sixth International Conference on Apicul- flowering tree species in disturbed seasonal ture in Tropical Climates, 152–56. Cardiff, Wales: dry forest, Costa Rica. Journal of the Kansas En- International Bee Research Association. tomological Society 70:281–96. Hammond, P. M. 1994. Practical approaches to Frankie, G. W., R. W. Thorp, L. E. Newstrom- the estimation of the extent of biodiversity in Lloyd, M. A. Rizzardi, J. F. Barthell, T. L. speciose groups. Philosophical Transactions of Griswold, J.-Y. Kim, and S. Kappagoda. 1998. the Royal Society of London, ser. B, 345:119–36. Monitoring solitary bees in modified wildland Hanson, P. E., and I. D. Gauld. 1995. The Hy- habitats: Implications for bee ecology and menoptera of Costa Rica. Oxford: Oxford Uni- conservation. Environmental Entomology 27: versity Press. 914 pp. 1137–48. Hölldobler, B., and E. O. Wilson. 1990. The Ants. Frankie, G. W., S. B. Vinson, S. O’Keefe, and M. A. Cambridge: Belknap Press, Harvard University Rizzardi. 2000. Assessing impacts of African- Press. 732 pp. ized honey bees in Costa Rica and California: Huryn, V. M. B. 1997. Ecological impacts of intro- A progress report. In Proceedings of the Sixth In- duced honeybees. Quarterly Review of Biology ternational Conference on Apiculture in Tropical 72:275–97. Climates, 157–61. Cardiff, Wales: International Huxley, C. R., and D. F. Cutler, eds. 1991. Ant-plant Bee Research Association. interactions. Oxford: Oxford University Press. Frankie, G. W., S. B. Vinson, R. W. Thorp, M. A. 601 pp. Rizzardi, N. Zamora, and P. S. Ronchi. 2002. Janzen, D. H., P. J. DeVries, M. L. Higgins, and Coexistence of Africanized honey bees and na- L. S. Kimsey. 1982. Seasonal and site variation tive bees in the Costa Rican seasonal dry forest. in Costa Rican euglossine bees at chemical baits In Proceedings of the Second International Con- in lowland deciduous and evergreen forests. ference on Africanized Honeybees and Their Mites, Ecology 63:66–74. ed. E. H. Erickson et al., 327–39. Medina, Ohio: Jennersten, O. 1988. Pollination in Dianthus del- A. I. Root Co. toides (Caryophyllaceae): Effects of habitat frag- Franks, N. A. 1982a. Ecology and population regu- mentation on visitation and seed set. Conserva- lation in the army ant Eciton burchelli. In The tion Biology 2:359–66. ecology of a tropical forest: Seasonal rhythms and Kaspari, M., and J. D. Majer. 2000. Using ants to long-term changes, ed. E. G. Leigh, A. S. Rand, monitor environmental change. In Ants: Stan- and D. M. Windsor, 389–95. Washington, D.C.: dard methods for measuring and monitoring bio- Smithsonian Institution Press. diversity, ed. D. Agosti, J. D. Majer, L. E. Alonso, ———. 1982b. A new method for censussing ani- and T. R. Schultz, 89–98. Washington, D.C.: mal populations: The number of Eciton burchelli Smithsonian Institution Press. army ant colonies on Barro Colorado Island, Kearns, C. A., and D. W. Inouye. 1993. Techniques Panama. Oecologia 52:266–68. for pollination biologists. Niwot: University Press Gess, F. W., and S. K. Gess. 1993. Aculate wasps of Colorado. 583 pp. and bees in semi-arid areas of Southern Africa. ———. 1997. Pollinators, flowering plants, and In Hymenoptera and biodiversity, ed. J. LaSalle conservation biology. BioScience 47:279–307. and I. D. Gauld, 83–114. Wallingford, England: Kevan, P. G. 2001. Pollinators as sensitive bio- CAB International. indicators of environmental conditions: Eco- Gottsberger, G. 1986. Some pollination strategies logical and practical concerns. In Tropical eco- in neotropical savannas forests. Plant System- systems: Structure, diversity and human welfare. atics and Evolution 152:29–45. Proceedings of the International Conference Gotwald, W. H. 1995. Army ants: The biology of on Tropical Ecosystems, ed. K. N. Ganeshaiah, social predation. Ithaca, N.Y.: Cornell University R. Uma Shaanker, and K. S. Bawa, 198–99. Press. 302 pp. New Delhi: Oxford-IBH.

CONSERVATION VALUES OF BEES AND ANTS 77 frankie chap 06 8/20/03 7:52 PM Page 78

Kevan, P. G., C. F. Greco, and S. Belaoussoff. 1997. from Guanacaste, Costa Rica. Journal of the New Log-normality of biodiversity and abundance York Entomological Society 105:190–92. in diagnosis and measuring of ecosystemic Oliver, J., and A. J. Beattie. 1996. Designing a cost- health: Pesticide stress on pollinators on blue- effective invertebrate survey: A test of methods berry heaths. Journal of Applied Ecology 34: for rapid assessment of biodiversity. Ecological 1122–36. Applications 6:594–607. Kruess, A., and T. Tscharntke. 1994. Habitat frag- Passera, L. 1994. Characteristics of tramp species. mentation, species loss, and biological control. In Exotic ants: Biology, impact, and control of Science 264:1581–84. introduced species, ed. D. F. Williams, 23–43. San Lambeck, R. J. 1997. Focal species: A multi-species Francisco: Westview Press. umbrella for nature conservation. Conservation Pearson, D. L. 1994. Selecting indicator taxa for the Biology 11:849–56. quantitative assessment of biodiversity. Philo- LaSalle, J., and I. D. Gauld. 1993a. Hymenoptera: sophical Transactions of the Royal Society of Lon- Their diversity, and their impact on the diver- don B 345:75–79. sity of other organisms. In Hymenoptera and Pe˛tal, J. 1978. The role of ants in ecosystems. In biodiversity, ed. J. LaSalle and I. D. Gauld, 1–26. Production ecology of ants and termites, ed. M. V. Wallingford, England: CAB International. Brian, 293–325. Cambridge: Cambridge Univer- ———, eds. 1993b. Hymenoptera and biodiversity Press. sity. Wallingford, England: CAB International. Richardson, D. H. S. 1992. Pollution monitor- 348 pp. ing with lichens. Naturalists’ Handbooks 19. Lawton, J. H., D. E. Bignel, B. Boulton, G. F. Bloe- Slough, England: Richmond Publishing Co. mers, P. Eggleton, P. M. Hammond, M. Hodda, 92 pp. R. D. Holt, T. B. Larsen, N. A. Mawdsley, and Roubik, D. W. 1989. Ecology and natural history of N. E. Stork. 1998. Biodiversity inventories, in- tropical bees. New York: Cambridge University dicator taxa and effects of habitat modification Press. 514 pp. in tropical forest. Nature (London) 391:72–76. ———. 1996. African honeybees as exotic pollina- Loeb, S. L., and A. Spacie, eds. 1994. Biological tors in French Guiana. In The conservation of bees, monitoring of aquatic systems. Ann Arbor, Mich.: ed. A. Matheson, S. L. Buchmann, C. O’Toole, CRC Press. 381 pp. P. Westrich, and I. H. Williams, 73–182. Lon- Longino, J. T., and P. E. Hanson. 1995. The ants don: Academic Press. (Formicidae). In The Hymenoptera of Costa Rica, ———. 2001. Ups and downs in pollinator pop- ed. P. E. Hanson and I. D. Gauld, 588–620. Ox- ulations: When is there a decline? Conserva- ford: Oxford University Press. tion Ecology 5(1):2. [On-line] URL: http://www. Lubin, Y. D. 1984. Changes in the native fauna of consecol.org/vol5/iss1/art2 the Galapagos Islands following invasion by the Saks, M., and C. R. Carroll. 1980. Ant foraging little red fire ant, Wasmannia auropunctata. Bi- activity in tropical agro-ecosystems. Agro- ological Journal of the Linnean Society 21:229–42. Ecosystems 6:177–88. MacKay, W. P., and S. B. Vinson. 1989. A guide Samways, M. J. 1994. Insect conservation biology. to the species identification of New World London: Chapman and Hall. 358 pp. ants (Hymenoptera: Formicidae). Sociobiology Stebbins, G. L. 1979. Rare species as examples of 16:3–47. plant evolution. In Great Basin Naturalist Mem- Majer, J. D. 1983. Ants: Bio-indicators of minesite oirs, no. 3: The Endangered Species: A Symposium; rehabilitation, land-use, and land conservation. 7–8 Dec. 1978, 113–18. Provo, Utah: Brigham Environmental Management 7:375–83. Young University. Michener, C. D. 2000. The bees of the world. Balti- Stradling, D. J. 1991. An introduction to the fungus- more: Johns Hopkins University Press. 872 pp. growing ants, Attini. In Ant-plant interactions, Michener, C. D., and R. W. Brooks. 1984. Compar- ed. C. R. Huxley and D. F. Cutler, 5–18. Oxford: ative study of the glossae of bees. Contributions Oxford University Press. of the American Entomology Institute, Ann Arbor Strickler, K., V. L. Scott, and R. L. Fisher. 1996. 22:1–73. Comparative nesting ecology of two sympatric New, R. T. 1999. Limits to species focusing in in- leafcutting bees that differ in body size (Hy- sect conservation. Annals of the Entomological menoptera: Megachilidae). Journal of the Kansas Society of America 92:853–60. Entomological Society 69:26–44. O’Keefe, S. T., and D. Agosti. 1998. A new species Sugden, E. A., R. W. Thorp, and S. L. Buchmann. of Probolomyrmex (Hymenoptera: Formicidae) 1995. Honeybee–native bee competition in Aus-

78 VINSON ET AL. frankie chap 06 8/20/03 7:52 PM Page 79

tralia: Focal point for environmental change Costa Rica. Journal of the Kansas Entomological and apicultural response. Bee World 77:26–44. Society 60:249–63. Tennant, L. E. 1994. The ecology of Wasmannia Vinson, S. B., G. W. Frankie, and J. Barthell. 1993. auropunctata in primary tropical rainforest in Threats to the diversity of solitary bees in a Costa Rica and Panama. In Exotic ants: Biology, neotropical dry forest in Central America. In impact, and control of introduced species, ed. D. F. Hymenoptera and biodiversity, ed. J. LaSalle and Williams, 80–90. San Francisco: Westview I. D. Gauld, 53–81. Wallingford, England: CAB Press. International. Tepedino, V. J. 1979. The importance of bees and Vinson, S. B., H. J. Williams, G. W. Frankie, and other insect pollinators in maintaining floral G. Shrum. 1997. Floral lipid chemistry of Byr- species composition. In Great Basin Naturalist sonima crassifolia (Malpigheaceae) and a use of Memoirs, no. 3: The Endangered Species: A Sym- floral lipids by Centris bees (Hymenoptera: Ap- posium; 7–8 Dec. 1978, 39–150. Provo, Utah: idae). Biotropica 29:76–83. Brigham Young University. Wcislo, W. T. 1987. The roles of seasonality, host Thorp, R. W. 1979. Structural, behavioral and synchrony, and behaviour in the evolutions and physiological adaptations of bees for collecting distributions of nest parasites in Hymenoptera, pollen. Annals of the Missouri Botanical Garden with special reference to bees. Biological Reviews. 66:788–813. Cambridge Philosophical Society 62:515–43. Tscharntke, T., A. Gathmann, and I. Steffan- Williams, D. F., ed. 1994. Exotic ants: Biology, im- Dewenter. 1998. Bioindication using trap- pact, and control of introduced species. San Fran- nesting bees and wasps and their natural ene- cisco: Westview Press. 332 pp. mies: Community structure and interactions. Williams, N. M., R. L. Minckley, and F. A. Silvera. Journal of Applied Ecology 35:708–19. 2001. Variation in native bee faunas and its Unruh, T. R., and R. H. Messing. 1993. Inter- implications for detecting community changes. specific biodiversity in Hymenoptera: Implica- Conservation Ecology 5(1):7. [On-line] URL: http:// tions for conservation and biological control. In www.consecol.org/vol5/iss1/art7 Hymenoptera and biodiversity, ed. J. LaSalle and Williams, P. H., and K. J. Gaston. 1994. Measuring I. D. Gauld, 27–52. Wallingford, England: CAB more biodiversity: Can higher-taxon richness International. predict wholesale species richness? Biological Vinson, S. B. 1997. Invasion of the red imported Conservation 67:211–17. fire ant (Hymenoptera: Formicidae): Spread, Willis, E. O., and Y. Oniki. 1978. Birds and army biology and impact. American Entomologist 43: ants. Annual Review of Ecology and Systematics 23–39. 9:243–63. Vinson, S. B., and G. W. Frankie. 2000. Nest se- Wilson, E. O. 1990. Success and dominance in eco- lection, usurpation, and a function for the nest systems: The case of the social insects. Oldendorf/ entrance plug of Centris bicornuta (Hymenop- Luhe, Germany: Ecology Institute. 104 pp. tera: Apidae). Annals of the Entomological Society Zenner-Polania, I. 1994. Impact of Paratrechina of America 93:254–60. fulva on other ant species. In Exotic ants: Biol- Vinson, S. B., G. W. Frankie, and R. E. Coville. ogy, impact, and control of introduced species, ed. 1987. Nesting habits of Centris flavofasciata D. F. Williams, 121–32. San Francisco: West- (Hymenoptera: Apoidea: Anthophoridae) in view Press.

CONSERVATION VALUES OF BEES AND ANTS 79 frankie chap 07 8/20/03 7:53 PM Page 80

chapter 7

Ecology of Dry-Forest Wildland Insects in the Area de Conservación Guanacaste

Daniel H. Janzen

ropical dry forest (Murphy and Lugo ration of tropical dry forest is still possible, hu- T1986; Bullock et al. 1995) once occupied at manity will not give the globe back to its wild- least 60 percent of the forested tropics. Today, land denizens, and old-growth tropical dry forest it is largely eliminated (Janzen 1988a). Where will never again cover large areas. present, it is almost entirely in some complex The ecology of wildland tropical dry-forest state of incomplete and iterative secondary suc- insects is therefore largely that of populations cession (e.g., Janzen 1974a, 1986a,b, 1988b,c, and individuals persisting in successional forests 1990, 2002; Holl 1999; Holl and Kappelle 1999; on the agroscape or in conserved wildlands that Toh et al. 1999). The elimination of tropical dry are “being restored.” The latter are ecological is- forest is largely due to its ease of removal and lands that will forever be under the inﬂuence of perturbation by timber mining and human- human-generated biotic and climatological forces facilitated ﬁre and its comparative ease of oc- sweeping through the agroscape (e.g., Janzen cupation by humans engaged in agriculture 1983d, 1986b, 1988a–d), forces that the com- and their domesticated animals and plants, as paratively small patches of conserved wildlands compared with lowland tropical rain forest and cannot escape. As the few conserved wildlands deserts (Janzen 1988a). Contemporary anthro- in tropical dry forests are gradually restored over pogenic drying and heating trends are accentu- the centuries into whatever ecosystems are able ating the process in dry forest and also render- to persist on their sites, future generations may ing former cloud forest and rain forest habitats be able to view a somewhat more “old-growth” more “dry forest-like” (e.g., Cochrane et al. 1999; version of what once existed. The larger the con- Goldammer 1999; Nepstad et al. 1999; Pounds served area and the less perturbed it was before et al. 1999; Still et al. 1999). Although the resto- restoration began, the closer it will be to the

80 frankie chap 07 8/20/03 7:53 PM Page 81

original. But it will still be substantially distorted bacteria. But if the island is small enough and from the prehuman state, owing to permanently the selection intense enough, the rate of the evo- altered circumstances for insects as well as for lutionary adjustment process may even begin to all the other organisms and their ecosystems. approximate that of the ecological adjustment Ironically, this permanently changed state process (e.g., Agrawal et al. 1999; Haukioja may not be as significant as an academic gener- 1999). Permanent habitat fragmentation by ation of evolutionary ecologists and conserva- humanity has the potential to generate specia- tionists have wanted to believe. At least on con- tion and impoverishment on a masssive scale, as tinental mainlands, the great majority of species though the world were abruptly converted into in any specific contemporary habitat did not thousands of archipelagoes rich in small islands. evolve in that particular place with that habi- Insect population structures and evolutionary tat’s particular interactants (e.g., Janzen 1986d); dynamics are such that they should be strongly rather, they evolved within some ecological is- affected by such a restructuring of the world. land or under other isolational circumstances In the past three decades, the scientific com- elsewhere and then immigrated to their place munity has learned much more about the insect and current habitat. ecology of tropical dry forest. However, these A wildland habitat is a collection of species new insights do not add substantially to what that persisted on their arrival and thus colonized is already generally understood about insects the habitat, and their interactions from that point and their interactions with the biotic and abiotic on are largely a product of the combination of world. Rather, they reveal that the insect ecology the traits they had on arrival and which species of tropical dry forest is a matter of proportional- were present when they arrived (Janzen 1985a, ity with respect to the multitude of ways that sets 1986d). When humans make massive and wide- of species do things. And if these years of peer- spread changes in a habitat, it becomes a new ing intently at one kind of tropical dry forest— habitat or ecosystem. In effect, every species is the lowland Mesoamerican coastal plain from yet again an immigrant by virtue of being thrust Veracruz, Mexico, to Pacific Panama—while into new ecological circumstances and biological glancing briefly at insects in other tropical sites relationships. Almost all the tropical dry forest (e.g., Janzen 1972, 1974b,c, 1976a, 1988d) have and its denizens studied today are in this state. taught me anything, it is that one needs to focus As a result, some species disappear, and others on one place in order to understand this pro- become more common, but those that remain portionality. Place-based ecological understand- adjust to a new habitat ecologically and largely ing is, however, very often subordinated in the with the traits they had before changes were in- pursuit of broadly applicable scientific principles. troduced, just as each of these species adjusted However, a local vision is essential to under- in the past as it spread from its place of origin. standing how and why organisms do what they In designating certain areas “conserved wild- do. Furthermore, place-based understanding lands,” humans are, in effect, creating ecologi- serves as a critical informational infrastructure cal island ecosystems. The species that eventu- for the very pragmatic process of integrating any ally populate these islands are those that are specific conserved wildland with its resident, inherently able to survive in the new habitat; for national, and international society, an integra- some species, however, the isolation of the con- tion that must sustain and develop with the conserved wildlands produces the circumstances served wildland if it is to be permanent (e.g., for further evolution. The new forms will adjust Janzen 1998a,b, 1999a,b, 2000a,b, 2001a,b). evolutionarily to the conserved wildland as well What follows is a brief comment on a few as fit into it ecologically. The time frame of this aspects of the insect ecology of the tropical dry process clearly varies among species: the pro- forest in the Area de Conservación Guanacaste cess affects jaguars differently than it affects (ACG) in northwestern Costa Rica. I focus on

ECOLOGY OF WILDLAND INSECTS 81 frankie chap 07 8/20/03 7:53 PM Page 82

MAP 7.1. Area de Conservación Guanacaste, northwestern Costa Rica.

the physical setting, the beginning of the rainy THE PHYSICAL SETTING season, and seasonal movements, and I largely refrain from redescribing what I have already The 153,000-ha ACG (see http://www.acguana- published on ACG dry-forest insect ecology per caste.ac.cr) (map 7.1) begins out in the Paciﬁc se (e.g., Janzen 1967a,b, 1973a,b, 1976b, 1978, Ocean, and its 60,000 ha of dry forests extend 1979, 1980, 1981a, 1982a–d, 1983a–c, 1984a–c, inland eastward across 15–40 km of lowland 1986a–d, 1987a–c, 1988e,f, 1989, 1993; Janzen coastal plain, mesas, and hills to the western and Schoener 1968; Janzen et al. 1980a; Janzen lower slopes of Volcán Orosí, Volcán Cacao, and et al. 1982; Janzen and Waterman 1984; Sharkey Volcán Rincón de la Vieja. Here, these extremely and Janzen 1995; Woodley and Janzen 1995; seasonal forests merge with the midelevation Janzen and Gauld 1997; Miller et al. 1997; Jan- wetter forests leading into upper-elevation cloud zen et al. 1998; Hunt et al. 1999). This is not forest (1,000–2,000 m) and thence down into meant to be a review of these topics. the Caribbean rain forest (400–600 m) in the

82 JANZEN frankie chap 07 8/20/03 7:53 PM Page 83

northern and eastern portions of the ACG. The The dry season is relatively cloud-free; the rainy ACG is bounded on the north, south, and east season alternates between heavy cloud cover and by agroscape. This 85-km-long conserved tran- frequent clear mornings. This meteorological sect contains the three widespread Neotropical pattern is also affected by strong trade winds vegetation types—dry forest, cloud forest, and from the east-northeast for the first half of the rain forest—and many intergrades (it contains dry season, bringing cooler air from the volca- nine Holdridge Life Zones and a plethora of noes to the northeast. Hurricanes have generally transitions). The underlying physical substrate lost much of their windy force, but not their rain, ranges from 85-million-year-old serpentine bar- by the time they arrive from the Caribbean and rens to marine limestone and volcanic deposits the Pacific. from tens of millions of years to only a few The dry-forest ecosystem of the ACG was days old. once continuous, stretching all the way north to The ACG dry-forest ecosystem contained be- the foothills of the mountains west of Mazatlán, tween the volcanic slope evergreen forests and Mexico (e.g., Martin et al. 1998), and south at the coastal mangroves is meteorologically char- least to the Canal Zone, broken by rain forest in acterized by five to seven continuously rain-free the Golfo Dulce region of southern Pacific Costa months (roughly December to mid-May), a total Rica and Pacific coastal Guatemala, a desert break rainfall of about 800–2,800 mm (contempo- at the Balsas River Basin in western Mexico, and rary 15-year average of about 1,500 mm), and a dry-forest extensions to the coastal lowlands of short (zero- to six-week) dry season (the Spanish- the Gulf of Mexico (Tampico to Yucatán and language veranillo, July–August) in the middle eastern Guatemala). of the rainy season. The ACG dry-forest meteor- Some physical characteristics of the ACG that ology is that of a rain shadow in the western lee are key for insect ecology are that of the volcanoes, affected by the rainy season brought by the twice-annual passage of the ther- . its dry forest lies within a few tens of kilo- mal equator (the ACG is about 10° N latitude). meters of both (cooler) high-elevation cloud The heating from the thermal equator pulls in forests and (wetter) low-elevation rain forests moist air from the Pacific, air that rises and and its margins intergrade with these other cools/condenses to create afternoon and evening ecosystems; thunderstorms. This moist air is occasionally . the dry season is severe enough to create overlain by masses of moist air pushed in from deciduous and semideciduous forest types hurricanes over the Caribbean, mostly in the coupled with very strong seasonal phenology rainy season. Nocturnal temperatures in this dry in plant growth and reproduction; forest generally range between 18°C and 22°C, . the internal rain-fed aquatic systems are se- with the lowest values on cloudless nights in the verely dried out by the long (and hot) rain-free dry season. The diurnal temperatures range be- season but the adjacent volcanoes create tween 26°C and 36°C, the hottest days occurring ever-flowing rivers that cross some of the dry during the relatively wind-free last two months forest as linear seasonal oases on their way of the dry season and the coldest days during to both the Atlantic and Pacific; the second half of the rainy season (October– November), when heavy rains produced by hur- . the diverse ages and chemistries of the ACG ricanes affect the area. The most abrupt tem- rock and soil substrates, combined with com- perature change is the sudden cooling that plex meteorological conditions, sustain a comes with the first rains at the end of the long complex array of plant species, phenologies, dry season (fig. 7.1) (Janzen 1984a, 1993; and see and life forms as food and shelter for insects; detailed weather records at http://www.acguana- . all fires in this forest are caused by humans caste.ac.cr/1999/indice_bases_datos/index.htm). and are therefore not part of the “natural”

ECOLOGY OF WILDLAND INSECTS 83 frankie chap 07 8/20/03 7:53 PM Page 84

FIGURE 7.1. The relationship between maximum daily temperature and rainfall during two consecutive years at the weather station in the administration area, Sector Santa Rosa, Area de Conservación Guanacaste.

events that the insect community has expe- been thoroughly cleaned of its vegetation rienced in the dry season prior to human and hence much of its insect fauna. entry; and, . despite four centuries of European-style The ACG old-growth forest remnants range agriculture and ranching and millennia of from several thousand hectares of dwarf forest indigenous use, the relatively low-quality on serpentine on the western tip of the Santa soils and long sociopolitical distances from Elena Peninsula to patches of a few tens of - Costa Rican and Nicaraguan social centers hectares on volcanic soils; the rest of the rem- have protected the general area from having nants are highly complex mosaics of secondary

84 JANZEN frankie chap 07 8/20/03 7:53 PM Page 85

successional stages that did not begin the long correlate exactly with the pulse of phytophagous transformation to old-growth status until the insects in May. Many dormant pupae are some- fires were eliminated or controlled in the mid- how able to perceive the oncoming rains, and 1980s. they initiate pupal development to the adult form, or even actually eclose, before the days of the first rains. When there is a single cohort of sib WHAT HAPPENS AT THE BEGINNING pupae, even sequestered in the monotonous OF THE RAINY SEASON ambient-air temperature of an ACG laboratory, The beginning of the rainy season, between late the rainy season initiation does not bring them April and late May, is the most conspicuous to eclose on a single day. Rather, their eclosion is aspect of ACG dry-forest insect ecology. The top- spread over one to several weeks, a phenomenon ics discussed later in this chapter—migration/ that creates overlapping distributions of abun- movements, seasonal effects, dry/wet contrasts— dance (e.g., Janzen 1984a, 1993). all relate in some way to this meteorological All this phenology varies within the year and event and its cascading biological consequences. between years. Even within a group of taxonom- Humans have long marked the beginning of the ically and ecologically related species, such as ACG rainy season entomologically by the masses the Saturniidae, each species perceives and re- of adult moths and beetles congregating at acts to the beginning of the rainy season some- lights, the conspicuous defoliation by caterpil- what differently, creating noncongruent peaks lars and beetles, and the nesting of insectivorous of adult species densities, followed by yet more birds. Other indications are the huge numbers variation in the peaks of caterpillar densities of Photinus fireflies on the first dead-calm nights, (Janzen 1993). In addition, each of the species’ the inescapable mating swarms of termites cued phenologies is affected somewhat differently by by the first rains, and, more poetically, the sen- gradients in intensity and the timing of the on- sation of a world “crawling with insects.” set of the rains, causing each year’s “beginning The massive “appearance” of large numbers of the rainy season” to be somewhat different of species and individuals of adult phytopha- from that of other years (fig. 7.1; see also Braby gous insects in the few days before and the 1995; Nijhout 1999; Roskam and Brakefield month following the first heavy rains is in fact 1999). In some species with short-lived adults, an extremely complex phenomenon. Some ar- there are even color morphs that match the spe- rive by migration from distant cloud forests cific background colors characteristic of a few and rain forests (see the section “Migration and days before the rains begin and a few days after, Other Movements” later in this chapter), others respectively. For example, Rothschildia lebeau are simply now-active adults that have been hid- eclosing during the weeks before the rains are ing throughout the dry season in more moist a rust red (which matches well the dry fallen habitats and microhabitats within the dry forest, leaves in the sun), whereas their sibs eclosing and still others are newly eclosed from dormant during the weeks after the rains are dark choco- prepupae and pupae in cocoons or the soil/litter late brown (which matches well the moist moldy (e.g., Janzen 1983a, 1987a,c, 1988e,f, 1993; see fallen leaves in the shade of new leaf crops) (e.g., also chapter 8). Strikingly, almost none of the re- Janzen 1984a). sultant juveniles, except for those of Orthoptera, Many univoltine species remain hidden as are from eggs that passed the dry season in a dormant prepupae and pupae from the second dormant state (see Janzen 1984b for the excep- month of the rainy season through the remain- tion by a saturniid moth, Hylesia lineata). ing four wet (and very leafy) months and the fol- The “beginning of the rains,” as indicated by lowing six-month dry season (see nearly year-long the increased amount of rainfall and the con- prepupal and pupal duration records in Janzen comitant drop in temperature (fig. 7.1), does not and Hallwachs 2003 for select Ichneumonidae,

ECOLOGY OF WILDLAND INSECTS 85 frankie chap 07 8/20/03 7:53 PM Page 86

Braconidae, Tachinidae, Saturniidae, Sphingidae, tions in the rainy season while some members Noctuidae, Geometridae, Notodontidae, Pyra- of the first generation remain dormant as pre- lidae, Lymantriidae, and Papilionidae). These pupae or pupae, others are active nonrepro- species are displaying a biology structured much ductive adults, and yet others even migrate (e.g., like that of classic univoltine extratropical and Xylophanes lefeburii and X. juanita in the Sphin- desert insects. By remaining dormant during gidae; Epargyreus in the Hesperiidae) to rain for- this part of the rainy season, they are apparently est. Other species are reproductively active in avoiding what would otherwise be a carnivore- the rainy season and either become dormant rich season generated by their own cadavers in the long dry season or leave (or are locally ex- (Janzen 1987c, 1988e, 1993). Their lack of on- tinguished). Some species are extremely syn- site reproduction is not due to more unpalatable chronized within the population, having virtu- or toxic foliage (Janzen and Waterman 1984). ally no overlap of rainy season generations, They do not attempt to reproduce during the whereas others have natural histories such that long dry season because of the physical chal- sometimes by the second, and definitely by the lenges and the scarcity of foliage. Most species fourth, generation virtually all life stages may be that pass this long dormancy as prepupae do found in the habitat at the same time (though so in extremely tough and desiccation-proof not in equal proportions). In general, the multi- cocoons (often underground, as well), and those voltine species with nonfeeding adults (e.g., that are dormant as pupae (in the litter or co- bombycoid Lepidoptera) maintain better syn- coons) have notably tough, thick, and presumably chrony of broods in the rainy season than do desiccation-resistant pupal cuticles. Some ex- those with feeding adults. It is particularily com- amples of the former are Thyreodon and Eni- monplace for Hesperiidae (skipper butterflies) cospilus (Ichneumonidae), Microplitis (Bracon- to have broadly overlapping generations during idae), Sylepta belialis (Pyralidae), Ethmia spp. the rainy season and no or few caterpillars in the (Ethmiidae), and many Limacodidae and Mega- dry season, leaving the parasitoids of the cater- lopygidae. Examples with notably tough pupal pillars to survive the dry season as equally re- cuticles are Xylophanes turbata, Manduca lefeburii productively inactive adults (e.g., Janzen et al. and M. dilucida (Sphingidae), Schausiella santa- 1998). The ease of finding hesperiid caterpillars rosensis (Saturniidae), Goacampa variabilis (Noto- in their conspicuous resting shelters greatly fa- dontidae), Holochroa (Geometridae), Euscirrhop- cilitates the tracking of this seasonal demogra- terus poeyi (Noctuidae), Eurytides epidaus and phy and interaction. E. philolaus (Papilionidae), and Metavoria (Ta- Many species that are full-time residents of chinidae). The univoltine (or facultatively bi- dry forest have only one generation of insects voltine) species that pass most or all the inter- when the rains begin, and they then live as non- vening months as reproductively inactive adults reproductive adults for most of the year; this trait (e.g., Eulepidotis, Melipotis, Letis, Coenipita, and creates havoc with classic attempts to map the many other Noctuidae) are particularly confus- phenology of their natural history by ordinary ing because collectors can find them in the collecting processes. The huge number of moths daytime or at night in almost any season (they at lights during the first week following the first may appear at lights on virtually any night of the rain, for example, has long been intuitively in- year) and therefore label them incorrectly as terpreted as “emergence following the rains” “multivoltine.” when in fact it is initially made up mostly of a Insect species of tropical dry forest have combination of incoming migrants and newly evolved a plethora of forms of multivoltinism. active adults that have been present for 6–11 In some species, phytophagous and carnivorous months but hiding in the foliage and in locally species alike, there are second to fourth genera- moist areas. The groups of moths at lights for

86 JANZEN frankie chap 07 8/20/03 7:53 PM Page 87

the next several weeks are made up largely of create food phenology, but the different reactions individuals newly eclosed from dormant pre- to the rains by various species that are food for pupae or pupae. others affect food availability. For a bird that During the first three months of the ACG brings only sphinx and saturniid moth caterpil- rainy season in dry forest, then, there is a mas- lars to its nestlings (Janzen and Joyce unpubl. sive pulse of reproductive activity by phyto- data), a year in which there are tons of noctuid phagous insects and their carnivores, followed caterpillars per hectare may be a year of nestling by an extremely variable interannual second, starvation. and greatly reduced, “breeding period” during Being large diurnal mammals, most humans the second half of the rainy season. This is fol- do not appreciate the microgeographical varia- lowed by a near absence of breeding insect pop- tion in tropical dry forest that is largely gener- ulations during the long dry season, except for a ated by the subtleties of differential drying of few examples (e.g., bees, bruchid beetles) dis- slopes, soil types, and wind exposures. However, cussed later in this chapter. Understanding the insects are very sensitive to these place-based phenologies of decomposers (e.g., Janzen 1983b), variations in their food, heat, and moisture. The pollinator systems (e.g., Janzen 1967b; Janzen et anticipation of the rainy season by adult insects al. 1980b, 1982), food for breeding birds (Janzen that are newly active, newly eclosed, and in- and Joyce, unpubl. data), and other major insect migrating begins near the ACG coast and spreads interactions needs to take this entomopheno- inland to the base of the volcanoes over a several- logical process into account. These systems all week period. A riparian tree that leafs out a week reinforce and detract from one another on both before its conspecifics on dry ridge soil may re- ecological and evolutionary time scales. ceive the great bulk of the egg load of the popu- Although there is a great burst of new foliage lation of a univoltine noctuid that has its only at the beginning of the rainy season, individual generation on new leaves “at the beginning of plants may refoliate as much as a month before the rains.” The plants on a dry ridge that become to a month after, depending on the species of totally deciduous in the long dry season may plant and even the specific age/ecological cir- have quite different herbivore loads than those cumstances of the individual (e.g., young plants of a slightly more moist site that contains a few often leaf out earlier than do adults, plants on evergreen (shady tree) sites ideal for estivation ridges often have quite different leaf phenol- (Janzen 1973a, 1976b). The cooler, shady under- ogies than do conspecifics in valley bottoms). story of more evergreen riparian vegetation Since many species of dry-forest herbivores are (e.g., Janzen 1976c) is an excellent habitat for host-specific to one or a few species of plants insects during the dry season but a poor one and often are even leaf-age specific (e.g., Janzen during the rainy season (owing to the lack of 1979, 1984b,c, 1985b; Janzen and Hallwachs photosynthesis), while a ridge a few meters away 2003), the interspecific phenological differences provides the opposite circumstance. are important and may be reflected in the details of abundance of a given insect species within MIGRATION AND OTHER MOVEMENTS and between years. The great bulk of the carnivores in the ACG Three types of migration occur in the ACG. The dry forest are also quite specific in their needs: first consists of extremely complex local move- the species of prey/hosts, the ecological circum- ments within the dry forest. Local seasonal move- stances in which they occur, or both (e.g., Janzen ments by dry-forest insects between sun and and Pond 1975; Janzen 1981b; Gauld et al. 1992; shade, moist and dry areas, warm and cool areas, Janzen and Gauld 1997; Janzen et al. 1998). Not and food-abundant and food-poor areas are only does the general beginning of the rains anticipated and obvious. The ACG is somewhat

ECOLOGY OF WILDLAND INSECTS 87 frankie chap 07 8/20/03 7:53 PM Page 88

atypical as a dry forest in this context because ture probably gave rise to classic monarch but- in addition to its seasonal rain-fed temporary terfly migration between highland Mexico and watercourses and pools, it has rivers flowing the United States, and the local resident ACG through portions of it from the cloud forest vol- population of monarch butterflies is one of the canoes to both oceans (northern ACG rivers flow rolling seasonal invaders of the ACG dry forest. into Lake Nicaragua and thence the Atlantic). The third type of migration is the movement These rivers offer more moist and more perma- of many species of ACG insects between the nent dry-season refugia (and more shady photo- ACG dry forest and more distant (but national) synthate-free habitats) than do the temporary ecosystems (also see chapter 8). (The appear- watercourses and also are wet corridors that ance of adults from dormant prepupae and pu- connect many kinds of dry forest. pae a few millimeters to centimeters below the However, in addition to the local seasonal soil is a very short distance variation on this ebb and flow of insect populations, there are theme.) There are at least two types of these two striking and generally unappreciated other migrants. local insect movements. First, and as yet poorly One type appears to migrate out to another documented, is what may be termed a “rolling ecosystem to be dormant there, to wait for the seasonal invasion” from the more evergreen next year’s rains. Social wasps such as Polistes and wetter forests to the east. These are insects variabilis (Vespidae) illustrate this well (Hunt et of the rain forest and wetter intergrade forests al. 1999), as do many other carnivores, such as whose resident populations fly into the adjacent tachinid flies, parasitic wasps, and dragonflies. newly green dry forest at the beginning of the In the last half of the rainy season, most but not rainy season and produce their first-generation all P. variabilis wasps abandon their small nests larvae. The second generation of these species in the dry forest in the ACG lowlands (in which then moves farther into the dry forest (toward caterpillar prey is sparse) and fly up into the cold the Pacific Ocean). The third and fourth genera- cloud forests on the volcano tops. There several tions, if any, move yet farther. By the end of a thousands (mostly female) of dormant individ- long and wet rainy season they may have in- uals sequester themselves in masses in hollow vaded the entire ACG dry forest wherever their trees. When the rains come, they descend to host plants (or prey) occur. The long dry season the dry-forest lowlands and establish many new then results in the local extinction of all (an un- (small) nests scattered throughout the dry for- known number of individuals retreat all the est. The relatedness of the individuals in these way back to the volcanoes). If there is a run of new nests is unknown. A skipper butterfly, Aguna relatively dry rainy seasons, many of these species asander (Hesperiidae), whose caterpillars feed are not encountered as “dry-forest species.” How- on Bauhinia ungulata (Fabaceae) leaves in the ever, if there is a run of relatively wet (and, in rainy season in the dry-forest lowlands, does the particular, long) rainy seasons, these species ap- same but roosts individually in the cold cloud pear in visiting scientists’ inventories as normal forest understory. dry-forest species. The much more conspicuous kind of distant The relevance of this phenomenon depends migration is displayed by the many species of on the proximity of dry forest to wetter forests, butterflies and moths that arrive with the first but the entire ACG dry forest is well within its rains, have one generation on ACG dry foliage, range. This phenomenon has its obvious paral- eclose from their pupae within several weeks, lels in the annual variable degree of summer and leave to have other generations in rain for- invasion of southern U.S. species into more est (in a small set of species, some individuals northern latitudes and invasion of arid Mexico remain and generate a second or third genera- from the Gulf Coast (e.g., Powell and Brown tion, which in turn leave at the end of the rainy 1990). Interestingly, a phenomenon of this na- season) (Janzen 1987c, 1988e, 1993; Powell and

88 JANZEN frankie chap 07 8/20/03 7:53 PM Page 89

Brown 1990; chapter 8). Many of their para- flies to evolve such that they do not even have a sitoids leave with them and may well have other generation on the rain forest side of Costa Rica. generations in the rain forest ecosystems to However, their relatively low capacity for full- which the adults migrate. It is likely that only year life spans as nonreproductive adult insects part of the adults from the rain forest genera- suggests that the evolution of univoltinism tions return to the dry forest at the beginning of would require the evolution of pupal dormancy the next rainy season. The numbers of individ- as displayed by some dry-forest nonmigrants. ual migrants of a given species appearing each It is unknown in which species of migrants year in the ACG dry forest vary dramatically the entire population moves back to the dry from virtually none to so many that the caterpil- forest and in which species only a portion of lars thoroughly defoliate their food plants (e.g., the population returns. In the ACG, each year Janzen 1985b, 1988f). Many but not all of these of caterpillar inventory (http://janzen.sas.upenn. species have caterpillars that specialize on leaves edu) has located additional species of Lepi- only a few days to weeks old, and virtually all are doptera with a very low density of caterpillars on species that feed as adults. the rain forest side throughout the year and a The Lepidoptera species that participate in large number of caterpillars in the first half of the east-west (wet-dry) seasonal migration are the rainy season in dry forest (e.g., Manduca species with very large geographical distributions occulta, M. sexta, Xylophanes pluto, X. chiron, Un- (from tropical Mexico to lower tropical South zela japix, Enyo ocypete, Pachylioides resumens, America). The migrating moth species can oc- Pachylia ficus, Neococytius cluentis, Cocytius an- cur in huge numbers at lights in passes at 600 teus, Eupyrrhoglossum sagra, Aellopos titan, all in to 2500 m elevation in Costa Rica’s mountain the Sphingidae; many species of Noctuidae, ranges that separate the rain forests from the dry Nymphalidae, Hesperiidae, and Papilionidae). forests (and see Powell and Brown 1990). It is Equally, every season of rearing parasitoids in very striking that there is roughly a 50:50 sex the rain forest locates more “dry forest” para- ratio in these migrants arriving at lights (pre- sitoids that have a large generation in the first sumably used for orientation), in contrast to the half of the rainy season but are also breeding at extremely heavy bias in favor of males arriving other times of the year in the lowland rain forest at lights in the lowland breeding areas (where on the eastern (Atlantic) side of the ACG (e.g., the light is used as part of the males’ orientation many species of Belvosia, Drino, Blepharipa, Hy- during search for females; Janzen 1983a). phantrophaga, Siphosturmia, Chetogena, Patelloa, Since the adults seem incapable of surviving and Atacta in the Tachinidae). the dry season in the dry forest (and the imma- It is likely that other insects are doing the tures by and large do not remain dormant dur- same thing as the Lepidoptera and their para- ing this time), they might be best thought of as sitoids, but to date the only truly spectacular ex- rain forest species that seasonally perform long- ample is the millions of dragonflies (Odonata) distance invasion of the dry forest during the that move from the ACG dry forest to the At- first half of its relatively carnivore-free rainy sea- lantic rain forests in August and January, using son. Later, rather than stay for another genera- the same low-elevation passes that the migrant tion in the second half of the rainy season, they moths and butterflies use. Their return, if they flee from the (rapidly burgeoning) carnivore com- return, is not conspicuous. munity, generated by the larvae of their genera- When one considers the dry season’s violent tion and the larvae of residents (Janzen 1987c). winds, dry soil and air, burning sun, leafless In this sense they are analogous to the many trees, lack of breeding birds and insects, and birds that migrate into extratropical latitudes to other seasonal “difficulties,” there is a strong have a single generation and then leave. It would tendency to label the rainy season as “good” for be only a short step for these moths and butter- insects and the long dry season as “bad” (thereby

ECOLOGY OF WILDLAND INSECTS 89 frankie chap 07 8/20/03 7:53 PM Page 90

equating the dry season with the northern win- aquatic beetles and bugs that will fly out across ter). However, as with the other dry-forest life the newly wet landscape six months later in forms of organisms, it all depends on natural search of temporary swamps rich in decompos- history. If food is present in the dry season, then ing vegetation and associated microbiota. The that may well be the good time of year. end of the short period of dry weather that oc- For Liomys salvini hispid pocket mice (Het- curs during the middle of the rainy season can eromyidae), which are major predators on ACG deceive pupae into behaving as if it is the end of dry-forest seeds and moth pupae in the litter the long dry season, and consequently the newly (Janzen 1986e), the rainy season becomes pro- eclosed adults (and their offspring) are thrust gressively more food-impoverished as the seeds into the carnivore-rich second half of the rainy germinate and the first-generation crop of moth season. The first half of the rainy season is the pupae ecloses or is harvested. Seeds for food time when densities of insect-eating carnivores are not abundant again until the following dry are at their annual minimum because of the rel- season. For bruchid beetles (Bruchidae), which ative absence of insects as food during the long have their one or two generations on these same dry season, and in the second half of the rainy seeds in the dry season (Janzen 1975, 1980), the season they should be at their maximum (unless last half of the dry season and the entire rainy they migrate). season are periods of population decline, and Newly fallen crops of dry-season fruits, rich the reproductively dormant adults hide in rolled in proteins and sugars for (extinct) large verte- leaves to avoid carnivores. Seed-eating weevils brates (Janzen and Martin 1982), offer a non- (Curculionidae) have to wait one and sometimes rotting food for generations of pyralid moth lar- two years as hidden and reproductively dormant vae (e.g., Anypsipyla univitella), which are then adults between (dry-season) fruit crops (e.g., deprived of these foods by fungi with the onset Janzen 1974a, 1976d, 1978, 1982c,d, 1989). For of the first rains (e.g., Janzen 1982a, 1983c). The many species of bees and other flower visitors, distribution of ant-lion nests in the dry season some portion of the dry season is the reproduc- (everywhere) and in the rainy season (only un- tive time of year (e.g., Frankie et al. 1997), and, der rain-free overhangs) is illustrative. The life- as for bruchid beetles, the rainy season is some- giving rain that falls in the forest canopy also thing to be tolerated in some form of dormancy. distributes lethal viruses, fungi, and bacteria This is even reflected in the beekeepers’ general among the leaves eaten by caterpillars. Many need to feed domestic Apis mellifera bees during caterpillars may eat the foliage at the very be- the first half of the rainy season in Guanacaste ginning of leaf flush because of its microbio- Province, the same bees from which a honey logical cleanliness as much as for some specific surplus can be harvested in the dry season. Even nutrient content or lack of internal chemical leaf-cutter ants (Atta) may actually be harvesting defenses. higher-quality food in the dry season (e.g., Wirth In contrast to what many have believed or et al. 1997), though the dry-season air is too dry wanted to believe, however, once the foliage has to permit foraging in the daytime. become several weeks old, it is basically the Each season brings some very specialized same food throughout the rainy season (Janzen kinds of habitats. The strong winds of the first et al. 1980a; Janzen and Waterman 1984). This two months of the dry season break living is even more so for the leaves of the evergreen branches, and the exposed wounds are ideal species sprinkled through the dry forest (e.g., oviposition sites for cerambycid beetles and Cos- Hymenaea courbaril, Swietenia macrophylla, sidae moths, reflected in a peak in numbers of Manilkara chicle, Swartzia cubensis, Inga vera, adult species and individuals at the lights in Jan- Hirtella racemosa, Ocotea veraguensis, Ouratea lu- uary. Drying streams have small puddles filled cens, Sloanea terniflora, Andira inermis, Licania with apparently reproductively dormant adult arborea).

90 JANZEN frankie chap 07 8/20/03 7:53 PM Page 91

What is “good” habitat for an insect also of being a haven for both dry-forest migrants shifts dramatically in microlocation with the and wet-forest elevational migrants. seasons. For example, in the long dry season, Second, because the ACG will always be an the cool and shady deep forest understory in ri- ecological island surrounded by a large and parian habitats and under individual evergreen complex agroscape, it will always be inﬂuenced tree species (and in relatively evergreen old- by distinctive edge effects generated by that growth forest as opposed to more deciduous agroscape, such as local climate change, inva- secondary succession) is a highly favored place sive biological control agents, and introduced for those species of insects that pass the dry sea- species (e.g., Janzen 1986b; Estrada et al. 1998; son as adults in reproductive dormancy. When Williams-Linera et al. 1998). Converting some the rains come, the photosynthate-poor, heavily portion of this neighboring agroscape to various shaded forest understory is quickly vacated in kinds of permanent secondary succession, such search of insolated canopies at all levels. Like- as semiwild timber plantations or hunting ar- wise, the friendly moist soil and litter of riparian eas, will offer suitable habitat for many sec- sites in the dry season may become an anaero- ondary successional species at high densities. bic swamp in the full rainy season. However, these in turn create a huge biotic edge effect penetrating kilometers into the (regenerating) old-growth forest; the secondary succes- CONCLUSION sional species invade the old growth at a density The ACG’s insect biodiversity, as is the case that would never be generated by the old growth with other taxa, has been affected by humanity itself (Janzen 1983d). Ironically, it may well be in four irreversible ways. Its actions continue to better for the ACG to be surrounded by a pol- have an impact even though the ACG is now be- ished and economically viable agroscape than by ing allowed to return to old-growth status over a so-called buffer zone of permanent secondary its entire 110,000 terrestrial hectares of dry for- succession and economically disadvantaged sub- est, rain forest, cloud forest, and intergrades. sistence agrarians. First, the ACG will always be an ecological is- Third, no matter how high the quality at- land of this size, with the attendant characteris- tained by the ACG’s regenerating old-growth tics of an island that was sliced off a mainland dry forest, it will never again receive the large (as opposed to an oceanic island whose biota ar- number of migrants that it once received from rives only by long-distance immigration and the huge area of lowland rain forest to the east, on-site evolution). It is unclear as to what equi- and hence it will also never generate the large librium level its species density and community numbers of migrants that it once did. There is structure will move. The dry-forest side of the no way to know which of these two processes ACG appears to be large enough to sustain a ma- will be more strongly affected. It will depend a jor portion of its original dry-forest biodiversity, great deal on which species are affected most though the destruction of complementary wetter severely by converting old-growth rain forest to areas outside the ACG—homes for seasonal patchy secondary succession. Similarly, the dry migrants—will take its toll. The volcano cloud forest can never again generate and receive the forest ecosystems are currently about the same same numbers of migrants to and from the cloud size they originally were, but climate change is forest (even if climate change is reversed and the shrinking them (Pounds et al. 1999). The rain cloud forests regain their territory). forest in the eastern ACG, now just a small frag- Fourth, the oncoming climate changes (dry- ment, was once thousands of times larger in an ing and warming for the ACG) will create mas- eastern direction. This rain forest fragment is sive changes in species’ distributions and inter- destined to go the way of all small islands with actions, both directly and through tiny shifts one exception: it has the peculiar characteristic in species’ natural histories, as are now being

ECOLOGY OF WILDLAND INSECTS 91 frankie chap 07 8/20/03 7:53 PM Page 92

documented outside the tropics (e.g., Petersen J.-M. Cadiou, J. Miller, J. Franclemont, J. Rawlins, 1998; Brown et al. 1999). P. Gentili, E. Munroe, R. Hodges, J. A. Powell, These four permanent changes and aspects, M. Epstein, J. Corrales, D. Whitehead, J. King- however, do not create an ecosystem that falls solver, P. J. DeVries, M. Scoble, L. Pitkin, outside “naturally” occurring circumstances for M. Schauff, M. Sharkey, I. Gauld, S. Ward, large old-growth dry forests that once occurred N. Zamora, R. Liesner, M. M. Chavarria, R. Es- elsewhere in the Neotropics. In other words, pinoza, M. Wood, N. Woodley, S. Shaw, J. Whit- there are dry-forest areas far from rain forest field, D. Quicke, S. Shaw, C. Darling, and D. Car- (and cloud forest) generators and receivers of mean for identifications, and for caterpillar and migrants and dry forests with no rain forests or parasitoid hunting and husbandry, R. Moraga, cloud forests on their margins. There are dry- G. Sihezar, G. Pereira, L. Rios, M. Pereira, forest islands and peninsulas surrounded by O. Espinosa, E. Cantillano, M. Pereira, R. Franco, deserts, oceans, and other unfavorable habitats. H. Ramirez, F. Chavarria, M. M. Chavarria, Invasion by new biological entities, species that E. Olson, C. Moraga, P. Rios, C. Cano, D. Gar- have arrived (by whatever means) and eventually cia, F. Quesada, E. Araya, E. Guadamuz, R. Es- become part of what is old growth, has been part pinosa, R. Blanco, A. Guadamuz, D. Perez, of the normal construction process for main- R. Blanco, F. Chavarria, C. Camargo, H. Kidono, land ecosystems as long as they have existed. A. Masis, W. Haber, and W. Hallwachs. Future generations will have at least some idea of what once was the ACG’s dry forest. In the end, the ACG will be what it will be, REFERENCES as long as we can cause it to be viewed as Agrawal, A. A., C. Laforsch, and J. Tollrian. 1999. sufficiently useful and important to society that Transgenerational induction of defences in it is accepted as a member with full rights animals and plants. Nature 401:60–62. (Janzen 1998a,b, 1999a,b, 2000a,b, 2001a,b). Braby, M. F. 1995. Reproductive seasonality in tropical satyrine butterflies: Strategies for the If we fail in this challenge, a challenge as much dry season. Ecological Entomology 20:5–17. sociological as scientific, neither the ACG nor its Brown, J. L., S.-H. Li, and N. Bhagabati. 1999. dry-forest ecosystem will survive. Long-term trend toward earlier breeding in an American bird: A response to global warming. Proceedings of the National Academy of Sciences 96:5565–69. ACKNOWLEDGMENTS Bullock, S. H., H. A. Mooney, and E. Medina. 1995. Seaonally dry tropical forests. Cambridge: Cam- This chapter, and the caterpillar/host/parasitoid bridge University Press. 450 pp. inventory that provided some of the informa- Cochrane, M. A., A. Alencar, M. D. Schulze, C. M. tion for it, has been supported by NSF grants Souza Jr., D. C. Nepstad, P. Lefebvre, and E. A. BSR 90-24770, DEB 93-06296, DEB-94-00829, Davidson. 1999. Positive feedbacks in the fire DEB-97-05072, and DEB-00-72730, as well as dynamics of closed canopy tropical forests. by financial, administrative, and logistic support Science 284:1832–35. Estrada, A., R. Coates-Estrada, D. A. Anzures, and from INBio, the government of Costa Rica, the P. Cammarano. 1998. Dung and carrion bee- Area de Conservación Guanacaste, CONICIT of tles in tropical rain forest fragments and agri- Costa Rica, Keidanren of Japan, the Children’s cultural habitats at Los Tuxtlas, Mexico. Journal Rainforest Japan, the Children’s Rainforest Swe- of Tropical Ecology 14:577–93. den, and the Children’s Rainforest UK. Many Frankie, G. W., S. B. Vinson, M. A. Rizzardi, T. L. Griswold, S. O’Keefe, and R. R. Snelling. 1997. individuals have supported the development of Diversity and abundance of bees visiting a mass the project in a multitude of ways. I especially flowering tree species in disturbed seasonal dry thank W. Hallwachs, A. Masis, R. Robbins, J. M. forest, Costa Rica. Journal of the Kansas Ento- Burns, D. Harvey, A. Solis, C. Lemaire, R. Poole, mological Society 70:281–96.

92 JANZEN frankie chap 07 8/20/03 7:53 PM Page 93

Gauld, I. D., K. J. Gaston, and D. H. Janzen. 1992. ———. 1976b. Sweep samples of tropical decidu- Plant allelochemicals, tritrophic interactions and ous forest foliage-inhabiting insects: Seasonal the anomalous diversity of tropical parasitoids: changes and inter-field differences in adult The “nasty” host hypothesis. Oikos 65:353–57. bugs and beetles. Revista de Biología Tropical Goldammer, J. G. 1999. Forests on fire. Science 24:149–61. 284:1782–83. ———. 1976c. The microclimate differences bete- Harvey, C. A., and W. A. Haber. 1999. Remnant ween a deciduous forest and adjacent riparian trees and the conservation of biodiversity in forest in Guanacaste Province, Costa Rica. Bre- Costa Rican pastures. Agroforestry Systems 44: nesia 8:29–33. 37–68. ———. 1976d. Two patterns of pre-dispersal seed Haukioja, E. 1999. Bite the mother, fight the predation by insects on Central American de- daughter. Nature 401:22–23. ciduous forest trees. In Tropical trees: Variation, Holl, K. 1999. Factors limiting tropical rain forest breeding and conservation, ed. J. Burley and B. T. regeneration in abandoned pasture: Seed rain, Styles, 179–88. London: Academic Press. seed germination, microclimate and soil. Bio- ———. 1978. Seeding patterns of tropical trees. In tropica 31:229–42. Tropical trees as living systems, ed. P. B. Tomlin- Holl, K., and M. Kappelle. 1999. Tropical forest son and M. H. Zimmerman, 83–128. New York: recovery and restoration. Tree 14:378–79. Cambridge University Press. Hunt, J. H., R. J. Brodie, T. P. Carithers, P. Z.Gold- ———. 1979. Natural history of Phelypera distigma stein, and D. H. Janzen. 1999. Dry season mi- (Boheman), Curculionidae, a Costa Rican defo- gration by Costa Rican lowland paper wasps to liator of Guazuma ulmifoliaLam. (Sterculiaceae). high elevation cold dormancy sites. Biotropica Brenesia 16:213–19. 31:192–96. ———. 1980. Specificity of seed-attacking beetles Janzen, D. H. 1967a. Fire, vegetation structure, and in a Costa Rican deciduous forest. Journal of the ant × acacia interaction in Central America. Ecology 68:929–52. Ecology 48:26–35. ———. 1981a. Patterns of herbivory in a tropical ———. 1967b. Synchronization of sexual repro- deciduous forest. Biotropica 13:271–82. duction of trees with the dry season in Central ———. 1981b. The peak in North American ich- America. Evolution 21:620–37. neumonid species richness lies between 38 and ———. 1972. Protection of Barteria (Passiflora- 42 degrees north latitude. Ecology 62:532–37. ceae) by Pachysima ants (Pseudomyrmecinae) ———. 1982a. Cenízero tree (Leguminosae: Pithe- in a Nigerian rain forest. Ecology 53:885–92. cellobium saman) delayed fruit development in ———. 1973a. Sweep samples of tropical foliage Costa Rican deciduous forests. American Journal insects: Effects of seasons, vegetation types, of Botany 69:1269–76. elevation, time of day, and insularity. Ecology ———. 1982b. Natural history of guacimo fruits 54:687–708. (Sterculiaceae: Guazuma ulmifolia) with respect ———. 1973b. Evolution of polygynous obligate to consumption by large mammals. American acacia-ants in western Mexico. Journal of Ani- Journal of Botany 69:1240–50. mal Ecology 42:727–50. ———. 1982c. Cleogonus weevil seed predation ———. 1974a. The deflowering of Central Amer- on Andira can be predicted by fruit punctures. ica. Natural History 83:48–53. Brenesia 19/20:591–93. ———. 1974b. Tropical blackwater rivers, animals, ———. 1982d. Simulation of Andira fruit pulp and mast fruiting by the Dipterocarpaceae. removal by bats reduces seed predation by Biotropica 6:69–103. Cleogonus weevils. Brenesia 19/20:165–70. ———. 1974c. Epiphytic myrmecophytes in Sara- ———. 1983a. Insects. In Costa Rican natural his- wak: Mutualism through the feeding of plants tory, ed. D. H. Janzen, 619–45. Chicago: Uni- by ants. Biotropica 6:237–59. versity of Chicago Press. ———. 1975. Interactions of seeds and their insect ———. 1983b. Seasonal change in abundance of predators/parasitoids in a tropical deciduous large nocturnal dung beetles (Scarabaeidae) in forest. In Evolutionary strategies of parasitic in- a Costa Rican deciduous forest and adjacent sects and mites, ed. P. W. Price, 154–86. New horse pasture. Oikos 41:274–83. York: Plenum Press. ———. 1983c. Larval biology of Ectomyelois muriscis ———. 1976a. The depression of reptile biomass (Pyralidae: Phycitinae), a Costa Rican fruit par- by large herbivores. American Naturalist 110: asite of Hymenaea courbaril (Leguminosae: 371–400. Caesalpinioideae). Brenesia 21:387–93.

ECOLOGY OF WILDLAND INSECTS 93 frankie chap 07 8/20/03 7:53 PM Page 94

———. 1983d. No park is an island: Increase in ———. 1988d. Complexity is in the eye of the interference from outside as park size decreases. beholder. In Tropical rainforests: Diversity and Oikos 41:402–10. conservation, ed. F. Almeda and C. M. Pringle, ———. 1984a. Weather-related color polymor- 29–51. San Francisco: California Academy of phism of Rothschildia lebeau (Saturniidae). Bul- Science and AAAS. letin of the Entomological Society of America 30 ———. 1988e. The migrant moths of Guanacaste. (2):16–20. Orion Nature Quarterly 7:38–41. ———. 1984b. Natural History of Hylesia lineata ———. 1988f. Ecological characterization of a (Saturniidae: Hemileucinae) in Santa Rosa Na- Costa Rican dry forest caterpillar fauna. Biotrop- tional Park, Costa Rica. Journal of the Kansas ica 20:120–35. Entomological Society 57:490–514. ———. 1989. Natural history of a wind-pollinated ———. 1984c. Two ways to be a tropical big moth: Central American dry forest legume tree (Ateleia Santa Rosa saturniids and sphingids. Oxford herbert-smithii Pittier). Monographs in System- Surveys in Evolutionary Biology 1:85–140. atic Botany from the Missouri Botanical Garden ———. 1985a. On ecological fitting. Oikos 45: 29:293–376. 308–10. ———. 1990. An abandoned field is not a tree fall ———. 1985b. A host plant is more than its chem- gap. Vida Silvestre Neotropical 2:64–67. istry. Illinois Natural History Bulletin 33:141–74. ———. 1993. Caterpillar seasonality in a Costa Ri- ———. 1986a. The future of tropical ecology. can dry forest. In Caterpillars: Ecological and evo- Annual Review of Ecology and Systematics 17: lutionary constraints on foraging, ed. N. E. Stamp 305–24. and T. M. Casey, 448–77. New York: Chapman ———. 1986b. The eternal external threat. In and Hall. Conservation biology: The science of scarcity and ———. 1998a. Gardenification of wildland nature diversity, ed. M. E. Soule, 286–303. Sunderland, and the human footprint. Science 279:1312–13. Mass.: Sinauer Associates. ———. 1998b. How to grow a wildland: The gar- ———. 1986c. Disruption and recovery of intra- denification of nature. Insect Science and Appli- crown fruiting synchrony in a Cassia grandis cation 17:269–76. (Leguminosae) tree. Brenesia 25/26:179–85. ———. 1999a. Gardenification of tropical con- ———. 1986d. Biogeography of an unexceptional served wildlands: Multitasking, multicropping, place: What determines the saturniid and and multiusers. Proceedings of the National Acad- sphingid moth fauna of Santa Rosa National emy of Sciences 96:5987–94. Park, Costa Rica, and what does it mean to ———. 1999b. La sobrevivencia de las areas sil- conservation biology? Brenesia 25/26:51–87. vestres de Costa Rica por medio de su jardini- ———. 1986e. Mice, big mammals, and seeds: It ficación. Ciencias Ambientales, no. 16:8–18. matters who defecates what where. In Frugi- ———. 2000a. Costa Rica’s Area de Conservacion vores and seed dispersal, ed. A. Estrada and T. H. Guanacaste: A long march to survival through Fleming, 251–71. Dordrecht, Holland: Dr. W. non-damaging biodevelopment. Biodiversity 1 Junk Publishers. (2):7–20. ———. 1987a. How moths pass the dry season in ———. 2000b. Wildlands as gardens. National a Costa Rican dry forest. Insect Science and Its Parks Magazine 74(11–12):50–51. Application 8:489–500. ———. 2001a. Latent extinctions—the living dead. ———. 1987b. Insect diversity of a Costa Rican dry In Encyclopedia of biodiversity, ed. S. A. Levin, forest: Why keep it, and how? Biological Journal 3:689–99. New York: Academic Press. of the Linnean Society 30:343–56. ———. 2001b. Good fences make good neighbors. ———. 1987c. When, and when not to leave. Oikos Parks 11(2):41–49. 49:241–43. ———. 2002. Tropical dry forest: Area de Conser- ———. 1988a. Tropical dry forests: The most en- vación Guanacaste, northwestern Costa Rica. dangered major tropical ecosystem. In Bio- In Handbook of ecological restoration, vol. 2, diversity, ed. E. O. Wilson, 130–37. Washington, Restoration in practice, ed. M. R. Perrow and A. J. D.C.: National Academy Press. Davy, 559–83. Cambridge: Cambridge Univer- ———. 1988b. Management of habitat fragments sity Press. in a tropical dry forest: Growth. Annals of the Janzen, D. H., and I. D. Gauld. 1997. Patterns of Missouri Botanical Garden 75:105–16. use of large moth caterpillars (Lepidoptera: Sat- ———. 1988c. Tropical ecological and biocultural urniidae and Sphingidae) by ichneumonid restoration. Science 239:243–44. parasitoids (Hymenoptera) in Costa Rican dry

94 JANZEN frankie chap 07 8/20/03 7:53 PM Page 95

forest. In Forests and insects, ed. A. D. Watt, Miller, J. S., D. H. Janzen, and J. G. Franclemont. N. E. Stork, and M. D. Hunter, 251–71. London: 1997. New species of Euhapigioides, new genus, Chapman and Hall. and Hapigiodes in Hapigiini, new tribe, from Janzen, D. H., and W. Hallwachs. 2003. Philoso- Costa Rica, with notes on their life history and phy, navigation and use of a dynamic database immatures (Lepidoptera: Notodontidae). Tropi- (“ACG Caterpillars SRNP”) for an inventory of cal Lepidoptera 8(2):81–99. the macrocaterpillar fauna, and its food plants Murphy, P. G., and A. Lugo. 1986. Ecology of trop- and parasitoids, of the Area de Conservación ical dry forest. Annual Review of Ecology and Sys- Guanacaste (ACG), northwestern Costa Rica. tematics 17:67–88. http://janzen.sas.upenn.edu. Nepstad, D. C., A. Verissimo, A. Alencar, C. Nobre, Janzen, D. H., and P. S. Martin. 1982. Neotropical E. Lima, P. Lefebvre, P. Schlesinger, C. Potter, anachronisms: The fruits the gomphotheres P. Moutinho, E. Mendoza, M. Cochrane, and ate. Science 215:19–27. V. Brooks. 1999. Large-scale impoverishment Janzen, D. H., and C. M. Pond. 1975. A compari- of Amazonian forests by logging and fire. Nature son, by sweep sampling, of the arthropod fauna 398:505–8. of secondary vegetation in Michigan, England Nijhout, H. F. 1999. Control mechanisms of and Costa Rica. Transactions of the Royal Ento- polyphenic development in insects. BioScience mological Society of London 127:33–50. 49:181–92. Janzen, D. H., and T. W. Schoener. 1968. Differ- Petersen, J. S. 1998. Notes on the effect of the El ences in insect abundance and diversity be- Niño phenomenon and forest fires on Lepi- tween wetter and drier sites during a tropical doptera populations of Palawan, the Philip- dry season. Ecology 49:96–110. pines, during 1997–1998. Lepidoptera News, Janzen, D. H., and P. G. Waterman. 1984. A sea- no. 3:4. sonal census of phenolics, fibre and alkaloids in Pounds, J. A., M. P. L. Fodgen, and J. H. Campbell. foliage of forest trees in Costa Rica: Some fac- 1999. Biological response to climate change on tors influencing their distribution and relation a tropical mountain. Nature 398:611–15. to host selection by Sphingidae and Saturni- Powell, J. A., and J. W. Brown. 1990. Concentra- idae. Biological Journal of the Linnean Society tions of lowland sphingid and noctuid moths 21:439–54. at high mountain passes in eastern Mexico. Janzen, D. H., S. T. Doerner, and E. E. Conn. Biotropica 22:316–19. 1980a. Seasonal constancy of intra-population Roskam, J. C., and P. M. Brakefield. 1999. Sea- variation of HCN content of Costa Rican Aca- sonal polyphenism in Bicyclus (Lepidoptera: cia farnesiana foliage. Phytochemistry 19:2022– Satyridae) butterflies: Different climates need 23. different clues. Biological Journal of the Linnean Janzen, D. H., P. DeVries, D. E. Gladstone, M. L. Society 66:345–56. Higgins, and T. M. Lewinson. 1980b. Self- and Sharkey, M. J., and D. H. Janzen. 1995. Review of cross-pollination of Encyclia cordigera (Orchi- the world species of Sigalaphus (Hymenoptera: daceae) in Santa Rosa National Park, Costa Braconidae: Sigalaphinae) and biology of Siga- Rica. Biotropica 12:72–74. laphus romeroi, new species. Journal of Hymen- Janzen, D. H., P. J. DeVries, M. L. Higgins, and optera Research 4:99–109. L. S. Kimsey. 1982. Seasonal and site variation Still, C. J., P. N. Foster, and S. H. Schneider. 1999. in Costa Rican euglossine bees at chemical baits Simulating the effects of climate change on in lowland deciduous and evergreen forests. tropical montane cloud forests. Nature 398: Ecology 63:66–74. 608–10. Janzen, D. H., M. J. Sharkey, and J. M. Burns. Toh, I., M. Gillespie, and D. Lamb. 1999. The role 1998. Parasitization biology of a new species of of isolated trees in facilitating tree seedling re- Braconidae (Hymenoptera) feeding on larvae of cruitment at a degraded sub-tropical rainforest Costa Rican dry forest skippers (Lepidoptera: site. Restoration Ecology 7:288–97. Hesperiidae: Pyrginae). Tropical Lepidoptera 9 Williams-Linera, G., V. Dominguez-Gastelu, and (suppl.):33–41. M. E. Garcia-Zurita. 1998. Microenvironment Martin, P. S., D. Yetman, T. R. Van Devender, and floristics of different edges in a fragmented P. Jenkins, M. Fishbein, and R. Wilson, eds. tropical rainforest. Conservation Biology 12:1091– 1998. Gentry’s Rio Mayo plants: The tropical de- 1102. ciduous forest and environs of northwest Mexico. Wirth, R., W. Beyschlag, R. J. Ryel, and B. Holl- Tucson: University of Arizona Press. 381 pp. dobler. 1997. Annual foraging of the leaf-cutting

ECOLOGY OF WILDLAND INSECTS 95 frankie chap 07 8/20/03 7:53 PM Page 96

ant Atta colombica in a semideciduous rain dae) mining leaves of Bromelia pinguin (Bro- forest in Panama. Journal of Tropical Ecology meliaceae) in a dry forest in Costa Rica. Journal 13:741–57. of Natural History 29:1329–37. Woodley, N. E., and D. H. Janzen. 1995. A new species of Melanagromyza (Diptera: Agromyzi-

96 JANZEN frankie chap 08 8/20/03 7:54 PM Page 97

Section B. Biotic Relationships with Other Costa Rican Forests frankie chap 08 8/20/03 7:54 PM Page 98 frankie chap 08 8/20/03 7:54 PM Page 99

chapter 8

Diversity, Migration, and Conservation of Butterﬂies in Northern Costa Rica

William A. Haber and Robert D. Stevenson

igration can be simply defined as a ual, and members of a succeeding generation Msustained, directional movement by an an- make the return trip (Haber 1993; Dingle 1996; imal that takes it out of one habitat and into chapter 7). another (Dingle 1996), and this is the definition Migration behavior has been recognized in used here. It distinguishes migrating behavior butterflies for centuries (Williams 1930; John- from local movements within an animal’s home son 1969; Baker 1978; Dingle 1996), although range that tend to be nonlinear, of short dura- it has been studied in detail for only a few species, tion, and confined to a single habitat (Dingle such as the monarch (Danaus plexippus) (Brower 1996). Seasonal migration typically involves a 1991, 1996) and the painted lady (Vanessa car- two-way trip between habitats, exemplified by dui) (Hansen 1997). Among butterflies, migrat- the annual back-and-forth flights of Neotropical ing individuals are generally easy to distinguish migrant songbirds between the North Temper- from nonmigrants. Migrants fly in a relatively ate Zone and the tropics (Levey 1994; Martin straight line, usually 1–2 m above the ground, and and Finch 1995). Two-way migration by individ- rise over objects in their path, such as houses or ual butterflies occurs in some longer-lived species forest patches, rather than flying around them. such as the monarch in North America (Brower In contrast, active butterflies that are not migrat- 1991, 1996) and the common Costa Rican mi- ing fly a crooked route, alighting often to bask or grant Manataria maculata (Satyrinae; Stevenson feed. Females stop frequently to check potential and Haber 2000a). However, the migration of host plants, and males perch to watch for pass- most insects, relatively short-lived compared with ing females. Their directed flights rarely take vertebrates, is a one-way trip for a given individ- them more than 20 to 30 m in a given direction.

99 frankie chap 08 8/20/03 7:54 PM Page 100

Most migration is a response to seasonal resources used by both immature and adult changes in the environmental resources that butterflies. A consequence of this pronounced affect the probability of reproductive success. climatic seasonality is the migration behavior so Butterflies of the North Temperate Zone that characteristic of the Costa Rican dry-forest fauna. migrate do so for the same reason that birds Migration among Costa Rican Lepidoptera do—to take advantage of the short-term abun- has been studied relatively little. DeVries (1983, dance of food that becomes available in early 1987, 1997) recorded many observations of ap- summer and to escape a harsh winter when food parent migration. In several papers Janzen (1984, is limited and reproduction becomes impossible 1987a–c, 1988) discussed migration by moths in (Levey 1994; Dingle 1996). Most species migrate relation to seasonality and the biology of cater- to the south in late summer or autumn and re- pillars in the dry forest of Parque Nacional Santa turn northward in summer (Walker 1978, 1980). Rosa. Opler (1989) described a pattern of mi- Often, movements over several generations are gration and diapause in Eurema daira (Pieridae). required for a population to reach the northern- Haber (1993) and Stevenson and Haber (2000b) most limits of its distribution during the sum- studied butterfly migration in the Monteverde mer (Walker 1978, 1980; Brower 1996). region, particularly censusing seasonal move- In the tropics, where a cold winter season is ments of many species across the mountains be- lacking, it is less obvious why animals migrate tween the dry Pacific slope and the wet Atlantic (Loiselle and Blake 1991; Levey and Stiles 1992; slope. Stevenson and Haber (2000a) described Powell and Bjork, 1995). Migrations have been the natural history of M. maculata, a long-lived most commonly observed to pass between sea- species that makes a two-way, elevational migra- sonally very dry habitats and less seasonal wet tion between the dry forest and the cloud forest habitats—for example, between the dry forest at Monteverde. of Costa Rica’s northwest and the evergreen for- In this chapter, we compare the butterfly est of the Atlantic plain (Janzen 1984, 1987a–c; faunas of the Costa Rican dry forest, the cloud Haber 1993; Stevenson and Haber 2000a,b). forest of the neighboring mountains, and wet Migrations are usually associated with seasonal forest on the adjacent Atlantic slope. We identify changes, and climatic factors apparently serve as the most conspicuous migrating taxa and de- proximate cues for initiating migrating behavior scribe the main patterns of migratory behavior (Dingle 1996). Ultimate selective causes respon- of dry-forest butterflies that we have observed sible for migration probably include a variety of and relate these responses to dry-forest season- factors among different species, such as poor ality. We also point out some gaps in research quality of host-plant leaves, an increase in pred- knowledge and make recommendations rele- ators and parasitoids, severe weather, and limited vant to conservation issues involving migrating availability of adult food resources (Janzen 1984, butterflies and their adaptation to a changing 1987a–c, 1988; Haber 1993; Dingle 1996). seasonal environment. The Costa Rican dry forest is characterized by strong seasonality, with a six-month dry sea- STUDY AREA AND METHODS son (mid-November to mid-May) creating the most obvious seasonal pattern (Janzen 1983; STUDY AREAS chapter 7). Seasonal rainfall results in a semi- The area discussed in this chapter centers on the deciduous forest that is shady in the wet season Cordillera de Tilarán in northern Costa Rica and and open in the dry season with much higher the adjacent Pacific and Atlantic slopes and low- temperatures and lower humidity (Janzen 1983). lands (map 8.1). On the Pacific slope, the study These changes in climate result in seasonal area includes northern Puntarenas Province and shifts in the quality and availability of most southern Guanacaste Province east to the Conti-

100 HABER AND STEVENSON frankie chap 08 8/20/03 7:54 PM Page 101

85°00´ 84°50´ 84°40´ 84°30´ 10°30´ 10°30´ Laguna de Volcán Arenal Arenal Tilarán Fortuna

El Dos 1500-m contour Florencia Monteverde Provincia La Tigra 10°20´ Reserve Guanacaste Santa Elena 10°20´ Monteverde Complex Las Juntas San Luis Cordillera de Tilarán 700-m contour Interamerican Highway Provincia Guacimal Continental Divide Alajuela 10°10´ Lagarto 10°10´

Sardinal 700-m contour Provincia San Ramón Puntarenas Pacific Ocean Forest fragments Extensive forest sites 0 5 10 15 20 km

85°00´ 84°50´ 84°40´ 84°30´ MAP 8.1. Monteverde Reserve Complex and study sites.

nental Divide at Monteverde. Although the veg- tween the villages of Cuatro Cruces, Lagarto, etation of the lowlands in this area is popularly and San Luis, ranging from 50 to 700 m in ele- referred to as dry forest, it actually consists of a vation (map 8.1). mosaic of different life zones (Tropical dry for- In the montane region near Monteverde, est, Tropical moist forest, and Premontane moist where counts of migrating butterflies as well as forest; Holdridge 1967; Bolaños and Watson forest censuses were carried out, we worked in 1993). All these vegetation types include decid- Premontane wet forest, Lower montane wet uous tree species, but the proportion of canopy forest, and Lower montane rain forest life zones that goes leafless in the dry season can vary from (all are evergreen forests from 700 to 1,800 m close to nothing to almost 100 percent depend- elevation). The two Lower montane zones are ing on the drainage, disturbance history, local popularly known as cloud forest (Hartshorn 1983; climate, and proximity to streams (Hartshorn Haber 2000). The Pacific lowlands in the study 1983; Haber 2000). Most research on the lower area receive 1,500 to 2,000 mm of rainfall annu- Pacific slope was conducted in forest fragments ally (mostly between May and November); rain- within the Tropical moist forest life zone be- fall increases to 2,500 mm (also with a strong

BUTTERFLY DIVERSITY AND MIGRATION 101 frankie chap 08 8/20/03 7:54 PM Page 102

seasonal distribution) on the upper Pacific slope at Monteverde, at 1,200–1,500 m elevation. With COUNTS OF MIGRATING BUTTERFLIES added moisture from cloud interception, the Butterflies were systematically counted as they cloud forest (1,500–1,850 m) receives more than passed through a roadcut that crosses a ridge near 3,000 mm of rain per year. the entrance to the Monteverde Cloud Forest On the Atlantic slope, research was concen- Preserve at Monteverde. At this site, known as trated in the foothills of the Cordillera de Tilarán Windy Corner because it is exposed to the north- and the adjacent plains near Fortuna, La Tigra, east trade winds, butterflies crossed the ridge and Florencia, ranging from 70 to 850 m. The close to the ground where the road cuts through vegetation of this area consists of a complex forest. Counts were made during sunny weather mix of Tropical moist forest, Tropical wet forest, between 0900 and 1200 hr CST (= local time). Premontane wet forest, and Premontane rain forest life zons and transition zone types (Bo- IDENTIFICATION RESOURCES laños and Watson 1993). These forest types can Identification resources included publications be characterized generally as rain forest, and (DeVries 1983, 1987, 1997), museum collections, they receive 3,000–7,000 mm of rainfall annu- and determinations by specialists. Within Costa ally (Instituto Meteorológico de Costa Rica; Clark Rica, we used collections at the Museo Nacional et al. 2000; Michael Fogden unpubl. data). The de Costa Rica and the Instituto Nacional de Bio- dry season (January–April) is milder and much diversidad. Identifications were also provided by more irregular on the Atlantic side than on the Isidro Chacón, German Vega, Philip DeVries, Pacific side. Daniel Janzen, Paul Opler, Robert Robbins, and Andrew Warren. CENSUS OF BUTTERFLIES IN FOREST SITES Adult butterflies were censused monthly in forest fragments and at sites within extensive forest RESULTS of the Monteverde reserve complex along a north- BUTTERFLY DIVERSITY easterly trajectory from the Pacific lowlands, through the cloud forest at Monteverde, and Regional Species Richness. Excluding skippers down to the adjacent Atlantic lowlands. A total (Hesperiidae), about 1,036 butterfly species are of 45 sites were sampled over a three-year period currently known from Costa Rica (table 8.1). A from 1994 to 1997. Each sample consisted of species list in progress for skippers currently the total individuals and species of butterflies includes more than 500 species (A. Warren pers. identified during a one-hour walk (approximately comm.), compared with 431 species listed by De- 0.6 km) along a trail or transect exclusively Vries (1983). Estimating from the numbers of either through forest interior or along the outer skippers known at Monteverde and Santa Rosa, forest edge. Butterflies within 5 m on either side we (W. Haber and D. Janzen) expect a total num- of the transect were recorded. The same tran- ber of skippers for the country to be 600–800 sects were sampled each month. These sites species. Estimating from known species and consisted mostly of primary forest, although ranges of species that overlap Costa Rica, 274 ly- virtually all lowland sites were disturbed by the caenid species are expected to occur in the coun- removal of some timber trees and most included try (R. Robbins pers. comm.). This provides a to- some secondary areas. The personnel conduct- tal of Costa Rican butterfly species ranging from ing the censuses included D. Brenes, C. Guin- a conservative 1,576 to a liberal 1,900. don, N. Obando, M. Ramírez, R. Rojas, M. Wain- Although the total number in the Costa Ri- wright, and W. Haber, who trained the others in can dry forest is also not precisely known, col- censusing techniques. lections from several northwestern sites provide

102 HABER AND STEVENSON frankie chap 08 8/20/03 7:54 PM Page 103

TABLE 8.1 Species Richness of Butterﬂies in Three Costa Rican Faunal Regions

family santa rosa monteverde la selva costa rica

Papilionidae 13 22 21 42 Pieridae 21 51 20 67 Nymphalidae 89 197 165 438 Riodinidae 32 24 84 255 Lycaenidae ca. 50 85 79 234 (estimated) (ca. 80) (ca. 100) (ca. 110) (274) Hesperiidae 281 157 175 500 (estimated) ca. 300 ca. 200 ca. 300 (ca. 700) Total 486 536 544 1,536 (estimated) (ca. 535) (594) (700) (ca. 1,776)

Sources: DeVries (1983, 1987, 1997); D. Janzen (1988, 2002, pers. comm.); W. Haber (1993 and unpubl. data); P. Opler (pers. comm.); R. Robbins (pers. comm.). Note: The three faunal areas were lowland dry forest (Santa Rosa National Park), montane wet forest (Monteverde, above 1,200 m), and lowland wet forest (La Selva Biological Station).

an estimate of about 535 species, or one-third of Dry-Forest Endemics. None of the butterflies the country’s fauna (table 8.1), with 32 percent that live in dry forest are endemic to Costa Rica recorded at Santa Rosa National Park. In com- (DeVries 1987, 1997; R. Robbins pers. comm.). parison, 35 percent of the known species have Although many of the rarest Costa Rican but- been recorded in the wet montane area at Mon- terflies belong to the Riodinidae (DeVries 1997), teverde and from the Atlantic lowland rain for- most of the 32 species known from the dry forest est at La Selva Biological Station. have broad distributions, ranging from Mexico Species Richness in Life Zones along an Elevational to Costa Rica or extending into South America Gradient. The census of butterflies along an (DeVries 1997). Only six of these are restricted elevational gradient from the Pacific lowlands to the dry forest (DeVries 1997); the others also across the mountains at Monteverde to the At- occur in moister habitats. Some of these species lantic lowlands showed species numbers increas- are extremely rare in Costa Rica, such as Pan- ing from the Pacific lowlands to midelevation demos godmanii, Riodinidae (collected only once forest at Monteverde but decreasing from the in Costa Rica, near Cañas), but are not rare every- Atlantic lowlands to the mountaintop (table 8.2). where in their range (DeVries 1997; P. Opler In addition, the Tropical wet forest of the At- pers. comm.). lantic lowlands (380 species) was much richer Although we know of no endemic dry-forest than the Tropical moist forest life zone of the butterflies, it seems probable that some species Pacific lowlands (265 species). In contrast, Pa- that were rare or periodic colonists from Atlantic cific slope (leeward) cloud forest at Monteverde slope wet forest are now missing from the dry (>1,500 m) had about twice as many species forest. Some of these species show up as rare (161 species) as Atlantic slope (windward) cloud individuals (seemingly strays) from time to time forest (85 species). (DeVries 1987; W. Haber pers. obs.; D. Janzen

BUTTERFLY DIVERSITY AND MIGRATION 103 frankie chap 08 8/20/03 7:54 PM Page 104

TABLE 8.2 Species Richness of Butterﬂies at Sites along an Elevational Gradient from the Paciﬁc Lowlands through Monteverde to the Atlantic Lowlands

paciﬁc slope atlantic slope

0–600 m 600–1,200 m 1,200–1,800 m 0–600 m 600–1,200 m 1,200–1,800 m

Papilionidae 19 15 22 14 4 3 Pieridae 29 30 51 17 19 21 Nymphalidae 116 135 197 162 126 82 Riodinidae 19 21 24 31 49 4 Lycaenidae 8 25 85 30 23 2 Hesperiidae 74 59 157 126 63 18 Total 265 285 536 380 284 130

Note: Numbers include both census data and collateral records.

pers. comm.). Many species known from the gen- stantially increased during the dry season. At a eral region of our study sites were apparently ab- site in the center of the extensive Monteverde resent from the small forest patches studied and serve complex at 800 m, the species richness are presumed to be locally extinct (DeVries 1983, increased by about 20 percent during the dry 1987, 1997). More extensive forest tracts in Gua- season, whereas it almost doubled in some low- nacaste could presumably support some species land fragments (fig. 8.1). The decrease in species that are now restricted to wet forest on the At- numbers in the dry forest during the dry season, lantic slope (see also chapter 7). at least in part, results from the migration of butterflies out of the Pacific lowlands to less SEASONALITY OF THE FAUNA seasonal habitats on the upper Pacific slope and Decrease in Species Richness during the Dry Season. on the Atlantic slope (see the section “Migration The number of butterfly species in the Pacific Patterns” later in this chapter). lowlands significantly decreases during the dry Some butterflies (e.g., Eurema dina and season. In censuses of forest fragments in the Phoebis sennae, Pieridae) that remained in the lowlands below Monteverde, butterfly species lowlands through the dry season continued to richness declined by 40–50 percent (fig. 8.1). reproduce, whereas others ceased reproductive The difference between wet and dry season activity and entered diapause (see the next sec- numbers decreased with elevation, as do the tion, “Reproductive Diapause”). Among these are length and severity of the dry season. Montane some species of open areas (pasture and second sites above 1,200 m on the Pacific slope had a growth) that moved into the forest in the dry sea- reduction in species during the dry season, but son, taking advantage of cool, humid conditions the decrease was less pronounced (20–30%). The there, especially in shaded ravines along streams upper Atlantic slope was similar to the Pacific and rivers. For example, many skippers rest in side. However, on the mid- and lower Atlantic the deep shade of overhanging stream banks, slope (below 1,200 m), species numbers sub- where they are virtually invisible until they ex-

104 HABER AND STEVENSON frankie chap 08 8/20/03 7:54 PM Page 105

FIGURE 8.1. Comparison of butterfly species during the FIGURE 8.2. Differences between wet season and dry sea- dry season and wet season at sites along a transect from the son in species numbers in forest and edge censuses from Pacific lowlands through Monteverde to the Atlantic low- four forest fragments in the Pacific lowlands; dry season lands. Species numbers in the middle of the dry season species numbers are expressed as a percentage of the wet (February–March) are divided by numbers in the early wet season totals. At each site the dry edge species were de- season (June–July). P < .01 for Pacific slope (r = .84) and pressed more than the forest species (mean = 13.5% fewer Atlantic slope (r = .80). species).

plode into flight when approached. In the middle periods, interrupted only by periodic feeding of the dry season (February–March) in the four activity. Pacific lowland sites, the forest edge censuses For some of these species, such as Mechanitis averaged 13 percent lower in species numbers polymnia (Ithomiinae), diapause occurred spo- than the forest interior censuses, supporting radically during the dry season, with some indi- our observations that some open-area butterflies viduals diapausing while others were reproduc- (e.g., Eurema daira, Pieridae; Urbanus dorantes, tive concurrently at the same site. The dry-season Hesperiidae) moved into forest habitat during diapause for most species began in December at the dry season (fig. 8.2). the start of dry season and ended in May with the Reproductive Diapause. Reproductive diapause onset of the first wet season showers. Some in- is a common strategy characteristic of butterflies dividuals of M. polymnia broke diapause and be- in highly seasonal habitats (Haber 1993; Braby came reproductive following an isolated shower 1995; Dingle 1996). Many butterflies that do not in one of the fragments in January 1995. Some emigrate from the Costa Rican dry forest in the species displayed a high degree of plasticity, dry season enter a state of reproductive diapause including both migrating and diapausing indi- that physiologically halts mating, egg maturation, viduals (e.g., E. daira). Some butterflies that re- and oviposition during part of the dry season mained reproductive throughout the dry season (DeVries 1987; Opler 1989; Haber 1993, un- include Itaballia demophile and E. dina (Pieridae), publ. data). By dissecting females collected dur- Cissia renata (Satyrinae), Papilio thoas and Parides ing the dry season, we documented 63 species iphidamus (Papilionidae). Yet individuals of these that enter reproductive diapause (38 nymphalids, same species were also occasionally recorded 14 pierids, 9 hesperiids, and 2 riodinids). Dia- while migrating through Monteverde—well pausing butterflies remained quiescent for long above their reproductive ranges. It is not yet

BUTTERFLY DIVERSITY AND MIGRATION 105 frankie chap 08 8/20/03 7:54 PM Page 106

clear whether migration and diapause were almost individual butterflies, the migration is a ternative strategies that occurred at the same one-way trip with reproduction occurring after site. migration. It is the offspring produced one or Diapausing butterflies were also found occa- more generations in the future that make the re- sionally in other areas that are both wetter and turn trip. We recorded more than 300 butterfly less seasonal in rainfall than the Pacific lowland species at the migration sampling sites in Mon- sites. Diapausing individuals of seven species teverde, most during the eastward leg of the (two pierids, five ithomiines) were found in the migration (see also Haber 1993). Atlantic lowlands during the exceptionally pro- The migration of any given species is dis- nounced dry season of 1994 near La Fortuna, tinctly seasonal and more or less predictable Provincia Alajuela. Nineteen diapausing species from year to year, although annual variation in were detected at Monteverde during the dry sea- climate greatly affected the numbers and timing son in various years in Premontane wet forest of migrants (e.g., rates of eastward migration at 1,400 m. These species included both cross- were very low in El Niño years). However, dif- country migrants that apparently suppress egg ferent species each have their own life history maturation during migration and elevational pattern, so that at any time of year a subset of migrants from the Pacific slope and lowlands species may be migrating. Peaks in eastward that spend the dry season in moister habitats in migration occurred late in the wet season to the mountains (see the section “Elevational Mi- early in the dry season (October–January) and gration” later in this chapter). during the veranillo or “little dry season” (late June through mid-August). Westward migration MIGRATION PATTERNS (from Atlantic to Pacific slope) occurred in April– We have identified three distinct patterns of May, during the transition from the dry to the migration among dry-forest butterflies. The pat- wet season. tern encompassing most species is a back-and- Cross-country migration of alternating gen- forth migration between the Pacific and Atlantic erations involves a substantial seasonal inter- sides of the country over the central mountain change among the butterfly faunas of the Pacific chain and Continental Divide (cross-country mi- and Atlantic slopes that results in a spectacular gration). A smaller number of dry-forest species annual increase in butterfly numbers in the dry migrate to cool, moist habitats in the nearby forest. Depending on the site, 40 to 75 percent mountains during the dry season (elevational mi- of the Pacific slope species migrate to the At- gration). In a third pattern, butterflies migrate lantic slope. Offspring of eastward migrants re- from the dry forest southeast along the Pacific turn to the Pacific side, usually preceding or coast to lowland wet forest in the southern coinciding with the start of the wet season, and Pacific region (coastal migration). they begin reproducing as soon as they arrive. Cross-Country Migration. Migrating butterflies Reproduction by this influx of westward mi- are especially visible as they pass over deforested grants, along with the breaking of diapause in ridges (flyways) near the Continental Divide at nonmigrants, results in a large flush of first Monteverde, where the strong northeast trade generation butterflies in late June and July. winds force them to fly near the ground (Haber Many of these newly emerged butterflies (e.g., 1993). These migrants originate in all the dif- Marpesia petreus and Smyrna blomfildia [Nympha- ferent habitats along the gradient from lowland linae], Aphrissa statira [Pieridae], and Manataria dry forest to the cloud forest. Even species (e.g., maculata [Satyrinae]) migrate back to the At- Actinote leucomelas [Acraeinae] and several itho- lantic side during the veranillo, while the others miines) that breed in cloud forest on the upper stay to reproduce on the Pacific side, completing Pacific slope above 1,500 m migrate seasonally one or more generations before the dry season to the Atlantic side of the mountain range. For arrives. Butterflies that reach the Atlantic slope

106 HABER AND STEVENSON frankie chap 08 8/20/03 7:54 PM Page 107

continue to reproduce there. With very few ex- Danaus gilippus and eresimus, the heliconiids ceptions (see the section “Reproductive Dia- Agraulis vanillae and Dryas iulia, and the pierids pause” earlier in this chapter), butterflies sam- Anteos clorinde, Aphrissa statira, Ascia monuste, and pled during migration through Monteverde and P. sennae. All these species were also observed on the Atlantic slope during the dry season were migrating through Monteverde. We observed reproductively active (see also Haber 1993). this southeasterly migration along the Pacific Elevational Migration. Elevational migration is coast during most years between 1992 and defined here as the seasonal movement of but- the present at Puntarenas, Jacó, Dominical, and terflies uphill from their breeding range to a Punta Uvita at various times, but most notice- different life zone or habitat type at a higher ably during August. At Puerto Caldera we ob- elevation. Comparatively few species of eleva- served butterflies coming to shore across the tional migrants have been observed compared Golfo de Nicoya, apparently having crossed the with cross-country migrants. Some of these gulf from the southern end of the Peninsula de species, such as Greta oto (Ithomiinae), M. mac- Nicoya. The behavior of these butterflies differed ulata, and Siproeta stelenes, entered reproductive sharply from that of the cross-country migrants. diapause in the high-elevation refuge sites where They held a bearing of about 90 degrees to they passed the dry season in a manner very sim- the northeast trade winds and maintained their ilar to that of Polistes instabilis (Vespidae) (Hunt southeasterly flight direction even as the sea et al. 1999; Stevenson and Haber 2000a,b). Af- breeze arose from the west during the day. In ter producing a single generation of offspring contrast, the flight path of cross-country migrants on the Pacific slope below Monteverde (at 50 to was directly into the trade winds. Nielsen and 800 m), the emerging adults of M. maculata Nielsen (1950) described a similar coastal mi- (June through August) immediately migrate up- gration of one of these same species, A. monuste, slope to cloud forest sites (at 1,400–1,600 m). In in response to changes in its host plant along the the montane forest, they spend 8–11 months as eastern coast of Florida. virgin adults in reproductive diapause, returning Proportion of Fauna That Migrates. About 60 to their breeding sites during the following April percent of the species found in the lowland and to May, cued by the first heavy shower of the wet intermediate sites up to 1,200 m on the Pacific season (Stevenson and Haber 2000a). A few slope were also observed to migrate through other species, such as D. plexippus (Danainae), Monteverde (fig. 8.3). Only 46 percent of the Anartia fatima, and Siproeta epaphus (Nymphali- species that are known or suspected to breed at nae), continue to reproduce during the dry sea- the upper elevation sites (1,200–1,800 m) also son as they move upslope to moister, evergreen migrated seasonally. In sharp contrast, 74 per- areas at midelevation, where effects of the dry cent of the species at sites above 1,200 m on season arrive later and are less pronounced. the Atlantic slope migrate. The highest site on the Coastal Migration. A small number of butterfly Atlantic slope (1,600 m) had 87 percent mi- species were observed repeatedly at several points grants. This proportion decreased to 41 percent along the Pacific coast from Puntarenas to Uvita, at intermediate elevation sites and to 51 per- migrating in a southeasterly direction closely cent at lowland sites on the Atlantic side. In gen- following the coastline or crossing the ocean eral, the proportion of migrant species was about from the Peninsula de Nicoya to the mainland. the same in lowland sites on each side but be- These butterflies were apparently moving from came increasingly disparate at higher elevations. dry forest in Guanacaste and northern Punta- About three-quarters of the 536 butterfly renas Provinces to less seasonal, tropical wet species recorded from the upper Pacific slope forest in the southern Pacific region. The most at Monteverde migrate to some extent, with common species were the nymphalids A. fatima, migrating species from lower elevations con- Euptoieta hegesia, and S. stelenes, the danaids tributing substantially to species richness of the

BUTTERFLY DIVERSITY AND MIGRATION 107 frankie chap 08 8/20/03 7:54 PM Page 108

FIGURE 8.3. Proportions of migrant species among the butterﬂies censused in forest fragments and extensive forest from the Paciﬁc lowlands through Monteverde to the Atlantic lowlands (bars). The sites are grouped into three elevational blocks (0–600 m, 600–1,200 m, 1,200–1,800 m) on each slope shown diagrammatically (open circles).

montane area. Many of these butterflies (e.g., sitoid wasps, and birds. Moths are also major Aguna asander [Hesperiidae], Adelpha iphiclus, players in the pollination of dry-forest plants Marpesia petreus, S. stelenes [Nymphalinae], (Haber and Frankie 1988; chapter 2). In addi- Aphrissa spp., Eurema spp. [Pieridae], and Eury- tion, with their diurnal behavior, sometimes bia elvina [Riodinidae]) appear in Monteverde massive numbers, and colorful aspect, migrat- only during their passage between the lowland ing butterflies offer one of the more spectacular habitats on either slope and do not reproduce natural history phenomena observable in Costa there. However, they contribute to the ecology Rica. However, butterflies are actually just the of the region while feeding along the way, and tip of the iceberg. They are diurnal and highly they often spend days or weeks in the area below visible compared with the largely unnoticed the cloud forest waiting for favorable weather in thousands of species of moths, flies, wasps, order to cross the divide. bugs, beetles, and dragonflies that also migrate seasonally (Haber 1993, unpubl. data; D. Janzen DISCUSSION pers. comm.). Lay people often notice migrating butterflies in Costa Rica, but, with the exception DRY-FOREST CONNECTIONS of the most spectacular migrations, they seldom As one of the most important herbivore taxa in recognize migrating behavior and often think tropical forests, Lepidoptera form a key compo- that these butterflies are just unusually abun- nent in the food chain as selective agents on dant (W. A. Haber pers. obs.). plants and as prey for parasitoids and predators, These migrating species form broad ecolog- such as tachinid flies, many families of para- ical and geographical connections among the

108 HABER AND STEVENSON frankie chap 08 8/20/03 7:54 PM Page 109

main biological regions of Costa Rica. Our stud- Some species are judged to be rare because ies of butterflies and moths (see chapter 7) can they are known from only a few individuals, of- identify indicators for the conservation status of ten at widely separated sites, such as an un- other insect groups and point to habitat needing described riodinid known from a total of three protection. Indeed, the realization that a major individuals at three widely separated sites in portion of the dry-forest moth fauna in the Area Costa Rica (P. J. DeVries, W. A. Haber, and I. Cha- de Conservación Guanacaste (ACG) seasonally con unpubl. data). A species that seems rare may migrates to the wet forests to the east was a ma- be abundant in the canopy, where observation is jor rationale in the mid-1980s for expanding the difficult except by specialized methods such as original dry forest of Parque Nacional Santa Rosa fruit trapping (DeVries 1988). In some cases, into the much larger ACG to include rain forest rare species may represent periodic colonists or and cloud forest (D. Janzen pers. comm.). In strays from neighboring habitats where they are addition, our emerging understanding of intra- more abundant and maintain resident popula- tropical migrating behavior of butterflies, sphin- tions. More detailed information is needed about gid moths, and birds within the Monteverde re- most rare species before we can assign them a gion provided motivation for increasing the conservation status. reserve complex at Monteverde from 4,500 ha to Long-term monitoring can provide data 25,000 ha during the same period. needed to assess the conservation status of rare species and their habitats. Monitoring pro- HOW MUCH WE KNOW grams have been conducted in various places Our knowledge of butterfly distribution, abun- in the world (Murphy and Weiss 1988; Pinheiro dance, and natural history may be too poor to and Ortiz 1992; Blair and Launer 1997), but the determine whether dry-forest butterfly species most notable long-term monitoring efforts have have become extinct or even to know if any are been in Great Britain (Pollard et al. 1995) and currently in danger of extinction. Although fairly Europe (van Swaay 1990), where amateur and complete species lists exist for several research professional butterfly enthusiasts have surveyed sites (Monteverde, La Selva, Santa Rosa), we do butterflies in a systematic way for many years not have a thorough knowledge of the butterfly (New et al. 1995). These studies were able to fauna of any research station or protected area detect annual changes in population size and in Costa Rica. All these sites have a profusion of distribution as well as immigration events. Peri- species that were observed only once or a few odic censuses of butterflies were also used suc- times. The ecological and conservation status cessfully in the Amazonian forest fragment study of such rare or transient species is essentially (Brown and Hutchings 1997) to follow changes unknown without data on larval host plants, sea- after fragmentation created small forest patches. sonality, annual fluctuations, and migration be- Systematic, long-term monitoring programs havior. Moreover, species may be recorded as for butterflies are nonexistent in Costa Rica, breeding at a site and then become extinct lo- although Dan Janzen’s caterpillar-rearing pro- cally. They may be recorded on a species list for gram is producing extremely valuable data on a site but are actually accidental visitors appear- the breeding biology of Lepidoptera of the ACG ing there only during periods of unusual cli- (see chapter 7). We carried out a sampling pro- mate, such as El Niño years. An example is the gram for three years at Monteverde, but this was influx of hundreds of Heliconius sara into Mon- an intensive and costly effort involving four peo- teverde during August 1998. Although this low- ple. The data have been used in setting conser- land species found no suitable host plants (Passi- vation priorities at the regional level, for defin- flora auriculata; DeVries 1987) and soon died ing research goals, and in mapping areas for out, it will remain on the Monteverde checklist protection, as well as in applied efforts such as forever. reforestation by the Monteverde Conservation

BUTTERFLY DIVERSITY AND MIGRATION 109 frankie chap 08 8/20/03 7:54 PM Page 110

League and the Monteverde Cloud Forest Pre- fragments isolated within agricultural land- serve. Haber continues to monitor migrant but- scapes. Species of mature forest are likely to be terflies with a permanent flight trap at Monte- those most sensitive to extinction because their verde. Despite previous intensive inventories, habitat tends to decrease with any kind of hu- this low-key effort has added about 20 species to man disturbance and recovers slowly (Brown the regional checklist. and Hutchings 1997). Even high grading can Our experience suggests that the diversity at result in the loss of deep forest butterflies, pre- most Costa Rican sites is too high for a casual sumably through increased canopy opening. monitoring effort to census the butterfly fauna In Indonesia, Hill et al. (1995) noted that six adequately. Often, the most interesting species species of endemic or narrowly distributed for- from a conservation standpoint are undescribed est butterflies were present in undisturbed forest or taxonomically confusing, so that they do not but absent from adjacent high-graded forest. appear in field guides or are recognizable only Unfortunately, we have no data to compare dry- by specialists (W. Haber pers. obs.; D. Janzen, forest butterflies before and after disturbance. P. Opler, A. Warren, R. Robbins pers. comm.). Studies comparing the butterfly populations of In addition, census results are too dependent on mature and secondary habitats with varying size weather and observer expertise to make butter- and disturbance regimes would show what re- flies as a group the best choice for monitoring. sources are needed to support different species. Taxa that can be trapped with baits or attracted For some species migration can function as to scents by standardized methods may be prefer- a hedge against regional extinctions as migrants able as practical indicators. colonize habitat fragments where small populations have died out. Reservoir populations on POTENTIAL CAUSES OF EXTINCTION the Atlantic side of the country that can inject a The most likely cause of butterfly extinction in new wave of immigrants into the Pacific low- the dry forest is loss of habitat, particularly ma- lands each season create a mosaic of extinction ture forest—perhaps exacerbated by fire and and recolonization over time, forming metapop- pesticides. Frankie et al. (1998) suggested that ulations at the regional level (Pollard and Yates agricultural pesticides have contributed to the 1992; Hanski et al. 1994, 1995; Neve et al. 1996). decreased abundance and diversity of solitary However, as forest fragments are lost on both bees in Guanacaste Province. However, no Atlantic and Pacific slopes, the migration system comparable data are available for butterflies. In will break down, leaving a higher proportion of Jamaica, Emmel and Garraway (1990) and Gar- open-area species as forest butterflies gradually raway et al. (1993) implicated both habitat loss drop out of the fauna. For this reason, a regional and pesticides as significant factors in reducing approach to conservation planning seems es- diversity and contracting the range of the en- sential, which means integrating information demic Papilio homerus (Papilionidae). Acting to- and action across habitat boundaries, as has been gether, they are possibly much more serious. the case in conservation of the Three-wattled Fragmented habitats leave populations with Bellbird and Resplendent Quetzal (Powell and highly reduced ranges, existing as small relicts, Bjork 1995). and often these habitat patches have been de- Superimposed on these human threats is the graded by timber removal, fire, and cattle graz- pattern of extreme climate fluctuations experi- ing. These isolated populations tend to become enced in Costa Rica over the past decade (Pounds extinct through annual population fluctuations et al. 1999; Lawton et al. 2001). With popula- (Hanski and Thomas 1994; Hanski et al. 1994, tions reduced in size by habitat fragmentation, 1995; New et al. 1995; Brown and Hutchings a severe drought year, such as several recent El 1997) and from loss of host plants. Pesticides Niño years, could wipe out species that maintain especially threaten small populations in forest only a marginal residence in the dry forest. As

110 HABER AND STEVENSON frankie chap 08 8/20/03 7:54 PM Page 111

an apparent result, during strong El Niño years, natural history observations in a database on the cross-country migration through Monteverde Internet using common Web browsers (Morris almost disappears (W. Haber pers. obs.). Each of and Stevenson 2002; Stevenson et al. 2002). our Pacific lowland forest fragments supported Encouraging an interest in butterflies among only a fraction of the species known from simi- amateurs has had a great effect in generating lar habitat on a regional scale. Thus, most deep distribution data in the United States, Great forest species already have highly contracted Britain, and Europe. Initiating annual butterfly ranges within the dry forest. counts (such as the July Butterfly Count conducted in the United States) at major research BIODIVERSITY LAW AND sites in Costa Rica would stimulate interest in COMMERCIAL THREATS butterfly observation and also serve as a training Costa Rica’s biodiversity law (Ley de Conserva- workshop for local guides and amateurs with a ción de la Vida Silvestre, No. 7317, 1992) claims natural history interest. The annual Christmas all biodiversity as property of the state. Anyone bird counts based at Monteverde and the La Selva interested in collecting, farming, or conducting Biological Station demonstrate the value of these research on Lepidoptera must apply for a permit group activities in building enthusiasm as well from the Ministry of Environment and Energy as for accruing long-term data. Organizing both (MINAE). In practice, the law is strictly applied a Christmas and July butterfly count would be only in national parks. In addition, a permit is ideal for taking a snapshot of butterfly popula- required for exporting specimens. Illegal col- tions in both the dry and the wet seasons. lecting of Lepidoptera in Costa Rica is not per- Our current limited knowledge of how thou- ceived as a problem as it has been in some sites sands of species of seasonally migrating inver- such as Jamaica, where targeted, commercial tebrates and dozens of vertebrate species (par- collecting threatens a relict population of the ticularly birds and bats) connect the ecology swallowtail Papilio homerus (Emmel and Gar- and conservation of dry forest, cloud forest, and raway 1990). No Costa Rican insects are listed rain forest in Costa Rica is beginning to reach by CITES. Costa Rica is at the point where ob- the consciousness of biologists, natural resource serving, monitoring, and collecting Lepidoptera managers, and even the general public. Promot- should be encouraged in the cause of generating ing a wider understanding of these connections information useful for conservation planning. will also motivate public support for biodiversity conservation. RECOMMENDATIONS FOR RESEARCH AND PUBLIC INVOLVEMENT ACKNOWLEDGMENTS Any kind of inventory work, especially observations of host-plant use by butterflies, is worth We thank the Monteverde Cloud Forest Preserve, recording. A sample of the host plant should al- the Monteverde Conservation League, and many ways be pressed and dried for identification by landowners for permission to work on their an expert. Researchers interested in butterflies property, as well as the Instituto Nacional de Bio- should be encouraged to make this a part of diversidad and Museo Nacional de Costa Rica ongoing work and education activities in their for providing access to their collections. We also study areas. An Internet-based site where field thank R. Rojas, N. Obando, D. Brenes, C. Guin- data could be posted and collated would stimu- don, M. Ramírez, and M. Wainwright for assis- late the capture of distribution and life history tance in the field; P. DeVries, D. Janzen, P. Opler, records and make it feasible to share them R. Robbins, D. Wagner, and A. Warren for shar- across the research and amateur communities. ing unpublished data; I. Chacón, P. DeVries, We are currently developing an electronic field D. Janzen, R. Robbins, G. Vega, and A. Warren guide with the potential for anyone to record for help with butterfly identifications; and

BUTTERFLY DIVERSITY AND MIGRATION 111 frankie chap 08 8/20/03 7:54 PM Page 112

G. Frankie, D. Janzen, P. Opler, and J. Powell for Dingle, H. 1996. Migration, the biology of life on reviewing the manuscript. The study was sup- the move. New York: Oxford University Press. ported by NSF grant A9400011 to R. Steven- 474 pp. Emmel, T. C., and E. Garraway. 1990. Ecology and son and NSF DBI-9808462 to R. Stevenson and conservation biology of the Homerus Swallow- R. Morris, as well as the National Fish and Wild- tail in Jamaica. Tropical Lepidoptera 1:63–76. life Foundation, the National Geographic Society, Frankie, G.W., S.B. Vinson, M.A. Rizzardi, T.L. and the Missouri Botanical Garden. Griswold, S. O’Keefe, and R. R. Snelling. 1998. Diversity and abundance of bees visiting a mass flowering tree species in disturbed sea- REFERENCES sonal dry forest, Costa Rica. Journal of the Kansas Entomological Society 70:281–96. Baker, R. R. 1978. The evolutionary ecology of ani- Garraway, E., A.J.A. Bailey, and T.C. Emmel. mal migrations. New York: Holmes and Meier. 1993. Contribution to the ecology and conser- 1012 pp. vation biology of the endangered Papilio home- Blair, R. B., and A. E. Launer. 1997. Butterfly diver- rus (Lepidoptera: Papilionidae). Tropical Lepidop- sity and human land use: Species assemblages tera 4:83–91. along an urban gradient. Biological Conservation Haber, W. A. 1993. Seasonal migration of mon- 80:113–25. archs and other butterflies in Costa Rica. In Bi- Bolaños, R. A., and V. Watson. 1993. Mapa ecoló- ology and conservation of the monarch butterfly, gico de Costa Rica. San José: Centro Científico ed. S. B. Malcolm and M. Zalucki, 201–7. Los Tropical. Angeles: Los Angeles County Museum of Nat- Braby, M. F. 1995. Reproductive seasonality in trop- ural History. ical satyrine butterflies: Strategies for the dry ———. 2000. Plants and vegetation. In Monte- season. Ecological Entomology 20:5–17. verde: Ecology and conservation of a tropical cloud Brower, L. P. 1991. Animal migrations: Endangered forest, ed. N. Nadkarni and N. Wheelwright, phenomena. American Zoologist 31:265–76. 39–94. New York: Oxford University Press. ———. 1996. Monarch butterfly orientation: Haber, W. A., and G. W. Frankie. 1988. A tropical Missing pieces of a magnificent puzzle. Journal hawkmoth community: Costa Rican dry forest of Experimental Biology 199:93–103. Sphingidae. Biotropica 21:155–72. Brown, K. S., and R. W. Hutchings. 1997. Dis- Hansen, M.D.D. 1997. Observations on the turbance, fragmentation, and the dynamics of migrations of the Painted Lady (Vanessa cardui diversity. In Tropical forest remnants: Ecology, (L.)) in Denmark in 1996 (Lepidoptera: management, and conservation of fragmented Nymphalidae). Entomologiske Meddelelser 65: communities, ed. W. F. Laurance and R. O. Bier- 165–73. regaard Jr., 91–110. Chicago: University of Chi- Hanski, I., and C. D. Thomas. 1994. Metapopula- cago Press. tion dynamics and conservation: A spatially ex- Clark, K. L., R. O. Lawton, and P. R. Butler. 2000. plicit model applied to butterflies. Biological The physical environment. In Monteverde: Ecol- Conservation 68:167–80. ogy and conservation of a tropical cloud forest, ed. Hanski, I., M. Kuussaari, and M. Nieminen. 1994. N. Nadkarni and N. Wheelwright, 15–38. New Metapopulation structure and migration in the York: Oxford University Press. butterfly Melitaea cinxia. Ecology 75:747–62. DeVries, P. J. 1983. Checklist of butterflies. In Costa Hanski, I., T. Pakkala, M. Kuussaari, and G. Lei. Rican natural history, ed. D. H. Janzen, 654–78. 1995. Metapopulation persistence of an endan- Chicago: University of Chicago Press. gered butterfly in a fragmented landscape. Oikos ———. 1987. The butterflies of Costa Rica and their 72:21–28. natural history. Vol. 1, Papilionidae, Pieridae, Hartshorn, G. S. 1983. Plants: Introduction. In Nymphalidae. Princeton, N.J.: Princeton Uni- Costa Rican natural history, ed. D. H. Janzen, versity Press. 327 pp. 118–57. Chicago: University of Chicago Press. ———. 1988. Stratification of fruit-feeding nym- Hill, J.K., K.C. Hamer, L.A. Lace, and W.M. Ban- phalid butterflies in a Costa Rican rainforest. ham. 1995. Effects of selective logging on trop- Journal of Research on the Lepidoptera 26:98–108. ical forest butterflies on Buru, Indonesia. Journal ———. 1997. The butterflies of Costa Rica and their of Applied Ecology 32:754–60. natural history. Vol. 2, Riodinidae. Princeton, N.J.: Holdridge, L. R. 1967. Life zone ecology. Rev. ed. San Princeton University Press. 288 pp. José: Tropical Science Center.

112 HABER AND STEVENSON frankie chap 08 8/20/03 7:54 PM Page 113

Hunt, J.H., R.J. Brodie, T.P. Carithers, P.Z. Gold- structure in the vulnerable butterfly Proclossiana stein, and D. H. Janzen. 1999. Dry season eunomia (Lepidoptera: Nymphalidae). Journal migration by Costa Rican lowland paper wasps of Applied Ecology 33:14–22. to high elevation cold dormancy sites. Biotrop- New, T. R., R. M. Pyle, J. A. Thomas, C. D. Thomas, ica 31:192–96. and P. C. Hammond. 1995. Butterfly conserva- Janzen, D. H. 1984. Two ways to be a tropical big tion management. Annual Review of Entomology moth: Santa Rosa saturniids and sphingids. Ox- 40:57–83. ford Surveys in Evolutionary Biology 1:85–140. Nielsen, E. T., and A. T. Nielsen. 1950. Contribu- ———. 1987a. Biogeography of an unexceptional tions toward the knowledge of the migration place: What determines the saturniid and sphin- of butterflies. American Museum Novitates 1471: gid moth fauna of Santa Rosa National Park, 1–29. Costa Rica, and what does it mean to conserva- Opler, P. A. 1989. Ecological and behavioral as- tion biology. Brenesia 25/26:51–87. pects of seasonal polyphenism in Eurema daira ———. 1987b. How moths pass the dry season in (Pieridae, Lepidoptera). In The evolutionary ecol- a Costa Rican dry forest. Insect Science and Its ogy of plants, ed. J. H. Brock and Y. B. Linhart, Application 8:489–500. 515–33. Boulder, Colo.: Westview Press. ———. 1987c. When, and when not to leave. Oikos Pinheiro, C.E.G., and J.V.C. Ortiz. 1992. Com- 49:241–43. munities of fruit-feeding butterflies along a ———. 1988. Ecological characterization of a vegetation gradient in central Brazil. Journal of Costa Rican dry forest caterpillar fauna. Biotrop- Biogeography 19:505–511. ica 20:120–35. Pollard, E., and T. J. Yates. 1992. The extinction ———. 2002. Caterpillar rearing voucher data- and foundation of local butterfly populations bases for the Area de Conservación (ACG) in in relation to population variability and other northwestern Costa Rica (http://janzen.sas. factors. Ecological Entomology 17:249–54. upenn.edu/caterpillars/database.htm). Pollard, E., D. Moss, and T. J. Yates. 1995. Population ———, ed. 1983. Costa Rican natural history. Chi- trends of common British butterflies at moni- cago: University of Chicago Press. 816 pp. tored sites. Journal of Applied Ecology 32:9–16. Johnson, C. G. 1969. Migration and dispersal of Pounds, J.A., M.P. Fogden, and J.H. Campbell. insects by flight. London: Methuen. 763 pp. 1999. Biological response to climate change on Lawton, R.O., U.S. Nair, R.A. Pielke, and R.M. a tropical mountain. Nature 398:611–15. Welch. 2001. Climatic impact of tropical low- Powell, G. V. N., and R. Bjork. 1995. Implications land deforestation on nearby montane cloud of intra-tropical migration on reserve design: A forests. Science 294:584–87. case study using Pharomachrus mocinno. Con- Levey, D. J. 1994. Why we should adopt a broader servation Biology 9:354–62. view of neotropical migrants. Auk 111:233–36. Stevenson, R. D., and W. A. Haber. 2000a. Man- Levey, D. J., and F. G. Stiles. 1992. Evolutionary ataria maculata (Nymphalidae: Satyrinae). In precursors of long-distance migration: Resource Monteverde: Ecology and conservation of a tropical availability and movement patterns in neotrop- cloud forest, ed. N. Nadkarni and N. Wheelwright, ical landbirds. American Naturalist 140:447–76. 119–20. New York: Oxford University Press. Loiselle, B. A., and J. G. Blake. 1991. Temporal ———. 2000b. Migration of butterflies through variation in birds and fruits along an elevational Monteverde. In Monteverde: Ecology and conser- gradient in Costa Rica. Ecology 72:180–93. vation of a tropical cloud forest, ed. N. Nadkarni Martin, T. E., and D. M. Finch, eds. 1995. Ecology and N. Wheelwright, 118–19. New York: Oxford and management of Neotropical migratory birds. University Press. New York: Oxford University Press. 489 pp. Stevenson, R.D., W.A. Haber, and R.A. Morris. Morris, R. A., and R. D. Stevenson. 2002. Elec- 2002. Electronic field guides and user commu- tronic field guide: An object oriented WWW nities in the eco-informatics revolution. Conser- database to identify species and record ecologi- vation Ecology 7(1). [On-line] URL: http://www. cal observations (http://www.cs.umb.edu/efg/). consecol.org/vol7/iss1/art3 Murphy, D. D., and S. B. Weiss. 1988. A long-term van Swaay, C.A.M. 1990. An assessment of the monitoring plan for a threatened butterfly. Con- changes in butterfly abundance in the Nether- servation Biology 2:367–74. lands during the 20th century. Biological Con- Neve, G., B. Barascud, R. Hughs, J. Aubert, H. De- servation 52:287–302. scimon, P. Lebrun, and M. Baguette. 1996. Dis- Walker, T. J. 1978. Migration and re-migration persal, colonization power and metapopulation of butterflies through north peninsular Florida:

BUTTERFLY DIVERSITY AND MIGRATION 113 frankie chap 08 8/20/03 7:54 PM Page 114

Quantification with malaise traps. Journal of the Williams, C. B. 1930. The migration of butterflies. Lepidopterists’ Society 32:178–190. London: Oliver and Boyd. 473 pp. ———. 1980. Migrating Lepidoptera: Are butterflies better than moths? Florida Entomology 63: 81–98.

114 HABER AND STEVENSON frankie chap 09 8/20/03 7:55 PM Page 115

chapter 9

Watershed Ecology and Conservation

HYDROLOGICAL RESOURCES IN THE NORTHWEST OF COSTA RICA

Alfonso Mata

quatic resources represent one of the (Quesada 1990). Perhaps the most critical area A most valuable and sought-after natural treas- is Guanacaste, the Costa Rican northwest, where ures, although in many areas of the planet they the dry season lasts six months. It is in this area are deteriorating. Playing a key role in climatic, that the largest extant territory of dry forest in ecological, and biogeochemical processes, the Mesoamerica is protected (Boza 1999). terrestrial water cycle is being destroyed at alarm- The existing hydrological connection between ing rates (Vörösmarty and Sahagian 2000). Hu- the different terrestrial, fluvial, and estuarine man populations have already appropriated half systems in that region obviously involves habi- of the accessible global freshwater runoff, and tats drained by the river web as well as all their this share will continue to rise to as much as ecologically related species (birds, bats, butter- 70 percent by the year 2025 if these populations flies, bees, mammals, fish, and humans, to name continue to grow as expected (Postel et al. 1996; a few). However, biological and ecological stud- Pringle 2000). In Costa Rica these resources are ies on the biota of dry-forest watersheds and con- still abundant in regions such as the Atlantic and necting uplands are scarce. Specific information southern basins but are scarce in others owing on the flora and fauna of riparian habitats is usu- to overuse or overpopulation, as in the Central ally limited and widely scattered in the litera- Valley of San José. This apparent abundance has ture. With the Guanacaste region as a frame, this allowed this country to neglect important pre- chapter analyzes the importance of the stream cautionary measures (strategic planning and corridors, current human affects, river research conservation) to protect and develop these re- and studies needed, and the actions that are be- sources as a fundamental element of sustain- ing or could be implemented to reduce or avoid able development and biodiversity protection major or irreversible damages.

115 frankie chap 09 8/20/03 7:55 PM Page 116

oxygen retention capacity. Forest shade THE ECOLOGICAL plays an important role. IMPORTANCE OF RIVERS 6. Biodiversity conservation: biological corri- “The biological diversity sustained by the fluvio- dor and wildlife species refuge in the dry riparian system is a critical step in the whole nat- and migratory seasons (Riley 1998: 96; ural food web of which human beings form a chapter 12). part” (Doppelt et al. 1993: 10). The fluvio-riparian habitat, also called a stream corridor, is one of Other benefits are derived from aesthetic qual- the most productive ecosystems. It is composed ities, shade, industrial and commercial trans- of a complex aquatic biological continuum nour- port, water provision for agriculture, and space ished by the bank and floodplain forest and the for outdoor sports, industry, and homes. Stream entire watershed. It provides nutrients reaching corridors are also important connections be- the river food web (Riley 1998: 97; Pringle and tween larger areas of natural habitats at differ- Scatena 1999: 112) from the headwaters to the es- ent elevational levels. They must be protected in tuary and ocean. Even under extreme conditions, and outside preserved areas (such as national the surface and subsoil riparian water discharges parks and refuges), from the headwaters to the (epirrhoeic and hyporrhoeic flow, respectively) seashore. always sustain prosperous vegetation and favor the associated fauna (see chapter 12). In general, RIVERS OF THE NICOYA PENINSULA these ecosystems have the following properties: AND TEMPISQUE RIVER BASIN

1. Protection and stabilization of soil that is Characteristics of the northwest region of Costa infiltrated by water, or the saturation zone. Rica are described in chapter 1. Map 1.1 shows This area is located upland on one or both the hydrological region of Guanacaste, featuring sides of the riverbank; it is a transitional at the center the estuarine, littoral, and terres- land edge between the river and surround- trial systems and watersheds that are all con- ing landscape or floodplain. This area is nected by the fluvial system. Rivers of the region most vulnerable to high surface runoff and can be grouped in two well-defined sectors: rivers nutrient leaching. of the northwestern coastal strip and those of the 2. The system works with multidimensional Gulf of Nicoya Basin (GNB). There are a few dynamics: longitudinal (displacement of others flowing to Lake Nicaragua (see also chap- aquatic species, with and against the flow); ter 1). to the lateral directions in the alluvial ter- RIVERS OF THE NORTHWESTERN races (movement of water and species to- COASTAL STRIP ward the floodplains and wetlands); and with vertical direction (groundwater current There are low coastal mountains (50–250 m and river channel current zone). above sea level) in the northwestern area of Guanacaste in which many creeks and a few 3. Retention of sediments, organic matter, small rivers originate, their microbasins empty- and pollutants that reach the river from ing directly into the Pacific Ocean. The region surrounding areas in mulch and under- presents the driest life zones of the country, such brush vegetation. as the tropical dry forest, the moist forest, and 4. Source of nutrients and species for other the premontane moist forest (chapter 1). These rivers (down the watershed), lagoons, lakes, rivers remain waterless during the six-month estuaries, and the ocean. dry season, but they rapidly flood with the per- 5. Control of water temperature of creeks, fa- sistent precipitation of the rainy season. The lo- voring aquatic life and increasing dissolved cal orography and winds from the south are

116 MATA frankie chap 09 8/20/03 7:55 PM Page 117

crucial in the establishment of heavier rainy tential. The area has witnessed development of conditions, which make the Nicoya Peninsula large agro-industries and hotels with good com- more humid. The Nosara River and Bongo River, munication networks. Its ethnographic and ar- the largest ones in the coastal strip, maintain a chaeological contributions are among the most perceptible base flow from March to May. relevant of the country. It is one of the main areas of Costa Rica, second in economic impor- RIVERS THAT FLOW TO THE GULF OF NICOYA tance only to the Central Valley (Mata and Blanco The GNB is an ample territory connected with a 1994). The watershed of the Tempisque River, net of rivers that have the gulf as their common with an area of 3,407.8 km2, constitutes the destiny (chapter 1). It covers around 12,000 km2, western part of the TRB. Its principal cities are almost half of the Pacific Ocean versant of Costa Liberia, capital of the province of Guanacaste; Rica. It is important not only because it covers Filadelfia; and Santa Cruz. The Bebedero water- almost 25 percent of the national territory (Mata shed has an area of 2,052.4 km2, and its principal and Blanco 1994) but also because it contains cities are Bagaces, Cañas, and Tilarán. almost all the life zones occurring in Costa Rica, Extensive agricultural areas, devoted prima- with a wide variety of terrestrial and aquatic rily to grains, sugarcane and, more recently, ecosystems. melon and livestock, dominate the physical and The Tempisque River Basin. The most extensive economic landscape. The TRB hosts virtually hydrological subregion in Guanacaste Province, half of the production of rice and sugarcane and the Tempisque River Basin (TRB), located in the a high percentage of the cattle ranching in Costa center of this region, has an average annual flow Rica. Other activities with impact on a national of 308 m3/s and has most of the seasonal dry for- level are melon production and forestry planta- est and related transitions in the country. It con- tions. Nevertheless, all these activities carry an sists of the watersheds of the Tempisque and environmental cost in terms of contamination, Bebedero Rivers. Variation in river flow is high, soil erosion, and decrease in water quality and reflecting wide seasonal climatic extremes with quantity. Clearly, the economic and natural sys- the potential for flooding as well as water scarcity tems are intimately linked, and it is necessary to (Ramirez and Jimenez 1998). Various habitats use the limited resources in a way that maxi- are crossed and drained by the stream network. mizes benefits to society. Three zones of the Na- The following are noteworthy: collections of hills tional System of Conservation Areas (SINAC) and mountains (coastal mountains and the Gua- have influence in this basin. Established by the nacaste Volcanic Mountain Range), small natu- Ministry of Environment and Energy (MINAE) ral reservoirs, wetlands (such as the floodplains for the protection and conservation of natural of the middle watershed of the Tempisque River resources, these areas are Area de Conservación and the mouth of the Bebedero River), and man- Guanacaste, Area de Conservación Tempisque, groves. This entire lowland region covers approx- and Area de Conservación Arenal. imately 1,300 km2. Eighty percent of it can be Other Basins and Subbasins. Other important considered a wetland, though the majority of the subbasins are that of the Abangares, Lagartos, area is traversed by agricultural activities (sugar- Guacimal, and Barranca Rivers, which empty cane, melon, rice). This river should have top into the Gulf of Nicoya. They have also suffered national priority in any decision regarding eco- extensive deforestation. The Lake Arenal sub- logical protection. basin (which includes Cote and Arenal Lakes The TRB has an area of 5,454 km2, equivalent and Chiquito River) originally emptied into the to 54 percent of Guanacaste Province and 10 per- Caribbean Sea through the San Juan River. How- cent of the national territory. It has primarily a ever, it was incorporated into the Pacific water- humid tropical climate, with assorted geological shed after the Arenal River was dammed, caus- formations of enormous beauty and tourist po- ing its volume to grow considerably and the lake

WATERSHED ECOLOGY AND CONSERVATION 117 frankie chap 09 8/20/03 7:55 PM Page 118

area to increase to 84 km2. This artificial con- (chapter 12). Amphibians, reptiles, and mam- nection, which supplies about 90 m3/s, estab- mals also make extensive use of riparian habitats lished the largest hydrological installation ever (chapters 5 and 12). Riparian areas are required assembled in the country, which is also called by several dry-forest bee species for nesting the Arenal-Tempisque Hydropower-Irrigation (Frankie et al. 1988, 1993), and larval host plants System. Hydroelectricity is produced through of butterflies and moths are commonly found three cascading power plants (Arenal, Corobici, in this habitat (chapters 7 and 8). and Sandillal), and irrigation from three prin- Nearly 15 percent of the 130 freshwater fish cipal channels supply the low flatlands of the species found in Costa Rica dwell in the fresh Bebedero River subbasin in the TRB. aquatic environment of Guanacaste rivers (Bus- sing 1998). Some fish of lowland dry-forest rivers (e.g., free-ranging convict cichclids) have received GROUNDWATER considerable study (Wisenden and Keenleyside Groundwater has been very important in Gua- 1994, 1995) over a period of several years. An nacaste, especially in the most isolated regions, important survey (Bussing and Lopez 1977) of where there has not been a municipal network fish associated with the interbasin transfer of of drinking water distribution, and because of the Arenal hydropower system was made before the six-month dry period. Even during the rainy the diversion of waters from the Caribbean to the season, wells are a necessity because of the ir- Pacific slope, but it has not been followed up. regularity of precipitation. The Zapandi area, on Finally, studies on aquatic invertebrates are the right side of the Tempisque River, has the po- rare, but it is believed that a relatively rich fauna tential to supply water for use in the intensive exists in many rivers that have not been subject agricultural months of summer, to some extent to extensive human disturbance. Overall, it is paralleling the irrigation system on the left side safe to say that riparian zones and stream corri- of this river, under the authority of the National dors need much more biological and ecological Irrigation and Drainage Service. study, especially in relation to associated human activities (Pringle et al. 2000a). STREAM CORRIDOR BIOTA AND THE DRY FOREST SYSTEM INTERACTION

Except for a few general documents, biological The area of the TRB, largest in the northwestern and ecological studies on the biota of dry-forest sector of the country (also known as the Choro- watersheds and connecting uplands are scarce. tega region), contains various aquatic and terres- Information on the flora and fauna of riparian trial ecosystems. They can be found in diverse habitats within watersheds is available for some environments, such as estuaries, mangroves, species, but it is usually limited and widely scat- floodplains, lagoons, prairies, and forests, and tered in the literature. Surveys in 1969–70 along all of them are located in different life zones that several permanent watercourses in the lowland characterize the region. The wetland ecosys- dry forest indicated that tree species are about tems, rich in flora and fauna, are located in the twice as diverse as compared with adjacent dry floodplains and are characterized by their geo- deciduous forest (Frankie et al. 1974). More re- graphical location and type of water (stagnant cent surveys indicate that a species-rich plant or flowing, fresh or brackish) that reaches them. understory forms a regular part of most riparian According to the World Conservation Union habitats. A wide variety of resident bird species classification for wetlands, the following can be are known to use riparian areas, and, as the dry found there (World Conservation Union 1992: season advances, resident as well as migratory 14): freshwater swamps, floodplains, estuaries, species seek riparian areas for food and cover coastal lagoons, and artificial wetlands (chap-

118 MATA frankie chap 09 8/20/03 7:55 PM Page 119

ter 11). The six principal life zones that occur on are densely populated and have historically suf- land indicate a variety of elevational ranges, even fered from strong human intervention (Vargas though the most important one corresponds to 1999: 88). In the case of the TRB and neigh- the tropical dry forest with its transition to moist boring areas, there are important population environments (see the section “Life Zones” in centers, such as Nicoya, Santa Cruz, Filadelfia, chapter 1). This life zone covers 10,400 km2, Cañas, Bagaces, Tilarán, and Liberia, that pro- representing roughly 10 percent of the national duce wastes. Others have been growing rapidly territory. It exists nowhere else in Costa Rica. and are creating problems that require prompt The amount of water that reaches the plains solutions, such as Cobano, Nandayure, Paquera, during the dry season depends entirely on the El Coco, Santa Cruz, and Las Juntas. base flow of headwaters in the highlands, created by the groundwater recharge during the rainy HUMAN IMPACT ON months. The larger rivers (Tempisque, Bolson, HYDROLOGICAL RESOURCES Bebedero, and a few of their principal tributaries) are perennial streams. The principal empty- Human effects on the environment, which lead ing stream is the estuary of the Tempisque River, to general ecological simplification of the rivers, where the main interaction between fresh water occur in various ways (Doppelt et al. 1993; Fed- and the Gulf of Nicoya saline water occurs. The eral Interagency Stream Restoration Working gulf is the most important estuary in Costa Rica Group 1998; Vörösmarty and Sahagian 2000). and one of the most attractive in tropical Amer- The main river ecological problems in the re- ica. It penetrates the northwestern part of the gion are related to extensive land use, fragmen- country, forming the Nicoya Peninsula (see map tation and alteration of stream corridors, fluvial 1.1 in chapter 1), with an area of approximately contamination, and recent new efforts to con- 1,530 km2. It arches from the Pacific Ocean to- struct levees in the floodplains. The final destina- ward the northwest, extending about 85 km to the tions of many pollutants from populated areas mouth of the Tempisque River (see chapter 10). are the Tempisque estuary and the Gulf of Nicoya. During the past 60 years, several agricultural (monocultures) and industrial activities have WATER EXTRACTION been developed throughout the GNB, which to- The most relevant water consumers in the basin gether have produced structural distortions in region are sugar industries, the Arenal-Temp- economic, social, and cultural activities. These ac- isque Irrigation System, fruit packaging plants tivities have had dramatic environmental effects (melon), aquaculture of tilapia, cities and towns, on the basin water, which has led to rapid reper- and, to a minor degree, the coffee factories of cussions downstream, all the way to the estuary. Tilarán. The Taboga, CATSA, and El Viejo sugar The Guanacaste Conservation Area and mills tap the Cañas, Liberia, and Tempisque Tempisque Conservation Area, which currently Rivers, respectively. These mills extract volumes include six national parks, five wildlife refuges, varying from 5 to 10 m3/s just for industrial use, two biological corridors, three wetlands, and a requiring another large quantity for irrigation. number of forestry reserves, would be better in- Since most of the harvest occurs during the dry tegrated ecologically if the fluvial-riparian zones season, water consumption dramatically alters (stream corridors) were restored and expanded. the volume of the Tempisque River and some This territory is extremely important because tributaries, which have already been depleted it protects a large part of the most extensive upstream. tropical-dry-forest zone that remains in Central America (Boza 1999: 54; Vargas 1999: 95). It MODIFICATION OF THE NATURAL CHANNEL must be noted that the driest regions of Central The slow expansion of sugarcane and water- America, which coincide with Pacific watersheds, melon cultivation needs more territories in the

WATERSHED ECOLOGY AND CONSERVATION 119 frankie chap 09 8/20/03 7:55 PM Page 120

form of reclamation works, channels, levees, indicating an absence of urban planning. These and fluvial detours. These alterations in the en- alterations have almost reached the shores of the vironment require strict regulation, especially rivers as they flow through the cities, where an- considering the vulnerability that these activities other component is added: contamination from introduce. Over the past few years, the time it organic wastes and trash. Several smaller rivers takes floodwater to withdraw has increased, a or creeks have only a few spaced trees with no condition attributable in part to the presence of understory vegetation. levees that hinder waters from returning to their During the 1940s several laws established normal course. So far, the main intervention the prohibition of tree cutting near banks of the consists of a 4-km channel and levees (built in country’s rivers and springs. Water Law No. 276 1983 by the National Service for Irrigation and of 1942 banned the cutting of trees within 50 m Drainage) circumventing 12 km of the Tem- of the riverbanks, sometimes corresponding pisque River that had supplied water to wetlands with the transitional upland fringe. The Forestry in this area. No riparian forests in the channel Law of 1996, No. 75757, indicates that a protec- have been restored after these developments. In tion area of 100 m around the springs must be fact, the service has plans to dam the river and respected. Furthermore, cutting is prohibited to continue dredging and canalization. Measures within 15 m of each side of the rivers and creeks should be considered in order to preserve river in rural areas. In urban areas, a 10-m strip on corridors, from the headwaters all the way to each side is protected if the terrain is flat, and the receiving water body (lake, estuary, or sea). 50 horizontal meters are protected if the terrain “Adopt a River” public action projects, similar to is inclined. The old Costa Rican farmer has gen- those implemented in Puerto Viejo, Sarapiquí, erally respected these ordinances, although it in the northern Caribbean slope of Costa Rica, is common to see domestic animals devastating will become increasingly important (Pringle et rich understory vegetation. Some authors (Wat- al. 2000b). son et al. 1998) consider the riparian forest to be a common element in the scenery across the FRAGMENTATION OF THE RIPARIAN SYSTEM country, but not everyone has complied with The riparian and fluvial continuum (the stream the established measures, particularly the large corridor) is essential to the maintenance of bio- monoculture industries such as pineapple, sug- diversity in terrestrial ecosystems. The fragmen- arcane, rice, and bananas. In many cases, their tation and ecological discontinuity produced by lands have been deforested all the way to the human activities result in a visible impact on the river edge, with the excuse that the canopies riparian and stream systems. These damages obstruct aerial fumigation operations. This out- are apparent in a large part of Guanacaste’s dated practice has slowly been changing, and rivers and in Costa Rica’s rivers in general. Each some industries have taken measures to restore bridge, road, or agricultural activity that is de- these forests. veloped at or near the river results in varying It is evident that 10 or 15 m of protection are degrees of riparian fragmentation. In many too few for a flat land and even less for an in- cases, these could be prevented by protecting the clined terrain. The appropriate measurement stream corridors by means of appropriate con- should be determined by practical and scientific struction design as well as enforcement of cur- observations of the stream’s conditions: its role rent laws (chapters 22 and 23). in local ecological processes, the amount of bio- Rivers of the Tempisque lower basin have diversity it contains, stream dynamics, and the also suffered great damages from the sugarcane equilibrium of related ecosystems. Appropriate fields. When one passes through Liberia, Cañas, limits are extremely important because in real- Filadelfia, and Santa Cruz, the devastating effect ity landowners will cut to that limit or more, in of the urban presence on the banks is evident, detriment to the stream corridor.

120 MATA frankie chap 09 8/20/03 7:55 PM Page 121

Protection of these areas must be considered systems, not only to maintain initial coverage immediately, as there are still rivers in the north- but also to increase it and, it is hoped, reach total western coastal strip (see map 1.1 in chapter 1) coverage. and important tracts of the thin Zapandi Ripar- Population Growth and Urban Planning. Although ian Protected Zone of the TRB (ca. 400 ha) that the waste treatment plant of Liberia was built in contain segments in very good condition, some an unpopulated area near the western sector of with trees that in nearby regions have been ex- the Pan American Highway following the basin’s tirpated or are becoming extinct in Guanacaste. slope, its southern and eastern perimeters have In fact, it is not too late to consider restoring the now been reached by heavily populated housing entire Tempisque River corridor. settlements. Another problem is flooding of the lowest sectors of the fast-growing town of CONTAMINATION FROM RESIDUAL WATERS Filadelfia, which are closest to the Tempisque Many pristine streams can still be found, but River and some of its tributaries (such as Palmas some streams suffer the impact of urban and River). When flooding occurs, latrines and sep- agricultural wastes. Rivers of the Pacific coast are tic tanks overflow, and their fermented content clean during the rainy season, but only when is dispersed, creating severe fecal pollution. storms and persistent rainfall are absent. When There is a lack of authority and coordination these events occur, rivers are loaded with sedi- among governing offices to give the respective ments resulting from land use in their basins. permits or to produce plans for appropriate ur- Further, owing to lack of regulation or enforce- ban development. ment of a corresponding law, it is common to observe small settlements or clusters of houses NUTRIENTS AND PESTICIDES and shacks where roads and bridges cross Liquid wastes from a large part of the industrial streams, which plays a role in polluting the sugarcane process should end up in the field as streams. Once rivers begin to pass through the irrigation or in large aeration lagoons. However, cities, they become contaminated. Even so, their if there are periods when there is no concerted waters are exploited in great volumes down- coordination of clearing, plowing, and irrigation stream by the sugar industry. Dissolved oxygen of the field, the principal channel carrying waste- contents are at near normal levels in most of the waters will empty into streams or associated Tempisque River course and tributaries, with wetlands. There have been reports of river fish the exception of tracts flowing through and after kills, due to depletion of normal dissolved oxygen main cities. of the stream, when oxidation pond contents are Sewage. The treatment system for domestic released. wastewaters in the main cities of Guanacaste, A three-year monitoring of waters from the installed 25 years ago, initially covered 70 per- drainage channels of the extensive irrigation cent of the urban areas. However, expansion of system, including the Bagatzi channel, near the sewage service has not kept pace with popula- Palo Verde National Park, indicates that the tion growth, and wastewater treatment now water coming out of rice fields that have a high covers only 33 percent of urban zones (Echever- input of fertilizers, although reflecting the inte- ría et al. 1998). Furthermore, this growth means gration of the inputs on the system, shows low a larger production of organic matter that the to moderate nutrient contents, such as nitrate environment will have to absorb, whether from (Rodríguez-Ulloa 1997). There is no soil pollu- oxidation in septic tanks or by direct dumping tion or appreciable changes in soil fertility into surface currents. As a result, surface con- (Cabalceta 1997). With respect to pesticide contamination of rivers passing through cities will tent in waterfowl and fish, studies (Rodríguez- increase noticeably in the coming years. A large Brenes 1997) showed very low levels of or-- investment is necessary to expand treatment ganochlorine compounds and other substances

WATERSHED ECOLOGY AND CONSERVATION 121 frankie chap 09 8/20/03 7:55 PM Page 122

(Lindane, Aldrin, Dieldrin, β-BHC, Heptachlor, Changes in the estuary’s water conditions α- and γ-chlordane, chlorpirifos, malathion, are highly dynamic in the upper sector. Water- etc.). These residues are much lower than those mixing processes are generally effective because encountered a decade ago in Isla Pajaros, in the of important geographical barriers, such as Toro lower Tempisque River (Hidalgo 1986). None of Island at the mouth of the Tempisque River, these chemicals exceed the drinking water stan- Chira Island in the middle part of the upper gulf, dards. Heavy metals in birds showed contents and other islands. Furthermore, the estuary is similar to natural levels. fairly shallow, with an average depth of 4 m in the upper sector and 30 m in the outer gulf (Mata SOLID WASTES and Blanco 1994). The established tidal cycles, Nowhere in the basin are wastes being treated winds, net surface and bottom currents system, appropriately (Echeverría et al. 1998). Despite stratification, mixing processes, and bottom some efforts made by the government, this prob- roughness efficiently mix the entire water body lem applies to the rest of the country as well. It (Voorhis et al. 1983). However, its capacity for is estimated that, at the municipal level, there diluting and alleviating contamination could be are no political measures to coordinate the con- surpassed by the intensity of future inputs. Chap- servation of the environment regionally, and ter 10 discusses these subjects. there is a significant lack of environmental education in all strata of society. Lack of financ- PRIORITIZING ENVIRONMENTAL ing, insufficient technical assistance, and an in- CONSERVATION MEASURES adequate database are also common problems. There is currently no strategy in Costa Rica to initiate restoration and appropriate conservation THREATS TO THE GULF OF NICOYA of its fluvial systems. No study has yet been done to determine which rivers nationwide are in pris- Since the biological production of this estuary tine condition, from their birth to their receiving is so important to the country, much attention body (river, lake, estuary, or sea). As a result, no should be paid to the large burden of contami- program can use this missing information for nants being poured into the Tempisque River. their special protection. This analysis is needed However, because the population of Guana- to connect the conservation needs of the fluvial- caste influencing the gulf directly represents riparian ecosystems with state policies and with only 10 percent of the national population (4 mil- conservationist, governmental, academic, and lion), it is necessary to consider the effects of the private activities. Urgent needs, priorities, and Tárcoles River middle basin, where a million and proper actions related to the conservation of these a half inhabitants are concentrated in the San extremely valuable ecosystems must be defined. José Valley (see chapter 10 for discussion). Sev- There are plenty of viable possibilities for con- eral efforts have been carried out by universi- servation, restoration, and urban planning. Such ties (Marine Science Center/University of Costa actions will not only preserve Costa Rica’s valu- Rica; Environmental Pollution Research Center/ able resources for future generations but, in the University of Costa Rica; Marine Sciences Labo- more immediate future, will decrease vulner- ratory/Universidad Nacional de Costa Rica) in ability to natural disasters triggered by natural order to establish levels of pollution in the gulf. hydrological phenomena. The following recom- There is evidence that this contamination is more mendations are offered for safeguarding hydro- significant in the upper gulf and at the mouth of logical resources and their associated biodiversity the Tempisque River. The largest concentration in Guanacaste: of contaminants is located in the Puntarenas estuary, but they are extensively diluted by daily tidal 1. Implementing protection zones for sources currents from the gulf (Mata and Blanco 1994). of drinking water from wells and springs.

122 MATA frankie chap 09 8/20/03 7:55 PM Page 123

Only half of these, administered by the ment systems, at least for the main cities Aqueducts and Sewage Institute, are sub- and tourism centers. ject to adequate bacteriological control of 8. Strict control of, and restrictions on, chan- water quality. The rest, managed by rural neling and construction of levees and a administration groups, have no integral clear environmental policy for the Insti- protection policies or strict regulations. tute for Irrigation and Drainage. 2. Selection of representative rivers and their 9. Development of environmental awareness corridors for protection as depositories of among citizens, students, nongovern- biodiversity and connectors of preserved mental organizations, and political leaders ecosystems or watersheds from the moun- through educational methods and other tains to receiving water systems (lagoon, strategies (see chapters 19 and 20). Educa- estuary, or ocean). tional programs can lead to local projects 3. Reinforcement of protection laws for for stream corridor protection and restora- stream corridors (perennial, intermittent, tion (slogan for the communities: Adopt a and ephemeral currents). Although forestry River; see Pringle et al. 2000b). law requires protection of certain zones on 10. Better selection and training of committed rivers’ edges, it is not rigorously applied personnel in the institutions in charge of because most of these strips continue to conserving resources. Although there are be private property. The actual protection good programs and dedicated public ser- strip width is not sufficient, and it seems vants, there is a lack of conservationist that the importance of this ecosystem is leadership and commitment at the highest not a priority for municipal and govern- levels of the administration and, in gen- mental authorities. The whole Tempisque eral, a weak and inefficient bureaucracy in River corridor should be declared of na- the Park Service (see chapters 7, 12, 15, 19, tional interest, extending the protected 21, 22, and 23). strips of wetlands, Zapandi Riparian Pro- 11. Sponsorship of more stream corridor tected Zone. biological studies, especially in relation to 4. Effective protection for aquifer recharges fish and riparian overstory and understory and soil conservation areas, which corre- vegetation. spond to the upper basins of rivers and creeks. Their protection would mitigate aquifer contamination and would diminish REFERENCES flood threats in the lowlands. 5. Strict regulations and enforcement of Boza, M. 1999. Biodiversity conservation in Meso- america. In Managed ecosystems: The Mesoamer- existing legal measures that limit coastal ican experience, ed. L. Hupton Hatch and M. E. zone groundwater extraction. Swisher, 51–60. New York: Oxford University 6. Extension and improvement of sewers, an Press. increase in wastewater treatment capacity Bussing, W. 1998. Peces de las aguas continentales de Costa Rica (Fresh water fishes of Costa Rica). for the principal urban centers of the re- 2d ed. San José: Editorial University of Costa gion (see the section “Population Growth Rica. 468 pp. and Urban Planning” earlier in the chap- Bussing, W., and M. Lopez. 1977. Distribución y as- ter), and better industrial wastewater con- pectos ecológicos de los peces de las cuencas trol (mainly sugar mills and aquaculture). hidrográficas de Arenal, Bebedero y Tempisque, Costa Rica. Revista de Biología Tropical 25(1): 7. New sanitary landfills, rehabilitation of 13–37. controlled solid waste dump sites, and es- Cabalceta, G. 1997. Presencia de residuos de especies tablishment of integrated waste manage- químicas clave en los suelos. Proyecto de Riego

WATERSHED ECOLOGY AND CONSERVATION 123 frankie chap 09 8/20/03 7:55 PM Page 124

Arenal-Tempisque. Final Report. San José: Envi- M. E. Swisher, 104–13. New York: Oxford Uni- ronmental Pollution Research Center/University versity Press. of Costa Rica. 99 pp. Pringle, C. M., M. C. Freeman, and B. J. Freeman. de la Rosa, C. 1999. Conservation and sustainable 2000a. Regional effects of hydrological alter- use of streams and rivers in Central America. ations on riverine macrobiota in the New World: In Managed ecosystems: The Mesoamerican expe- Tropical-temperate comparisons. BioScience 50: rience, ed. L. Hupton Hatch and M. E. Swisher, 807–23. 122–32. New York: Oxford University Press. Pringle, C. M., F. N. Scatena, P. Paaby, and Doppelt, B., M. Scurlock, C. Frissell, and J. Karr. M. Nuñez-Ferrera. 2000b. River conservation 1993. Entering the watershed: A new approach to in Latin America and the Caribbean. In Global save Americas’s river ecosystems. Washington, D.C.: perspectives on river conservation: Science, policy Island Press. 463 pp. and practice, ed. P. J. Boon, B. R. Davies, and Echeverría, J., A. Echeverría, and A. Mata. 1998. G. N. Petts, 41–47. New York: John Wiley and Plan de acción para la Cuenca del Río Tempisque: Sons. Antecedentes del estudio y resumen ejecutivo. San Quesada, C., ed. 1990. Estrategia de conservación José: Centro Científico Tropical/Asociación para para el desarrollo sostenible. San José: Servicios el Manejo de la Cuenca del Tempisque. 39 pp. Litograficos. 180 pp. Federal Interagency Stream Restoration Working Ramirez, G., and J. Jimenez. 1998. Diagnóstico de Group. 1998. Stream corridor restoration: la situación actual de los recursos hídricos en la Principles, processes and practices. Washing- Provincia de Guanacaste, Costa Rica. San José: ton, D.C.: Federal Interagency Stream Restora- Instituto Costarricense de Acueductos y Alcan- tion Working Group. 568 pp (CD). tarillado. 19 pp. Frankie, G. W., H. G. Baker, and P. A. Opler. 1974. Riley, Ann L. 1998. Restoring streams in cities. Wash- Comparative phenological studies of trees in ington, D.C.: Island Press. 426 pp. tropical wet and dry forests in the lowlands of Rodríguez-Brenes, O. M. 1997. Residuos de plagui- Costa Rica. Journal of Ecology 62:881–919. cidas en la zona de influencia del Proyecto Frankie, G. W., S. B. Vinson, L. Newstrom, and de Riego Arenal-Tempisque. Final Report. San J. F. Barthell. 1988. Nest site and habitat pref- José: Environmental Pollution Research Center/ erences of Centris bees in the Costa Rican dry University of Costa Rica. 85 pp. forest. Biotropica 20:301–10. Rodríguez-Ulloa, A. 1997. Vigilancia de la calidad Frankie, G. W., L. Newstrom, S. B. Vinson, and de las aguas superficiales en la zona de influen- J. F. Barthell. 1993. Nesting-habitat preferences cia del Proyecto de Riego Arenal-Tempisque. of selected Centris bee species in Costa Rican Final Report. San José: Environmental Pollu- dry forest. Biotropica 25:322–33. tion Research Center/University of Costa Rica. Hidalgo, C. 1986. Determinación de residuos de 85 pp. plaguicidas organoclorados en huevos de ocho Vargas, G. 1999. The geography of dry land plant especies de aves acuáticas que anidan en la Isla formations in Central America. In Managed eco- Pájaros, Guanacaste, Costa Rica. Graduate the- systems: The Mesoamerican experience, ed. L. Hup- sis, Postgraduate Studies System, University of ton Hatch and M. E. Swisher, 88–97. New York: Costa Rica. 89 pp. Oxford University Press. Mata, A., and O. Blanco. 1994. La Cuenca del Golfo Voorhis, A., C. Epifanio, C. Maurer, A. Dittel, and de Nicoya: Reto al desarrollo sostenible. San J. A. Vargas. 1983. The estuarine character of the José: Editorial de la Universidad de Costa Rica. Gulf of Nicoya, an embayment on the Pacific 235 pp. coast of Central America. Hydrobiologia 99: Postel, S. L., G. C. Daily, and P. R. Ehrlich. 1996. 225–37. Human appropriation of renewable freshwater. Vörösmarty, C. J., and D. Sahagian. 2000. Anthro- Science 271:785–88. pogenic disturbance of the terrestrial water Pringle, C. M. 2000. Threats to U.S. public lands cycle. BioScience 50:753–65. from cumulative hydrologic alterations outside Watson, V., S. Cervantes, C. Castro, L. Mora, of their boundaries. Ecological Applications 10 M. Solís, I. Porras, and B. Conejo. 1998. Mak- (4):971–89. ing space for better forestry. Series 6, Policy that Pringle, C. M., and F. Scatena. 1999. Aquatic eco- works for forests and people. Edited by J. Mayers. system deterioration in Latin America and the International Institute for Environment and Caribbean. In Managed ecosystems: The Meso- Development. Nottingham, England: Russell american experience, ed. L. Hupton Hatch and Press. 110 pp.

124 MATA frankie chap 09 8/20/03 7:55 PM Page 125

Wisenden, B. D., and M. H. A. Keenleyside. 1994. parental cichlid ﬁsh. Environmental Biology of The dilution effect and differential predation Fishes 43:145–51. following brood adoption in free-ranging con- World Conservation Union. 1992. Conservación vict cichlids (Cichlasoma nigrofasciatum). Ethol- de humedales. Edited by P. J. Dugan. Paris: Im- ogy 96:203–12. primerie Sadag. 100 pp. ———. 1995. Brood size and economy of brood defence: Examining Lack’s hypothesis in a bi-

WATERSHED ECOLOGY AND CONSERVATION 125 frankie chap 10 8/20/03 7:56 PM Page 126

chapter 10

Where the Dry Forest Feeds the Sea

THE GULF OF NICOYA ESTUARY

José A. Vargas and Alfonso Mata

he gulf of nicoya (gn) is an estuary lo- in the dry-forest watershed has effects on the es- Tcated in the northwestern part of the Pacific tuary via the rivers. This chapter focuses on the coast of Costa Rica (10° N, 85° W). The extensive main physical, chemical, and biological features hydrological connection between the GN and of the estuary, as well as the main human activ- watersheds draining into it gives rise to one of ities that endanger its sustainable development. the most prominent ecological and geographical systems of Costa Rica (maps 1.1 and 1.2 in chap- HYDROLOGICAL REGIME ter 1; chapter 9). Half of this large estuarine- OF THE GN BASIN terrestrial basin consists of the Tempisque River Basin (TRB), with a variety of ecological, cultural, The GN receives waters collected by a group of and socioeconomic resources, some of them con- watersheds associated mainly with the TRB in flicting. If the Tárcoles River watershed, which the upper sector of the gulf and the Tárcoles drains the lower sector of the GN, is included, River watershed (from the Western Central Val- the whole area represents about 25 percent of ley), together with other minor areas in the lower the country. It has become clear over the past sector of the GN (see map 1.1 in chapter 1). Half decades that deforestation of the watershed has of the entire drainage system corresponds to the led to a reduction in average flow of the Temp- Tempisque Valley, with a total surface area of isque River (TR) in the dry season and to an 5,965 km2. Input from the Caribbean versant increase in the magnitude of floods during the through the Arenal reservoir discharges nearly rainy season (Lizano 1998). The dry-forest– 90 m3/s of fresh water from the Arenal- TR–GN system can no longer be considered as Tempisque hydroelectric generating system. Part composed of separate units because what occurs of this flow is used during the dry season for irri-

126 frankie chap 10 8/20/03 7:56 PM Page 127

gation of floodplains. Properties of this hydro- the Revista de Biología Tropical (vol. 44, Suppl. 3, logical system are discussed in chapter 9. 1996; vol. 46, Suppl. 6, 1998). Because the GNEEP was developed after several centuries of land use that resulted in the conversion of dry DEFINITION OF THE GN AS AN ESTUARY forests into pasturelands, it is unknown whether The GN is a shallow system (around 20 m) in its the effects observed in the biota in recent years upper reaches, but it deepens steadily to nearly started years before commercial exploitation of 500 m at its mouth. Its estuarine character is the resource (in the 1950s) and were later en- provided mainly by the freshwater input from hanced by increased fishing that reached its peak the Tempisque, Bebedero, and Tárcoles Rivers. in the 1970s and 1980s. The most important of these is the TR, whose According to Voorhis et al. (1983), fresh water watershed irrigates most of the dry-forest land discharged in the upper GN by the TR (and its of Guanacaste. This river is also important in large tributary Bebedero) and other rivers results defining strong horizontal and vertical salinity in a southward flow through the constriction gradients in the middle of the upper gulf, a re- between San Lucas Island and the Puntarenas gion characterized by mangrove forests, sand Peninsula in the middle section of the estuary. and mudflats, sandy beaches, and islands. This flow is compensated for by a northward flow of more saline water into the upper estuary along the bottom, which is termed estuarine circula- MAIN FEATURES OF THE tion. Flushing time (amount of time required to GN AS AN ESTUARY replace all existing fresh water at a rate equal to The GN is the country’s main fishing ground for the runoff discharge) was estimated at one to two finfish, penaeid shrimp, and shellfish. The de- months, a key figure in pollution studies. Look- cline of these resources, together with the lack ing at the GN from a chemical oceanography per- of ecological information to aid in policy mak- spective, Epifanio et al. (1983) showed that the ing, prompted several institutions, including the GN is similar to other tropical estuaries because University of Costa Rica, National University, and it is subject to extreme seasonal variations in the Costarrican Institute of Fishing and Aqua- riverine flow. Recent work by Chaves and Birkicht culture, to develop research programs in the GN (1996) on the Victor Hensen provided additional region. Some miscellaneous studies were also information on offshore sources of nitrogen for conducted by the Costarrican Institute of Water- the GN, in addition to that entering from rivers, works and Sewage and by nongovernmental or- which has yet to be measured. Using a simple box ganizations such as the Tropical Science Center model, Lizano (1998) illustrated the relationship and the Organization for Tropical Studies. The between freshwater input from the TR and other University of Costa Rica established the Gulf of rivers and the estuary’s salinity. The model shows Nicoya Ecological Evaluation Research Program that the more freshwater input into the system, (GNEEP) in 1979 at its Center for Research in the farther upstream the estuarine bottom salin- Marine Science and Limnology (CIMAR), and ity flow moves. Deforestation of watersheds dur- more than 100 scientific papers on the ecology ing the past centuries (see chapters 1 and 9) and and oceanography of the estuary have been pub- more recent use of river water for irrigation in- lished, making the GN the best-known tropical creased floods during heavy rains and may have estuary in the world. Cooperative research be- altered the average flow of the TR. tween local and foreign scientists allowed the development of complex sampling programs BIOTA OF THE GN aboard research vessels, the most recent of which was the expedition of the German research ves- A great deal of work under CIMAR’s GNEEP was sel Victor Hensen. Results have been published in focused on studying the GN from the biological

WHERE THE DRY FOREST FEEDS THE SEA 127 frankie chap 10 8/20/03 7:56 PM Page 128

oceanography point of view. A list of relevant phyla Brachiopoda, Sipuncula, Chaetognatha, publications prior to 1994 is included in Vargas and Echiura were reported for the first time by (1995). Among these, Bartels et al. (1983, 1984) Emig and Vargas (1990), Cutler et al. (1992), provided a description of finfish populations, Hossfeld (1996), and Dean and Cutler (1998), and others focused on fauna living in (Maurer respectively. New flatworm species parasitic in and Vargas 1984) or on sediments (Maurer et al. stingrays have been described by Marques et al. 1984, 1987). A total of 214 fish species were iden- (1995). tified, many of which were already under heavy Ecological studies in the GN that have been exploitation by fishers, such as corvinas (sea conducted as part of the GNEEP include those trout), snappers, groupers, sea-catfish, snooks, on the impact of shrimp trawling (Campos et and sharks. An update on fish biodiversity in the al. 1984; Rostad and Hansen 1999); trophic dy- GN is presented in Bussing (1994) and Wolff namics (Campos and Corrales 1986; Dittel 1993; (1996). The recent description of a new species Dittel et al. 1997); fish biology (Lai et al. 1993; of stingray (Urotrygon cimar) by López and Duncan and Szelistowski 1998); fish larval Bussing (1998) indicates that there is still work ecology (López 1983; López and Arias, 1987; to be done on the GN’s fish diversity, particularly Ramírez et al. 1989, 1990; Molina-Ureña 1996); in its rocky shores and intertidal areas. artificial reefs (Campos and Gamboa 1989; Sediment faunal surveys yielded about 100 Thorne et al. 1989); and the ecology of man- species collected in trawl samples. Decapod crus- groves (Gocke et al. 1981; Villalobos et al. 1985; taceans were the dominant taxon, with 54 species. Perry 1988; Soto 1988; Borjesson and Szelis- Recent updates on this fauna are provided by towski 1989; Wehrtmann and Dittel 1990; Koch Dean (1996, 1998, 2001a,b), Vargas et al. (1996), and Wolff 1996). and Cruz (1996) for marine worms (Polychaeta), The first reports on primary production in crustaceans, and mollusks, respectively. Among the GN waters come from Gocke et al. (1990) crustaceans, particular attention was given to and Córdoba-Muñoz (1998). They establish the the blue crab Callinectes arcuatus, a species of GN as one of the most productive (in terms of potential commercial importance (Epifanio and organic carbon fixed by primary producers) es- Dittel 1982; DeVries et al. 1983; Dittel and Epi- tuaries in the world. Only recently has there been fanio 1984; Dittel et al. 1985). All these studies an effort (Wolff et al. 1998) to integrate most of also emphasized that the GN is an ecosystem the available environmental information into a rich in species composition. Voucher specimens holistic ecological model of the GN. The result- from most species that were identified were de- ing model represents a steady-state approach in posited in the Museum of Zoology of the Uni- which biomass production of—and imports versity of Costa Rica. Despite significant attempts to—the system compartments are balanced by to identify all specimens collected, many of these consumption and exports. According to the remain unidentified or undescribed to date be- model, shrimp and demersal fish are the most cause of lack of expertise or updated taxonomic prominent benthic groups in terms of biomass, keys, or both. whereas plankton, small pelagic fish, carangids, Study of the GN’s marine biodiversity con- and squids dominate the pelagic biomass. The tinues at CIMAR and includes the work of Var- key role of mangroves is emphasized, as these gas (1987, 1988, 1989), Szelistowski (1990), Dit- forests contribute more than 75 percent of the tel (1991), Mielke (1992), Berman and Brooks system’s biomass (but only about 1 percent of (1994), Marques et al. (1995), Morales and Var- the primary production). The model shows that gas (1995), Vargas (1996), and Dean (1996, shrimp occupy a central role in the GN’s trophic 1998, 2001a,b). The first report on the presence dynamics as converters of much of the system’s of phylum Kinorhyncha in the GN was made by detritus and other bentho-pelagic matter into De la Cruz and Vargas (1986), and species of food biomass. Overexploitation of shrimp pop-

128 VARGAS AND MATA frankie chap 10 8/20/03 7:56 PM Page 129

ulations means a significant reduction of food the gulf (De la Cruz 1989; Robinson 1993), and stock for some predators. Some key topics still the presence of organic wastes from the sugar- need to be considered to improve the predictive cane industry, which also depletes the water’s power of the model, such as accounting for oxygen content (see chapter 9). Sugar factories intra-annual changes associated with seasonal have recently improved their antipollution meas- conditions (dry versus rainy) in the gulf; quanti- ures, but the fast-growing aquaculture of tilapia fying the sediment-nutrient-detritus input from fish in artificial ponds poses new and large de- rivers; the relative contribution of microorgan- mands for water and the disposal of residues. isms, phytoplankton, and phytobenthos to the The presence of organochlorine pesticides system’s productivity; and an evaluation of the in eggs of eight species of aquatic birds at Isla microbial loop. These data will provide a better Pajaros, near the mouth of the TR, has been de- picture of the GN as a system for development scribed by Hidalgo (1986). Sum of the residues of a management plan of the estuary and its of DDT detected was around 10 µg/g of dry watershed (Wolff et al. 1998). An improved sample; the heavy elements also showed very model will depend on these data and will further low levels. Mercury was found in normal back- emphasize the strong link between the GN and ground concentrations (De la Cruz 1989). All the drainage basin of its rivers, particularly the these studies are not conclusive; no systematic TR. A similar model is also available for Golfo sampling was conducted, including water and Dulce, a fjordlike embayment also in the Pacific animal tissue samples. Rice production is grow- coast of Costa Rica (Wolff et al. 1996). ing in lowland areas, and the entire situation calls for urgent and thorough research on the biota and limnological characteristics of the TR, IMPACTS ON THE GN particularly in the dynamic zone where the river A pioneer study in environmental economics meets the estuary. showed a clear decline of the GN’s fishing re- The problem of urban garbage management sources (Solórzano et al. 1991). Since 1983 there has not been solved yet, mainly because of un- has been a significant and steady decline in total controlled urban development, lack of public fish capture in the GN, with a concomitant in- awareness and education, and poor govern- crease in fishing efforts necessary to obtain the mental assistance for appropriate waste man- same capture. agement (see chapter 9). An ambitious project (“Let’s Save the Estuary”) was recently launched TR DISCHARGES by the government, and results thus far have Most of the population of Guanacaste Province been acceptable. The concern now is whether lives in the Tempisque Valley, concentrated in government support will continue. the cities of Cañas, Bagaces, Liberia (capital), Filadelfia, and Santa Cruz, as well as in many THE PUNTARENAS ESTUARY: A CASE STUDY towns (see maps in chapter 1). Old sewage treat- This estuary, located in the central eastern part ment plants of larger cities were built about of the GN, is a small-scale model of the gulf. Liq- 30 years ago, and actual coverage is only around uid wastes reach this estuary from the densely 30 percent of the cities’ needs (see chapter 1). populated port area of Puntarenas. Metabolic The rest of the towns use septic tanks and la- activity of this enclosed body of water permits trines. Massive deaths of shrimp and fish (e.g., absorption of a large amount of organic matter. sea bass) in this river are observed every year. Biochemical oxygen demand from effluents emit- This mortality rate is associated with the pres- ted by the sugar industry was greater at the head ence of xenobiotic substances (Szelistowski and of the estuary but was rapidly reduced toward its Garita 1989), pollutants introduced by washings mouth (Mata and Blanco 1994), where dissolved and land erosion from agricultural activity around oxygen values increase, approaching acceptable

WHERE THE DRY FOREST FEEDS THE SEA 129 frankie chap 10 8/20/03 7:56 PM Page 130

levels for aquatic wildlife. There is an important in those of the GN estuary (Dean et al. 1986; mixing of waters from inland with those from Fuller et al. 1990). The sediment load coming the GN, with different patterns of vertical ho- out of the highlands, which is due to destructive mogeneity and partial mixing along the estuary agricultural practices and land movement for (Acuña et al. 1998). These waters show measur- construction of buildings and roads, also has to able stratification and a definite biodegradation be considered. of organic wastes. Although solid waste recycling and landfills may have improved over the past decade, there DISCHARGES OF THE is still much to do. At the beginning of 1999 the WESTERN CENTRAL VALLEY municipal service collected around 90 tons of The Great Metropolitan Area (GMA) of Costa garbage per day from the streets of San José. It Rica is located in the Central Valley at an ele- was also estimated that careless people threw vation of around 1,000 m. The GMA has no about 250 tons per day in rivers. The impact on sewage treatment, and as of today a significant the GN has diminished very recently with the part of these waters are left to natural river self- reconstruction of the Brazil Dam on the Virilla purification. The coliform concentrations vary River. Cost for the extraction of 200 tons of from several dozens of thousands per 100 ml garbage per day (mainly wood and plastic ob- outside the GMA (Pacheco 1987) to millions jects) was $200,000 in 1999; when hours lost in several creeks and rivers of the densely pop- to cleaning generators are added, total cost for ulated central zone. Most of the nutrient in- that year was $360,000. This means that cost of organic load, sediments and resilient organic the lack of education was at least $30,000 per matter, enters, via the Tárcoles River, the lower month, without taking into account damage to sector of the GN, which is a much deeper water the river biota and its habitats. body in which dynamic mixing and strong tidal current conditions favor dilution of this load. EFFECTS OF POPULATION GROWTH However, an effective and definitive solution to these urban and industrial wastewaters is long The natural inertia of the demographic growth overdue. of Costa Rica is a well manifested urban all- Until a few years ago, coffee and sugar mills directional growth factor. Cities located in the produced organic pollution in the Central Valley watersheds of the GN greater basin are growing equivalent to 4.2 million inhabitants (Ramírez with doubling periods as short as 20 years. De- 1999). However, with new processes and the spite recent advances in the coffee industry and installation of waste treatment plants, pollution garbage management, the number of waste loads have dropped by more than 50 percent, and treatment plant systems available for the GMA another 50 percent improvement is expected is still unacceptable. The fast pace of urban during the next five years. Population growth growth in the Central Valley volcanic slopes on continues at a fast pace, however, and no mu- the outskirts of the GMA is resulting in the loss nicipal treatment is visible on the horizon. of options for development or conservation (or Although heavy metals (Cr, Cu, Pb) are found both) of the area’s extant resources. The National at very low concentrations in running waters Strategy of Conservation for Sustainable Devel- (Ramírez 1987), in view of current industrial opment (Quesada 1990) proposed a general plan growth it will be necessary to monitor possible for conservation of watersheds and urban devel- increases of metal waste discharges. Heavy metal opment that should have been developed by the contents in sediments are more significant. For beginning of the present century. Although pos- example, chromium from leather industries in itive results have been less than expected, inter- the Central Valley is 83 times higher in fine sed- est in the rescue of this beautiful valley is still iments of the Virilla-Tárcoles river system than alive (Echeverría et al. 1998).

130 VARGAS AND MATA frankie chap 10 8/20/03 7:56 PM Page 131

THE IMPACT OF CONTROL INITIATIVES 2. Reinforcement of selected laws and regulations to protect fishing grounds from over- In order to protect and promote the rational use exploitation, along with measures such as of resources, several important and well-coordi- fishing bans, fines, and net impoundment. nated steps must be taken into account. Needed actions include 3. Implementation of coffee, sugar mill, and pig-farming industrial agreements for pol- 1. the more efficient implementation and re- lution abatement, particularly in the TR inforcement of various standing laws, in- and Tárcoles River Basins. cluding strict application of management 4. Various actions for the recovery and re- plans for protected areas (chapters 21–24); cycling of solid wastes, mainly in the GMA. 2. improvement of laws and bylaws for the rational use and protection of resources; However, something missing in the whole sce- 3. strict compliance with international agree- nario of actions is an integral management plan ments and conventions; of the GN and the TR and Tárcoles River Basins. 4. increased interinstitutional coordination This plan would be multidisciplinary and would and collaboration; represent a great opportunity for terrestrial 5. support for environmental education ecologists, limnologists, oceanographers, and (especially in fishing communities and planners to come together and find appropriate schools and among consumers); mechanisms for the sustainable use of this 6. reduction of unnecessary habitat invaluable area of the country. destruction; 7. control of coastal and inland pollution FINAL COMMENTS sources (input by rivers); In recent years there has been a trend among 8. enlargement and updating of the sewage some economists to leave territorial ordering to treatment capacity of the Guanacaste re- the laws of supply and demand. It is clear that gion, particularly the Western Central this approach benefits only those rich sectors of Valley; society who could profit from opportunism and 9. support for marine and fish catch research; reflects the lack of formal courses in environ- and mental economics in most training programs 10. support for ecological research on the TR that prepare professionals at undergraduate and and at the dynamic zone where the river graduate levels. If problems associated with the meets the estuary. dry forest and the GN are to be approached from the point of view of sustainable development, The following are some actions that are being an integral management plan, organized by the implemented: principles of Integrated Coastal Management 1. “Let’s Save the Estuary,” a program created (ICM) (GESAMP 1996), for the entire region is by decree (Gaceta Oficial 23/10/95) and co- urgently needed. It should involve (a) public par- ordinated by the Costarrican Institute of ticipation such that the values, concerns, and as- Fishing and Aquaculture, with the partici- pirations of affected communities are discussed pation of various governmental institutions, and future directions are negotiated for appro- municipalities, and community members. priate guidance of community-based develop- The Gulf of Nicoya Contingency Plan is a ment; (b) steps by which relevant policies, leg- strategy being designed and implemented islation, and institutional arrangements can be by this institute with the help of two public developed and implemented to meet local needs universities. and circumstances while recognizing national

WHERE THE DRY FOREST FEEDS THE SEA 131 frankie chap 10 8/20/03 7:56 PM Page 132

priorities; and (c) collaboration between man- forcing the government to meet short-term agers and scientists in all stages of developing needs, a well-coordinated effort of all agencies management policy and programs (see chap- dealing directly or indirectly with the GN estu- ter 18), especially in the design, conduction, ary in the country is absolutely necessary. interpretation, and application of research and monitoring. In the case of the GN and its dry-forest water- REFERENCES shed, development of an ICM plan should first Acuña, J., V. García, and J. Mondragon. 1998. Com- bring to the table representatives from the main paración de algunos aspectos físico-químicos y governmental sectors involved, including the calidad sanitaria del Estero de Puntarenas, Costa Rica. Revista de Biología Tropical 46(Suppl. 6): Ministry of the Environment, the Costarrican 1–10. Institute of Tourism, and the Costarrican Insti- Baltz, C., and J. Campos. 1996. Hydrophone iden- tute of Fishing and Aquaculture. A crucial role tification and characterization of Cynoscion squa- has to be played by municipal governments, mipinnis (Perciformes: Scianidae) spawning which are the main actors in the granting of sites in the Gulf of Nicoya, Costa Rica. Revista permits for land use. To gain experience in this de Biología Tropical 44:743–51. Bartels, C., K. Price, M. López, and W. Bussing. process, the design of an ICM plan for the estu- 1983. Occurrence, distribution, abundance and ary of Puntarenas and its relatively small water- diversity of fishes in the Gulf of Nicoya, Costa shed should be the first priority. It can then be Rica. Revista de Biología Tropical 31:75–101. used as a reference model for the more complex ———. 1984. Ecological assessment of finfish as GN estuary. Finally, it is well known that world indicators of habitats in the Gulf of Nicoya, Costa Rica. Hydrobiologia 112:197–207. fish production is stagnated (Ludicello et al. Berman, R., and D. Brooks. 1994. Escherbothrium 1999) and is showing signs of decline, as is the molinae n. gen. et n. sp. (Eucestoda: Tetraphyll- case of the GN (Solórzano et al. 1991). It has idae: Triloculaiidae) in Urotrygon chilensis (Chon- also been estimated that world fishing fleets are drichthyes: Myliobatiformes: Urolophidae) from 30 percent larger than necessary to catch maxi- the Gulf of Nicoya, Costa Rica. Journal of Para- mum sustainable yield of global fish stocks (Mc- sitology 80(5): 775–80. Borjesson, D. L., and W. A. Szelistowski. 1989. Guinn 1998). These findings have created an Shell selection, utilization, and predation in the uncertain atmosphere in conservation and gov- hermit crab Clibanarius panamensis Stimpson ernmental sectors of the country, and it is urgent in a tropical mangrove estuary. Journal of Exper- to understand the causal factors. imental Marine Biology and Ecology 133:213–28. The GN’s production capacity has a theoret- Bussing, W. 1994. Demersal and pelagic inshore fishes of the Pacific coast of lower Central Amer- ical limit (Wolff et al. 1998) supported by an ica. Special publication, Revista de Biología Trop- important input of nutrients from the ocean by ical. 163 pp. bottom currents and from the mainland via the Cameron, W. M., and D. W. Pritchard. 1963. Estu- rivers. High rates of dispersion and absorption aries. In The Sea, ed. M. N. Hill, 2:306–24. New of biodegradable pollutants are characteristics of York: Wiley Interscience. this estuary; however, more quantitative studies Campos, J., and A. Corrales. 1986. Preliminary results of the trophic dynamics of the Gulf of are needed about pollution and its possible ef- Nicoya, Costa Rica. Anales del Instituto de Cien- fects on the marine biota. Studies are still needed cias del Mar y Limnología (Universidad Nacional to determine the fishery and aquacultural po- Autonoma de México) 13:329–34. tential of the GN’s 215 identified fish species Campos, J., and C. Gamboa. 1989. An artificial tire (Bartels et al. 1983). Considering that fisheries reef in a tropical marine system: A management tool. Bulletin of Marine Science 44:757–66. in the GN have reached their capacity, fishing Campos, J., B. Burgos, and C. Gamboa. 1984. Ef- prohibitions and enforcement are insufficient, fect of shrimp trawling on the commercial the existing fishing capacity is larger than nec- ichthyofauna of the Gulf of Nicoya, Costa Rica. essary, and the population is growing relentlessly Revista de Biología Tropical 32:203–7.

132 VARGAS AND MATA frankie chap 10 8/20/03 7:56 PM Page 133

Chaves, J., and M. Birkicht. 1996. Equatorial sub- ———. 1993. Cambios en los hábitos alimentarios surface water and the nutrient seasonality dis- de Callinectes arcuatus (Crustacea: Decapoda) en tribution of the Gulf of Nicoya, Costa Rica. Re- el Golfo de Nicoya, Costa Rica. Revista de Biología vista de Biología Tropical 44(Suppl. 3):41–47. Tropical 41:639–46. Córdoba-Muñoz, R. 1998. Primary productivity Dittel, A., and C. Epifanio. 1984. Growth and de- in the water column of Estero Morales, a man- velopment of the portunid crab Callinectes arcu- grove system in the Gulf of Nicoya. Revista de atus Ordway: Zoeae, megalopae and juveniles. Biología Tropical 46(Suppl. 6):257–62. Journal of Crustacean Biology 4:491–94. Cruz, R. A. 1996. Annotated list of marine mol- Dittel, A., C. E. Epifanio, and J. B. Chavarría. 1985. lusks collected during the R.V. Victor Hensen Population biology of the portunid crab Call- Costa Rica Expedition, 1993–1994. Revista de inectes arcuatus Ordway in the Gulf of Nicoya, Biología Tropical 44(Suppl. 3):59–67. Costa Rica; Central America. Estuarine, Coastal, Cutler, N., E. Cutler, and J. A. Vargas. 1992. Peanut and Shelf Science 20:593–602. worms (Phylum Sipuncula) from Costa Rica. Dittel, A., C. E. Epifanio, L. A. Cifuentes, and Revista de Biología Tropical 40:153–58. D. Kirchman. 1997. Carbon and nitrogen Dean, H. K. 1996. Subtidal benthic polychaetes sources from shrimp postlarvae fed natural di- (Annelida) of the Gulf of Nicoya, Costa Rica. ets from a tropical mangrove system. Estuarine, Revista de Biología Tropical 44(Suppl. 3):69–80. Coastal, and Shelf Science 45:629–37. ———. 1998. The Pilargidae (Polychaeta) of the Duncan, R. S., and W. A. Szelistowski. 1998. Pacific Coast of Costa Rica. Revista de Biología Influence of puffer predation on vertical distri- Tropical 46(Suppl. 6):47–62. bution of mangrove littorinids in the Gulf of ———. 2001a. Some Nereididae (Annelida: Poly- Nicoya, Costa Rica. Oecologia 117:433–42. chaeta) from the Pacific coast of Costa Rica. Echeverría, J., A. Echeverría, and A. Mata. 1998. Revista de Biología Tropical 49(Suppl. 2):37–67. Plan de acción para la Cuenca del Río Tempisque: ———2001b. Capitellidae (Annelida: Polychaeta) Antecedentes del estudio y resumen ejecutivo. San from the Pacific coast of Costa Rica. Revista de José: Tropical Science Center/Association for Biología Tropical 49(Suppl. 2):69–84. the Tempisque River Management. 39 pp. Dean, H. K., and E. B. Cutler. 1998. Range exten- Emig, C., and J. A. Vargas. 1990. Glottidia audebarti sion of Nephasoma pellucidum, and new records (Broderip) (Brachiopoda, Lingulidae) from the of Apionsoma (Edmonsius) pectinatum, Sipun- Gulf of Nicoya, Costa Rica. Revista de Biología cula and Thalassema steinbecki (Echiura) from Tropical 38:251–58. the Pacific of Costa Rica. Revista de Biología Epifanio, C. E., and A. Dittel. 1982. Comparison of Tropical 46(Suppl. 6):279–80. dispersal of crab larvae in Delaware Bay (USA) Dean, H. K., D. Maurer, J. A. Vargas, and C. H. Tins- and Gulf of Nicoya (Central America). In Estu- man. 1986. Trace metal concentrations in sed- arine comparisons, ed. V. Kennedy, 447–87. New iment and invertebrates from the Gulf of Nicoya, York: Academic Press. Costa Rica. Marine Pollution Bulletin 17:128–31. Epifanio, C. E., D. Maurer, and A. I. Dittel. 1983. De la Cruz, E. 1989. Organochlorine and total Seasonal changes in nutrients and dissolved mercury in the Anadara tuberculosa bivalve and oxygen in the Gulf of Nicoya, a tropical estuary other species from the Nicoya Gulf, Costa Rica. on the Pacific coast of Central America. Hydro- Master’s thesis, Free University of Brussels, biologia 101:231–328 Belgium. 58 pp. Fuller, C. C., J. A. Davis, D. J. Cain, P. J. Lamothe, De la Cruz, E., and J. A. Vargas. 1986. Estudio pre- T. L. Fries, G. Fernandez, J. A.Vargas, and M. M. liminar de la meiofauna de la playa fangosa de Murillo. 1990. Distribution and transport of Punta Morales, Golfo de Nicoya, Costa Rica. Bre- sediment bound metal contaminants in the Río nesia 25–26:89–97. Grande de Tárcoles, Costa Rica. Water Research DeVries, M., C. E. Epifanio, and A. Dittel. 1983. Re- 24:805–12. productive periodicity of the tropical crab Call- GESAMP (IMO/FAO/UNESCO-IOC/WMO/WHO/ inectes arcuatus Ordway in the Gulf of Nicoya, IAEA/UN/UNEP Joint Group of Experts on Costa Rica. Estuarine, Coastal, and Shelf Science the Scientific Aspects of Marine Environmental 17:709–16. Protection). 1996. The contributions of science to Dittel, A. 1991. Distribution, abundance and sexual coastal zone management. Rep. Stud. GESAMP composition of stomatopod Crustacea in the 61. 66 pp. Gulf of Nicoya, Costa Rica. Journal of Crustacean Gocke, K., M. Vitola, and G. Rojas. 1981. Oxygen Biology 11:269–76. consumption patterns in a mangrove swamp

WHERE THE DRY FOREST FEEDS THE SEA 133 frankie chap 10 8/20/03 7:56 PM Page 134

on the Pacific coast of Costa Rica. Revista de Maurer, D., and J. A. Vargas. 1984. Diversity of soft- Biología Tropical 29:143–54. bottom benthos in a tropical estuary: Gulf of Gocke, K., J. Cortes, and C. Villalobos. 1990. Ef- Nicoya, Costa Rica. Marine Biology 81:97–106. fects of red tides on oxygen concentration and Maurer, D., C. E. Epifanio, H. K. Dean, S. Howe, distribution in the Gulf of Nicoya, Costa Rica. J. A. Vargas, A. Dittel, and M. Murillo. 1984. Revista de Biología Tropical 38:401–7. Benthic invertebrates of a tropical estuary: Gulf Hidalgo, C. C. 1986. Determinación de residuos of Nicoya, Costa Rica. Journal of Natural History de plaguicidas organoclorados en huevos de 18:47–61. ocho especies de aves acuáticas, colectados du- Maurer, D., H. K. Dean, and J. A. Vargas. 1987. Soft rante 1983–1984 en la Isla Pájaros, Guanacaste, bottom invertebrate communities from the Costa Rica. Master of Science thesis, Universi- Gulf of Nicoya, Costa Rica. Memorias del V Sim- dad de Costa Rica, Costa Rica. 120 pp. posio de Biología Marina, Universidad Autonoma Hossfeld, B. 1996. Distribution and biomass of de Baja California, Sur: 107–12. arrow worms (Chaetognatha) in the Golfo de McGuinn, A. P. 1998. Rocking the boat: Conserv- Nicoya and Golfo Dulce, Costa Rica. Revista de ing fisheries and protecting jobs. Worldwatch Biología Tropical 44(Suppl. 3):157–72. Paper No. 142. Washington, D.C.: Worldwatch Koch, V., and M. Wolff. 1996. The mangrove snail Institute. Thais kioskiformis (Duclos): A case of life history Mielke, W. 1992. Six representatives of the Tetrag- adaptation to an extreme environment. Journal onicipitidae (Copepoda) from Costa Rica. Micro- of Shellfish Research 15:421–32. fauna Marina 7:101–46. Lai, H. L., M. Mug-Villanueva, and V. F. Gallucci. Molina-Ureña, H. 1996. Ichthyoplankton assem- 1993. Management strategies for the tropical blages in the Gulf of Nicoya and Golfo Dulce corvina reina, Cynoscion albus, in a multi-mesh embayments, Pacific coast of Costa Rica. Revista size gillnet artisanal fishery. In Proceedings of the de Biología Tropical 44(Suppl. 3):173–82. International Symposium on Management Strate- Morales, R. A., and J. A. Vargas. 1995. Especies gies for Exploited Fish Populations, 21–38. Alaska comunes de copepodos (Crustacea: Copepoda) Sea Grant College Program, AK-SG-93-02. pelágicos del Golfo de Nicoya. Revista de Bio- Lizano, O. 1998. Dinámica de las aguas en la parte logía Tropical 43:207–18. interna del Golfo de Nicoya ante las descargas Pacheco, A. V. 1987. Evaluación preliminar del del Río Tempisque. Revista de Biología Tropical Río Tiribí, 1981–82. Tecnología en Marcha (Costa 46(Suppl. 6):11–20. Rica) 2–3:5–10. López, M. 1983. Lycodontis verrilli (Pisces: Murae- Perry, D. 1988. Effects of associated fauna on growth nidae) descripción de su larva leptocéfala del and productivity in the red mangrove. Ecology Golfo de Nicoya, Costa Rica. Revista de Biología 69:1064–75. Tropical 31:343–44. Peterson, C. L. 1960. The physical oceanography López, M., and C. Arias. 1987. Distribución tem- of the Gulf of Nicoya, Costa Rica, a tropical es- poral y espacial del ictioplancton en el estuario tuary. Bulletin of the Interamerican Tropical Tuna de Pochote, Bahía Ballena, Pacífico de Costa Commission 4:139–216. Rica. Revista de Biología Tropical 35:121–26. Quesada, C., ed. 1990. Estrategia de conservación López, M., and W. Bussing. 1998. Urotrygon cimar, para el desarrollo sostenible de Costa Rica. Minis- a new eastern Pacific stingray (Pisces: Urolo- terio de Recursos Naturales Energía y Minas phidae). Revista de Biología Tropical 46(Suppl. 6): (MIRENEM). San José: Servicios Litográficos. 371–77. 163 pp. Ludicello, Z., M. Weber, and R. Wieland. 1999. Ramírez, A. R., W. A. Szelistowski, and M. López. Fish, markets, and fishermen: The economics of over- 1989. Spawning pattern and larval recruitment fishing. Washington, D.C.: Island Press. 193 pp. in Gulf of Nicoya anchovies (Pisces: Engrauli- Marques, F., D. R. Brooks, and S. Monks. 1995. dae). Revista de Biología Tropical 37:55–62. Five new species of Acanthobothrium van Bene- Ramírez, A. R., M. López, and W. A. Szelistowski. den, 1849 (Eucestoda: Tetraphylidea: Oncho- 1990. Composition and abundance of ichthyo- bothiidae) in stingrays from the Gulf of Nicoya, plankton in a Gulf of Nicoya mangrove estuary. Costa Rica. Journal of Parasitology 81:942–51. Revista de Biología Tropical 38:463–66. Mata, A., and O. Blanco. 1994. La Cuenca del Golfo Ramírez, J. M. 1987. Metales pesados en los ríos de Nicoya: Reto al desarrollo sostenible. San José: Virilla y Grande de Tárcoles. Ingeniería y Cien- Editorial de la Universidad de Costa Rica. 235 pp. cia Química 11(2–3):57–59.

134 VARGAS AND MATA frankie chap 10 8/20/03 7:56 PM Page 135

———. 1999. Contaminación de ríos en la Cuenca mud flat in the Gulf of Nicoya, Costa Rica. Re- Virilla-Tárcoles. In San José: Gloria, ocaso y revista de Biología Tropical 37:207–11. cate, ed. E. Flores, 50–59. Memoria de la Acad- ———. 1995. The Gulf of Nicoya estuary, Costa emia Nacional de Ciencias, vol. 3. San José: Rica: Past, present and future cooperative re- Academia Nacional de Ciencias. search. Helgoländer Meeresunters 49:821–28. Robinson, T. 1993. Fate and transport of agricultural ———. 1996. Ecological dynamics of a tropical in- contaminants from rice paddies, Guanacaste, Costa tertidal mudflat community. In Estuarine shores: Rica. Servicio Nacional de Riego y Avenamiento, Evolution, environments and human alterations, Costa Rica. Santa Barbara: University of Cali- ed. K. F. Nordstrom and C. T. Roman, 355–71. fornia. 48 pp. London: John Wiley and Sons. Rostad, T., and K. L. Hansen. 1999. The effects Vargas, R. C., S. Jesse, and M. Castro. 1996. Check- of trawling on the benthic fauna of the Gulf of list of crustaceans (Decapoda and Stomatopoda) Nicoya, Costa Rica. Revista de Biología Tropical collected during the Victor Hensen Expedition 49(Suppl. 2):91–95. (1993/1994). Revista de Biología Tropical 44 Solórzano, R., R. de Camino, R. Woodward, J. Tosi, (Suppl. 3):97–102. V. Watson, A.Vazquez, C. Villalobos, J. Jimenez, Villalobos, C., G. Cruz, and R. A. Cruz. 1985. Notas R. Repetto, and W. Cruz. 1991. Accounts overdue: sobre la biología de Sphaeroma terebrans Bate Natural resource depreciation in Costa Rica. San 1966 (Sphaeromatidae, Isopoda) en el manglar Pedro and Washington, D.C.: Centro Científico de Pochote, Provincia de Puntarenas, Costa Rica. Tropical and World Resources Institute. 110 pp. Brenesia 24:287–96. Soto, R. 1988. Geometry, biomass allocation and Voorhis, A., C. E. Epifanio, D. Maurer, A. I. Dittel, leaf demography of Avicennia germinans (L.) and J. A. Vargas. 1983. The estuarine charac- L. (Avicenniaceae) along a salinity gradient in ter of Gulf of Nicoya, an embayment on the Salinas, Puntarenas, Costa Rica. Revista de Bio- Pacific Coast of Central America. Hydrobiologia logía Tropical 36:309–23. 99:225–37. Szelistowski, W. A. 1990. A new clingfish (Tele- Wehrtmann, I., and A. Dittel. 1990. Utilization of ostei: Gobiesocidae) from mangroves of Costa floating mangrove leaves as a transport mecha- Rica, with notes on its ecology and early devel- nism of estuarine organisms, with emphasis opment. Copeia 1990:500–507. on Decapod Crustacea. Marine Ecology Progress Szelistowski, W. A., and J. Garita. 1989. Mass mor- Series 60:67–73. tality of sciaenid fishes in the Gulf of Nicoya, Wolff, M. 1996. Demersal fish assemblages along Costa Rica. Fishery Bulletin 87:363–65. the Pacific coast of Costa Rica: A quantitative Thorne, R. E., J. B. Hedgepeth, and J. Campos. and multivariate assessment based on the Vic- 1989. Hydroacoustic observations of fish and tor Hensen Costa Rica Expedition (1993–1994). behavior around an artificial reef in Costa Rica. Revista de Biología Tropical 44(Suppl. 3):187–214. Bulletin of Marine Science 44:1058–64. Wolff, M., H. J. Hartmann, and V. Koch. 1996. A Vargas, J. A. 1987. The benthic community of an pilot trophic model for Golfo Dulce, a fjord-like intertidal mud flat in the Gulf of Nicoya, Costa tropical embayment, Costa Rica. Revista de Bio- Rica: Description of the community. Revista de logía Tropical 44(Suppl. 3):215–231. Biología Tropical 35:229–316. Wolff, M., J. B. Chavarría, V. Koch, and J. A. Vargas. ———. 1988. A survey of the meiofauna of an 1998. A trophic flow model of the Golfo de eastern tropical Pacific intertidal mud flat. Re- Nicoya, Costa Rica. Revista de Biología Tropical vista de Biología Tropical 36:541–44. 46(Suppl. 6):63–79. ———. 1989. Seasonal abundance of Coricuma nicoyensis (Crustacea: Cumacea) on an intertidal

WHERE THE DRY FOREST FEEDS THE SEA 135 frankie chap 11 8/20/03 7:57 PM Page 136

chapter 11

Mangrove Forests under Dry Seasonal Climates in Costa Rica

Jorge A. Jiménez

long the 6,600-km coastline of Central further reduces the amount of appropriate habi- AAmerica, mangrove forests are a distinctive tats. As a result, mangrove areas on this coast are element in the landscape. More than 340,000 ha limited to small isolated patches in some river of mangroves exist along the Pacific coast, and mouths and coastal lagoons. 225,000 ha are found on the Caribbean coast This chapter summarizes available informa- of the isthmus. These forests represent around tion on biological characteristics and conserva- 7 percent of Central America’s natural forest tion issues associated with mangrove forests coverage and 8 percent of the world’s mangroves under dry seasonal climates in Costa Rica. I em- (CCAD 1998). phasize the impact that dry seasonal climates More than 20 percent of the human popu- have on diversity patterns and the structural and lation of Central America lives near coastal functional attributes of these ecosystems. I also areas, and more than 200,000 people directly discuss the strong dependence of these forests depend on coastal fishery industries in Central and their associated fauna on the surrounding America (CCAD 1998). Mangrove forests have, ecosystems. Finally, I offer recommendations therefore, high socioeconomic relevance in the on key areas needing research and integrative region. approaches to coastline management. In Costa Rica, mangrove forests are almost exclusively located on the Pacific coast of the MANGROVES ON THE country. The Caribbean coastline, lacking inlets, PACIFIC COAST OF COSTA RICA bays, and other wave-protected areas, does not provide suitable habitats for mangrove develop- Within the Pacific region, current climatic pat- ment. In addition, a tidal range of only 30 cm terns have produced two different associations:

136 frankie chap 11 8/20/03 7:57 PM Page 137

mangroves growing under rainy, less seasonal In the internal zone of the forest, runoff and climates and mangroves growing under dry sea- seepage are absent for six to eight months of the sonal climates (Jiménez 1994). Mangrove forests year. Soil salinity therefore tends to increase af- in and north of the Gulf of Nicoya are influenced ter the rainy season and with distance away from by strongly seasonal precipitation averaging be- the tidal channels. Such spatial gradients can tween 1,500 and 2,000 mm per year, whereas vary from 35 parts per thousand (ppt) close to those south of the gulf receive between 2,500 the tidal channel to 163 ppt in the innermost ar- and 4,000 mm per year in a less seasonal pat- eas of the forest (Soto and Corrales 1987). The tern. Floristic differentiation of these two asso- innermost sections are not only the ones with ciations is evident at the internal edge of the the highest soil salinities during the dry season forests. Here floral elements are more diverse but also the ones with highest variations in soil under rainy climates. Differences in floristic di- salinity between dry and wet seasons. When soil versity are also the result of climatic influence salinity reaches above 100 ppt, vegetation disap- on hydrological regimes within mangrove forests. pears, giving way to salt flats; some of these are Surface water running into mangroves (runoff) quite extensive, such as the 583 ha of salt flats at and subsurface water infiltrating mangrove soils the Gulf of Nicoya (Solórzano et al. 1991). (seepage) modify interstitial salinity patterns. During dry months, runoff and seepage are re- STRUCTURAL ATTRIBUTES duced, resulting in an increase in soil salinity As with species diversity, structural attributes that is due to salt concentration, which in turn is exhibit the largest variations along the internal a result of evapotranspiration processes. During zone. Under dry seasonal climates most struc- rainy months soil salinity is reduced when salt tural attributes (such as height, diameter, basal is washed away by higher runoff and seepage. area, wood volume, and bark volume) show a re- Species diversity (especially in the interior of duction with distance away from the channel. For mangrove forests) is higher where fresh water is Rhizophora species it is not uncommon to find abundant and constant. reductions in wood volume (from 95 m3/ha to 20 m3/ha) and in bark volumes (from 8 m3/ha PLANT SPECIES DISTRIBUTION to 2 m3/ha) in the first 100 m away from the channel (Jiménez 1994). As soil salinity increases About 46 percent (19,000 ha) of the mangrove landward, plants need to lock more amino acids forests in Costa Rica are growing under dry sea- and other organic compounds into their cyto- sonal climates. The dominant floristic elements plasm to achieve proper osmotic balance (Tom- in these forests are the trees: Rhizophora race- linson 1986). These compounds are therefore mosa, R. mangle, Laguncularia racemosa, Avicen- not utilized in tissue growth, which results in nia germinans, and A. bicolor. lower structural development of the forest (Scho- Some floristic elements are restricted to man- lander et al. 1965). groves under dry seasonal climates. A. bicolor is In areas where the internal zone receives limited to these mangroves, where it can be a abundant runoff and seepage, the structural de- dominant forest component. Clerodendrum pit- velopment is higher, particularly in the A. bicolor tieri and Sesuvium portulacastrum are herbs also stands, located in the landward edge. Basal areas restricted to this type of mangrove, but neither of around 40 m2/ha can be found in these is a dominant element in the forest. Some plant stands (Jiménez 1990; Jiménez and Sauter 1991). species (Caparis odoratissima, Heliotropium curassavicum, and Philoxerus vermicularis) associated FAUNAL ELEMENTS with the internal zone of the forest are also restricted to dry seasonal mangrove sites (Jiménez Because they constitute an ecotone between ter- and Soto 1985). restrial and coastal ecosystems, mangrove forests

MANGROVE FORESTS 137 frankie chap 11 8/20/03 7:57 PM Page 138

show faunal elements from both systems. In Costa Rica, where it is common in Pelliciera rhi- mangrove forests, the avian community is less zophorae stands. Dendroica erithacorides, an in- diverse than in other forest types. Nevertheless, sectivore warbler, and the Mangrove Vireo (Vireo more than 200 species of birds are associated pallens) are restricted to mangrove habitats both with mangrove forests in Costa Rica (Jiménez in rainy and dry climates and can be found 1994). In highly seasonal mangroves, however, throughout the eastern Pacific mangroves. bird diversity is lower; about 90 species use these Insects in mangrove forests have received forests. Mudflats and salt evaporation ponds at little study. The moth Lygropia erythrobathrum the external edge of mangroves are used as feed- (Pyralidae) eats the mesophyl of Avicennia leaves, ing grounds by more than 35 species of birds leaving only the translucent cuticle in many of that feed on benthic invertebrates (Alvarado- the tree crowns. Avicennia seedlings are eaten Quesada and Moreno 1997). Around 64 bird by massive groups of Junonia evarete (Nymphal- species (mainly insectivores) use Avicennia stands idae; Turner and Parnell 1985). In A. germinans (A. germinans and A. bicolor alike) at the internal leaf galls are induced by three different species edge of the forest (Warkentin and Hernández of flies (Cecidomyiidae), by Telmapsylla minuta 1995). (Homoptera: Psylloidea), and by an unidentified Migratory species are an important compo- mite (P. Hanson pers. comm.). Interestingly, nent of mangroves. Of the 64 bird species re- galls seem to be restricted to A. germinans indi- ported only in Avicennia stands, 31 percent are viduals, whereas A. bicolor adjacent plants are Nearctic migrants. This percentage is higher than gall-free. In Laguncularia racemosa leaf galls are the 24 percent for migrants present in the total produced by Limbopsylla lagunculariae (Homop- avifauna of Costa Rica. Migratory species can tera: Psylloidea). account for 64 percent of the bird individuals Mammals in dry seasonal mangrove forests, found in a mangrove forest (Warkentin and Her- although not present in large populations, in- nández 1995). The high proportion of migrants clude frequently observed species such as the emphasizes the value of mangrove forests as collared anteater (Tamanduamexicana), the white- stopover habitats for migrating birds. They use faced monkey (Cebus capuchinus), the water this habitat largely as foraging areas or as night- opossum (Chironectes minimus), and the north- time roosts during their annual migrations ern raccoon (Procyon lotor). Occasionally, large (chapter 12). mammals such as the white-tailed deer (Odocoil- Local species of high relevance to conserva- eus virginianus) and the jaguarundi (Felis jaguar- tion are also associated with mangrove habitats. undi) are observed within Avicennia forests. Some are permanent residents, whereas others Crustaceans, mollusks, and fishes dominate are only seasonal or occasional visitors. The the estuarine component of the mangrove macro- Scarlet Macao (Ara macao) at the Carara Biolog- fauna. More than 50 species of crabs are found ical Reserve roosts at night in the Guacalillo throughout the mangrove forest (Jiménez 1994). mangrove forest while feeding during the day in Several species, such as Callinectes arcuatus and the nearby rain forest (Vaughan et al. 1991). The C. toxotes, extend their distribution into adjacent Yellow-naped Parrot (Amazona ochrocephala), an estuarine environments, where their swimming endangered species, uses hollow trunks of Avi- capabilities allow them to thrive. Others crab cennia forests during the nesting season and species are ground-dwellers (e.g., Sesarma sul- spends the rest of the year outside the mangrove catum) restricted to the forest floor and decom- forest. posing trunks. Land crabs (e.g., Cardisoma cras- No endemic bird species is reported for dry sum and Gecarcinus quadratus) have developed seasonal mangroves in Costa Rica, although the physiological and anatomical properties that hummingbird Amazilia boucardii is endemic to allow them to thrive in more terrestrial environ- the rainy nonseasonal mangroves of southern ments away from water bodies. Massive migra-

138 JIMÉNEZ frankie chap 11 8/20/03 7:57 PM Page 139

tion between adjacent dry forests and mangrove other groups of species, such as the gastropods areas is part of their reproductive cycle. Other Theodoxus luteofasciatus and Melampus carolianus, crabs are highly specialized for arboreal life, such as well as several Uca crab species are associated as Aratus pisonii, which leave mangrove trees with the pneumatophores of Avicennia species. and move into coastal waters during the mating season. SEASONAL PATTERNS Regarding mollusks, 78 species of bivalves and 46 species of gastropods are reported in Sharp seasonality in water availability in dry sea- mangroves of the Pacific coast of Central Amer- sonal mangrove forests regulates many of its ica (Cruz and Jiménez 1994). Several, such as functional attributes. Growth patterns are clearly Grandiarca grandis, Anadara tuberculosa, Chione seasonal in these forests. Height and diameter subrugosa, Protothaca asperrima, and Mytella increase during June–October, when ground- guyanensis, have high commercial value. Many water recharge is highest and soil salinities low- species are restricted to specific habitats. Some est. This growth seasonality results in lower an- gastropods, such as Detracia graminea and El- nual growth rates than those exhibited by trees lobium stagnalis, are restricted to decomposing under more rainy climates (Jiménez 1988, 1990). trunks. Other species, such as the gastropods Leaf fall increases at the beginning of the dry Marinula concinna and Cerithidea pulchra, are season (December–January), followed by a sec- associated with decomposing litterfall in areas ond leaf drop event at the onset of the rainy dominated by Rhizophora species. More than season (May), just before a flush of new leaves 125 species of fish have been reported from man- occurs. Herbivorous populations closely follow grove areas in Costa Rica (Szelistowski 1990; this seasonal pattern. The moth Junonia evarete Jiménez 1994). Estuarine fish families such as exhibits high densities around September, which Gerreidae, Engraulidae, Tetraodontidae, Ather- is closely linked to the seasonal establishment of inidae, Ariidae, and Centropomidae dominate a new cohort of Avicennia in the seedling bank. the ichthyofauna. The often cited migration of Moth density declines by November. fish species in early stages of development be- Flowering in A. bicolor forests starts in De- tween coastal waters and mangrove areas holds cember and extends to late January, during the for most fish species in the Gulf of Nicoya man- onset of dry conditions. The hummingbird (Ama- groves, underscoring the ecological connection zilia boucardii) populations visiting and (sup- between mangroves and coastal waters. A large posedly) pollinating Pelliciera rhizophorae may percentage of these fish species support arti- have seasonal visitation patterns, since P. rhizo- sanal and subsistence fisheries within the gulf phorae flowering is clearly seasonal. Climatic and adjacent coastal waters. Faunal elements anomalies such as El Niño result in significant within mangrove forest are clearly associated alterations of phenological cycles. Massive flower with specific zones along the flooding gradient. abortion has been associated with severe drought Bivalves such as Melongena patula, Natica chem- conditions (Jiménez 1990). initzii, and Chione subrugosa, most fish species, When runoff from the onset of the rains and crabs such as Xanthidae or Portunidae starts in May–June, fruit development sharply species are largely limited to the seaward areas increases, reaching maturity in mid-August, of the forest in close contact with estuarine en- when water levels in the forest are highest. vironments under daily tidal influence. Mollusk Propagule crops in Avicennia bicolor vary dra- species such as A. tuberculosa, Polymesoda inflata, matically from year to year, from 83 to 380 Littoraria zebra, Cerithium stercusmuscarum, and propagules per square meter (Jiménez 1990). Cerithidea montagnei are restricted to the exter- Crab populations feeding on propagules seem nal edge of the forest among prop roots of Rhi- to follow these seasonal patterns of propagule zophora trees. In interior sections of the forest, production.

MANGROVE FORESTS 139 frankie chap 11 8/20/03 7:57 PM Page 140

Seedling establishment for Avicennia species the canopy, migratory bird species escaping the occurs within two weeks after propagule drop. boreal winter become abundant during the dry Established seedlings show a clear seasonal season, sometimes surpassing local residents pattern of mortality, with about 35 percent of a in number, whereas the white-faced monkeys cohort dying during the following dry season. migrate to areas of the forest adjacent to fresh- Both establishment and mortality levels vary water courses. In water channels fish and crus- along the flooding gradient. The more severe taceans show seasonal migration patterns be- conditions in the innermost edge of the forest tween mangroves and adjacent coastal areas result in higher mortality levels during the dry (Szelistowski 1990). These relationships high- season. Drought seasonality along the flooding light the close dependency that mangroves have gradient also regulates differential survival of on adjacent habitats. Avicennia and Rhizophora seedlings, reinforcing zonation patterns within the forest (Jiménez CONSERVATION STATUS and Sauter 1991). The seasonal nature of mangrove forests is Mangrove forests in Costa Rica have been pro- also reflected in the faunal communities. Many tected under legislation such as the Coastal- species, especially in the internal edge of the Maritime Law (1977), the Wildlife Law (1992), forest, sharply reduce their densities during the and more recently the Environment Law (1995). dry season, when the ground becomes drier and A series of executive decrees during the past three saltier owing to a decrease in runoff and rainfall. decades have also regulated specific aspects of The gastropod Theodoxus luteofasciatus reduces mangrove conservation and management. its population densities from more than 300 in- This legal framework has proved effective in dividuals per square meter to fewer than 1 indi- preventing mangrove destruction in most areas. vidual per square meter. Similar crashes in pop- However, reclamation has occurred at specific ulation density are observed in gastropods such sites where pressure for land or resources is high as Melampus carolianus and many fiddler-crab and enforcement has been limited. The Gulf species (Uca spp.). Because tidal inundation is of Nicoya mangroves (around 14,000 ha) have reduced in accordance with a reduction in river been the most affected; around 7 percent of the flow during the dry season, large sections of the gulf’s original coverage has been lost as a result mangrove floor become drier, even in the exter- of shrimp cultivation ponds (630 ha) and salt nal edge of the forest. Ground-dweller mollusks evaporation ponds (350 ha). This reclamation such as Cerithidea pulchra, Littoraria zebra, and has occurred mainly at the expense of Avicennia the bivalve Polymesoda inflata also reduce their germinans stands in the internal zone of the numbers. With the onset of the wet season, forests (Solórzano et al. 1991). The shrimp op- insect populations and ground-dweller mollusks erations at Chomes and salt evaporation ponds (e.g., T. luteofasciatus) and crustaceans (e.g., around Jicaral are responsible for the greatest Sesarma spp.) rapidly colonize the forest again. impact. Mating and nesting of canopy birds such as Small areas (1–6 ha each) have been devel- Dendroica erithacorides and Amazona ochrocephala oped for hotel and marina sites in the Nicoya coincide with the start of the rainy season when Peninsula in a piecemeal fashion. Landfilling food availability is higher (Barrantes 1998). of mangroves in Samara, Conchal, and Playa Seasonal faunal movements between man- Grande for hotel construction is an example of groves and surrounding environments are com- this destructive process. In these cases, however, mon. Some mangrove crabs (Gecarcinus quad- the total number of hectares destroyed has been ratus, Cardisoma crassum, Ucides occidentalis) relatively low (20–30). show seasonal migrations between mangroves Although total area destroyed is compara- and nearby land forests for mating purposes. In tively low, mangrove forests, water quality, and

140 JIMÉNEZ frankie chap 11 8/20/03 7:57 PM Page 141

associated faunal communities have been more for sugarcane irrigation in the lower watershed. extensively degraded. Existing legislation pro- Ecological consequences of these alterations are vides little support to prevent this environ- largely unknown. mental degradation. Socioeconomic processes The quality of freshwater discharges is also are closely associated with this problem. Eco- severely deteriorated in most river basins. nomical recession and armed conflicts in a Sediment-bound metal contaminants such as neighboring country have resulted in heavy chromium, originating from leather tanneries human immigration to the mangrove areas dur- in the Central Valley, find their way into man- ing the past two decades. A higher number of grove sediments at the Gulf of Nicoya via the people living at subsistence levels seek man- Grande de Tárcoles River (or Tárcoles River) grove areas as sources of firewood, poles, char- (Fuller et al. 1990). Coliforms from untreated coal, shellfish, crustaceans, and other products. sewage disposals also are affecting areas such The slower growth rates of mangrove forests as the Guacalillo and Chacarita mangroves (see in dry seasonal climates make their conserva- chapters 9 and 10). Large amounts of solid waste tion harder in areas of increasing demand for (especially plastic debris) are transported by the wood resources. Areas such as Estero Pitahaya, Tárcoles River and distributed by longshore drift Costa de Pajaros, and Puerto Thiel exhibit forest currents into nearby Guacalillo, Tivives, and Ba- degradation as a result of wood extraction for jamar mangrove areas. poles, firewood, and charcoal. Overexploitation Agroindustrial processes in the associated of faunal resources is notorious and widespread. basins are also a main source of pollution. A species of mangrove cockle (Grandiarca gran- Sugarcane processing has resulted in strong dis) is listed as endangered as a result of over- seasonal variations in water quality within the exploitation and has almost disappeared within Chacarita estuary. Dissolved oxygen can drop to the Gulf of Nicoya. Around 8 million individuals zero during harvest season, resulting in massive of another seacockle species (Anadara tubercu- mortality of fish, crustaceans, and other associ- losa) are harvested yearly from mangroves in the ated fauna in the mangrove estuary (Blanco and Gulf of Nicoya (Solórzano et al. 1991). A histori- Mata 1994). The quality of some faunal resources cal reduction in the average size of captures in- (such as bivalves, crabs, and fishes) extracted from dicates overexploitation of the resource. Shrimp mangrove areas for human consumption is likely species (such as Penaeus stylirostris) also show a affected. The absence of long-term monitoring reduction in the average size of captured indi- programs on water quality is lamentable. viduals (Palacios et al. 1996). Although legislation forbids the use of gill nets within mangrove RECOMMENDATIONS areas, illegal fishing is widespread, despite analysis showing overexploitation of fish populations. Interactions between mangroves and surround- Artisanal fisheries within the Gulf of Nicoya ing environments highlight the need for further have been reduced by 40 percent, and the num- analysis of the processes regulating these inter- ber of fishing boats has doubled (Foer and Olsen actions. Little is understood about the depend- 1992). ence of mangrove organisms on surrounding The close linkage between dry seasonal man- environments. Several elements of the avian groves and associated river basins as sources of community (such as the Mangrove Humming- freshwater discharges poses a major conserva- bird and the Mangrove Warbler) seem to depend tion challenge. No programs are, however, being on adjacent forests for seasonal foraging. En- implemented to achieve integrated management. dangered species (such as Ara macao and Ama- The Tempisque River, with its 3,700 ha of asso- zona ocrocephala) may have a strong dependence ciated mangroves, is literally dried out during on mangrove areas. How alterations in the the dry season. Pumping stations extract water surrounding landscape affect mangrove avian

MANGROVE FORESTS 141 frankie chap 11 8/20/03 7:57 PM Page 142

populations is in need of immediate study. Man- A more integral approach to mangrove manage- grove areas seem to function as refuges for ment is required in light of the close depend- fauna displaced from deforested coastal plains. ence of these systems to biophysical processes Some bird species of Dendrocolapidae, the silk in surrounding systems. Closer coordination anteater (Cyclopes dydactilus), and the tree ant- among different isolated government agencies eater (Tamandua mexicana), among others, seem and regional NGOs is urgently needed. The in- to have increased their use of mangrove areas volvement of local communities and the agro- (J. Jiménez pers. obs). More research is needed industrial sector in decision-making processes to elucidate the processes that regulate mammal and management plans is insufficient. The weak use of mangroves and movements between enforcement of existing legislation needs to be coastal plain ecosystems and mangroves. Sea- strengthened. These deficiencies have until now sonal rhythms followed by insect, crab, and mol- prevented an integrated management of man- lusk populations have been studied only super- grove areas. ficially. The strong seasonality in dry seasonal mangrove areas highlights the dependence of many of their communities on freshwater sup- REFERENCES ply. Alteration of hydrological and climatic Alvarado-Quesada, G. M., and L. I. Moreno. 1997. regimes might have a severe impact on these Use of salt evaporation ponds by aquatic birds seasonal patterns. Pollination processes (in Avi- and bird abundance. In Proceedings of the ATB- cennia, Lagunculara, and Pelliciera) are unknown. OTS Symposium: Tropical Biodiversity: Origins and Maintenance, 35. San José: Organization for Besides massive flower abortions, what is the Tropical Studies. impact of El Niño/La Niña phenomena on pol- Barrantes, G. 1998. Biologia y comportamiento de linators and propagule-feeding crabs? Bottom- Dendroica petechia xanthotera. Brenesia 49–50: dwelling mollusks have a close dependence on 61–69. runoff regimes, but no information is available Blanco, O., and A. Mata. 1994. La Cuenca del Golfo on the impact of water diversion on this seasonal de Nicoya: Reto al desarrollo sostenible. San José: Editorial de la Universidad de Costa Rica. fauna. 336 pp. In addition to more information on ecolog- CCAD (Comisión Centroamericana de Ambiente ical interactions, there is a need for long-term y Desarrollo). 1998. Estado del ambiente y los programs that monitor water flows and water recursos naturales en Centroamerica. San José: quality reaching mangrove forests. This type of Comisión Centroamericana de Abiente y De- sarrollo. 179 pp. information will allow a better understanding Cruz, R. A., and J. A. Jiménez. 1994. Moluscos aso- of the linkages between mangrove areas and up- ciados a las áreas de manglar de la costa Pacífica stream ecosystems and processes. The impact of de América Central. Heredia, Costa Rica: Edito- changes in patterns and quality of freshwater dis- rial Fundación UNA. 182 pp. charges on the aquatic and benthic fauna within Delgado, P., J. A. Jiménez, and D. Justic. 1999. Pop- mangrove forests is also largely unknown. ulation dynamics of mangrove Avicennia bicolor on the Pacific coast of Costa Rica. Wetlands Ecol- Mangrove reclamation by tourism infrastruc- ogy and Management 7:113–20. ture, salt evaporation ponds, shrimp farms, and Ellison, A. M., E. J. Farnsworth, and R. E. Merkt. urban development is a threat in many man- 1999. Origins of mangrove ecosystems and the grove areas of northwestern Costa Rica. How- mangrove biodiversity anomaly. Global Ecology ever, deterioration of forest resources, alteration and Biogeography 8(2):95–115. Foer, G., and S. Olsen. 1992. Las costas de Cen- of water quality, overexploitation of faunal re- troAmerica. Washington, D.C.: U.S. Agency for sources, and the diversion of freshwater dis- International Development. 290 pp. charges seem to be more relevant menaces to the Fuller, C. C., J. A. Davis, D. J. Cain, T. L. Fries, conservation of dry-seasonal mangrove forests. G. Fernández, J. A. Vargas, and M. M. Murillo.

142 JIMÉNEZ frankie chap 11 8/20/03 7:57 PM Page 143

1990. Distribution and transport of sediment- on the Caribbean coast of Nicaragua. Rhodora bound metal contaminants in the Río Grande de 93:183–86. Tárcoles, Costa Rica. Water Research 24:805–12. Scholander, P. F., H. T. Ha, E. E. Bradstreet, and Graham, A. 1995. Diversification of gulf/ E. A. Hemmingsen. 1965. Sap pressure in vas- Caribbean mangrove communities through cular plants. Science 148:339–45. Cenozoic time. Biotropica 27:20–27. Solórzano, R., R. de Camino, R. Woodward, J. Tosi, Jiménez, J. A. 1984. A hypothesis to explain the re- V. Watson, A. Vasquez, C. Villalobos, R. Repetto, duced distribution of the mangrove Pelliciera and W. Cruz. 1991. Accounts overdue: Natural re- rhizophorae Tr. & Pl. Biotropica 16:304–8. source depreciation in Costa Rica. Washington, ———. 1988. The dynamics of Rhizophora race- D.C.: World Resources Institute. 111 pp. mosa forests on the Pacific coast of Costa Rica. Soto, R., and L. F. Corrales. 1987. Variación de al- Brenesia 30:1–12. gunas características foliares de Avicennia ger- ———. 1990. The structure and function of dry minans en un grdiente climático y de salinidad. weather mangroves on the Pacific coast of Revista de Biología Tropical 35:245–56. Central America, with emphasis in Avicennia Szelistowski, W. A. 1990. Importance of mangrove bicolor forests. Estuaries 13(2):182–92. plant litter in fish food webs and as temporary, ———. 1994. Los manglares del Pacífico Centro- floating habitat in the Gulf of Nicoya, Costa americano. Heredia, Costa Rica: Editorial FUNA. Rica. Ph.D. diss., University of Southern Cali- 336 pp. fornia. 78 pp. Jiménez, J. A., and K. M. Sauter. 1991. Structure Tomlinson, P. B. 1986. The botany of mangroves. and dynamics of mangrove forests along a flood- Cambridge Tropical Biology Series. Cambridge: ing gradient. Estuaries 14(1):49–56. Cambridge University Press. 419 pp. Jiménez, J. A., and R. Soto. 1985. Patrones re- Turner, T. W., and J. R. Parnell. 1985. The identi- gionales en la estructura y composición florís- fication of two species of Junonia Hubner (Lep- tica de los manglares de la costa Pacífica de idoptera: Nymphalidae): J. evarete and J. genoveva Costa Rica. Revista de Biología Tropical 33:25–37. in Jamaica. Journal of Research on the Lepidop- Palacios, J. A., R. Angulo, and J. A. Rodríguez. tera 24:142–53. 1996. La pesqueria de Penaeus stylirostris en el Vaughan, C., M. McCoy, and M. B. Liske. 1991. Golfo de Nicoya. Revista de Biología Tropical 44: Scarlet macaw (Ara macao) ecology and man- 225–31. agement perspectives in Carara Biological Re- Ramírez, A. R., M. I. López-Sánchez, and W. A. serve, Costa Rica. Miscellaneous Publications from Szelistowski. 1990. Composition and abun- the Center for Study of Tropical Birds 1:23–34. dance of ichthyoplankton in a Gulf of Nicoya Warkentin, I. G., and D. Hernández. 1995. Avi- mangrove estuary. Revista de Biología Tropical faunal composition of a Costa Rican mangrove 38:463–66. forest during the boreal winter. Vida Silvestre Roth, L. C., and A. Grijalva. 1991. New record of Neotropical 4(2):140–43. the mangrove Pelliciera rhizophorae (Theaceae)

MANGROVE FORESTS 143 frankie chap 11 8/20/03 7:57 PM Page 144 frankie chap 12 8/20/03 7:58 PM Page 145

Section C. Biotic Relationships with Other Geographical Areas frankie chap 12 8/20/03 7:58 PM Page 146 frankie chap 12 8/20/03 7:58 PM Page 147

chapter 12

Geographical Distribution, Ecology, and Conservation Status of Costa Rican Dry-Forest Avifauna

Gilbert Barrantes and Julio E. Sánchez

n costa rica the dry forest covers the Santa The average temperature of the region is IElena Peninsula and adjacent areas and some 27.6°C ± 0.7°C and the annual precipitation small areas around the Gulf of Nicoya (Slud 1,619 mm (Gómez 1986). Temperature varies 1980; Gómez 1986; see maps in chapter 1). little throughout year. Precipitation, however, Large tracts of semideciduous forest also occur fluctuates dramatically, with a well-defined dry below 500 m in the northwest of Costa Rica. For season from December to May (some years to the purpose of this chapter, we consider decidu- June) (see chapter 1). Seasonality in the dry-forest ous, semideciduous, and other associated habi- ecosystem largely influences species composi- tats as the dry-forest ecosystem. The ecosystem tion, food resource availability, habitat use, and, includes both terrestrial and aquatic habitats. to a large extent, mutualistic interactions that Deciduous and riparian forests and savanna occur in this life zone. are the most extensive terrestrial habitats, and mangroves, mudflats, lagoons, and swamps are SOURCES OF INFORMATION the most common aquatic habitats. Mangroves and mudflats cover large areas in the Gulf of Most of this chapter’s information on avian Nicoya, and lagoons and swamps are part of species richness, migratory status, habitat use, the complex hydrological system that occurs at and reproduction comes from years of fieldwork the lower basin of the Tempisque River. We by the authors. We complemented our data on excluded more humid areas of the Nicoya Penin- some aspects of the reproduction of some species sula (e.g., Reserva Absoluta Cabo Blanco; Gómez with information from Slud (1964) and Stiles 1986). and Skutch (1989). We also used part of the

147 frankie chap 12 8/20/03 7:58 PM Page 148

TABLE 12.1 Number of Dry-Forest Species (Residents, Migrants, and Species with Populations of Both) Using Aquatic and Terrestrial Habitats

number of species per habitat

aquatic terrestrial total (%)

Residents 37 163 200 (58) Migrants 58 66 124 (36) Resident/migratory 13 8 21 (6) Total 108 237 345

Note: Resident species may or may not reproduce in dry forest.

information recorded and compiled by Stotz et mangroves, lagoons, seasonal marshes, and al. (1996), which helped to limit the Costa Rican rivers. Large extensions of mudflats become ex- dry-forest avifauna to a more regional scale and posed during low tides within the Gulf of Nicoya identify the endemic species. (Barrantes and Pereira 1992). These muddy ex- We also used information on the phenolog- tensions are important stopover and wintering ical patterns of dry-forest plants (Frankie et al. grounds for a large group of migratory aquatic 1974; Opler et al. 1980; see chapter 2) and birds, particularly shorebirds and their relatives. changes in insect abundance (Janzen 1973) as a Invertebrates present in these substrates are the baseline to explain the response of dry-forest food source for thousands of these birds, which, birds to seasonal changes in food abundance. having different prey preferences and using dif- Such information provided a general pattern ferent foraging strategies, coexist in the Gulf of of fluctuation of insects and flower and fruit Nicoya for several months every year (Barrantes abundance. and Pereira 1992). Mangrove forests cover relatively large areas along the coast within the Gulf of Nicoya and GENERAL ASPECTS OF COSTA particularly at the mouth of large rivers (e.g., RICAN DRY-FOREST AVIFAUNA Barranca, Tempisque). This seashore forest, de- The dry-forest avifauna of Costa Rica includes spite its low plant species richness (Jiménez and 345 species. Of this total, 58 percent of the species Soto 1985), is used by a relatively large number are resident, 36 percent are latitudinal migrants, of aquatic birds (Barrantes 1998). Herons, egrets, and about 6 percent have resident and migra- ibises, and some large shorebirds are common tory populations (table 12.1). A large percentage users of mangroves. This forest is the most of dry-forest species (31%) are associated with important habitat for foraging, roosting, and aquatic environments. Yet the proportion of reproduction of several herons (e.g., Black- aquatic birds varies substantially between mi- crowned Night Heron, Yellow-crowned Night grants and residents (table 12.1). About 50 per- Heron, Boat-billed Heron). Other species, al- cent of the latitudinal migrants are aquatic though not as closely associated with man- species, whereas only 19 percent of the resident groves as night herons or the Boat-billed Heron, species use aquatic habitats (table 12.1). rely largely on it to reproduce (e.g., Green- There are five main aquatic environments in backed Heron) and/or roost (e.g., Cattle Egret, the Costa Rican dry-forest ecosystem: mudflats, Little Blue Heron, White Ibis).

148 BARRANTES AND SÁNCHEZ frankie chap 12 8/20/03 7:58 PM Page 149

Freshwater habitats (i.e., lagoons, seasonal species may be considered exclusive to either marshes, rivers) are also of great importance one (table 12.2). A larger trophic diversity is also for numerous resident and migratory aquatic found in these forests (table 12.2). For instance, species. Lagoons and seasonal marshes serve as nectar feeders (e.g., Blue-throated Goldentail, foraging and breeding habitat for some resident Streaked-backed Oriole), frugivores (e.g., Long- species. Some aquatic species, such as the Pin- tailed Manakin), raptors (e.g., Collared Forest nated Bittern and Least Bittern, breed and feed Falcon, Spectacle Owl), and seed eaters (e.g., exclusively in lagoons and seasonal marshes in Yellow-naped Parrot) are all well represented in the Costa Rican dry forest. Many other resident both forests. birds (e.g., Bared-throated Tiger Heron, Wood Mangrove forest has lower bird species rich- Stork, Jabiru, Northern Jacana, Black-bellied ness in comparison with deciduous and riparian Whistling Duck) use these freshwater environ- forests. This is partially explained by the low ments intensively as their foraging areas. Such abundance of fleshy fruits and bird-pollinated sites are especially important during the breed- plants in mangroves (Tomlinson 1986). Con- ing season, as they become primary sources of sequently, nectar- and fruit-feeding birds are food for the birds’ offspring. These two fresh- poorly represented. Insectivores are the group of water habitats are also heavily used as foraging terrestrial birds best represented in mangroves. and resting areas by latitudinal migrants (e.g., This habitat is intensively used by a large group some shorebirds, egrets, herons), especially of insectivorous migrants, particularly warblers, ducks (e.g., Blue-winged Teal). Few species, which in some areas surpass the resident birds either migratory or resident, use rivers as their in abundance (G. Barrantes pers. obs.) primary habitat. The most extensive habitats for terrestrial ORIGIN AND DISTINCTIVENESS OF birds in the dry forest are savanna and riparian, DRY-FOREST AVIFAUNA OF COSTA RICA deciduous, and mangrove forests. Savanna com- bines pasturelands and some extensions that are The avifauna of the Pacific Arid Slope of Meso- in different early successional stages as a result america (PAS) consists of a group of species of deforestation, cattle ranching, and a variety of with different origins that share the capacity to agricultural systems (e.g., sorghum, sugarcane, cope with the environmental conditions of the rice plantations) (Hartshorn 1983; Janzen 1988). dry-forest ecosystem. In Costa Rica, of 345 bird The majority of birds present in this habitat are species that have been recorded in the dry forest resident, and all of them are also present in ad- (J. E. Sánchez unpubl. data), 87 have South Amer- jacent forests. Raptors (e.g., Roadside Hawk, Bay- ican affinity, 38 North American, 26 Central winged Hawk), granivores (e.g., White-collared American, 13 pantropical, 37 Old World, and 144 Seedeater, Inca Dove), and insectivores (e.g., with unknown origin (following Karr 1976). A Lesser Ground Cuckoo, Tropical Kingbird) are large proportion of this avifauna involves either common species in the savanna. species with broad distributions (e.g., Northern Deciduous forest, although greatly destroyed Jacana), which are also present in tropical rain and fragmented by human activities (chapter 1), forests, or seasonal (e.g., Black-hooded Antshrike) still covers relatively large areas in the north- and occasional visitors (e.g., Ornate Hawk-Eagle) western region of Costa Rica, particularly on that enter into dry forest during the rainy season hills and calcareous soils (Janzen 1988). Ripar- (G. Barrantes and J. E. Sánchez pers. obs.). Yet, ian forest, however, is naturally restricted to small even when more typical dry-forest birds are contracts along seasonal and perennial watercourses sidered, the broad biogeographical origin still (Hartshorn 1983). The majority of terrestrial remains as an emergent characteristic of the resident and migratory species occur in both Costa Rican dry-forest avifauna and, in general, riparian and deciduous forests. Thus, very few of the avifauna of the entire PAS region.

AVIFAUNA DISTRIBUTION AND ECOLOGY 149 frankie chap 12 8/20/03 7:58 PM Page 150

TABLE 12.2 Distribution, Habitat, and Forest Dependence of Dry-Forest Bird Species in Costa Rica

forest distribution habitat dependence

Agelaius phoeniceus U.S.–Costa Rica lag None Aimophila botterii U.S.–Costa Rica sav/dec None A. ruficauda Mexico–Costa Rica sav/dec None Amazilia rutila Mexico–Costa Rica dec Low Amazona albifrons Mexico–Costa Rica sav/rip/dec/man Moderate A. auropalliata Mexico–Costa Rica sav/rip/dec/man Moderate Ammodramus savannarum Canada–Ecuador sav None Aratinga canicularis Mexico–Costa Rica sav/rip/dec/man None Arremonops rufivirgatus Mexico–Costa Rica sav/dec None Asturina nitida plagiada U.S.–Costa Rica sav/rip/dec Low Burhinus bistriatus Mexico–Brazil sav None Buteo albicaudatus U.S.–Argentina sav None B. albonotatus U.S.–Argentina sav Low Buteogallus subtilis Mexico–Peru man Moderate Calocitta formosa Mexico–Costa Rica sav/rip/dec/man Low Camptostoma imberbe U.S.–Costa Rica dec Low Campylorhynchus rufinucha Mexico–Costa Rica sav/rip/dec/man Low Caracara plancus U.S.–Argentina sav/rip/dec None Chiroxiphia linearis Mexico–Costa Rica rip/dec Moderate Colinus cristatus leucopogon Guatemala–Costa Rica sav/dec None Columbina inca U.S.–Costa Rica sav/dec/man None Crypturellus cinnamomeus Mexico–Costa Rica rip/dec Moderate Dendroica petechia xanthotera Guatemala–Costa Rica man Moderate Eumomota superciliosa Mexico–Costa Rica rip/dec/man Low Euphonia affinis Mexico–Costa Rica sav/dec Low Glaucidium brasilianum U.S.–Argentina sav/rip/dec Low Guiraca caeruleaa U.S.–Costa Rica sav/dec None Heliomaster constantii Mexico–Brazil sav/dec Low Icterus pectoralis Mexico–Costa Rica dec/man Low I. pustulatus Mexico–Costa Rica dec/rip Low Laterallus ruber Mexico–Costa Rica lag None Melanerpes hoffmannii Honduras–Costa Rica sav/rip/dec/man Low frankie chap 12 8/20/03 7:58 PM Page 151

TABLE 12.2 (continued)

forest distribution habitat dependence

Morococcyx erythropygius Mexico–Costa Rica sav/dec Low Myiarchus nuttingi Mexico–Costa Rica rip Moderate M. tyrannulus brachyurus Nicaragua–Costa Rica sav/rip/dec/man Low Myiodynastes maculatusa Mexico–Brazil rip/dec/man Moderate Ortalis vetula U.S.–Costa Rica rip/dec High Otus cooperi Mexico–Costa Rica dec/man Moderate Pachyramphus aglaiae U.S.–Costa Rica sav/rip/dec Low Parabuteo unicinctus U.S.–Argentina sav Low Passerina cirisb U.S.–Panama sav/dec Low Platyrinchus cancrominus Mexico–Costa Rica rip/dec Very high Plegadis falcinellus U.S.–Costa Rica rip/man Moderate Polioptila albiloris Mexico–Costa Rica dec/man Low Quiscalus mexicanus U.S.–Venezuela sav/man None Rosthramus sociabilis U.S.–Argentina lag None Salpinctes obsoletus Canada–Costa Rica sav None Sporophila torqueola U.S.–Panama sav/dec None Sublegatus arenarum arenarum Costa Rica–Panama man Moderate Thryothorus pleurostictus Mexico–Costa Rica rip/dec Moderate T. rufalbus Mexico–Colombia rip High Trogon elegans U.S.–Costa Rica rip High T. melanocephalus Mexico–Costa Rica dec Low Tyrannus forﬁcatus U.S.–Costa Rica sav/man None Vireo pallens Mexico–Costa Rica man Moderate Xiphorhynchus ﬂavigaster Mexico–Costa Rica rip/dec Moderate Zenaida asiaticaa U.S.–Panama rip/dec/man None Z. macrouraa Canada–Panama rip/dec/man None

Note: Only species exclusive to the Mesoamerican dry-forest region or typical of Costa Rican dry forest are included. Habitat: rip = riparian forest; dec = deciduous forest; sav = savanna; man = mangrove; lag = lagoon. aSpecies with migratory and resident populations. bMigratory species. frankie chap 12 8/20/03 7:58 PM Page 152

Two components of the dry-forest avifauna, may allow movements of individuals between the Mesoamerican and North American, may distant localities. be considered autochthonous to the PAS. These species may have become adapted to climatic and ENDEMISM physiognomic conditions of the dry forest along with the historical climatic changes that have Endemism in dry-forest and arid scrub regions occurred in this region since the Tertiary (Haf- is as high as (or higher than) that in lowland fer 1974). South American species (or ancestors), humid forests (Stotz et al. 1996). Endemism in however, likely arrived in Mesoamerica during the PAS region, and in all dry-forest regions the Pleistocene, after formation of the first bridge in general, is more common in terrestrial than that connected South America with North Amer- in aquatic birds. The PAS embraces seven en- ica (Haffer 1974; Webb and Rancy 1996). Thus, demic species, four of which are present in the historical events, such as climatic changes and Costa Rican dry forest (Lesser Ground Cuckoo, geological events, have likely shaped the current Pacific Screech Owl, Long-tailed Manakin, and composition of the avifauna of Mesoamerica’s White-throated Magpie-Jay). None of these dry forest, the third most diverse dry-forest re- species are restricted to Costa Rica, although all gion of the Neotropics (Stotz et al. 1996). four are abundant in the NPR. Species richness in the PAS increases slightly from Costa Rica to Mexico (Stiles 1983). This in- REPRODUCTIVE ACTIVITY crement in species number with latitude may be explained by the larger area of dry forest in Mex- Reproductive activity of dry-forest birds occurs ico and northern Central America (Murphy and year-round (fig. 12.1), with a peak between March Lugo 1995). Of all the bird species that inhabit and July (reproductive information is given for the Pacific Mesoamerican region, only 32 out of 183 species that are known to breed in the NPR). the 71 (typical) deciduous forest species are shared Breeding periods differ between terrestrial and with the Yucatán Peninsula region (separated aquatic species and within and between some from the PAS by some hundreds of kilometers) taxonomic and/or ecological groups in both and 12 with the northern South American dry- categories, as well. Reproduction of terrestrial forest region (Stotz et al. 1996). This indicates species is concentrated in the period from March that large tracts of more humid forests that sur- to July (Barrantes 1998). In a large number of round dry-forest areas may function as efficient small terrestrial birds (e.g., wrens, woodcreepers, barriers for many dry-forest species. warblers, and most flycatchers and their rela- In Costa Rica the northern Pacific region tives), this activity begins at the onset of the rainy (NPR), which is a small area that constitutes season (April–May) and extends to the end of the southernmost segment of the PAS, contains June/beginning of July. Insects, particularly lar- more than half (58 species) of the Pacific Meso- vae, are present in great numbers after the first american typical dry-forest birds (table 12.2). heavy rains, and this abundance is exploited by The small differences in species composition terrestrial birds in rearing their offspring at this throughout the PAS region indicate the absence time (Janzen 1973; see also chapters 7 and 8). of geographical barriers and lack of habitat Raptors (i.e., Falconiformes and Strigiformes), specialization in dry-forest birds. Lack of spe- some hummingbirds, and a few other species, cialization in dry-forest birds may have originated however, reproduce during the dry season. from adaptation to the contrasting conditions Greater prey availability (e.g., mice) during this (e.g., flower and fruit production) that occur season is perhaps the most important factor in in dry forests between rainy and dry seasons determining the reproductive period for birds (Frankie et al. 1974; Opler et al. 1980; chapter 2). of prey (Janzen and Stiles 1983). Furthermore, Thus, the low degree of habitat specialization leafless vegetation in the dry season makes ro-

152 BARRANTES AND SÁNCHEZ frankie chap 12 8/20/03 7:58 PM Page 153

FIGURE 12.1. Number of breeding species per month for terrestrial, aquatic, and total species in Costa Rican dry forest. The 183 species for which reproduction information is available are included.

dents and other prey much more conspicuous, climate inside nests is responsible for determin- and this greater detectability of prey probably in- ing the reproduction timing in these birds. Cav- fluences time of reproduction in raptors, as well. ity nests maintain constant temperature (about Likewise, reproduction of dry-forest humming- 20°C) and low (but not extremely low) relative birds is largely determined by nectar availability. humidity, even during the driest and hottest Large numbers of plants bloom, including those periods of the dry season, thereby creating an used by hummingbirds (e.g., Combretum fari- appropriate microclimate that protects embryos nosum, Aphelandra spp.), during the dry season and nestlings from desiccation and pathogens (Frankie et al. 1974; chapters 1 and 2) and at the (Stiles 1983; Collias and Collias 1984). beginning of rainy season (Opler et al. 1976). The reproduction of waterbirds begins to- Hence, this resource seems to be an important ward the second half of the year. The majority factor in timing the reproduction of these birds. of aquatic birds reproduce during the rainy sea- It is possible that the dry and windy conditions son, between April and October (fig. 12.1). Food prevalent in this period may increase reproduc- availability (e.g., fish, tadpoles, frogs, snails, tive success of hummingbirds because insect aquatic insects) increases dramatically when parasites and vectors of blood parasites decrease water levels in lagoons and seasonal swamps are during this season (Ricklefs 1992). high. Numerous groups of birds, such as herons, The reproduction of parrots, trogons, wood- ducks, rails, cranes, jacanas, and some hawks, peckers, and other cavity nesters also occurs benefit from the increasing availability of food during the dry season (Stiles 1983). Timing in resources to raise their progeny. A small group these species may be determined by the rela- of aquatic birds (e.g., Ciconiidae, Threski- tively high abundance of fruits and seeds during ornithidae) breed outside the rainy season. These the second half of the dry season (Frankie et species feed mostly on fish, crustaceans, and al. 1974). Yet Stiles (1983) proposed that micro- other invertebrates trapped in small shallow

AVIFAUNA DISTRIBUTION AND ECOLOGY 153 frankie chap 12 8/20/03 7:58 PM Page 154

ponds. These prey become more available when dance of these species decreases greatly, as water levels begin to decline after the rainy expected. Other species seem to be affected season. not only by food availability but also by extreme conditions of the dry season. For example, populations of Elegant Trogon DIET AND HABITAT USE: decrease notably during the driest periods COPING WITH SEASONALITY in riparian forests (J. E. Sánchez pers. obs.). LAND BIRDS It is likely that this species moves between different locations to avoid extreme condi- Diet and the habitat used by birds are greatly tions. This behavior differs from that of the influenced by environmental changes. Tropical Black-headed Trogon, despite the similarity dry forests are sharply seasonal as a consequence in their diets. The Black-headed Trogon is of changes in rainfall, temperature, and wind a common inhabitant of deciduous forest pattern. These changes affect habitat conditions year-round. This suggests that severity in (e.g., foliage cover) and food abundance (Karr environmental conditions, more than food 1976). Consequently, resident birds in tropical availability, defines the pattern of move- dry forest need to adjust to these changes in or- ment for the Elegant Trogon. Field obser- der to meet the energy needs, not only for their vations suggest that some species might growth and maintenance but also for molting move either to middle and high elevations and reproduction (King 1973). Seasonal varia- on the cordilleras de Guanacaste and tion in food resources poses a challenge to birds, Tilarán (following the riparian corridors) which are likely “adapted” to respond to limi- or to evergreen forests of the Nicoya tation of food in different ways (e.g., changing Península. habitats, shifting diet). Insect composition and abundance (Janzen 2. Changes in forest type. The large number 1973, 1985), as well as plant resources (Frankie of phytophagous insects decreases dramati- et al. 1974; Borchert 1983), change drastically cally as foliage availability declines with on a seasonal basis in Costa Rican dry forests. progression of the dry season (Janzen Yet because dry forest consists of a mosaic of 1973). Some groups of insects migrate, but different habitats, with deciduous and ever- many others enter into reproductive dia- green forests dominating, the magnitude and pause. Consequently, during this season direction of such changes vary with the particu- insect larvae are practically absent, and lar conditions in each habitat (e.g., insect abun- adults concentrate in riparian forests. With dance varies from one forest to another). Con- the onset of the rainy season many plants sequently, birds may use different strategies to sprout leaves, and insects, particularly cope with such changes: moth and butterfly larvae, increase dramatically in deciduous forests (Janzen 1973; 1. Movements to localities with milder environ- chapters 7 and 8). Seasonal changes in mental conditions and/or higher resource abundance and distribution of arthropods abundance. Presence and abundance of force some insectivorous birds (e.g., North- numerous dry-forest birds depend, to a ern-barred Woodcreeper, Streaked-headed large extent, on availability of food re- Woodcreeper, Banded Wren) to follow sources. Thus, abundance and spatial dis- these fluctuations to fulfill their energy tribution of species such as Long-tailed requirements. Manakins, tityras, and some flycatchers are largely determined by abundance of the 3. Shifting diet on a seasonal basis. Like food resource. In periods of scarcity, abun- arthropods, blooming and fruiting are also

154 BARRANTES AND SÁNCHEZ frankie chap 12 8/20/03 7:58 PM Page 155

seasonal in dry forests. A large proportion HABITAT DETERIORATION AND of trees flower during the dry season and CONSERVATION OF THE produce their fruits between the second DRY-FOREST AVIFAUNA half of this season and beginning of the rainy season (Frankie et al. 1974; Opler Destruction of natural habitats in Costa Rica et al. 1976). As a consequence of this phe- caused by expansion of the agricultural frontier, nological pattern, fruits and bird-pollinated cattle raising, and logging activities has been flowers may become scarce during part devastating (Janzen 1986; Sader and Joyce of the year. Such fluctuations in food re- 1988; chapters 1, 21, and 23). Forest management source availability may influence some strategies that have been recently implemented species (e.g., Rose-throated Becard, Masked by the Costa Rican government have done little Tityra, Bright-rumped Attila, Streak-backed to reduce the high rate of deforestation in the Oriole) to change their diet on a seasonal country (Barrantes et al. 1999). Lack of knowl- basis. edge in basic reproductive biology of tropical trees (e.g., phenology, sex ratio, pollination, seed A direct consequence of dry-forest seasonal- dispersal, germination) makes forest manage- ity on the avifauna composition is the absence of ment unsustainable in Costa Rica. Despite this frugivores and specialized insectivorous groups. situation, during the past several decades, the For example, families with a predominance of combination of cattle ranching, expansion of frugivorous species (e.g., Ramphastidae, Cotin- agricultural lands (e.g., rice, sugarcane), and lum- gidae, Tharupidae) are poorly represented (or ab- bering activities has greatly reduced this eco- sent) in dry forests. Similarly, specialized groups system, the most threatened in Costa Rica (Janzen of insectivores, such as furnarids, that feed 1988), to small patches surrounded by second- mostly in moss, vines, and tangles of vegetation growth areas and large extensions of pasture- and formicarids, the army ant followers, are prac- land. The establishment of national reserves such tically absent in dry forests (table 12.2). as Parque Nacional Santa Rosa, Parque Nacional Palo Verde, and Parque Nacional Guanacaste has, AQUATIC BIRDS however, gradually permitted some regeneration, Some aquatic habitats, such as lagoons and at least in these protected areas. swamps, are seasonal in Costa Rican dry for- The massive destruction of this ecosystem ests and practically disappear after two or three has obviously affected populations of many bird months of drought. Consequently, the majority species in the region and has already caused lo- of birds that require these habitats must mi- cal extinction of at least one species (the White- grate to avoid food shortages. Thus, presence of faced Whistling Duck [Dendrocygna viduata]; waterbirds in the dry-forest region is largely de- table 12.3). A total of 26 species show a notable termined by availability of their habitats and decline in population size (the population of food resources. Dependence of aquatic birds most of these species has decreased along their on specific habitats increases because very few whole distribution, but we refer to Costa Rican species are inherently able to shift habitats or dry forest in particular) (Rodríguez-Ramírez diet (or both) when drought makes these habi- and Hernández-Benavidez 1999). Both aquatic tats unsuitable. Hence, the relatively low flexi- and terrestrial species are affected by human bility in habitat use and diet of aquatic birds alteration of natural habitats, although the ef- makes a large proportion of these species more fect of such changes is proportionally higher on susceptible to even slight environmental distur- waterbirds (table 12.3). More terrestrial species bances than terrestrial species in terms of their can overcome the alteration of primary habitats reproduction and survival. and subsist in altered or areas in regeneration

AVIFAUNA DISTRIBUTION AND ECOLOGY 155 frankie chap 12 8/20/03 7:58 PM Page 156

TABLE 12.3 Conservation Status, Habitat, and Possible Causes for Changes in Population Size and Survivorship of Dry-Forest Birds

conservation status habitat possible causes

ardeidae (herons, egrets) Botaurus pinnatus T a Habitat destruction Ixobrychus exilis T a Habitat destruction

threskiornithidae (ibises, spoonbills) Ajaia ajaja ID a Habitat destruction, nesting predation?, water pollution?

ciconiidae (storks) Jabiru mycteria ID a Habitat destruction Sarcoramphus papa ID t Habitat destruction

anatidae (ducks) Dendrocygna viduata* E a Unknown, habitat destruction? Cairina moschata T a Hunting, habitat destruction Nomonyx dominicus T a Habitat destruction, hunting

accipitridae (hawks, eagles) Pandion haliaetus* T a Habitat destruction (mainly in breeding areas) Leptodon cayanensis* T t Habitat destruction (mainly in breeding areas) Chondrohierax uncinatus T t Habitat destruction Rostrhamus sociabilis T a Habitat destruction Busarellus nigricollis T a Habitat destruction, water pollution? Circus cyaneus* T a Habitat destruction (breeding and wintering areas) Geranospiza caerulescens T t Habitat destruction Buteogallus urubitinga T t Habitat destruction Parabuteo unicinctus* ID t Unknown, habitat destruction? Buteo albicaudatus T t Habitat destruction Spizaetus ornatus T t Habitat destruction

falconidae (falcons, caracaras) Micrastur semitorquatus T t Habitat destruction Falco peregrinus ID t Habitat destruction, water pollution (breeding and wintering areas) frankie chap 12 8/20/03 7:58 PM Page 157

TABLE 12.3 (continued)

conservation status habitat possible causes

psittacidae (parrots) Aratinga canicularis T t Habitat destruction, poaching Ara macao ID t Habitat destruction, poaching Amazona auropalliata* ID t Habitat destruction, poaching

trochilidae (hummingbirds) Amazilia boucardi T t Habitat destruction

vireonidae (vireos, greenlets) Vireo pallens* T t Habitat destruction

icteridae (orioles, blackbirds) Icterus pectoralis* T t Unknown, poaching?, habitat destruction?

Source: Following Rodríguez-Ramírez and Hernández-Benavidez (1999). Note: Conservation status: T = threatened; ID = in danger; E = locally extinct. Habitat: a = aquatic; t = terrestrial. * = Redeﬁned status based on our observations.

than aquatic birds. As expected, destruction of 1. Protection of aquatic habitats. Despite in- natural environments is the main factor that ternational agreements and treaties, many affects population size and increases risk of swamps and seasonal lagoons in the north- extinction in many species (Table 12.3). Yet other western region of Costa Rica are illegally factors, such as water pollution, hunting, and being drained and transformed into rice poaching, also affect the survival of birds. Un- fields. This clearly reflects the passive atti- fortunately, the available habitat for many threat- tude of Costa Rican government authorities ened or endangered species seems insufficient toward this issue and their unwillingness to for their populations to recover. Therefore, if an deal with it. To solve this problem, national intensive program of restoration in unprotected laws and international treaties need to be areas is not implemented in the near future, enforced immediately (chapters 21–23). local extinction will be the most likely fate for 2. Connection between dry-forest patches. several of these species, as in the case of the The majority of dry-forest tracts, including White-faced Whistling Duck. some protected areas (e.g., Parque Nacional Palo Verde), are interspersed within a deforested area. It is extremely important to RECOMMENDATIONS FOR THE connect these patches to lessen the loss of CONSERVATION OF DRY-FOREST BIRDS diversity caused by isolation. The first step The extinction of a large number of dry-forest is to consolidate the system of conservation birds seems imminent if current practices in areas in northwestern Costa Rica; the Tem- land development and use continue without pisque Conservation Area, in particular, appropriate measures to conserve all wildlife in would benefit from a change in the way the region. Thus, to reverse or at least lessen this these conservation areas are administered. trend, it is necessary to take several actions: Since its creation, this area has been ruled

AVIFAUNA DISTRIBUTION AND ECOLOGY 157 frankie chap 12 8/20/03 7:58 PM Page 158

more by political interests than by the bio- Barrantes, G., and A. Pereira. 1992. Abundancia logical interest of preserving its unique y fluctuaciones de aves limnícolas (Charadri- natural ecosystem. Unfortunately, the solu- iformes) en una playa fangosa de Chomes, Costa Rica. Revista de Biología Tropical 40:303–8. tion to this problem also depends on deci- Barrantes, G., Q. Jiménez, J. Lobo, T. Maldonado, sions of the government, whose interest in M. Quesada, and R. Quesada. 1999. Evaluación the conservation of natural ecosystems has de los planes de manejo forestal autorizados en steadily decreased over the past several el periodo 1997–1999 en la Península de Osa. years. We consider it important for national Cumplimiento de normas técnicas ambientales e impacto sobre el bosque natural. Unpublished and international nongovernmental organ- report. Funcación Cecropia. 96 pp. izations interested in the conservation of Borchert, R. 1983. Phenology and control flower- dry forests to assume more leadership in ing in tropical trees. Biotropica 15:81–89. the future and have a positive influence Collias, N. E., and E. C. Collias. 1984. Nest building on decision makers in the Costa Rican and bird behavior. Princeton, N.J.: Princeton Uni- government. versity Press. Frankie, G. W., H. G. Baker, and P. A. Opler. 1974. 3. Elimination of fire. Annually, thousands of Comparative phenological studies in tropical hectares of dry forest are destroyed by fires wet and dry forests in the lowlands of Costa that are caused by humans (chapters 19 and Rica. Journal of Ecology 62:881–919. 21). In addition to resulting in the deaths of Gómez, L. D. 1986. Vegetación de Costa Rica. Apuntes para una biogeografía costarricense. San many individual birds, fires affect most (if José: Editorial Universidad Estatal a Distancia. not all) species indirectly because food re- 327 pp. source and reproductive substrates decrease Haffer, J. 1974. Avian speciation in tropical South drastically. Thus, a very aggressive campaign America. Publications of the Nuttall Ornitho- and environmental education program are logical Club, No. 14. Cambridge, Mass.: Nuttall Ornithological Club. 390 pp. imperative to eliminate fires in the dry forest Hartshorn, G. S. 1983. Plants: Introduction. In Costa (chapter 19). In the Guanacaste Conservation Rican natural history, ed. D. H. Janzen, 118–57. Area, a program of environmental education Chicago: University of Chicago Press. with a focus on local primary schools has Janzen, D. H. 1973. Sweep samples of tropical fo- been in progress for more than ten years. liage insects: Effects of seasons, vegetation types, Similar programs for children and adults elevation, time of day, and insularity. Ecology 54: 687–708. need to be implemented in the whole re- ———. 1985. Heterogeneity of potential food gion if positive results in the conservation abundance for tropical small landbirds. In Mi- and protection of dry forests are expected. grant birds in the Neotropics: Ecology, behavior, distribution, and conservation, ed. A. Keast and Other activities, such as deforestation, hunt- E. S. Morton, 545–52. Washington, D.C.: Smith- ing, poaching, and chemical contamination in sonian Institution Press. ———. 1986. Guanacaste National Park: Tropical rivers, also increase the risk of local extinction of ecological and cultural restoration. San José: Edi- many aquatic and terrestrial dry-forest birds. torial Universidad Estatal a Distancia. 103 pp. Unfortunately, these problems also receive little ———. 1988. Management of habitat fragmenta- attention from the Costa Rican government, tion in a tropical dry forest: Growth. Annals of the particularly the Ministry of Environment and Missouri Botanical Garden 75:105–16. Energy (MINAE). Janzen, D. H., and F. G. Stiles. 1983. Buteo mag- nirostris (gavilán chapulinero, roadside hawk). In Costa Rican natural history, ed. D. H. Janzen, REFERENCES 551–52. Chicago: University of Chicago Press. Jiménez, J. A., and R. Soto. 1985. Patrones re- Barrantes, G. 1998. Reproductive activity of birds gionales en la estructura y composición florís- in a mangrove swamp in northwest Costa Rica. tica de los manglares de la costa pacífica de Costa Revista de Biología Tropical 46:1163–66. Rica. Revista de Biología Tropical 33:25–37.

158 BARRANTES AND SÁNCHEZ frankie chap 12 8/20/03 7:58 PM Page 159

Karr, J. R. 1976. Birds of Panama: Biogeography Energía y Sistema de Areas de Conservación. and ecological dynamics. In The botany and nat- 28 pp. ural history of Panamá, ed. W. G. D’Arcy and Sader, A., and A. T. Joyce. 1988. Deforestation rates M. Correa, 77–93. Monograph Systematics in and trends in Costa Rica, 1940 to 1983. Biotrop- Botany 10. St. Louis: Missouri Botanical Garden. ica 20:11–19. King, J. R. 1973. Energetics of reproduction in birds. Slud, P. 1964. The birds of Costa Rica: Distribution In Breeding biology of birds, ed. D. S. Farner, and ecology. Bulletin of the American Museum of 78–107. Washington, D.C.: National Academy Natural History 128:1–430. of Sciences. ———. 1980. The birds of Hacienda Palo Verde, Murphy, P. G., and A. E. Lugo. 1995. Dry forests of Guanacaste, Costa Rica. Smithsonian Contribu- Central America and the Caribbean. In Season- tions to Zoology 292:1–92. ally dry tropical forests, ed. S. H. Bullock, H. A. Stiles, F. G. 1983. Birds: Introduction. In Costa Ri- Mooney, and E. Medina, 9–34. Cambridge: Cam- can natural history, ed. D. H. Janzen, 502–30. bridge University Press. Chicago: University of Chicago Press. Opler, P. A., G. W. Frankie, and H. G. Baker. 1976. Stiles, F. G., and A. F. Skutch. 1989. A guide to the Rainfall as a factor in the release, timing, and birds of Costa Rica. Ithaca, N.Y.: Cornell Univer- synchronization of anthesis by tropical trees and sity Press. 511 pp. shrubs. Journal of Biogeography 3:231–36. Stotz, D. F., J. W. Fitzpatrick, T. A. Parker III, and ———. 1980. Comparative phenological studies D. K. Moskovits. 1996. Neotropical birds: Ecology of treelet and shrub species in tropical wet and and conservation. Chicago: University of Chicago dry forests in the lowlands of Costa Rica. Jour- Press. 478 pp. nal of Ecology 68:167–88. Tomlinson, P. B. 1986. The botany of mangroves. Ricklefs, R. E. 1992. Embryonic development pe- Cambridge Tropical Biology Series. New York: riod and the prevalence of avian blood parasites. Cambridge University Press. 413 pp. Proceedings of the National Academy of Sciences Webb, S. D., and A. Rancy. 1996. Late Cenozoic 89:4722–25. evolution of the Neotropical mammal fauna. In Rodríguez-Ramírez, J., and J. Hernández-Benavidez. Evolution and environment in tropical America, 1999. Especies de ﬂora y fauna silvestre con ed. J. B. C. Jackson, A. F. Budd, and A. G. poblaciones reducidas y en peligro de extinción. Coates, 335–58. Chicago: University of Chicago Unpublished report. Ministerio del Ambiente y Press.

AVIFAUNA DISTRIBUTION AND ECOLOGY 159 frankie chap 13 8/20/03 7:59 PM Page 160

chapter 13

An Ultrasonically Silent Night

THE TROPICAL DRY FOREST WITHOUT BATS

Richard K. LaVal

ithout bats, in their varied roles as pol- verted to nonforest land uses, primarily agricul- Wlinators of commercially important trees, tural (Janzen 1988). Although a large portion seed dispersers critical for forest succession, and of the land area included would be classified as consummate predators of nocturnal insects, the tropical dry forest under the Holdridge Life Zone tropical dry-forest life zone would be a vastly dif- System (Holdridge 1967), there are areas of moist ferent place. Despite the unquestioned impor- forest (to the south) and very dry forest (to the tance of these common mammals, few scientific north) that I include in the dry-forest life zone papers have dealt specifically with bat conserva- because of floral and faunal similarities. tion until very recently. Fenton (1997) has clarified the relationship between recent advances in DIVERSITY AND ABUNDANCE OF bat biology and bat conservation, and a book ed- BATS IN THE DRY FOREST ited by Kunz and Racey (1998) addresses bat conservation issues worldwide. Ceballos (1995) Ninety-two bat species occur in the dry forest discusses vertebrate conservation in the dry for- as defined in the previous paragraph (based pri- est, with special emphasis on the importance marily on maps in Hall 1981). Of these, 15 are of bats. endemic to the dry forest or nearly so. Sixteen For the purposes of this chapter, “dry forest” are restricted to Mexico, 67 occur in both Mex- refers to a deciduous-forested life zone that ex- ico and Central America, and 9 are found only in tends at low elevations along the entire Pacific Central America. Most Central American species coast from west-central Mexico to northwest- also occur in South America. In comparison, ern Costa Rica, with rainfall mostly less than there are 93 species in the tropical wet forest 1,500 mm. Much of this forest has been con- of the Caribbean lowlands, extending from east-

160 frankie chap 13 8/20/03 7:59 PM Page 161

TABLE 13.1 Conservation Priorities for Dry-Forest Bats Based on Distribution and Abundance

priority category number of species

1 Locally scarce, with restricted distribution 12 2 Locally abundant, with restricted distribution 5 3 Locally scarce, with widespread distribution 53 4 Locally abundant, with widespread distribution 22

central Mexico to Costa Rica. Only 3 of these are CONSERVATION ISSUES endemic. If, as noted by Arita and Ortega (1998), FOR DRY-FOREST BATS there are a total of 166 bat species in Meso- america (Mexico plus Central America), then SPECIES OF SPECIAL more than half of those occur in the dry forest. CONSERVATION CONCERN In Costa Rica, where detailed surveys of bat Using a scheme proposed by Marinho-Filho and faunas have been carried out at several sites, an Sazima (1998) for illustrating the relative rarity interesting comparison of species numbers in of bats for conservation purposes, I categorized dry forest, wet forest, and premontane wet forest/ the 92 species of Mesoamerican dry-forest bats lower montane wet forest can be made. Studies as shown in table 13.1. Assignment to these cat- in Guanacaste Province at Santa Rosa National egories was by necessity rather subjective because Park, Lomas Barbudal Biological Reserve, and data on abundance or distribution (or both) are Hacienda La Pacifica (Fleming et al. 1972; LaVal sparse or lacking for many dry-forest species. and Fitch 1977; Wilson 1983; Fleming 1988) have However, to the extent that the focus in conser- documented a total of 55 dry-forest species. In vation efforts reflects concerns for individual spe- the wet forest of La Selva 68 species have been cies, the species in categories 1 and 2 (table 13.1) captured (Timm and LaVal 1998). As expected, should be given the most attention. This in- fewer species (46) are known in the premontane cludes most of the 15 species that are endemic and lower montane forests of Monteverde (Timm (or mostly endemic) to North American dry for- and LaVal 2000). In summary, dry forest in est. The most seriously threatened species are Costa Rica is only slightly less diverse than the those that have the most restricted distribution. wet forest (see maps in chapter 1). These include Musonycteris harrisoni, Myotis Although actual bat abundance is difficult to findleyi, M. carteri, and Rhogeessa mira, all of estimate, it is possible at least to compare mist- which seem to be uncommon and restricted to net capture rates. One such comparison can be very small areas of western Mexico. Artibeus in- made using selected data from LaVal and Fitch opinatus is in the same category but is restricted (1977) and Fleming (1988): bats were caught at to a small portion of western Central America. a rate of 1.05 bats per net-hour at La Selva in wet Other scarce dry-forest species with restricted forest and 1.41 per net-hour at Santa Rosa in dry distributions include Micronycteris schmidtorum, forest. Over a period of one year, I mist-netted Choeronycteris mexicana, Rhogeessa genowaysi, 1,208 bats at La Selva, and over a ten-year period, R. alleni, Nyctinomops femorosaccus, N. macrotis, Fleming (1988) netted 11,627 at Santa Rosa. and Eumops underwoodi. Five additional species Clearly, these are abundant animals, which, be- have restricted distributions but are relatively cause of the ecological roles they play, strongly abundant, including Balantiopteryx plicata, Glos- affect the dry-forest ecosystem. sophaga alticola, G. morenoi, Artibeus hirsutus,

ULTRASONICALLY SILENT NIGHT WITHOUT BATS 161 frankie chap 13 8/20/03 7:59 PM Page 162

and Rhogeessa parvula. The others are somewhat Musonycteris harrisoni, Myotis findleyi, and Rho- buffered from local extinction events by having geessa mira. Those of special concern included a larger distributional area. Artibeus hirsutus, Choeronycteris mexicana, Hy- Using a different set of criteria, Wilson (1996: lonycteris underwoodi, Leptonycteris nivalis, Myotis app. 9.1, pp. 172–77) listed 14 dry-forest species elegans, and Rhogeessa alleni (H. underwoodi and as being endangered, vulnerable, or potentially M. elegans are not dry-forest species as here de- vulnerable. “Endangered” means that “the taxon fined). In a subsequent report Arita and Santos- is in danger of extinction, and its survival is un- del-Prado (1999) pointed out that 11 species of likely if the causal factors continue operating.” Mexican nectar-feeding bats occur mostly in dry “Vulnerable” signifies “that the taxon likely will forest and are particularly at risk because of their be relisted under the endangered category in the specialized diets. near future, if causal factors do not change.” It is clear that, irrespective of the different “Potentially vulnerable” indicates that “the taxon criteria used to compile the three lists, many has small world populations, occupies only a species appear on all three, some on two, and a small percentage of its preferred habitat, or oc- few only on one. Unless new data show that cupies a very specific, limited habitat, thus is some of these species have large, relatively safe faced with present or foreseeable pressures that populations, it would be prudent to consider all could lead to its becoming vulnerable.” Of the listed species as subject to some level of threat 14 listed dry-forest species, he considered only and therefore deserving of special conservation Myotis velifer as endangered (M. velifer is a com- efforts. mon species in some parts of its range, for example, in western Texas; further, it does not MIGRATION occur in dry forest as it is defined herein). Lep- Although long-distance migration of certain tonycteris curasoae, L. nivilis, and Eumops glauci- species, most notably Leptonycteris curasoae nus were classified as vulnerable, and the rest and Tadarida brasiliensis, has been well docu- (Noctilio leporinus, Mormoops megalophyla, Chro- mented (Cockrum 1969, 1991; Fleming 1991, topterus auritus, Lonchorhina aurita, Micronycteris 1997; Rojas-Martinez et al. 1999) and must be minuta, M. sylvestris, Vampyrum spectrum, Cho- carefully considered in any conservation strategy, eronycteris mexicana, Musonycteris harrisoni, and short-range seasonal movements into and out Rhogeessa mira) are potentially vulnerable. of the dry-forest zone have been little studied. Using yet another set of criteria, Arita and These seasonal movements could be primarily Ortega (1998) listed ten species of dry-forest elevational migrations, as have been amply doc- bats as being of critical or special concern. The umented for birds (in Costa Rica by Stiles 1988) critical species are scarce with restricted distri- and butterflies (also in Costa Rica; see chap- bution and are dry-forest endemics. Bats of ter 8). In western Mexico, Alvarez and González- special concern have traits that make them im- Quintero (1970) and Herrera-Montalvo (1997) portant for conservation or have habitats suscep- presented convincing evidence that glosso- tible to perturbation. Included in this category phagine bats displayed elevational movements. are species that are endemic or quasi-endemic Timm and LaVal (2000) demonstrated that two to Mesoamerica (as defined earlier) and that show common dry-forest species (Sturnira lilium and one or more of the following characteristics: Artibeus lituratus) are seasonally absent from (1) widespread, but with very low population lev- Monteverde (premontane wet forest) and prob- els; (2) habitat specialists; (3) ecological special- ably migrate elevationally in search of preferred ists; (4) migratory; (5) potential keystone species; fruiting plants. Stoner (1997) and Stoner and and/or (6) roost in caves used by vampire bats. Timm (chapter 5) suggested, based on data from Among those Arita and Ortega (1998) consid- Palo Verde National Park (Tempisque Conserva- ered of critical status were Artibeus inopinatus, tion Area), that several frugivorous and nectariv-

162 LAVAL frankie chap 13 8/20/03 7:59 PM Page 163

TABLE 13.2 Numbers of Bat-Serviced Trees and Shrubs in the Dry Forest (Santa Rosa National Park), the Wet Forest of La Selva, and the Premontane Wet and Lower Montane Rain Forests of Monteverde

premontane and lower dry forest wet forest montane forests

Number pollinated 23 21a 14 Number seed-dispersed 32 >60 59 Total bar-serviced 57 >81 73 Percentage of total species 21.9 >13.3 8.4 Total species of 260 608 874 trees and shrubs

aIncludes all growth forms, not just trees and shrubs.

orous species migrate out of the dry forest sea- bats dispersed more seeds than birds (primarily sonally. In western Mexico, Iñiguez (1997) noted to disturbed areas and consisting primarily of that the most common species in his cloud for- pioneer species), they are likely to play an im- est study area, Sturnira ludovici, migrated out portant role in successional and restoration proof the area seasonally. He believes that they cesses among habitats as structurally and vege- migrate into the dry forest (pers. comm), and tationally different as cornfields, old fields, cacao indeed, that species is known in dry forest in plantations, and forest.” western Mexico south to El Salvador (Ramírez- Detailed lists of dispersers have been com- Pulido et al. 1977). There is reason to believe that piled only for certain localities in Costa Rica’s dry future studies will document extensive move- forest. In Santa Rosa National Park about 23 tree ment of bats into and out of the dry forest. and shrub species are bat-pollinated, and an- From a conservation perspective this means other 32 species have bat-dispersed fruits (Heit- that some species will not be adequately pro- haus et al. 1975; Fleming 1988; W. A. Haber pers. tected until all habitats used within their migra- comm.). Thus, about 55 of a total of 260 species tory pathways are safeguarded. In the case of (B. Hammel and N. Zamora pers. comm.) of long-distance migrants, habitat protection at both trees and shrubs are currently known to be serv- ends of the migratory route is crucial but is made iced by bats (table 13.2). Among these are both more difficult for Leptonycteris and Tadarida, as common primary forest trees and successional noted earlier, because they cross international trees and shrubs that are often among the first boundaries. Fortunately, both Mexico and the colonists following human disturbances such United States have recently taken initial steps as fire or agricultural clearing. Commercially toward protecting migratory bats (“Binational important dry-forest trees, such as Pochota quin- partnership” 1994; Walker 1995). ata (Bombacaceae), are also bat-pollinated. At Chamela Biological Station (Estación de Investi- BATS AS DISPERSERS OF SEEDS AND POLLEN: gación, Experimentación, y Difusión “Chamela,” THE MAINTENANCE OF FOREST DIVERSITY Jalisco, Mexico), perhaps 25 to 35 species have Bats are commonly regarded as among the most seeds that are bat-dispersed (A. Pérez pers. important of all dispersers. In a recent study, comm.), suggesting that bats are important conducted in Chiapas, Mexico (Medellín and seed dispersers throughout the Mesoamerican Gaona 1999: 478), the authors stated, “Since dry-forest zone. Because phyllostomid fruit bats

ULTRASONICALLY SILENT NIGHT WITHOUT BATS 163 frankie chap 13 2003-09-11 12:14 Page 164

FIGURE 13.1. Sturnira ludovici (Stenoderminae: Phyllostomidae) about to pick a fruit of Solanum umbellatum (Solanaceae), a widespread pioneer species with small seeds that are ingested and later defecated by the bat.

often defecate in ﬂight, they are more likely than bat-pollinated, but 59 have seeds that are dis- birds (which habitually defecate while perched) persed by bats (Murray et al. 2000). As evident to drop seeds into deforested areas. from table 13.2, a higher percentage of trees and In the tropical wet forest of La Selva, which shrubs are bat-serviced in the dry forest. Bats has much higher plant diversity, fewer species not only service many kinds of plants but also (21 trees, shrubs, and epiphytes) are known to handle/visit a vast number of fruits/ﬂowers. For be bat-pollinated (M. Tschapka pers. comm.). example, Kalko (1997a) calculated that fruit bats An unknown number have seeds that are bat- on 15-km2 Barro Colorado Island, Panama, dis- dispersed, but it must be a fairly large number persed seeds of 16 million to 32 million fruits (>60 and perhaps >75), considering that most per year. Disappearance of bats would likely pre- species of Piper have seeds that are bat-dispersed cipitate major changes in the structure of the and 44 species occur at La Selva and also that dry-forest community. 17 species of Ficus (whose seeds are often bat- Two widespread groups of dry-forest plants dispersed) are found at La Selva. López (1996), largely pollinated by bats deserve special men- examining bat fecal samples at La Selva, found tion. The columnar cacti are composed of vari- seeds from 42 species of plants during the pe- ous genera, and many species occur conspic- riod January–August 1995. His count is likely an uously throughout the dry forest. The highest underestimate, as he would have missed species rates of species diversity and abundance are fruiting later in the year as well as species with found in the drier portions of western Mexico. large seeds that are dropped without being In fact, in many areas these are the dominant passed through the digestive tract. In the pre- plants, reaching abundance levels of 1,650 plants montane wet and lower montane rain forests of per hectare, with 70 species in Mexico alone. Monteverde, also with much higher plant diver- Two-thirds of these species show characteristic sity than dry forest, only 14 trees and shrubs are adaptations for bat pollination, and several spe-

164 LAVAL frankie chap 13 2003-09-11 12:14 Page 165

FIGURE 13.2. Glossophaga commissarisi (Glossophaginae: Phyllostomidae) pollinating the ﬂowers of a common tropical forest liana, Mucuna urens (Fabaceae). Apparently, only bats can open the ﬂowers and successfully pollinate this plant.

cies studied in western Mexico depend entirely insect orders include those insects most damag- or primarily on bats as pollinators (Valiente- ing to crops and human health. Banuet et al. 1996). Another abundant group of Although few dietary studies exist for Neo- dry-forest plants (occurring in other life zones, as tropical insectivores, several have appeared for well) pollinated by bats are Agave species, some T. brasiliensis, which is common in parts of the of which are widely cultivated commercially for dry forest and in the southern United States. fiber and for distilling tequila, especially in Mex- T. brasiliensis in central Texas averaged nightly ico. Arita and Wilson (1987) reported that seed feeding rates of 39 percent to 73 percent of the set in agaves drops by a factor of 3,000 if not bats’ body weight (Kunz et al. 1995). These bats pollinated by bats. Bats responsible for pollina- have a mean weight of 11.5 g. Because they live tion of columnar cacti and agaves may well be in cave colonies of millions, they must have a regarded as keystone species in many parts of significant impact on insect populations, as was the dry forest. suggested by McCracken (1996). In a large colony of mormoopid bats (as many as 800,000) BATS AS A BIOLOGICAL in western Mexico, a quantity estimated as high CONTROL OF INSECT PESTS as 3,805 kg of insects per night was consumed Aerial insectivores feed on a large variety of flying (Bateman and Vaughan 1974). Kalko (1997a) cal- insects, but in most species studied to date, lep- culated that one of 28 species of insectivorous idopterans and coleopterans predominate in bats’ bats on Barro Colorado Island, Panama, might diets. Dipterans (and sometimes other orders) eat 1.3 million large insects per year. One species are also important to some species (for exam- of Rhogeessa, a genus that reaches maximum ples, see Black 1974; Gardner 1977; Humphrey species diversity in the dry forest of Mexico, et al. 1983; Willig et al. 1993). Members of these was recently studied by Sosa et al. (1996) in

ULTRASONICALLY SILENT NIGHT WITHOUT BATS 165 frankie chap 13 8/20/03 7:59 PM Page 166

Venezuela. They found that the diet of R. minu- fragments of land ranging in size from less tilla, in what is probably very dry forest, was than 5 ha to 40 ha (Granjon et al. 1996; Cosson composed of 42.5 percent Diptera, 13.3 percent et al. 1999). Bats were sampled in the fragments Lepidoptera, 10.9 percent Coleoptera, and 18.2 before the reservoir filled, and one year after percent Hymenoptera. the fragments became islands. The data demon- Because many species of insectivorous bats strated that the islands, even though separated tend to be environmentally stratified, having pre- from the “mainland” by relatively short dis- ferred foraging spaces that often differ from those tances, contained a significantly less diverse of other species (Fenton 1982, 1997; Findley bat community, with a concomitant decrease in 1993), the impact of foraging insectivorous bats bat abundance. From this study, it appears that is felt in almost all dry-forest microhabitats ac- bats are less inclined to cross a barrier of water cessible to volant mammals. Since in-flight defe- than a barrier composed of disturbed terrestrial cation is common, insectivorous bats may be im- habitats. portant in nutrient transfer in ecosystems, acting Although no equivalent studies have been as “nutrient pepper shakers” (Pierson 1998). done in dry forest, I suggest that a habitat mosaic similar to that described by Estrada et al. EFFECTS OF LARGE-SCALE (1993), but in dry forest, may support a large HABITAT ALTERATION percentage of the original bat species. Much de- As pointed out by Janzen (1988), only about pends on how the habitat has been modified. If 2 percent of the original Central American dry most of it has been converted to pasture, only forest remains, and only about 0.1 percent has edges, steep ravines, and riparian strips remain protected status (Maass 1995). According to Ce- to serve as bat habitat, and species richness and ballos and García (1995), the largest remaining abundance will be low. Often, however, patches dry forests north of the equator are in western of clear-cut forest are allowed to regenerate, at Mexico, although no specific figures are given. least temporarily. Parklands often contain siz- However, these authors give a deforestation rate able areas of secondary forest in various stages of 300,000 ha per year, or 2 percent of total sur- of regeneration. For example, in Guanacaste viving Mexican dry forest per year. Although this Conservation Area in Costa Rica (104,000 ha), is an appalling rate of deforestation, their data most of the protected area is regenerating dry also imply that millions of hectares remain. Pro- forest, much of it having been pasture ten years tected and otherwise undisturbed patches of ago. Because the conservation area contains dry forest are widely separated by a mosaic of patches of primary forest and older secondary disturbed habitats, mainly agricultural. These dis- forest, conservationists hope that most of the turbed habitats vary widely in their usefulness species diversity has been preserved. to bats. In a study carried out in tropical wet/rain Nevertheless, results of several studies sug- forest in southern Veracruz, sampling sites var- gest a significant change in the structure of ied from those supporting essentially no bats bat faunas in disturbed versus primary forest, (pastures) to mixed agricultural habitats (coffee, including the disappearance of certain groups cocoa, citrus, and banana) whose species richness of bats in deforested areas (LaVal and Fitch and abundance were only slightly less than that 1977; Fenton et al. 1992; Brosset et al. 1996; of nearby small patches of primary forest. About Kalko 1998; Simmons and Voss 1998; Wilson et 80 percent of bat species known from large tracts al. 1998; Ochoa 2000). Notable is that in these of undisturbed forest in the region were cap- studies populations and species diversity in the tured in the study, with highest species diversity subfamily Phyllostominae were greatly reduced in primary forest patches (Estrada et al. 1993). in disturbed habitats, making this a potential In another relevant investigation in French indicator group for measuring the degree of Guiana, a dam project resulted in the creation of disturbance.

166 LAVAL frankie chap 13 8/20/03 7:59 PM Page 167

As pointed out by Brosset et al. (1996), habi- In another European study, Eptesicus serotinus tats disturbed by humans often offer bats an were observed following corridors of woodland abundance of bat-dispersed fruits, such as species from their colonies to foraging areas, which were of Piper, Solanum, Cecropia, and Vismia, which increasingly fragmented, resulting in higher occur in early stage successional forest in most commuting costs to bats (Robinson and Steb- Neotropical life zones. The ﬁrst three of these bings 1997). plant genera are widespread in dry forest, but Bats, being small mammals, can probably Piper and Cecropia seem to be restricted to mesic maintain adequate populations in smaller habi- habitats such as riparian strips in western Mex- tat patches than can larger mammals. However, ico, according to Lott (1993) and Alfredo Pérez rare bats, of which there are many species in the (pers. comm.). They found bats to be much more tropics, may require sizable protected areas to abundant in these disturbed areas, with capture survive. This problem certainly deserves investi- rates of 1.68 to 3.6 bats per net-hour, as con- gation. Without becoming embroiled in issues trasted with 0.16 to 0.45 per net-hour in un- of minimum patch sizes required to protect disturbed primary forest. Their data show that viable ecosystems of organisms, it is clear that higher capture rates of a few species of fruit the destruction of most Central American dry bats (Stenoderminae), especially of the genera forest, and the concomitant survival of only Artibeus, Carollia, and Sturnira, were respon- small isolated parcels, bode ill for the survival of sible for the increase in capture rates in dis- dry-forest endemic and rare bat species. turbed habitats. Studies in various areas, including dry for- GENERAL CONSERVATION ISSUES est, support these conclusions (Fleming et al. RELATING TO NEOTROPICAL BATS 1972; Heithaus et al. 1975; LaVal and Fitch 1977; Ramírez-Pulido and Armella 1987; Estrada et THE VAMPIRE BAT AND RABIES PROBLEMS al. 1993; Ochoa 2000). In general these studies In the past two decades a gradual improvement reported relatively low numbers and diversity in the public image of bats, and thus increased of aerial insectivores, owing to the inadequacy public support for bat conservation, have oc- of prevailing mist-net capture techniques. How- curred in Europe, the United States, and Canada. ever, Kalko et al. (1996) and Kalko (1997b), using This is primarily due to the leadership of organ- acoustic monitoring in Panama (tropical moist izations such as Bat Conservation International forest), demonstrated that insectivorous bats can (BCI) of Austin, Texas. However, fear and dislike be both common and diverse. My preliminary of bats continue to be a major problem for con- data from acoustic monitoring in Guanacaste servation in Latin America and other parts of the suggest that insectivorous bats are also abun- world. dant there. Rural Latin Americans tend to be very toler- Species that roost in human habitations are ant of wild animals as long as they have no effect frequently abundant in urban areas. This is evi- that local people interpret as negative. Thus, most dent in almost any town in the dry forest, where bats would be ignored and left in peace (even roosts of molossids in buildings can be easily odoriferous roost colonies are rarely molested) if located by odor, vocalizations, or exit ﬂight ob- not for the common vampire, Desmodus rotun- servations. Insectivorous bats, however, may be dus. Because livestock, mainly cattle, are kept in affected in either a positive or negative manner large numbers throughout the Neotropics, vam- by human constructs. For example, in Europe pires have multiplied and become abundant in a study demonstrated that “some species of most disturbed areas. They are regarded as pests bats avoid open habitats for foraging, and these because they can transmit the rabies virus to species may be negatively affected by patchiness cattle during epidemics. An unchecked rabies epi- or habitat fragmentation” (de Jong 1995: 246). demic may cause 30–50 percent herd mortality

ULTRASONICALLY SILENT NIGHT WITHOUT BATS 167 frankie chap 13 8/20/03 7:59 PM Page 168

(Constantine 1970). Vampire bats also may Thus, when a vampire problem exists, especially transmit rabies to humans, although normally if rabies is involved, the usual reaction is to few humans are bitten. However, in some areas, destroy any bats that can be found. The most significant numbers of human rabies cases are easily located roosting bats are in caves and hol- attributed to vampire bats. During the period low trees, sites also used by vampires. Unfortu- 1931–90, a total of 330 cases were documented nately, bats commonly found in such situations as being bat-transmitted, mainly in Peru, Trini- tend to be highly beneficial insectivores, frugi- dad, Brazil, and Mexico (Schneider and Santos- vores, and nectarivores, which then become the Burgoa 1995). object of extermination efforts. As stated by Lord Even where rabies is rare or unknown, cattle (1988: 217), “For many years vampire bats have farmers dislike vampires because they inflict been destroyed by burning, gassing, and even wounds that sometimes attract damaging screw- dynamiting their roosts. Clearly, all bats, not just worms or bacteria. (The recent depression of vampires, were killed.” In recent decades gov- screwworm [Cochliomyia hominivora] populations ernment campaigns to eradicate vampires have by the sterile male release technique may actu- resulted in the slaughter of countless thousands ally have a beneficial effect on the reputation of of harmless beneficial bats, yet vampires have vampires by reducing or eliminating this prob- remained common (Tuttle 1988). The answer lem.) It is also widely believed that vampires is better environmental education among Latin weaken cattle by extracting large amounts of Americans, especially in rural areas, with the blood. A vampire may consume more than 20 ml help of international and national conservation nightly, and if an individual cow was attacked organizations. each night by multiple vampires, the animal might be significantly weakened (Lord 1988). THREATS TO PREFERRED ROOST SITES However, a single bite resulting in a 20-ml blood Nearly three decades ago Humphrey (1975: 321) loss would hardly be noticeable to such a large pointed out that “the distribution and abundance mammal. Effective control measures are now of colonial Nearctic bats is determined largely by available, although they are neither simple nor the availability of suitable roosts.” He further inexpensive. The commonly used method in- emphasized that “the importance of roosts indi- volves applying “vampiricide” (an anticoagulant) cates that roosts should be managed as a key fea- to the fur of captured vampires. They are then ture of the habitat of bats whose populations are released, and, on returning to their home colony, in need of protection.” It seems likely that a they are groomed by fellow colony members, similar situation prevails in the Neotropics. How- thus ensuring their demise. This method has the ever, although certain kinds of roost sites (caves, advantage of causing little or no harm to other most obviously) are probably limiting for popu- bat species (Lord 1988), assuming workers can lations of some bat species, other types of roost recognize vampires so that other species are not sites (for example, among twigs and leaves of treated with the toxic salve. Unfortunately, this trees) are unlikely to be limiting. Findley (1993) is often not the case. For example, in Venezuela provides a concise discussion of resource limi- government technicians implementing a vam- tation. In some cases the availability of adequate pire control program were reported to apply foraging habitat may be the important limiting the paste to all bats captured, including an en- factor (Pierson 1998). Fortunately for those cases dangered species (Ochoa 1992). in which roost sites are especially important to a Regrettably, most rural Latin Americans are bat species, conservation efforts may be focused poorly informed about bats. A widely held belief on these more easily and with less conservation is that all bats are vampires. The Spanish word investment than would be required to protect for vampire (vampiro) is commonly used inter- large areas of foraging habitat. As pointed out changeably with the word for bat (murciélago). by McCracken (1988), roost site losses account

168 LAVAL frankie chap 13 8/20/03 7:59 PM Page 169

for the most serious threats to bat populations. (“Binational partnership” 1994; Walker 1995). Sheffield et al. (1992) have provided a valuable But hundreds of caves deserving protection have set of guidelines for the protection of bat roosts. received little or no attention (in Mexico alone The impact of bat roost disturbance depends on there are at least 215 in which bats are known to three factors: the type of roost, the bats’ fidelity live, according to Arita [1993]), including most to that roost, and the bats’ propensity to exploit caves in Central America. Abandoned mines new roosts (Fenton and Rautenbach 1998). as bat habitat may be more common or impor- Clearly, without adequate biological knowledge tant (or both) than caves in some areas to some of a species it is difficult to determine how its species. Most such mines have yet to receive roost sites should best be protected or even if it conservation attention in Mesoamerica, although is possible or practical to do so. a movement to protect mines used by bats has Caves. Arita (1993) demonstrated that 60 species gained momentum in North America (Tuttle and (more than half) of Mexican bats used cave Taylor 1998). roosts. Of these, 29 used caves as their main Crevices. Crevices in rocks and cliffs provide roost; the remaining species used caves as al- shelter for many bats, especially in the insectiv- ternate roosts. The author also noted that indi- orous family Molossidae (Kunz 1982). One dividual caves are used by as many as 13 species. rect threat to these colonies is rock-quarrying Although caves are widespread and locally operations, which are often short-lived and may abundant, many parts of Central America and still provide roosting habitat when abandoned. western Mexico have few or no caves because of Only a much higher conservation awareness than tectonic and igneous development of mountain now exists could provide protection to these bats; ranges. In some areas, such as Costa Rica’s fortunately, many are situated in inaccessible Nicoya Peninsula, interiors of caves are so in- sites that provide excellent natural protection. accessible that bats have apparently been little Tree Cavities. Hollow spaces within tree trunks disturbed. Fortunately, most of those caves are are used by many species of bats, especially those within a national park. However, in a survey of in the common and diverse Neotropical family Mexican caves used or formerly used by T. bra- Phyllostomidae. At least 28 species in this fam- siliensis, more than half of the ten most important ily roost in tree cavities (Tuttle 1976). In exten- overwintering caves have lost 95 to 100 percent sive areas of primary forest, tree cavities may be of their bats (“Binational partnership” 1994). the only available sites for bats that require this Some of these caves have been rendered unsuit- type of roosting habitat. Only primary forest pro- able for future use by bats. vides trees of adequate size and age. Thus, both Arita (1993: 697) suggested that “an effective clear-cutting and selective logging tend to de- plan for the conservation of Mexican cave bats stroy most roost trees. Even in protected areas would require a double strategy: the protection with mature trees, fires, a serious dry-forest prob- of caves with unusually high diversity and multi- lem, can result in death for those trees (pers. species populations and the management of obs.). Few protected areas in the dry forest have cave bats of special concern (fragile, vulnerable, adequate fire control programs, and fires burn- and endemic species).” As demonstrated in the ing into protected areas from adjoining private United States, organizations of cave explorers land during the dry season will gradually elimi- working in conjunction with government agen- nate trees that provide cavity roosts for many bat cies can cooperate to provide protection for im- species. A significant exception is the Guana- portant bat caves (Thorne 1990; McCabe 1995). caste Conservation Area in Costa Rica, which Joint U.S.-Mexican efforts to protect remaining uses firebreaks, firefighting teams, and local T. brasiliensis caves in Mexico have involved var- environmental education to reduce fire damage ious government agencies in both countries, (G. Frankie pers. comm.). Because deforestation Mexican caving organizations, and bat biologists is so widespread in the dry forest, any forest patch

ULTRASONICALLY SILENT NIGHT WITHOUT BATS 169 frankie chap 13 8/20/03 7:59 PM Page 170

or riparian strip with mature trees provides im- roost structures (“bat houses”), of which thou- portant habitat to cavity-roosting bats and should sands have now been erected. In one North be protected. It is likely that cavity roost space is American campaign, occupancy rates of up to limiting to many dry-forest bat species. 50 percent have been reported (Tuttle and Hens- Foliage, Tree Limbs, and Similar Roosts. Many ley 1993). A recent news note in Bats magazine species of Neotropical bats roost on the bark of (April 1999: 14) mentions occupancy rates of limbs and trunks, as well as within dense vege- 75 percent. When these bat houses are placed in tation, rolled leaves, or tents constructed from sites where roosting space is limiting, they could leaves by the bats themselves (Kunz 1982). Ade- be a boon for some species. However, bat houses quate roost sites can often be found in second- as currently designed may not help those species ary forest or even in mixed agricultural habitats. with the most restrictive roost requirements, These species are the least threatened from the and there has been little or no experimentation standpoint of roosting habitat loss. with artificial roost structures in the dry forest Anthropogenic Roosts. Humans have created a or elsewhere in the Neotropics. Still, the poten- myriad of structures for many uses. These usu- tial exists for the implementation of successful ally occupy space that was once available as bat designs. foraging or roosting habitat. Nevertheless, in OTHER ANTHROPOGENIC many of these structures new types of roosting EFFECTS ON BAT HABITAT spaces become available to bats, especially cave-, crevice-, and cavity-roosting species (Kunz 1982). Habitat Destruction. Perhaps the most serious In North America Eptesicus fuscus, Myotis lucifu- problem to be addressed is the replacement of gus, and M. yumanensis are found almost entirely natural habitats with disturbed habitats. It is in buildings during the summer (Kunz 1982), almost certain that the dry-forest zone of Meso- and several dry-forest species, especially in the america was covered with forest when Euro- genus Molossus, are usually found in spaces peans first arrived (about 500 years ago), even under metal roofs. Tunnels, culverts, mines, and though small areas have been cleared for agri- similar structures are frequently used by cave culture for at least 5,000 years (Murphy and bats (Kunz 1982). Finally, introduced plants have Lugo 1995). Today approximately 2 percent of frequently proved to be suitable roost sites for Central American dry forest remains in pristine foliage- and surface-roosting bats. condition (Janzen 1988). A much larger but Thus, human activity has unwittingly pro- unquantified area remains in western Mexico vided roost sites to replace many of those that (Ceballos and García 1995). Surviving forest is a have disappeared with forest clearing and dis- patchwork of small tracts separated by various turbance of caves. Unfortunately, appropriate kinds of disturbed habitat. Much of this dis- foraging resources may not exist in proximity to turbed habitat supports an abundance of bats these potential roost sites, especially for more of low diversity. Species that are largely absent specialized species. As noted, species diversity from disturbed areas may all be considered at seems to decrease rapidly as levels of distur- risk, especially if most of the remaining Central bance increase. Further, bats are often consid- American dry forest is to be cut. Relatively small ered unwelcome guests, notwithstanding that protected zones in parks and reserves, mostly many of the reasons for their poor reputation in the Guanacaste Conservation Area and the are invalid. Pierson (1998) opined that negative Chamela-Cuixmala, Manantlán, and La Sepul- effects of human alterations, in general, out- tura Biosphere Reserves, already exist, but there weigh positive effects. is potential to protect much more (Ceballos and As people have finally begun to realize that García 1995, 1996). If only these areas are pro- bats are worthwhile neighbors, mainly in North tected in the long run, those “at risk” species that America and Europe, they have designed artificial do not occur in one of the protected zones will

170 LAVAL frankie chap 13 8/20/03 7:59 PM Page 171

likely become extinct, unless they are also found such studies have been relatively few. One ex- in large protected tracts in other life zones. ample is the destruction of a cave colony of en- With approximately 2,700 km separating dangered Myotis grisescens in Missouri by toxic western Mexico and Guanacaste, and dry-forest levels of pesticide residues (Clark et al. 1978; patches in general being small and widely sepa- Clark et al. 1983). Residues of toxic compounds rated, there is little chance of establishing pro- were taken from tissues of 23 species of bats by tected corridors between patches. As stated by Clark (1981). Samples of bat guano from nine Janzen (1983) and documented for bats by Es- major bat caves in Mexico all showed residues of trada et al. (1993) in Mexico, primary forest dangerous pesticides, but they were not yet at patches have lower diversity than large tracts of toxic levels (Clark et al. 1995). It is safe to say undisturbed forest, and the smaller the patch, that few dead and dying bats are found and taken the lower the diversity. Because Mesoamerican to laboratories for testing of toxic chemical res- dry forest extends through six relatively poor idues. Therefore, although the extent of this third world countries, each with differing con- threat remains unknown, I suggest that exten- servation priorities and resources, the outlook sive spraying of toxic chemicals for agricultural for any broadly based program to protect, much purposes must have locally serious effects on bat less to increase, dry forest is dim. populations in the dry-forest zone, if for no other Toxic Chemicals. A clear but hardly documented reason than the destruction of the bats’ prey. A threat to Neotropical bats is the poorly controlled ray of hope is that current progress in the use use of agricultural and other toxic chemicals. It of biological controls in place of chemicals may is generally believed that control of these chem- eventually lead to reduced mortality from this icals is more lax in Latin America than in North cause. America and Europe. However, this kind of in- Human Population Growth. The overriding formation is rarely published, and in the opin- threat to the dry forest and its biota is human ion of two biologists (D. R. Clark and M. Mora population pressure, which tends to exacerbate pers. comm.) who study effects of pesticides on all other negative effects of human activity on wildlife, there is no ﬁrm evidence that this is bats. Murphy and Lugo (1995) emphasized that true. On the contrary, the total amount of pesti- the population density of humans in the dry- cides applied per hectare has been and remains forest zone of Guatemala, Honduras, Nicaragua, much greater in the southwestern United States and Costa Rica was more than twice that of (where migratory bats are found) than in Latin moist forest and 6.6 times as dense as in the America, according to Clark and Mora. Clark’s wet forest. There is no evidence that population analyses of T. brasiliensis (a migratory species) growth in this part of the world will be halted in from Carlsbad Caverns, New Mexico, indicates the near future. Fortunately for bats as a group, that bats accumulate more DDE during their disturbed habitats will probably continue to pro- summer stay in the United States than during vide refuge for many species, some of which will their winter in Mexico. Clark and Mora also remain abundant. pointed out that many studies of migratory birds (with the exception of one study of peregrine fal- POTENTIAL EFFECTS OF cons) have failed to demonstrate that birds ac- GLOBAL CLIMATE CHANGE cumulate more pesticide residues while in Latin To date, most known effects of global warming America than in the United States. Finally, they are physical changes, indicated mainly by tem- note an almost complete lack of scientiﬁc studies perature records. Biologically it can be predicted of the effect of pesticide applications on wildlife that under warming conditions species whose in Latin America. distribution is limited by latitude and elevation Pesticides have sometimes been connected will move northward and to higher elevations, with bat mortality in the United States, although with loss of range at lower elevations and latitudes

ULTRASONICALLY SILENT NIGHT WITHOUT BATS 171 frankie chap 13 8/20/03 7:59 PM Page 172

(Parmesan 1996). By looking at a long-studied tions such as BCI, have gradually improved the species of butterfly found in the western United public image of bats (as chronicled in Bats mag- States, Canada, and Baja California, Parmesan azine over the past decade). (1996) was able to document these predicted Similar positive developments are being movements. Although comparable data do not initiated in Mesoamerica. BCI distributes bro- yet exist for bats, Scheel et al. (1996) constructed chures, videos, and slide shows in Spanish a model to predict the effect of global warming throughout the region, although relatively few on the habitat of all 27 bat species in Texas. Al- people have been reached so far, especially in though they predicted no extinctions within the Central America. In Mexico impressive progress state due to global warming, they did predict ma- has been made. An organization called Programa jor shifts in distribution and abundance, mainly para la Conservacíon de Murciélagos Migrato- because the model predicts major changes in rios (PCMM), a binational coalition of BCI, gov- vegetation types in the state. Cave- and crevice- ernment agencies, museums, universities, and roosting bats may have enhanced or decreased professional organizations, has been formed to foraging potential around areas where these protect bats that migrate between the United roost sites are available, thus affecting their pop- States and Mexico. A major goal of PCMM is to ulations. Scheel et al. also predicted that some educate children and adults in communities near species of tropical Mexican bats would be added major bat caves, and substantial progress has to the Texas bat fauna. been reported (Navarro et al. 1996). Recently Although no similar studies have yet been PCMM received a large grant for bat conserva- published for Mesoamerican bats, there is evi- tion from the Mexican Nature Conservation Fund dence that changes in temperature, rainfall pat- (Walker 1997). If this program proves to be a terns, and vegetation types affect bat populations success (as suggested by preliminary analysis of (LaVal in press), as has been documented for results; L. Navarro and J. Arroyo pers. comm.), birds, lizards, and frogs at Monteverde, Costa it could serve as a model for similar environ- Rica (Pounds et al. 1999). For example, some mental education programs in Central America. life zones on mountaintops and ridgetops may Another BCI program, an interactive museum cease to exist, thus eliminating habitat for species exhibit called “Masters of the Night,” has just that occur only in those life zones. Studies at toured Mexico after a successful tour of U.S. and Monteverde (LaVal in press) demonstrate that Canadian museums (Keleher 1996). the relative abundance of the two most com- Another avenue of education in bat conser- mon bat species has changed drastically over a vation is to take advantage of scientific studies 27-year period and that in general species from of trained bat biologists, both foreign and local. lower elevations are increasingly common. How- In Mesoamerica, only Mexico has a significant ever, with respect to the dry forest, this life zone number of local bat researchers, although there will likely occur at increasingly higher elevations are also several in Costa Rica. In these two coun- under global warming conditions, possibly aug- tries there has been substantial research on bats, menting the habitat available to dry-forest bats often carried out by foreign biologists. These (see also chapter 4). researchers are a priceless resource and should be involved in conservation efforts by local or- APPROPRIATE ENVIRONMENTAL EDUCATION ganizations and agencies. For example, in March As pointed out earlier, increased levels of bat 1997 Bernal Rodríguez, a Costa Rican bat spe- awareness in North America have led to greater cialist, and I organized and taught a workshop interest in conservation. Wider and more posi- on bat biology and conservation for employees tive media coverage of bats, books for children of Guanacaste and other conservation areas and the general public, videos, slide shows, and throughout the northern half of the country museum exhibits, often sponsored by organiza- (mostly from dry-forest parks). Subsequently

172 LAVAL frankie chap 13 8/20/03 7:59 PM Page 173

Rodríguez constructed and installed an excel- dresses, and facsimile copies of relevant papers. lent interactive bat exhibit, including a bat cave Bob Timm read and commented on an early with live bats, at the National Museum of Costa draft, and Dixie Pierson critically read and im- Rica. Thousands of Costa Ricans, especially proved a later draft. I thank my wife, Margaret, as large groups of school children, responded with well as Bernal Rodríguez for being continuous great enthusiasm to the exhibit, which lasted for sources of inspiration. two months and resulted in very favorable press coverage. REFERENCES In early 2001 a new program, called the Pro- grama para la Conservación de Murciélagos de Alvarez, T., and L. González-Quintero. 1970. Análi- sis polínico del contenido gástrico de murcié- Costa Rica, was initiated with the help of BCI lagos Glossophaginae de México. Anales Es- and PCMM. The focus of this program was cuela Nacional de Ciencias Biológicas, México educating school children living in or near the 18:137–165. conservation areas of Guanacaste about the bene- Arita, H. T. 1993. Conservation biology of the ﬁts of bats. A set of original children’s books in cave bats of Mexico. Journal of Mammalogy Spanish and other excellent teaching materials 74:693–702. Arita, H. T., and J. Ortega. 1998. The middle- have been provided by PCMM for use in the American bat fauna: Conservation in the Costa Rican program. Over the course of several Neotropical-Nearctic border. In Bat biology and years, I have given more than a hundred slide conservation, ed. T. H. Kunz and P. A. Racey, lectures on bats to groups of students and adults, 295–308. Washington, D.C.: Smithsonian In- usually followed by hands-on experience in mist- stitution Press. Arita, H. T., and K. Santos-del-Prado. 1999. Con- netting bats (stressing conservation) in Monte- servation biology of nectar-feeding bats in Mex- verde and Lomas Barbudal. No doubt other bi- ico. Journal of Mammalogy 80:31–41. ologists have carried out similar activities, and Arita, H. T., and D. E. Wilson. 1987. Long-nosed more should be encouraged to do so. Partic- bats and agaves: The tequila connection. Bats ular emphasis should be given to including 5(4):3–5. government employees charged with control- Bateman, G. C., and T. A. Vaughan. 1974. Nightly activities of mormoopid bats. Journal of Mam- ling vampires or interpreting park wildlife to malogy 55:45–65. the public. A binational partnership to protect Mexican free- tailed bats. 1994. Bats 12(4):6–7. Black, H. L. 1974. A north temperate bat commu- ACKNOWLEDGMENTS nity: Structure and prey populations. Journal of Mammalogy 55:138–57. The following people sent me literature, materi- Brosset, A., P. Charles-Dominique, A. Cockle, J.-F. als, and useful contacts with E-mail addresses; Cosson, and D. Masson. 1996. Bat communities answered my questions about their research; and deforestation in French Guiana. Canadian and in many cases allowed me to cite unpub- Journal of Zoology 74:1974–82. lished data: Gordon Frankie, Jeff Cosson, Ale- Ceballos, G. 1995. Vertebrate diversity, ecology, jandro Estrada, Hector Arita, Marco Tschapka, and conservation in Neotropical dry forests. In Seasonally dry tropical forests, ed. S. H. Bullock, Alfredo Pérez, Gerardo Ceballos, Manuel Maass, H. A. Mooney, and E. Medina, 195–220. Cam- Roger Blanco, Barry Hammel, Nelson Zamora, bridge: Cambridge University Press. Don Clark, and Miguel Mora. Book editors Tom Ceballos, G., and A. García. 1995. Conserving Kunz, Paul Racey, Nalini Nadkarni, and Nat Neotropical biodiversity: The role of dry for- Wheelwright gave me access to manuscripts ests in western Mexico. Conservation Biology 9:1349–53. of chapters in books not yet published as I was ———. 1996. La selva baja: Biodiversidad única en writing this chapter. Barbara French of BCI peligro. Ocelotl 5:4–9. was especially helpful in supplying me with in- Clark, D. R., Jr. 1981. Bats and environmental con- formation, names of publications, E-mail ad- taminants: A review. Special Science Report—

ULTRASONICALLY SILENT NIGHT WITHOUT BATS 173 frankie chap 13 8/20/03 7:59 PM Page 174

Wildlife No. 235. Washington, D.C.: U.S. De- Fenton, M. B., L. Acharya, D. Audet, M. B. C. partment of the Interior, Fish and Wildlife Ser- Hickey, C. Merriman, M. K. Obrist, and D. M. vice. 27 pp. Syme. 1992. Phyllostomid bats (Chiroptera: Clark, D. R., Jr., R. K. LaVal, and D. M. Swineford. Phyllostomidae) as indicators of habitat disrup- 1978. Dieldrin-induced mortality in an endan- tion in the Neotropics. Biotropica 24:440–46. gered species, the gray bat (Myotis grisescens). Findley, J. S. 1993. Bats: A community perspective. Science 199:1357–59. Cambridge: Cambridge University Press. 167 pp. Clark, D. R., Jr., C. M. Bunck, E. Cromartie, and Fleming, T. H. 1988. The short-tailed fruit bat: A R. K. LaVal. 1983. Year and age effects on study in plant-animal interactions. Chicago: Uni- residues of dieldrin and heptachlor in dead gray versity of Chicago Press. 365 pp. bats, Franklin County, Missouri—1976, 1977, ———. 1991. Following the nectar trail. Bats 9(4): and 1978. Environmental Toxicology and Chem- 4–7. istry 2:387–93. ———. 1997. The long and short of it: Compara- Clark, D. R., Jr., A. Moreno-Valdez, and M. A. Mora. tive ecology of two plant-visiting phyllostomid 1995. Organochlorine residues in bat guano bats. Abstract from the annual meeting of the from nine Mexican caves, 1991. Ecotoxicology 4: Association for Tropical Biology. Bat Research 258–65. News 39(2):40. Cockrum, E. L. 1969. Migration in the guano bat, Fleming, T. H., E. T. Hooper, and D. E. Wilson. Tadarida brasiliensis. University of Kansas Mu- 1972. Three Central American bat communities: seum of Natural History Miscellaneous Publica- Structure, reproductive cycles, and movement tions 31:303–36. patterns. Ecology 53:555–69. ———. 1991. Seasonal distribution of northwest- Gardner, A. L. 1977. Feeding habits. In Biology of ern populations of the long nosed bats, Lepto- bats of the New World family Phyllostomatidae, nycteris sanborni Family Phyllostomidae. Anales pt. 2, ed. R. J. Baker, J. K. Jones Jr., and D. C. Instituto de Biología Universidad Nacional Autó- Carter, 293–350. Lubbock: Texas Tech Press. nomo México. 62:181–202. Granjon, L., J. F. Cosson, J. Judas, and S. Ringuet. Constantine, D. G. 1970. Bats in relation to the 1996. Inﬂuence of tropical rainforest frag- health, welfare, and economy of man. In Biol- mentation on mammal communities in French ogy of bats, ed. W. A. Wimsatt, 2:320–449. New Guiana: Short-term effects. Acta Oecologica 17: York: Academic Press. 673–84. Cosson, J. F., J.-M. Pons, and D. Masson. 1999. Ef- Hall, E. R. 1981. The mammals of North America. fects of forest fragmentation on frugivorous 2 vols. New York: John Wiley and Sons. 1,181 pp. and nectarivorous bats in French Guiana. Jour- Heithaus, E. R., T. H. Fleming, and P. A. Opler. nal of Tropical Ecology 15:515–34. 1975. Foraging patterns and resource utilization de Jong, J. 1995. Habitat use and species richness in seven species of bats in a seasonal tropical of bats in a patchy landscape. Acta Theriologica forest. Ecology 56:841–54. 40:237–48. Herrera-Montalvo, L. G. 1997. Evidence of altitu- Estrada, A., R. Coates-Estrada, and D. Merritt Jr. dinal movements of Leptonycteris curasoae 1993. Bat species richness and abundance in (Chiroptera: Phyllostomidae) in Central Mexico. tropical rainforest fragments and in agricul- Revista Mexicana de Mastozoología 2:116–18. tural habitats at Los Tuxtlas, Mexico. Ecography Holdridge, L. R. 1967. Life zone ecology. San José: 16:309–18. Tropical Science Center. 206 pp. Fenton, M. B. 1982. Echolocation, insect hearing, Humphrey, S. R. 1975. Nursery roosts and commu- and feeding ecology of insectivorous bats. In nity diversity of Nearctic bats. Journal of Mam- Ecology of bats, ed. T. H. Kunz, 261–80. New malogy 56:321–46. York: Plenum Press. Humphrey, S. R., F. J. Bonaccorso, and T. L. Zinn. ———. 1997. Science and the conservation of bats. 1983. Guild structure of surface-gleaning bats Journal of Mammalogy 78:1–14. in Panamá. Ecology 64:284–94. Fenton, M. B., and I. L. Rautenbach. 1998. The Iñiguez, L. I. 1997. Ecology of Sturnira ludovici in impacts of ignorance and human and elephant the Sierra de Manatlan, Jalisco, Mexico. Abstract populations on the conservation of bats in from the annual meeting of the Association for African woodlands. In Bat biology and conser- Tropical Biology. Bat Research News 38(2):39. vation, ed. T. H. Kunz and P. A. Racey, 261– Janzen, D. H. 1983. No park is an island: Increase 81. Washington, D.C.: Smithsonian Institution in interference from outside as park size de- Press. creases. Oikos 41:402–10.

174 LAVAL frankie chap 13 8/20/03 7:59 PM Page 175

———. 1988. Tropical dry forests: The most en- Lott, E. J. 1993. Annotated checklist of the vascular dangered major tropical ecosystem. In Bio- ﬂora of the Chamela Bay Region, Jalisco, Mex- diversity, ed. E. O. Wilson, 130–37. Washington, ico. Occasional Papers of the California Academy D.C.: National Academy Press. of Science 148:1–60. Kalko, E. K. V. 1997a. Diversity in tropical bats. In Maass, J. M. 1995. Conversion of tropical dry for- Tropical biodiversity and systematics: Proceedings est to pasture and agriculture. In Seasonally dry of the International Symposium on Biodiversity and tropical forests, ed. S. H. Bullock, H. A. Mooney, Systematics in Tropical Ecosystems, ed. H. Ulrich, and E. Medina, 399–418. Cambridge: Cam- 13–43. Bonn: Zoologisches Forschungsinstitut bridge University Press. und Museum Alexander Koenig. Marinho-Filho, J., and I. Sazima. 1998. Brazilian ———. 1997b. Conservation of Neotropical bats: bats and conservation biology: A ﬁrst survey. In Monitoring techniques, community compar- Bat biology and conservation, ed. T. H. Kunz and isons, and long-term studies. Abstract from the P. A. Racey, 282–94. Washington, D.C.: Smith- annual meeting of the Association for Tropical sonian Institution Press. Biology. Bat Research News 39(2):40. McCabe, S. 1995. Bob Currie: The quiet conserva- ———. 1998. Organisation and diversity of tropi- tionist. Bats 13(3):12–15. cal bat communities through space and time. McCracken, G. F. 1988. Who’s endangered and Zoology 101:281–97. what can we do? Bats 6(3):5–9. Kalko, E. K. V., C. O. Handley Jr., and D. Handley. ———. 1996. Bats aloft: A study of high-altitude 1996. Organization, diversity, and long-term feeding. Bats 14(3):7–10. dynamics of a Neotropical bat community. In Medellín, R. A., and O. Gaona. 1999. Seed dis- Long-term studies of vertebrate communities, ed. persal by bats and birds in forest and disturbed M. L. Cody and J. A. Smallwood, 503–53. New habitats in Chiapas, México. Biotropica 31: York: Academic Press. 478–85. Keleher, S. 1996. The Masters of the Night Exhibit: Murphy, P. G., and A. E. Lugo. 1995. Dry forests of A winner for bats. Bats 14(2):3–6. Central America and the Caribbean. In Season- Kunz, T. H. 1982. Roosting ecology of bats. In Ecol- ally dry tropical forests, ed. S. H. Bullock, H. A. ogy of bats, ed. T. H. Kunz, 1–55. New York: Mooney, and E. Medina, 9–34. Cambridge: Plenum Press. Cambridge University Press. Kunz, T. H., and P. A. Racey, eds. 1998. Bat biology Murray, K. G., S. Kinsman, and J. L. Bronstein. and conservation. Washington, D.C.: Smithson- 2000. Plant-animal interactions. In Monte- ian Institution Press. verde: Ecology and conservation of a tropical cloud Kunz, T. H., J. O. Whitaker Jr., and M. D. Wadanoli. forest, ed. N. M. Nadkarni and N. T. Wheel- 1995. Dietary energetics of the insectivorous wright, 245–302. New York: Oxford University Mexican free-tailed bat (Tadarida brasiliensis) Press. during pregnancy and lactation. Oecologia 101: Navarro, L., J. Arroyo, and R. Medellín. 1996. Bat 407–15. awareness in Mexico begins with children. Bats LaVal, R. K. In press. Impact of global warming 14(3):3–7. and locally changing climate on tropical cloud Ochoa, J. 1992. Venezuela’s bats: A case for con- forest bats. Journal of Mammalogy. servation. Bats 10(3):10–13. LaVal, R. K., and H. S. Fitch. 1977. Structure, move- ———. 2000. Efectos de la extracción de maderas ments and reproduction in three Costa Rican bat sobre la diversidad de mamíferos pequeños en communities. Occasional Papers of the Museum bosques de tierras bajas de la Guayana Vene- of Natural History, University of Kansas 69:1–28. zolana. Biotropica 32:146–64. López, J. 1996. Habitos alimentarios de murciéla- Parmesan, C. 1996. Climate and species range. gos frugívoros y su participación en la disper- Nature 382:765–66. sión de semillas, en bosques secundarias Pierson, E. D. 1998. Tall trees, deep holes, and húmedos de Costa Rica. Thesis. Programa Re- scarred landscapes: Conservation biology of gional en Manejo de Vida Silvestre para Meso- North American bats. In Bat biology and conser- america y el Caribe—Universidad Nacional, vation, ed. T. H. Kunz and P. A. Racey, 309– Heredia, Costa Rica. 73 pp. 25. Washington, D.C.: Smithsonian Institution Lord, R. D. 1988. Control of vampire bats. In Nat- Press. ural history of vampire bats, ed. A. M. Greenhall Pierson, E. D., and P. A. Racey. 1998. Introduction and U. Schmidt, 216–26. Boca Raton, Fla.: CRC to pt. 4, Conservation biology. In Bat biology and Press. conservation, ed. T. H. Kunz and P. A. Racey,

ULTRASONICALLY SILENT NIGHT WITHOUT BATS 175 frankie chap 13 8/20/03 7:59 PM Page 176

247–48. Washington, D.C.: Smithsonian Insti- meeting of the Association for Tropical Biology. tution Press. Bat Research News 38(2):39. Pounds, J. A., M. P. L. Fogden, and J. H. Campbell. Thorne, J. 1990. Bats and cavers. Bats 8(1):10–14. 1999. Biological response to climate change on Timm, R. M. 1994. The mammal fauna. In La a tropical mountain. Nature 398:611–15. Selva: Ecology and natural history of a Neotropical Ramírez-Pulido, J., and M. A. Armella. 1987. Ac- rain forest, ed. L. A. McDade, K. S. Bawa, H. A. tivity patterns of Neotropical bats (Chiroptera: Hespenheide, and G. S. Hartshorn, 229–37. Chi- Phyllostomidae) in Guerrero, Mexico. South- cago: University of Chicago Press. western Naturalist 32:363–70. Timm, R. M., and R. K. LaVal. 1998. A ﬁeld key to Ramírez-Pulido, J., A. Martínez, and G. Urbano. the bats of Costa Rica. Occasional Publications 1977. Mamíferos de la Costa Grande de Guer- Series, Center for Latin American Studies, Univer- rero, México. Anales Instituto de Biología Uni- sity of Kansas 22:1–30. versidad Nacional Autónomo México 48:243–92. ———. 2000. Mammals. In Monteverde: Ecology Robinson, M. F., and R. E. Stebbings. 1997. Home and conservation of a tropical cloud forest, ed. N. M. range and habitat use by the serotine bat, Eptesi- Nadkarni and N. T. Wheelwright, 223–44. New cus serotinus, in England. Journal of Zoology, York: Oxford University Press. London 243:117–36. Tuttle, M. D. 1976. Collecting techniques. In Biol- Rojas-Martinez, A., A. Valiente-Banuet, M. del ogy of bats of the New World family Phyllostom- Coro Arizmendi, A. Alcántara-Eguren, and H. T. atidae, pt. 1, ed. R. J. Baker, J. K. Jones Jr., and Arita. 1999. Seasonal distribution of the long- D. C. Carter, 71–88. Lubbock: Texas Tech Press. nosed bat (Leptonycteris curasoae) in North Amer- ———. 1988. Introduction to the natural history of ica: Does a generalized migration pattern really vampire bats. In Natural history of vampire bats, exist? Journal of Biogeography 26:1065–77. ed. A. M. Greenhall and U. Schmidt, 1–6. Boca Scheel, D., T. L. S. Vincent, and G. N. Cameron. Raton, Fla.: CRC Press. 1996. Global warming and the species richness Tuttle, M. D., and D. Hensley. 1993. Bat houses: of bats in Texas. Conservation Biology 10:452–64. The secrets of success. Bats 11(1):3–14. Schneider, M. C., and C. Santos-Burgoa. 1995. Al- Tuttle, M. D., and D. A. R. Taylor. 1998. Bats and gunas consideraciones sobre la rabia humana mines. Bat Conservation International, Resource transmitida por murciélago. Salud Pública Méx- Publications 3:1–50. ico 37:354–62. Valiente-Banuet, A., M.del C. Arizmendi, A. Rojas- Shefﬁeld, S. R., J. H. Shaw, G. A. Heidt, and L. R. Martínez, and L. Domínguez-Canseco. 1996. McClenaghan. 1992. Guidelines for the protec- Ecological relationships between columnar tion of bat roosts. Journal of Mammalogy 73: cacti and nectar-feeding bats in Mexico. Journal 707–10. of Tropical Ecology 12:103–19. Simmons, N. B., and R. S. Voss. 1998. The mam- Walker, S. 1995. Mexico-U.S. partnership makes mals of Paracou, French Guiana: A Neotropical gains for migratory bats. Bats 13(3):3–5. lowland rainforest fauna. Pt. 1, Bats. Bulletin of ———. 1997. Murciélagos de Nuevo León. Bats the American Museum of Natural History 237: 15(1):16. 1–219. Willig, M. R., G. R. Camilo, and S. J. Noble. 1993. Sosa, M., A. De Ascençào, and P. J. Soriano. 1996. Dietary overlap in frugivorous and insectivorous Dieta y patrón reproductivo de Rhogeessa minu- bats from edaphic cerrado habitats of Brazil. tila (Chiroptera: Vespertilionidae) en una zona Journal of Mammalogy 74:117–28. árida de Los Andes de Venezuela. Revista de Bio- Wilson, D. E. 1983. Checklist of mammals. In Costa logía Tropical 44:867–75. Rican natural history, ed. D. H. Janzen, 443–47. Stiles, F. G. 1988. Altitudinal movements of birds Chicago: University of Chicago Press. on the Caribbean slope of Costa Rica: Implica- ———. 1996. Neotropical bats: A checklist with tions for conservation. In Tropical rainforests: conservation status. In Neotropical biodiversity Diversity and conservation, ed. F. Almeda and and conservation, ed. A. C. Gibson, 167–77. Los C. M. Pringle, 243–58. San Francisco: California Angeles: Mildred E. Mathias Botanical Garden, Academy of Science and American Association University of California, Los Angeles. for the Advancement of Science. Wilson, D. E., C. F. Ascorra, and S. Solari. 1998. Stoner, K. E. 1997. Changes of abundance and sex Bats as indicators of habitat disturbance. In ratio of frugivorous and nectarivorous bats in Manu: The biodiversity of southeastern Peru, ed. tropical dry forest and their implications for D. E. Wilson and A. Sandoval, 613–26. Wash- seasonal migration. Abstract from the annual ington: Smithsonian Institution Press.

176 LAVAL frankie chap 14 8/20/03 8:00 PM Page 177

chapter 14

Biodiversity and Conservation of Mesoamerican Dry-Forest Herpetofauna

Mahmood Sasa and Federico Bolaños

he herpetofauna of Mesoamerica (de- studies of the amphibians and reptiles of north- Tﬁned as the region running south from about western Costa Rica (Sasa and Solórzano 1995) central Mexico through Panama) is undoubtedly and on extensive material recently collected from one of the richest and most complex vertebrate the dry forests of Mexico (Oaxaca, Chiapas), Gua- faunas of the New World. It involves more than temala, Honduras, and Nicaragua. We begin by 210 genera, comprising approximately 693 spe- summarizing biodiversity and biogeographical cies of reptiles and 598 species of amphibians. aspects of reptiles and amphibians associated Such high diversity results from the divergence with dry environments in Mesoamerica. We then of species that evolved in situ in Mesoamerica, as describe general patterns of resource use of dry- well as from the interchange of species between forest reptiles and amphibians and their adapta- North and South America (Duellman 1966; Sav- tion to arid conditions, emphasizing the roles age 1966). that some species have in the community and Based on the distribution of reptiles and the threat that human activities might pose to amphibians in Mesoamerica, Duellman (1966) them. Finally, we discuss the importance of recognized ﬁve major ecological assemblages: conservation of this herpetofaunal assemblage humid tropical, humid mountain, high moun- in terms of area selection and conservation. tain, arid mountain, and arid tropical. This last ecological unit, the tropical arid assemblage, in- THE DRY-FOREST ENVIRONMENTS cludes species distributed in subhumid lowlands OF MESOAMERICA covered by scrub forest, savannas, and deciduous vegetation and is the primary focus of this Approximately 26 percent of the surface of Meso- chapter. We based our account on preliminary america is seasonally dry, covered by different

177 frankie chap 14 8/20/03 8:00 PM Page 178

MAP 14.1. Lowland tropical dry forest in Mesoamerica. Numbers refer to the localities listed in table 14.1; locality 1 is out of the range of the map. Modiﬁed from Wilson and McCranie (1998).

xerophytic and deciduous vegetation types (Ce- weg and Oliver 1940; Duellman 1960), north- ballos 1995; Murphy and Lugo, 1995). These veg- western nuclear Central America (Johnson 1989), etation associations occur in five geographical the Yucatán and dry savannas in Petén (Stuart regions: (1) the Pacific coastal plains and foot- 1935; Duellman 1965a; Lee 1996; Campbell hills northwest of Tapachula (Mexico), in the 1998), and interior dry valleys (Schmidt and Stu- Isthmus of Tehuantepec, (2) the Pacific Coast of art 1941; Stuart 1954a; Campbell and Vannini Central America, (3) the lowlands of the outer- 1988; Wilson and McCranie 1998). Also see most portion of the Yucatán Peninsula, (4) the Mertens (1952), Stuart (1954b), Rand (1957), Central Depression of Chiapas, and (5) the series Savage and Villa (1986), Campbell and Vannini of interior rain shadow valleys of Guatemala and (1989), Sasa and Solórzano (1995), and Stafford Honduras (map 14.1). A disjunct dry-forest area (1998) for data on the herpetological commu- also occurs in the Azuero Peninsula, Panama nities along the Pacific versant from southern (Holdridge and Budowski 1959; Murphy and Guatemala to Costa Rica. Lugo, 1995), but has been poorly studied. Overall, 266 species of amphibians and reptiles are known to occur in Mesoamerica’s arid environments. The total number of species in- SPECIES COMPOSITION AND RICHNESS cludes 2 caecilians, 5 salamanders, 50 anurans, Species accounts for amphibians and reptiles are 13 turtles, 3 crocodilians, 76 lizards, and 117 available for several localities in Mesoamerican snakes. arid regions. Localities that have been surveyed Species compositions for 20 tropical dry- include Jalisco (García and Ceballos 1994; forest localities of Mesoamerica are presented Ramírez-Bautista 1994), Michoacán (Duellman in table 14.1. We also include data from three 1961, 1965b), the Isthmus of Tehuantepec (Hart- wet-forest localities (La Selva, Osa Peninsula,

178 SASA AND BOLAÑOS frankie chap 14 8/20/03 8:00 PM Page 179

and Barro Colorado Island), as well as a list for Likewise, general aspects of ecology and natural Sonora, located in extratropical Mexico. We quan- history of the majority of dry-forest reptile and tiﬁed the number of species of amphibians, amphibian species are poorly known or are re- turtles, crocodilians, saurians, and snakes rela- stricted primarily to the rainy season. tive to the total number of species at each locality and then compared the resulting proportion DIEL ACTIVITY of species per locality. No differences were ob- Activity patterns within a species can vary de- served in the proportion of species inhabiting pending on locality and season, but general- tropical dry localities (table 14.1, G-statistic = izations can be offered regarding reptiles and 49.8, d.f. = 69, P = 0.961). However, the pro- amphibians of dry forests. Duellman (1965b) portion of species of each taxonomical category recognized two patterns of diel activity in Micho- differs between dry and wet localities (G- acán: diurnal (51.1% of reported species) and noc- statistic = 120.1, d.f. = 3, P < 0.001). Furthermore, turnal (48.9% of species). In Sector Santa Rosa the mean number of species in dry localities is (Guanacaste Conservation Area, Costa Rica), 69, whereas localities in wet lowland forest have 37 percent of the species can be considered di- a mean of 123 species. The most visible differ- urnal, in particular lizards and their snake pred- ence in species composition between dry and ators, whereas 34 percent of the species (almost wet-forest localities is in amphibian richness. all geckos, all frogs, and most of their snake pred- For the 20 dry-forest sites, amphibians repre- ators) are nocturnal (Sasa and Solórzano 1995). sent merely 26.2 percent of the total number of The remaining percentage includes species with species (percentage range: 14.1–28.0), whereas unknown patterns or that are active during both in wet-forest sites they account for 39 percent night and day. (percentage range: 37.1–39.4). MICROHABITAT USE As in other ecosystems, reptiles and amphibians COSTA RICAN that inhabit seasonal dry forests use a variety DRY-FOREST HERPETOFAUNA of microhabitats, depending on species and sea- Costa Rica is a country with a well-known her- son. Amphibians are usually observed above petofauna composed of a total of 397 species, ground only during the rainy season, and we 174 amphibians and 223 reptiles (table 14.2). assume that most species enter a state of estiva- The dry forest contains 21 percent of the coun- tion and burrow in the ground during the arid try’s total number of species. With one caecilian, months (see section “Adaptations to Highly Sea- no salamanders, and few species of frogs and sonal Climate” later in this chapter). Dry-forest toads, the dry forest differs greatly in amphibian frogs, lizards, and snakes are found in ﬁve gen- composition from the country as a whole. The erally exclusive microhabitats (table 14.3). Over- proportion of reptiles, on the other hand, is sim- all, 18 percent of the species are fossorial or live ilar for both Costa Rica as a whole and areas of under the leaf-litter, 29 percent are found exclu- dry forest (table 14.2, G-statistic = 5.733, d.f. = 3, sively on the ground, 7 percent are associated P = 0.126), varying from 4 species of turtles to with rivers and ponds, 14 percent inhabit bushes 34 species of snakes. and tree trunks, and the remaining 30 percent are likely to be found both on the ground and in bushes. The majority of anurans and snakes in RESOURCE USE dry forests are ground dwellers, whereas lizards To our knowledge, a quantitative study of the rel- tend to use more bushes and tree trunks. This ative abundance of species in an arid herpeto- pattern differs from the patterns in other eco- fauna community has never been performed. systems. In extratropical arid regions, almost a

HERPETOFAUNA BIODIVERSITY AND CONSERVATION 179 frankie chap148/20/038:00PMPage180

TABLE 14.1 Comparison of Herpetofaunas at 24 Mesoamerican Localities

caecilians and anurans turtles and sauria serpentes locality salamanders (frogs/toads) crocodilians (lizards) (snakes) total mexico 1. Sonora 2 26 9 89 61 187 2. Michoacán 0 20 4 28 28 80 3. Jalisco 0 19 4 21 37 81 4. Guerrero 1 19 7 24 47 98 5. Tehuantepec 1 22 9 27 46 105 6. Yucatán 1 15 7 21 47 91 7. Grijalva Valley 1 19 3 27 36 86 guatemala 8. Río Negro 0 13 0 10 15 38 9. Motagua Valley 1 12 3 15 22 53 10. Pacific Guatemala 2 15 6 18 27 68 el salvador 11. Pacific El Salvador 2 17 5 21 34 79 honduras 12. Pacific Honduras 2 10 3 13 21 49 13. Choluteca Valley 1 13 2 16 23 55 14. Comayagua Valley 0 11 2 8 15 36 15. Otoro Valley 0 11 1 8 3 23 frankie chap148/20/038:00PMPage181

16. Sula Plain 3 15 7 20 45 90 17. Aguán Valley 1 10 0 15 11 37 18. Guaype Valley 1 16 5 14 14 50 nicaragua 19. Paciﬁc Nicaragua 2 18 5 20 35 80 costa rica 20. North Paciﬁc Costa Rica 1 19 5 17 34 76 21. Cañas 1 22 6 16 34 79 22. La Selva 4 41 6 23 45 119 23. Osa Peninsula 6 40 5 22 44 117 panama 24. Barro Colorado 3 50 7 26 47 133 Total 17 124 27 193 210 571

Note: Locality 1 is extratropical, localities 2–21 correspond to tropical dry forest, and localities 22–24 correspond to lowland wet forest in isthmian Central America. frankie chap 14 8/20/03 8:00 PM Page 182

TABLE 14.2 third of the snake species are fossorial, whereas Number of Species of Amphibians and Reptiles in most lizards are ground dwellers (table 14.3). Costa Rica and in the Costa Rican Dry Forest Conversely, in more mesic forest, a signiﬁcantly costa rica dry forest higher number of reptiles and amphibians live in the canopy and tree trunks (Guyer 1990; amphibians table 14.3). The lack of canopy dwellers in dry Cecilians 4 1 forest might be related to the great proportion (close to 100% in nonriparian habitats) of trees Salamanders 37 — that are deciduous and to differences in density, Frogs and toads 133 24 height, and productivity of dry-forest associations (Gentry 1995). reptiles Lizards 73 17 FEEDING ECOLOGY Snakes 134 34 Diet analyses for the majority of amphibians Turtles 14 4 and reptiles inhabiting dry regions of Meso- america are scarce, and valuable information Crocodiles 2 2 on prey types, foraging behavior, and temporal Total 397 81 and spatial feeding activity for most species is still lacking. In general, a predator can follow two major feeding strategies: (1) diet specialization toward one (or a few) prey items, or (2) an increase in the variety of diet items accepted. These two modes should be considered as ex-

TABLE 14.3 Percentage of Species of Anurans, Lizards, and Snakes Found in Five Microhabitats in Three Mesoamerican Ecosystems

fossorial ground ground and leaf and aquatic arboreal (exclusive) litter bushes

tropical dry forest Anurans 10.46 7.38 48.70 9.86 23.51 Lizards 0.00 38.10 28.34 4.29 28.27 Snakes 2.52 9.99 47.40 17.66 22.40

extratropical dry forest Anurans 16.67 8.33 58.33 12.51 4.16 Lizards 0.00 25.00 65.00 3.33 6.67 Snakes 1.81 1.81 50.90 30.90 14.54

tropical wet forest Anurans 15.15 33.33 38.63 3.03 23.48 Lizards 0.00 53.12 26.56 12.50 7.81 Snakes 0.77 17.05 45.73 16.27 20.15

182 SASA AND BOLAÑOS frankie chap 14 8/20/03 8:00 PM Page 183

tremes in a continuum of possible feeding ingly, prey types from this wet-forest community strategies. Pianka (1986) indicates that in cir- do not differ substantially from those observed in cumstances of scant food supply, mean search Santa Rosa by Sasa and Solórzano (1995). How- time per item encountered is long, and therefore ever, with a less seasonal regimen, anurans and the rate of prey encounter is low. In a tropical dry slugs are the most common prey items of snakes. forest, food abundance varies both in space and At Izabal, 30 percent of the snake species prey time. Most anurans seem to have generalist di- on frogs, whereas in Santa Rosa only 12 percent ets, preying on insects and other invertebrates, of the snake fauna feeds on them. but some are specialists (i.e., Hypopachus variolosus and Rhynophrynus dorsalis, which prey on REPRODUCTION termites). Lizards also feed on small invertebrates, with some exceptions being the herbivo- In dry forest, the majority of species have a rous green iguana (Iguana iguana), the omni- marked breeding season. In lizards and snakes, vorous black iguana (Ctenosaura), and basilisk mating starts early in the year, during the lizard (Basiliscus). In contrast to amphibians and middle of the dry season. In some snakes (e.g., lizards, snakes are dietary specialists. Sasa and Crotalus durissus; Solórzano and Cerdas 1988) Solórzano (1995) recognized three foraging females are reproductively active during the be- modes in Santa Rosa National Park’s dry forest ginning of the rainy season (May to October). It (Costa Rica): is not clear what factor or factors trigger reproductive activity; however, some indirect evidence 1.Absolute specialist: snakes with a low num- suggests that seasonal changes in precipitation ber of prey species. This category is com- have an effect on reproductive cycles of the posed of 31 percent of the species with lizards Norops sagrei (Brown and Sexton 1973) known diets for that locality and includes, and N. cupreus (Fleming and Hooker 1975). among others, Leptotyphlops goudotii, which Reproductive aspects of amphibians are feeds on termites; Scolecophis atrocinctus, highly diverse, and some reproductive modes which preys on myriapods; and Sibon an- and life cycles found in dry forest are obvious thracops, which feeds on slugs and snails. adaptations to subhumid conditions. In anurans, 2.Specialists on morpho-types (40%): speciesreproduction is highly synchronized and de- that restrict themselves to prey of the same pends on water availability and relative humid- type but whose prey comprises several ity. Several burrowing frogs (e.g., Rhynophrynus species (e.g., ophiophagous Micrurus nigro- dorsalis, Hypopachus variolosus) are explosive cinctus and the spider eater Stenorrhina breeders, and their reproduction is restricted to freminvilli). a few days at the beginning of the rainy season. 3.Generalists (29%): species whose diets in-In 1993, for example, only during three nights clude a variety of species of diverse types were R. dorsalis heard calling in Santa Rosa Na- (e.g., Drymarchon corais, Boa constrictor, tional Park. Conophis lineatus). Five reproductive modes can be found in amphibians inhabiting dry forest: (1) viviparity, The majority of snakes inhabiting dry forest (2) direct development, (3) eggs deposited in in Santa Rosa are stenophagous (categories 1 water, (4) eggs deposited in vegetation, and and 2), which may represent an adaptation to re- (5) foam nest. The ﬁrst mode is found only in duce competition. High food partitioning has caecilians and thus is the least common strategy. also been observed in other snake communities Among amphibians that lay eggs, direct devel- (Vitt 1987). In the rain forests of Izabal (eastern opment is also a rare mode in dry forests, found Guatemala), Smith (1994) found that snake diets only in Plethodontid salamanders and the frog do not overlap much among species. Interest- genus Eleutherodactylus. This strategy usually

HERPETOFAUNA BIODIVERSITY AND CONSERVATION 183 frankie chap 14 8/20/03 8:00 PM Page 184

involves large eggs that take a long period to de- 1968), which affects ecological aspects of the velop, hence increasing the risk of desiccation. species. In general, adaptations to drought in- Direct development has been mentioned as an clude seasonal reproductive cycles (discussed adaptation for dry conditions in Australian desert previously), seasonal changes in niche and be- species (Tyler 1994), but in the dry forest of Meso- havior, and estivation. america, direct development occurs only in pe- Seasonal changes in diet, habitat, and activ- ripheral species whose closest relatives inhabit ity are evident in the lizard Norops cupreus when wet forests. The great majority of species (58%) niche overlap is greater during the dry season deposit their eggs in water, usually in shallow (especially for perch sites) than during the rainy places (temporal or permanent ponds) where season (Fleming and Hooker 1975). This species predators are scarce. In those species, eggs and also exhibits variation in the mean prey size all stages of metamorphosis occur in the water. consumed by males and females, a feature that Other species, in particular some leptodactylids might reduce intersexual competition for food (Leptodactylus labialis, L. melanonotus, and Physa- during the dry season (Fleming and Hooker laemus pustulosus), construct foam nests to avoid 1975). Seasonal diet shifts have also been reported desiccation of eggs and larvae. In this reproduc- for Basiliscus basiliscus (Van Devender 1983). tive mode, early stages of metamorphosis might Behavioral responses are among the most im- occur in the foam, but the larvae finish their portant means for dealing with hot, dry seasons. development in water. Thus, the foam nest is Reptiles and amphibians can restrict their activ- an adaptation to dry environments that allow the ity to the cooler parts of the day, move to shaded use of temporal bodies of water. Finally, few frogs areas for protection, or seek shelter and become (Agalychnis callidryas, Hyla ebraccata) lay their inactive during periods of hot, dry weather. eggs in the vegetation, but the larvae develop Turtles and crocodilians cool down in water dur- in water. As in the case of species with direct ing the hottest hours of the day. In Jalisco and development, these species face the problem of the Yucatán, frogs of the genus Triprion plug desiccation of the egg mass. the openings of their tree holes with their head, There are remarkable differences between a behavior that may prevent predators from the reproductive modes used by dry-forest and capturing them (Duellman 1960). This behavior, wet-forest amphibians. In wet forest, a great termed phragmosis, may also help in preventing proportion (54%) of amphibians lay their eggs desiccation. out of water (direct development and eggs de- In arid regions, weight loss and an increase posited in vegetation), and there are relatively in electrolytes in the plasma have been reported fewer foam-nest anurans. Furthermore, in wet for several reptiles and amphibians (Ruibal et al. forest a sixth reproductive strategy can be iden- 1969; Minnich 1977). In those cases, feeding may tified: eggs deposited on land and larvae trans- cease, apparently to prevent further increase in ported to water (frog family Dendrobatidae), a plasma electrolytes when there is insufficient mode that is absent in dry forests. water to excrete them (Gregory 1982). Under these conditions, animals become inactive and enter estivation, a physiological state that paral- ADAPTATIONS TO HIGHLY lels hibernation. Factors stimulating onset of SEASONAL CLIMATE estivation are unknown, although high temper- Like many other animals living in arid areas, ature, drought, and dehydration of food have amphibians and reptiles in Mesoamerican dry been mentioned. Estivation can be obligate or forests must be able to tolerate high tempera- facultative depending on the species and area. For tures, low precipitation, and low relative humid- dry-forest reptiles, estivation has been reported ity. In addition, resource availability is highly sea- for Heloderma horridum (Bogert and Martín del sonal in tropical dry forests (Janzen and Schoener Campo 1993).

184 SASA AND BOLAÑOS frankie chap 14 8/20/03 8:00 PM Page 185

Amphibians also avoid periods of dehydra- body temperatures or keeping an adequate mois- tion by burrowing into a moist substrate and ture level in the body during dry season. entering a state of dormancy. The permeability of amphibian skin to water—a characteristic THREATENED AND EXTINCT SPECIES that always has been viewed as a disadvantage for their distribution in arid environments— Information on population biology for most actually might be an adaptive advantage because species inhabiting arid regions of Mesoamerica it allows effective absorption of water from the is almost nonexistent. This precludes a direct soil (Ruibal et al. 1969). Secretions of the skin analysis of the status of those populations, and are also important adaptations. For example, in therefore an examination of the human impact the hylid Phrynohyas venulosa, the skin glands over the species must be accounted for indirectly. become more developed during the dry season There is still considerable debate about the (Duellman 1956; McDiarmid 1968). These glands quality of natural ecosystems in arid regions of secrete a mucous excretion that might confer Mesoamerica. Moreover, it is not yet clear if the some protection against desiccation. Other spe- reduced forest in northwestern Costa Rica con- cies of frogs, including desert dwellers (Lee and stitutes an example of the magnificent forest that Mercer 1967) and the dry-forest-inhabiting once extended from south Guatemala to Costa Smilisca baudinii (McDiarmid and Foster 1987) Rica or if the savannas in arid regions of nuclear and Pternohyla fodiens (Ruibal and Hillman 1981), Central America are natural or the result of form estivation cocoons. These cocoons are human activities (Johnson 1989). It is clear, how- formed from numerous layers of the epidermal ever, that human activities have substantially stratumcorneum separated by subcorneal spaces modified the biotic and abiotic composition of filled with mucuslike secretions (Ruibal and Hill- arid areas. The process might have started 3,000– man 1981). McDiarmid and Foster (1987) sug- 3,500 years ago, when the distinctive cultures gest that cocoon-forming frogs might burrow and civilizations of Mesoamerica emerged (Wil- near the surface and thus are stimulated by the ley et al. 1964). There are two direct threats to start of the wet season’s first rains. In this way, the herpetofauna of the region: hunting pres- they become active earlier than more deeply bur- sures and habitat loss (or its modification). In rowing species. general, reptiles and amphibians are considered Frogs that do not form cocoons have other by most modern locals as dangerous and dirty strategies. In Bufo marinus, shallow burrows are animals, some of them (snakes) dramatically so. created and the nares are usually exposed to the However, in the past (as well as in the present) surface, enabling the toad to avoid periods of some species have played an important role in unfavorable conditions on the surface. In a hy- tradition, medicine, and diet. Thus, it is neces- drated B. marinus, most carbon dioxide of res- sary to address the relationship of the local her- piration is eliminated by cutaneous diffusion. petofauna with human societies from a historical When the toad has burrowed (and when the cu- perspective. A comprehensive account on ethno- taneous moisture is low), skin gas exchange is herpetology of Mesoamerican civilizations is reduced, and the toad responds with a reduction beyond the scope of this chapter, but the role in breathing rate (Boutilier et al. 1979). Codeco that some species of reptiles and amphibians (1999) found that breathing frequency in B. ma- have played in the Mayan-Nahual societies is ad- rinus could be reduced to enter long periods of dressed in J.C. Lee’s (1996) chapter on the eth- apnea, lasting up to eight hours. This long pe- noherpetology of the Yucatán and in Gutiérrez- riod of apnea, together with torpidity, suggests González’s (1993) account on the vertebrate that metabolism is reduced during burrowing. remains at Nacascole (Nicoya, Costa Rica). All of these mechanisms lower the risk of death Mayan civilization extended from the Isthmus from overheating and desiccation by reducing of Tehuantepec to the western part of Honduras

HERPETOFAUNA BIODIVERSITY AND CONSERVATION 185 frankie chap 14 8/20/03 8:00 PM Page 186

and El Salvador and was characterized by a great stitute a major source of protein for many human architectural development in Petén and the Yu- populations along coastal areas (see chapter 15). catán Peninsula. To the south, a Nahual cultural Snakes are usually considered dangerous ani- region extended along the Pacific coast of Hon- mals, and although most people know that not duras, Nicaragua, and northern Costa Rica. From all snakes are venomous, they are commonly these two cultures, it is common to find repre- killed. This “just in case” attitude derives from sentations of amphibians and reptiles in ar- the lack of information on how to discriminate chaeological artifacts, especially in monuments, between venomous and harmless snakes. One paintings, stelae, and ceramics (Ferrero 1987). species, the rattlesnake Crotalus durissus, is as According to Lee (1996), increasing demand for important in popular medicine of modern cul- crop production would have reduced populations ture as it was in the Mayan-Nahual tradition. This of game animals, thus increasing consumption species is hunted for its putative curative prop- of animals such as amphibians and reptiles. In erties, for it is a common Mesoamerican belief support of this idea, there are massive remains that the dry carcass of a rattlesnake, once pul- of reptiles in archaeological sites throughout the verized, is a potent supplement to the diet. The lowland Mayan area, from Cozumel to Lubaan- powder of rattlesnake is also considered to be tun. Common remains include bones of fresh- effective against any type of cancer and anemia water turtles (Rhinoclemmys areolata, Trachemys (Campbell 1998), a belief that is deeply rooted scripta, Kinosternon sp., Dermatemys mawii), main rural communities along Central America. rine turtles (Chelonia mydas), lizards (Ctenosaura In recent years, several species of reptiles similis, Iguana iguana), and crocodiles (Crocody- and amphibians in tropical Mesoamerica have lus moreletii). The iguanas Ctenosaura similis and been hunted for the pet trade. There is increas- I. iguana were also extensively consumed along ing interest in domestically keeping tropical the Pacific coast of Nicaragua and the Nicoya amphibians and reptiles, especially in North Peninsula (Gutiérrez-González 1993). Another America, Europe, and some Asiatic countries. list of uses reflects medicinal and religious be- Breeding amphibians and reptiles in captivity liefs. The fat bodies of Iguana and Ctenosaura has yielded information about their reproduc- were, and still are, commonly used as remedies tion and other aspects of their biology, as well as for a list long of illnesses. Some toads and frogs created some renewable resources. Although less (Bufo marinus, Rhinophrynus dorsalis) were used quantifiable, another positive aspect of the pet to disinfect wounds or in ceremonies. Snakes trade is the exposure of reptiles and amphibians played a very important role in the mythological to the general public, allowing a positive attitude tradition of Mayas and Nahuals, in particular toward conservation to be nurtured. However, the vipers Crotalus durissus, Bothrops asper, and the pet trade has also produced some undesir- Agkistrodon bilineatus. These snakes are also com- able consequences, such as the high rate of mor- mon in the artistic tradition of these people. tality during transportation. The effect of this Hunting pressure for some reptiles and trade on populations (if any) is unknown. Fur- amphibians prevails today. The green and black thermore, in an attempt to regulate the trade, iguanas (I. iguana, Ctenosaura similis), and the governments and conservation agencies have basilisk lizards (Basiliscus spp.) are still an im- restricted permits to study and collect speci- portant source of protein in arid environments. mens for scientific purposes (Campbell and Throughout the dry forest, residents hunt C. sim- Frost 1993). ilis for entertainment, for important dietary sup- Without a doubt, habitat destruction is the plements, and to feed dogs and other domestic most important factor affecting the herpetofauna animals. Turtles are also hunted for their meat of dry forests. Archaeological evidence, and the in several localities. Eggs of marine turtles con- fact that in some areas extensive crops were

186 SASA AND BOLAÑOS frankie chap 14 8/20/03 8:00 PM Page 187

established, might suggest that pre-Columbian Honduras and inhabits only dense forest. When societies had a tremendous impact on the envi- examining the effects of land-use changes, the ronment and local herpetofauna. However, a sub- case of the cantil Agkistrodon bilineatus deserves stantial modification has occurred in the past further consideration. This pit viper occurs in 500 years, with the opening of forest for cattle dry forests along the Pacific coast and foothills production. By the beginning of the previous cen- from southern Sonora (Mexico) southward to tury, extensive cattle ranches were firmly estab- Nicaragua and northwestern Costa Rica (Camp- lished in most dry areas of Mesoamerica (chap- bell and Lamar 1989). Despite recent efforts to ter 5), leading to decreased herpetofauna habitat. locate A. bilineatus in many parts of its range, Habitat destruction in Mesoamerican dry only in Sector Santa Rosa (Guanacaste Conserva- environments includes extensive deforestation, tion Area) are specimens easily found (Solórzano the creation of grasslands for cattle breeding, et al. 1999). The cantil seems to be associated change to ponds and rivers for soil irrigation, with mature forest, and given strong deforesta- contamination, and fires to open areas. Each of tion throughout most of its range, we suspect these aspects affects a particular group of or- that over much of its distribution A. bilineatus ganisms in a specific way. is probably extinct. Human-caused fires are common during Amphibians provide the most dramatic ex- the dry season, and their effects on amphibian ample of threatened populations and local dis- and reptile populations are not clear. Seasonal appearances. Throughout tropical Mesoamerica burning has modified the ecosystem through- there are several documented cases of declining out most of tropical Mesoamerica, resulting in a populations or population and species extinc- reduction of critical habitat for some amphib- tion, or both; the most notorious are the golden ians and reptiles. Species that are slow moving toad (Bufo periglenes) and harlequin frog (Atelo- or that cannot burrow deep enough during fires pus varius) of Monteverde, Costa Rica (Pounds are particularly susceptible. Acuña (1990) stud- and Crump 1994). Other species have also dis- ied the mortality of mud turtles (Kinosternon appeared mysteriously in that area (Pounds et al. scorpioides) during a dry season in Laguna de 1997). The extinction of an A. varius population Palo Verde (northwestern Costa Rica). Of a total in the Osa Peninsula (South Pacific) has been of 164 observed casualties, 140 were attributed hypothesized in recent years, providing the first to the effect of a three-day fire, 38 percent of Costa Rican example of a lowland population in them adult females. decline (F. Bolaños unpubl. data). We are not Loss of tropical dry forest by logging is a cru- aware of any reports on amphibian declines in cial problem along the Pacific coast of Central lowland dry forests of Mesoamerica. Therefore, America. Species that are associated with gallery an effort to monitor such populations is crucial, forests and secondary growth are clearly more not only to detect possible declines but, more susceptible to logging and clearance of dry forest important, to produce adequate information to than those inhabiting open areas. Amphibians be used as a reference for comparative studies in such as Bolitoglossa mexicana, Oedipina taylori, the future. Eleutherodactylus stejnegerianus, and E. rhodopis only occur in riparian forest. Other species such PROBLEMS OF CONSERVATION as Gymnopis multiplicata, Ctenosaura palearis, C. quinquecarinata, Norops pentaprion, Sibon spp., Perhaps the greatest concern related to the con- and Agkistrodon bilineatus are strongly associated servation of amphibians and reptiles is the lack with deciduous and semideciduous forest. The of baseline information. Although we have a good iguana C. palearis has a restricted distribution idea of which species are distributed in the dry in few interior dry valleys of Guatemala and forest of Mesoamerica, we are still ignorant about

HERPETOFAUNA BIODIVERSITY AND CONSERVATION 187 frankie chap 14 8/20/03 8:00 PM Page 188

important aspects of behavior, ecology, popula- the protection of dry-forest herpetofauna, focus tion dynamics, population genetics, home ranges, must be brought to bear on two aspects: location and mobility for the majority of species. To help of the reserves and their size. alleviate this information gap, governments and conservation agencies should collaborate with LOCATION OF RESERVES researchers, acting as facilitators for studies in At least five major geographical regions are evi- natural areas. Permits for research and scientific dent from a biogeographical analysis of the her- collection should be accessible to both national petofauna of Mesoamerica’s dry environments and international scientists, a situation that, un- (fig. 14.1): (a) Mexico’s western (Pacific) coast fortunately, is not common at present. In Costa (Jalisco/Michoacán), (b) Tehuantepec Plains, Rica, for example, research fees in national parks (c) Yucatán Peninsula, (d) Central America’s Pa- are very high for local biologists and students, a cific coast (plus the Motagua Valley), and (e) in- situation that has forced many study sites to be terior dry valleys of Honduras. Furthermore, moved to private protected reserves. several snake populations distributed in these Conservation efforts on a species-specific regions are genetically distinct (M. Sasa unpubl. basis will be costly, if not impossible, and a good data). Hence, these dry regions contain much of representation of the overall community might the genetic diversity and evolutionary differenti- not be assured. Alternatively, we can focus on the ation of the herpetofauna populations under pro- protection of areas and ecosystems, thus main- tection. In an effort to create reserves with an taining a multitude of species and their habitats. adequate representation of reptiles and amphib- However, these efforts may be of little help if the ians (and their genetic diversity) occurring in arid home range and mobility of those species that Mesoamerica, it can be argued that these con- we aim to protect are unknown. servation areas should at least be located in each Obviously, it is not possible to preserve the biogeographical region. When the current dry- entire dry forest of Mesoamerica. We can, how- forest reserves in Mesoamerica (table 14.4) are ever, protect a portion of that ecosystem, thus compared with these biogeographical regions, attempting to save some species as ecologically however, the areas are not found to overlap. functional members of the community. To this day, protection of dry forests and associated AREA OF RESERVES ecosystems in Mesoamerica occurs only in 15 The next heuristic step is to investigate if the reserves scattered in the Yucatán and along the current reserves are large enough to sustain vi- Pacific coast of Mexico and Costa Rica (table 14.4). able populations of the dry-forest herpetofauna. In contrast, the efforts to preserve natural areas One could approach this question by estimating in the humid Caribbean coast have resulted in the area necessary to maintain predetermined the creation of a Mesoamerican biological corri- numbers of individuals of the larger predators dor that connects several reserves from the Yu- in the herpetofaunal community. Large preda- catán to Panama (Rodríguez and Montero 1999). tors are likely to require more space than other Envisioning a similar effort to protect dry forests species and therefore may serve as “umbrella” along the Pacific coast of Mesoamerica implies species for other vertebrates. Unfortunately, the creation of new protected areas in Guate- we do not have direct estimates of population mala, El Salvador, Honduras, and Nicaragua. This densities of the largest reptilian predators in indeed requires a serious commitment from tropical dry forests: crocodilians and the snakes the governments and conservationists of those Crotalus durissus, Boa constrictor, and Drymarchon countries. corais. Based on minimum viable population To evaluate whether the reserves listed in (MVP) of 500 individuals, a reasonable size given table 14.4 give some measure of assurance for theoretical and empirical considerations, Jorge

188 SASA AND BOLAÑOS frankie chap 14 8/20/03 8:00 PM Page 189

TABLE 14.4 Federal and State Reserves That Protect Dry-Forest Herpetofauna in Mesoamerica

reserve state country area (ha) vegetation types

Chamela, Chísmala Jalisco Mexico 13,142 Lowland deciduous forest, spiny shrub Sierra de Manantlána Jalisco, Colima Mexico 139,570 Savanna, pine-oak forest El Veladero Guerrero Mexico 3,159 Lowland deciduous forest Laguna de Chacahuaa Oaxaca Mexico 14,187 Mangrove, evergreen forest Benito Juareza Oaxaca Mexico 2,737 Pine-oak forest, lowland deciduous forest La Encrucijadaa Chiapas Mexico 144,868 Mangrove La Sepultura Chiapas Mexico 167,309 Lowland deciduous forest, spiny shrub Río Lagartos Yucatán Mexico 47,840 Lowland deciduous forest, mangrove Dzibilchaltun Yucatán Mexico 533 Lowland deciduous forest Isla Contoy Quintana Roo Mexico 176 Coastal vegetation, mangrove Sian Ka’an Quintana Roo Mexico 528,147 Savanna Tuluma Quintana Roo Mexico 664 Coastal vegetation, mangrove Sierra Gordaa Queretaro Mexico 383,567 Deciduous forest, evergreen forest El Cimatarioa Queretaro Mexico 2,447 Xerophytic vegetation, savanna Guanacaste Guanacaste Costa Rica 110,000 Lowland deciduous forest Tempisque Guanacaste Costa Rica 20,500 Riparian forest, lowland deciduous forest

Source: Unidad Coordinadora de Areas Naturales Protegidas (Mexico) and Sistema Nacional de Areas de Conservación (Costa Rica). aReserves that also include ecosystems other than dry forest.

da Silva and Sites (1995) assessed the potential ably only El Veladero, La Sepultura, Río Lagartos, of Brazilian reserves for the conservation of and Guanacaste Conservation Area have areas of three large snake species. They suggest that dry forest as large as recommended by Jorge da in order to sustain that MVP for large snakes, Silva and Sites (1995). reserves should range between 19,700 and From these data, it is clear that protected ar- 162,700 ha. If we adopt a similar range, only eas should be established in the Yucatán and 53 percent of the reserves currently established on the Paciﬁc coast of northern Central Amer- (table 14.4) have areas big enough to maintain ica. Moreover, some taxa—in particular those large herpetofaunal species. Furthermore, prob- distributed in dry valleys of Guatemala and

HERPETOFAUNA BIODIVERSITY AND CONSERVATION 189 frankie chap 14 8/20/03 8:00 PM Page 190

FIGURE 14.1. Herpetofaunal similarities among 19 arid localities in Mesoamerica. Pairwise similarities estimated using Dice (1945) coefﬁcient. Amalgamation rule is unweighted pair group average (UPGA).

Honduras—are not currently included in any It is also recommended that biologists work protected areas (e.g., Ctenosaura palearis, Helo- more closely with public and private conserva- derma horridum charlesbogartii), and unless some tionists to conserve more habitat for herpeto- efforts are made, they will probably perish as fauna. This will require proposal writing and col- habitat modiﬁcation proceeds. laboration. In the case of public land in Costa Rica, we recommend that the two large dry-forest conservation areas, Guanacaste Conservation Area RECOMMENDATIONS and Tempisque Conservation Area, be expanded Although the preservation of dry-forest herpeto- to include more herpetofauna habitat. This may fauna is not an easy task, we believe that positive be achieved through new land acquisitions or results can be achieved through increased re- possibly through private conservation easements. search, land conservation, and, most important, Parallel to the efforts in area conservation, educational outreach. an intense environmental education campaign Even though species diversity and geograph- should be established in the region. Education ical distributions of the herpetofauna of Meso- should occur not only on a university level but american dry forests are well known, little in- administratively, as well. In order to study the formation is currently available on the ecology, herpetofauna, permits must often be secured, behavior, and natural history of most species. and this process can be cumbersome. We rec- We recommend that more studies be conducted ommend that more knowledgeable and better immediately on representative species of the educated government personnel be assigned to various herpetofauna groups. The urgency is issue permits to study high-risk organisms such related, in part, to the enormous habitat distur- as some reptiles. Also needed are responsible bances that are in progress throughout the Meso- land managers, public and private, who can re- american region. This research is especially im- ceive and use the results to improve on the pro- portant in the case of amphibians, as species in tection of herpetofauna, which will require that this group are being lost with little understand- biologists invest some of their research time in ing of the causes of extinction. a variety of outreach efforts (see chapter 19).

190 SASA AND BOLAÑOS frankie chap 14 8/20/03 8:00 PM Page 191

Educating people about the important roles biologists must invest a greater amount of their of reptiles and amphibians in the ecosystem is time in outreach projects. particularly significant given the poor reputation that some species (particularly snakes) have ACKNOWLEDGMENTS among humans. The challenge for herpetologists is to support teachers and update environmen- We are grateful for the valuable revisions to this tal education programs with current information chapter made by Jonathan Campbell, Paul T. on the biology and conservation of local herpeto- Chippindale, Lisa Patrick, William W. Lamar, fauna. It is necessary to direct more efforts to the and Gerardo Chaves. This is a contribution from education and information of the general popu- the Museo de Zoología, Escuela de Biología, lation, especially children, through the integra- Universidad de Costa Rica. Gustavo Serrano tion of scientific authorities and conservation drew map 14.1. Fieldwork for this chapter was programs. Some interesting efforts have been financed partially through a grant from the conducted in Costa Rica, through an extensive Vicerrectoría de Investigación (Universidad de program at the Instituto Clodomiro Picado Costa Rica) and through National Science Foun- (Universidad de Costa Rica). Since 1985, biolo- dation Grant 003656001 to J.A. Campbell. gists of this institute, led by Rodrigo Aymerich, have been giving speeches to a wide variety of REFERENCES audiences, lecturing on the biology of venomous snakes and prevention of snakebites. Although Acuña, R. A. 1990. El impacto del fuego y la sequía conservation has not been its major goal, this sobre la estructura de la población de Kinoster- non scorpioides (Testudines: Kirnosternidae) en program has helped to demystify snakes as evil Palo Verde, Guanacaste, Costa Rica. Brenesia creatures, and a new generation of teachers and 33:85–97. students have acquired more positive attitudes Baroni-Urbani, C., and M. W. Buser. 1976. Simi- toward snakes. The achievements of this pro- larity of binary data. Systematic Zoology 25: gram have never been quantified, but it is obvi- 251–59. ous that in Costa Rica more people (especially Bogert, C. M., and R. Martín del Campo. 1993. The Gila monster and its allies. 2d ed. Oxford, Ohio: young people) are currently more interested in Society for the Study of Amphibians and Rep- the topic than few decades ago. The reinforce- tiles. 262 pp. ment of this and similar programs and the in- Boutilier, R. G., D. J. Randall, G. Shelton, and D. P. corporation of biodiversity conservation as a for- Towens. 1979. Acid-base relationships in the mal topic in the school curriculum are necessary blood of the toad Bufo marinus. III. The effects of burrowing. Journal of Experimental Biololgy components if we want to create awareness to 82:357–65. preserve a good sample of the rich herpetofauna Brown, C. M., and O. J. Sexton. 1973. Stimulation of the region. of reproductive activity of females Anolis sagrei It is easy for most people to appreciate charis- by moisture. Physiological Zoology 46:168–172. matic animals such as birds, mammals, sea tur- Campbell, J. A. 1998. Amphibians and reptiles of tles, butterflies, or flashy beetles. For one reason northern Guatemala, the Yucatán, and Belize. Nor- man: University of Oklahoma Press. 400 pp. or another, it is not as easy to appreciate snakes, Campbell, J. A., and D. R. Frost. 1993. Anguide salamanders, lizards, and frogs. In summary, the lizards of the genus Abronia: Revisionary notes, only way these often overlooked groups can gain description of four new species, a phylogenetic improved status in Mesoamerica is through ap- analysis, and key. Bulletin of the Museum of Nat- propriate environmental education. More ag- ural History 216:1–121. Campbell, J. A., and W. W. Lamar. 1989. The ven- gressive, larger campaigns must be initiated to omous reptiles of Latin America. Ithaca, N.Y.: Cor- improve the understanding of the ecological nell University Press. 425 pp. services that herpetofauna provide for nature, Campbell, J. A., and J. P. Vannini. 1988. A new sub- as well as humans. In order to reach this goal, species of beaded lizard Heloderma horridum,

HERPETOFAUNA BIODIVERSITY AND CONSERVATION 191 frankie chap 14 8/20/03 8:00 PM Page 192

from the Motagua Valley of Guatemala. Journal vol. 13, Physiological ecology, 53–154. New York: of Herpetology 22:457–68. Academic Press. ———. 1989. Distribution of amphibians and rep- Gutiérrez-González, M. 1993. El aprovechamiento tiles in Guatemala and Belize. Proceedings of the de la fauna en el sitio arqueológico Nacascole, Western Foundation of Vertebrate Zoology 4:1–21. Bahía Culebra, Guanacaste. Tesis de Licen- Ceballos, G. 1995. Vertebrate diversity, ecology, ciatura, Universidad de Costa Rica. and conservation in neotropical dry forests. In Guyer, C. 1990. The herpetofauna of La Selva, Seasonally dry tropical forests, ed. S. H. Bullock, Costa Rica. In Four Neotropical rainforests, ed. H. A. Mooney, and E. Medina, 195–220. New A. H. Gentry, 371–85. New Haven: Yale Univer- York: Cambridge University Press. sity Press. Codeco, F. 1999. Afferent signal oscillations and Hartweg, N., and J. A. Oliver. 1940. A contribution the control of breathing in Bufo marinus. Ph.D. to the herpetology of the Isthmus of Tehuante- diss., University of Texas. 96 pp. pec. IV. An annotated list of the amphibians and Dice, L. R. 1945. Measures of the amount of eco- reptiles collected on the Pacific slope during logical association between species. Ecology 26: the summer of 1936. Miscellaneous Publications 297–302. of the Museum of Zoology, University of Michigan Duellman, W. E. 1956. The frogs of the hylid genus 47:1–29. Phrynohyas Fitzinger, 1843. Miscellaneous Publi- Holdridge, L. R., and G. Budowski. 1959. Mapa cations of the Museum of Zoology, University of ecológico de Panamá. Instituto Interamericano Michigan 96:1–47. de Ciencias Agrícolas (IICA), San José, Costa ———. 1960. A distributional study of the am- Rica. phibians of the Isthmus of Tehuantepec. Uni- Janzen, D. H., and T. W. Schoener. 1968. Differ- versity of Kansas Publications of the Museum of ences in insect abundance and diversity be- Natural History 13:19–72. tween wetter and drier sites during a tropical ———. 1961. The amphibians and reptiles of dry season. Ecology 49:96–110. Michoacán, Mexico. University of Kansas Publi- Johnson, J. D. 1989. A biogeographic analysis of cations of the Museum of Natural History 15:1–148. the herpetofauna of northwestern Central Amer- ———. 1965a. Amphibians and reptiles from the ica. Contributions in Biology and Geology Milwau- Yucatán Peninsula, Mexico. University of Kansas kee Public Museum 76:1–66. Publications of the Museum of Natural History 15: Jorge da Silva, N., and J. W. Sites. 1995. Patterns of 577–614. diversity of Neotropical squamate reptile species ———.1965b. A biogeographic account of the with emphasis on the Brazilian Amazon and herpetofauna of Michoacán, Mexico. University the conservation potential of indigenous re- of Kansas Publications of the Museum of Natural serves. Conservation Biology 9:873–901. History 15:627–709. Lee, A. K., and E. H. Mercer. 1967. Cocoon sur- ———.1966. The Central American herpetofauna: rounding desert-dwelling frogs. Science 157: An ecological perspective. Copeia 1966:700– 87–88. 719. Lee, J. C. 1996. The amphibians and reptiles of the Ferrero, L. 1987. Costa Rica Precolombina. San José: Yucatán Peninsula. Ithaca, N.Y.: Cornell Uni- Editorial Costa Rica. 89 pp. versity Press. 512 pp. Fleming, T. H., and R. S. Hooker. 1975. Anolis McDiarmid, R. W. 1968. Population variation in cupreus: The response of a lizard to tropical the frog genus Phrynohyas in Middle America. seasonality. Ecology 56:1243–61. Contributions in Science Natural History Museum García, A., and G. Ceballos. 1994. Guía de campo of Los Angeles County 134:1–25. de los anfibios de la costa de Jalisco, México. Mex- McDiarmid, R. W., and M. S. Foster. 1987. Cocoon ico City: Fundación Ecológica de Cuixmala. formation in another hylid frog, Smilisca bau- 184 pp. dinii. Journal of Herpetology 21:352–55. Gentry, A. H. 1995. Diversity and floristic compo- Mertens, R. 1952. Die amphibien und reptilien von sition of Neotropical dry forest. In Seasonally dry El Salvador. Abhandlungen der Senckenbergischen tropical forests, ed. S. H. Bullock, H. A. Mooney, Naturforschenden Gesellschaft 487:1–120. and E. Medina, 46–194. New York: Cambridge Minnich, J. E. 1977. Adaptive responses in water University Press. and electrolyte budgets of native and captive Gregory, P. T. 1982. Reptilian hibernation. In Biol- desert tortoises, Gopherus agassizii, to chronic ogy of the Reptilia, ed. C. Gans and F. H. Pough, drought. In Desert Tortoise Council, Proceedings

192 SASA AND BOLAÑOS frankie chap 14 8/20/03 8:00 PM Page 193

of 1977 Symposium, ed. M. Trotter and C. G. Smith, E. N. 1994. Biology of the snake fauna of the Jackson Jr., 102–29. San Diego: Desert Tortoise Caribbean rainforest of Guatemala. Master’s Council. thesis, University of Texas. Murphy, P. G., and A. E. Lugo. 1995. Dry forest of Solórzano, A., and L. Cerdas. 1988. Biología repro- Central America and the Caribbean Islands. In ductiva de la cascabel Centroamericana Crotalus Seasonally dry tropical forests, ed. S. H. Bullock, durissus durissus (Serpentes: Viperidae) in Costa H. A. Mooney, and E. Medina, 9–34. New York: Rica. Revista de Biología Tropical 36:221–26. Cambridge University Press. Solórzano, A., M. Romero, J. M. Gutiérrez, and Pianka, E. R. 1986. Ecology and natural history of M. Sasa. 1999. Natural history and venom desert lizards: Analyses of the ecological niche and composition of the cantil Agkistrodon bilineatus community structure. Princeton: Princeton Uni- howardgloydi (Serpentes: Viperidae): Venom versity Press. 222 pp. composition and diet. Southwestern Naturalist Pounds, J. A., and M. L. Crump. 1994. Amphibian 44(4):478–83. declines and climate disturbance: The case of Stafford, P. J. 1998. Amphibians and reptiles of the the golden toad and the harlequin frog. Conser- Cordillera de Guanacaste, Costa Rica: A ﬁeld vation Biology 8:72–85. list with notes on color patterns and other ob- Pounds, J.A., M.L. Fogden, J.M. Savage, and servations. British Herpetology Society Bulletin G. C. Gorman. 1997. Test of null models for 62:9–19. amphibian declines on a tropical mountain. Stuart, L. C. 1935. A contribution to the knowledge Conservation Biology 11:1307–22. of the herpetology of a portion of the savanna Ramírez-Bautista, A. 1994. Manual y claves ilus- region of central Peten, Guatemala. Miscella- tradas de los anﬁbios y reptiles de la región de neous Publications of the Museum of Zoology, Chamela, Jalisco, México. Mexico DF: Instituto University of Michigan 29:1–59. de Biología, Universidad Autónoma de México. ———. 1954a. A description of a subhumid corri- 127 pp. dor across northern Central America, with com- Rand, A. S. 1957. Notes on amphibians and reptiles ments on its herpetofaunal indicators. Contri- from El Salvador. Fieldiana: Zoology 34:505–34. butions of the Laboratory of Vertebrate Biology, Rodríguez, E., and V. V. Montero. 1999. Corredor University of Michigan 65:1–26. biológico mesoamericano: Sección Costa Rica. San ———. 1954b. Herpetofauna of the southeastern José: Ministerio de Ambiente y Energía. highlands of Guatemala. Contributions of the Ruibal, R., and S. Hillman. 1981. Cocoon struc- Laboratory of Vertebrate Biology, University of Mich- ture and function in the burrowing hylid frog igan 69:1–65. Pternohyla fodiens. Journal of Herpetology 15: Tyler, M. J. 1994. Australian frogs: A natural history. 403–8. 2d ed. Auckland, New Zealand: Reed Books. Ruibal, R., L. Tevis Jr., and V. Roig. 1969. The ter- 192 pp. restrial ecology of the spadefoot toad Scaphio- Van Devender, R. W. 1983. Basiliscus basiliscus. In pus hammondii. Copeia 1969:571–84. Costa Rican natural history, ed. D. H. Janzen, Sasa, M., and A. Solórzano. 1995. The reptiles and 379–80. Chicago: University of Chicago Press. amphibians of Santa Rosa National Park, Costa Vitt, L. J. 1987. Communities. In Snakes: Ecology Rica, with comments about the herpetofauna of and evolutionary biology, ed. R. S. Seigel, J. T. xerophytic areas. Herpetological Natural History Collins, and S. S. Novak, 335–65. New York: 3:113–26. Macmillan. 529 pp. Savage, J. M. 1966. The origins and history of Willey, G. R., G. F. Ekholm, and R. F. Millon. 1964. the Central America herpetofauna. Copeia 1966: The patterns of farming life and civilization. In 719–66. Natural environment and early cultures, ed. R. C. Savage, J. M., and J. R. Villa 1986. Introduction to West and R. Waunchope, 446–98. Austin: Uni- the Herpetofauna of Costa Rica. Society for the versity of Texas Press. Study of Amphibians and Reptiles Contributions Wilson, L. D., and J. R. McCranie. 1998. The bio- in Herpetology 3:1–207. geography of the herpetofauna of the subhu- Schmidt, K. P., and L. C. Stuart. 1941. The herpeto- mid forest of Middle America (Isthmus of logical fauna of the Salama Basin, Baja Vera- Tehuantepec to northwestern Costa Rica). paz, Guatemala. Field Museum of Natural His- Royal Ontario Museum, Life Science Contribu- tory, Zoology Series. 24:233–47. tions 163:1–50.

HERPETOFAUNA BIODIVERSITY AND CONSERVATION 193 frankie chap 15 8/20/03 8:02 PM Page 194

chapter 15

Parque Marino Las Baulas

CONSERVATION LESSONS FROM A NEW NATIONAL PARK AND FROM 45 YEARS OF CONSERVATION OF SEA TURTLES IN COSTA RICA

James R. Spotila and Frank V. Paladino

he edge of the sea marks one boundary make occasional cameo appearances to attract Tof the tropical dry forest in Costa Rica. Just the attention of tourists and potential buyers of as the ocean draws Costa Ricans and foreigners land and houses before disappearing from the to vacation spots along the Pacific coast in the scene. It is not at all clear that this play will have summer, the beaches are a magnet for biolo- a happy ending. gists, conservationists, developers, and politi- The star of this play is the flagship species cians. This is because the beaches are critical for this ecotone, the sea turtle. Unfortunately for habitat for all these people and a focal point for the star and its ecosystem, it is not clear whether one of the greatest dramas in conservation at the in the future we will look back on this play as beginning of the twenty-first century. Biologists high drama, a comedy of errors, or a tragedy. come to study sea turtles and other exotic flora The stage is one of the newest parks in the Costa and fauna, conservationists come to save species Rican system of natural areas, Parque Marino from extinction, developers come with inter- Las Baulas. The park is situated along the Gua- national financing to foster coastal development nacaste coast and protects the largest surviving as an economic boom, and politicians chant the nesting population of leatherback turtles in the mantra of “sustainable development” to justify Pacific Ocean. the uncontrolled spread of tourist and residen- In this chapter we focus on the biology of sea tial facilities along the coast. In one sense, all turtles that nest on tropical dry forest beaches these people interact in a play that will determine of Costa Rica; give the history of conservation the future of this ecotone between the forest efforts for these species in Costa Rica; discuss and the marine ecosystems. Local residents are local, national, and international aspects of this on stage as minor characters, and exotic species conservation activity; present the status of Par-

194 frankie chap 15 8/20/03 8:02 PM Page 195

que Marino Las Baulas; detail the process that other beaches on the Paciﬁc coast of Costa Rica led to its formation; consider ﬂaws and limita- during most months of the year, but especially tions of the human and turtle actors; and pres- from August to February. There are no ridley ent recommendations for the future. turtles on the Caribbean coast of Costa Rica.

BLACK TURTLE BIOLOGY OF SEA TURTLES The black turtle is a subspecies of the green Four species of sea turtles nest on the Pacific turtle and is 100 cm long, compared with 120 cm coast of Costa Rica. Leatherback turtles, Der- for the green. Eastern Pacific Chelonia popu- mochelys coriacea, nest primarily in Guanacaste lations are distinct but are not taxonomically Province. Olive ridley turtles, Lepidochelys oli- distinct at the species level (Karl and Bowen vacea, nest in large flotillas or arribadas at Playa 1999). On the Pacific coast black turtles nest Nancite in the Santa Rosa Sector of Parque Na- primarily from August to February, although cional Guanacaste and at Ostional on the Nicoya some nesting occurs throughout the year. On Peninsula and singly on other beaches. Black the Caribbean coast green turtles nest from June turtles, Chelonia mydas agassizii, nest in small to September. numbers at several beaches from Parque Na- cional Guanacaste to the Osa Peninsula, in- HAWKSBILL TURTLE cluding Playa Naranjo, Playa Cabuyal, and Playa Hawksbill turtles are rare on the Pacific coast of Naranjo in Guanacaste and Playa Carate and Costa Rica. They are 90 cm long and are distin- Playa Río Oro in Osa. Hawksbill turtles, Eretmo- guished by their birdlike beak and beautiful “tor- chelys imbricata, nest on Playa Carate and Playa toise shell”–covered carapace. In general, hawks- Río Oro on the Osa Peninsula (Drake 1996). bill populations are only a small fraction of their presettlement numbers throughout their range LEATHERBACK TURTLE (Bjorndal 1999). They are now seen primarily as The leatherback turtle is the largest sea turtle, solitary nesters, but in areas where they are pro- reaching a length of 2.8 m from head to tail and tected, such as Buck Island Reef National Mon- a mass of 916 kg. The carapace (upper shell) is ument in the U.S. Virgin Islands and Jumby Bay 130 to 190 cm long and elongate with seven Island off Antigua, they nest in groups just like sharply peaked longitudinal ridges. It tapers to green and black turtles (Meylan 1999). They a point in the rear called the pygal process. The probably did so in the past before they were thin, leathery skin is black with white spots, hunted to near extinction for their shells. There and the head is large with an irregular pink spot is little scientific information about their biology above the pineal gland. On the Pacific coast in Costa Rica, but hawksbills are in the Golfo leatherbacks nest primarily from October to Dulce and nest on the beaches of the Osa Penin- February, whereas on the Caribbean coast they sula (Drake 1996). Hawksbills also nest along nest from March to June. This coincides with the Caribbean coast of Costa Rica. the dry season on each coast. Between 70 and 90 percent of leatherbacks on the Pacific coast REGIONAL AND GLOBAL CONTEXT of Costa Rica nest at Parque Marino Las Baulas. Sea turtles are an excellent example of the need OLIVE RIDLEY TURTLE for conservation efforts to focus on local, na- The olive ridley is the smallest sea turtle, with tional, and international levels in order to es- an adult size of 36 kg and length of 76 cm. It tablish effective protection plans and policies. has a hard shell and is olive drab in color. It nests Because these species travel great distances in in large arribadas at Playa Nancite and Playa Os- the ocean, they live in waters controlled by many tional and as solitary individuals at these and nations and in international waters, as well.

SEA TURTLE CONSERVATION 195 frankie chap 15 8/20/03 8:02 PM Page 196

Hundreds of leatherback turtles nest on the from these waters within the next few years Caribbean coast of Costa Rica and Panama. Pro- (Spotila et al. 1996, 2000). The decline is due tection is provided in Tortuguero National Park, primarily to the incidental catch of turtles in where the Caribbean Conservation Corporation longline, gill net, and trawl fisheries. Harvesting (CCC) has an ongoing study. Farther south, John of eggs from nesting beaches and some killing Denham’s Pacuare Reserve protects 6 km of of adults on nesting beaches and at sea by in- beach where in some years 100–200 leather- digenous peoples for food and medicinal oils backs nest. His rangers patrol the beach and face and ointments also all have an impact. poachers who are armed with weapons ranging Olive ridley turtles that nest in Costa Rica from machetes to automatic weapons. The ag- traverse great distances in the eastern Pacific gressive program of patrolling and arrests deters (Plotkin et al. 1995, 1996). In contrast to leather- poachers and keeps nest loss to a minimum. backs (Morreale et al. 1996), olive ridleys do not Denham’s rangers also use hatcheries to protect follow migration corridors but rather swim in nests from more remote portions of the beach. diverse patterns in waters ranging from Mexico South of Limón, Didier Chacón and his volun- to Peru and the Galápagos. These turtles live in teers of Asociación ANAI protect beaches in the waters under the jurisdiction of many coun- Refugio Nacional de Fauna Silvestre Gandoca- tries and in international waters. Any successful Manzanillo near the Panamanian border. In Pan- conservation plan must address the problem of ama leatherbacks are unprotected, and poachers interactions of sea turtles with the fishing in- kill adults to get their eggs in addition to digging dustry of the Americas and Asia, as well. up nests in the Bocas del Torro region. John In addition to the two arribada beaches in Denham and Clara Padilla of the Wildlife Con- Costa Rica, olive ridleys nest at one arribada beach servation Society began a protection program in Pacific Mexico and two arribada beaches in In- there in cooperation with Panamanian author- dia. Other arribada beaches in Mexico were de- ities in 2002. stroyed by the 1980s through extensive killing There are other major nesting beaches for of adult females for the leather trade. There are leatherbacks in Mexico. The Mexican population some minor arribada beaches on the Pacific coast was 70,000 females in 1980 (Pritchard 1982) of Nicaragua and Guatemala and scattered nest- but declined to fewer than 200 in 1999 (Spotila ing along Pacific Mesoamerica. et al. 1996; L. Sarti and D. Dutton pers. comm.). Black turtles nest in Nicaragua, El Salvador, Although biologists monitor many beaches and Guatemala, and Mexico, as well as Costa Rica. provide protection at Mexiquillo in Michoacán, Conservation takes place at many beaches in there is no protection on beaches to the south these countries, but the poaching of eggs takes such as Tierra Colorado in Guerrero and Bahía place on most beaches and is a serious problem Chacahua and Barra de la Cruz in Oaxaca. for survival of this species. In Nicaragua conser- Poachers take essentially all the eggs on these vationists working with members of local com- beaches and kill some adults (E. Possardt pers. munities protect the main beaches of Choco- comm.). In 1998 and 1999 Georgita Ruiz’s fed- cente and La Flor. Soldiers provide protection for eral rangers in Oaxaca confiscated trailer truck the beaches, and hatcheries protect some nests. loads of turtle eggs from middlemen who pur- Development threatens these beaches. In Gua- chased eggs from poachers. These included tens temala the Asociación de Rescate y Conserva- of thousands of eggs from olive ridley, black, ción de Vida Silvestre (ARCAS) protects several and leatherback turtles. A few leatherbacks also small beaches and places eggs in hatcheries. Lo- nest in Nicaragua. cal people are paid in flour, corn, and other foods The leatherback turtle is declining in num- in return for ten eggs from every nest of eggs bers at a catastrophic rate in the Pacific Ocean, that they take. That means that only 10 percent and if current trends continue, it will disappear of the eggs at most are being saved. Although

196 SPOTILA AND PALADINO frankie chap 15 8/20/03 8:02 PM Page 197

FIGURE 15.1. Numbers of leatherback turtles nesting on Playa Grande from the 1988/89 nesting season to the 1998/99 nesting season. Data from 1988 to 1992/93 are based on counts of the number of nests and assume a clutch frequency of 7. Numbers from 1993/94 to 1998/99 are based on Passive Integrated Transponder (PIT) tag identiﬁcation of individual nesting female turtles. The population has undergone an exponential decline during this period.

this is a noble effort, it is doomed to failure. adult female leatherbacks, compared with about Studies on freshwater turtles (Congdon et al. 115,000 in 1980 (Pritchard 1982). The greatest 1994) and modeling of life history characteris- declines occurred in Malaysia and the Pacific tics of sea turtles (Spotila et al. 1996) demon- coast of Mexico. Both of these colonies experi- strate that this level of predation on nests cannot enced an exponential decline. We are now see- be sustained by a turtle population. Owing to ing an exponential decline at the beaches of Las the high rate of mortality of the few turtle hatch- Baulas National Park, as well (fig. 15.1). The east- lings released into the ocean, insufficient ani- ern Pacific population is now about 1,750, and mals are being produced to ensure a return of the world population is about 27,600 (table 15.1). adults when this cohort would be maturing and Stable leatherback populations cannot with- returning to nest. stand an increase in adult mortality above natural In Mexico considerable effort is being made background levels without decreasing. However, to monitor black turtle beaches in Michoacán. protection of eggs and hatchlings during the In addition, Mexican marines have provided first day of life can have a significant effect on protection for biologists, turtles, and their nests. overall stability of leatherback populations with However, poaching is still rampant, and despite a moderate increase in adult mortality (5%). efforts with local communities, these populations Leatherback populations in the Pacific and In- are still being heavily exploited. dian Oceans cannot survive the current levels of adult mortality from fisheries in these oceans. Atlantic populations are also being exploited at a EXTINCTION OF LEATHERBACK TURTLES? rate that cannot be sustained. Thus, leatherbacks In 1996 (Spotila et al. 1996) we estimated the are on the verge of extinction (Spotila et al. 1996, number of leatherbacks nesting on 28 beaches 2000). around the world from the literature and from If our models are right and protection of communications with investigators studying nests and hatchlings can offset moderate levels those beaches. In 1995 there were about 34,500 of adult mortality, then perhaps an all-out

SEA TURTLE CONSERVATION 197 frankie chap 15 8/20/03 8:02 PM Page 198

TABLE 15.1 Spoonbill, White Ibis, Boatbill Heron, Wood Regional Population Estimates for Stork, Great Blue Heron, egrets, howler mon- Nesting Leatherback Turtles, Dermochelys coriacea keys, kinkajous, jaguars, and raccoons. estimated number of The town of Tamarindo is built behind what region nesting females once was a leatherback nesting beach. Only rarely does a leatherback come ashore to nest Western Atlantic 15,000 in Tamarindo now. This area is not in the park. Lights of the town are a major problem for both Eastern Atlantic 4,700 adult and hatchling leatherbacks because the Caribbean 4,000 animals are attracted to the lights. Adults get Eastern Pacific 1,750 entangled in boat lines in the anchorage off the town, and hatchlings wash ashore after swim- Western Pacific 1,700 ming to the light sources. There is no law pro- Indian Ocean 450 hibiting lights from shining onto a turtle beach. Total 27,600 Thus, unchecked development undermines protection of the park because of the lack of a light- Source: Numbers are based on data reported in Spotila et al. ing ordinance. (1996) and recent information from nesting beaches. Just offshore from Tamarindo the Isla Capi- tan provides habitat for many marine birds. conservation effort in Costa Rica may save the Tamarindo Bay is filled with fish and shellfish, leatherback in the Pacific Ocean. Certainly nest- but there has not been a scientific survey of the ing beaches must be preserved and nests pro- species living there. Although this is a protected tected or there will be no “next generation.” zone of the park and turtles are protected by law However, this effort will not be sufficient in the within Costa Rican waters, commercial trawlers face of large-scale mortality in the fishery. ply the bay at will, since the park does not have a large patrol boat. The Estero San Francisco is another large es- PARQUE MARINO LAS BAULAS tuary with mangrove forests and abundant bird Parque Marino Las Baulas (map 15.1) on the life; it isolates Playa Langosta from Tamarindo. Guanacaste coast includes the Bahía de Tama- The beach is set aside as a biological reserve rindo, adjacent beaches, and mangrove estuar- and is not open to tourism or visitation of nest- ies. The park has three beaches that are used by ing leatherbacks. Behind the beach and estuary leatherbacks as nesting sites, Playa Ventanas, is Finca Pinilla, which is being developed with Playa Grande, and Playa Langosta. homes, a golf course, and hotels. A new hotel on Playa Grande is a crescent-shaped white sand the rocky shore north of Playa Langosta and the beach composed of finely broken shell rubble estuary poses a threat to this nesting beach from and sand and is regarded as one of the world’s its lights, noise, and activity. Workers filled in finest surfing areas. The other beaches are of part of the estuary near the hotel site. Despite similar composition but smaller in length. Eighty the filing of a legal action (denuncio) against the to 90 percent of leatherbacks nest on Playa hotel by park officials, the estuary has not been Grande. There is increasing development along restored as of 2003. The hotel now blocks access all three beaches just outside the 50-m public to Playa Langosta and uses the beach and estu- zone. The Estero Tamarindo is the largest estuary as a private location for guests despite their ary in dry Pacific Central America and hosts reserve status. Again, another denuncio has had many species of plants and animals. Flagship no effect on this activity. species include the red, white, black, buttonwood, In the 1988/89 season, 1,367 leatherbacks and tea mangroves, American crocodile, Roseate nested on Playa Grande. In contrast, during the

198 SPOTILA AND PALADINO frankie chap 15 8/20/03 8:02 PM Page 199

MAP 15.1. Parque Nacional Las Baulas. Redrawn from Steyermark et al. (1996) with permission.

1998/99 season only 117 leatherbacks nested Leatherbacks are long-lived like other large there (fig. 15.1). Average annual mortality at sea marine organisms. Populations of long-lived for these turtles is 28 percent. In a given year organisms cannot withstand heavy human ex- 75 percent of leatherbacks are first-time nesters, ploitation (Congdon et al. 1994; Crouse 1999; and 25 percent are repeat migrants. Most sea Heppell et al. 1999; Musick 1999). The short turtle populations have a smaller percentage lives of leatherbacks that we observe at Las of repeat migrants than do freshwater turtles Baulas today are not normal and reflect intense (Congdon et al. 1993, 1994) owing to some level fishing pressure imposed on them. Thousands of exploitation by humans. However, these per- of leatherbacks died each year in oceanic long- centages are especially low for the Las Baulas line and gill net fisheries in the 1980s and 1990s population. (Frazier and Brito Montero 1990; Nishimura and

SEA TURTLE CONSERVATION 199 frankie chap 15 8/20/03 8:02 PM Page 200

Nakahigashi 1990; Wetherall et al. 1993; Eckert portion of the beach without any environmental and Sarti 1997). assessment or permits. Another person built a hotel at the southern end of the beach by cutting mangroves and filling in part of the Tamarindo THE MAKING OF A PARK: THE LONG estuary. Plans also existed for a major develop- JOURNEY TO AN UNCERTAIN FUTURE ment of 250 houses behind Playa Grande and Parque Marino Las Baulas exists primarily be- more than 50 behind Playa Ventanas. A major cause of the efforts of Maria Teresa Koberg, Mario luxury resort for 5,000 with a casino, night- Boza, Clara Padilla, and Peter Pritchard. Much club, hotel, condominiums, and a yacht club has been written in the popular press, maga- was planned for Playa Grande. zines, and books about this park, and much is Maria Teresa Koberg led the early efforts to incorrect. We worked at Las Baulas since 1990 obtain protection for the leatherbacks and was a and participated in or observed most of the pioneer in converting poachers to guides and activity surrounding the park. We were involved rangers. Her most famous convert was Esper- in research and conservation efforts at the park anza Rodríguez, the matriarch of a family that and worked with officials of the Ministry of was long the major source of poaching on Playa Environment and Energy (MINAE) and its Na- Grande. From 1988 to 1996 Rodríguez made tional Park Service (SPN) in efforts to consoli- daily nest counts on Playa Grande while riding date the park. Between 1990 and 1999 we raised the beach at sunrise on her horse. These data more than $788,000 in grants and donations to were critical to our understanding of the popu- support these efforts from ten organizations (in- lation trends of leatherbacks there. Koberg, not cluding Earthwatch, the U.S. National Science Wilson and Pastor (Honey 1999), mounted a pro- Foundation, the National Geographic Society, gram of education and protection. She brought Drexel University, World Wildlife Fund, Guin- in Boy Scouts from Costa Rica and Minnesota. ness, Ltd., and the U.S. National Marine Fish- They patrolled the beach and protected nests eries Service) and many individuals. Here we and turtles by persuasion. She obtained national relate the history of the park as we participated media attention and promoted the idea of a park. in it. In 1990 a report by Pritchard and members In the late 1980s the Estero de Tamarindo and of the local community (Pritchard et al. 1990) Playa Grande were designated as the Tamarindo recommended the formation of a park. In 1991 Wildlife Refuge, a Refugio Nacional de Vida Sil- President Rafael Angel Calderón issued a decree vestre, and the estuary was protected as a wetland that established the park as Las Baulas de Gua- under the RAMSAR Convention on wetlands. nacaste (Executive Decree No. 20518, 5 June The beach was protected up to 50 m above the 1991). Even though the limit to the park was only high-tide mark. However, the refuge designa- 125 m inland from the high tide, it did protect tion did not improve on the nationwide legal pro- the ocean offshore to 19.2 km. tection given to mangroves, ocean beaches, and Koberg was the first director of the park and turtles. established a protection program involving Under its refuge status Playa Grande was the leaders and members of the Playa Grande and scene of uncontrolled poaching, tourism, and Tamarindo communities. She brought in a young beachfront development. The beach was bright biologist, Randall Arauz, who began a program with camera flashes during the height of the of education in nearby villages. In 1992 he be- nesting season, and many people rode on the came the second director of the park and car- backs of leatherbacks as they crawled up and ried out a vigorous program of research and con- down the beach. Poachers stole almost every egg servation on Playa Grande (Arauz and Naranjo laid. One person built a hotel along the northern 1994). Gradually residents began to accept the

200 SPOTILA AND PALADINO frankie chap 15 8/20/03 8:02 PM Page 201

park as they became turtle guides and realized From 1994/95 to 1997/98 Sergio Obando that their concerns were being heard (Naranjo was director and did an excellent job in begin- and Arauz 1994). ning the consolidation of the park. Outreach in In 1991–92 we raised funds for rural guards local villages increased, and we established a and obtained grant support for Anny Chaves, program of conservation education in the Playa who, with her students, conducted a study of the Grande community and Matapalo school. In nesting ecology on nearby Playa Langosta and a 1995 the National Assembly finally passed a law program of environmental education in local making the park a permanent entity. Unfortu- villages (Chaves et al. 1996). On Playa Grande a nately, an apparent “clerical error” established 1,300-m restricted zone that tourists could not the park as 125 m under the sea instead of 125 m enter provided a haven for nesting leatherbacks inland from high tide. As of 2002 this error has (Herzog 1992). Significant progress was made not been corrected. with local residents through a cooperative pro- During the Obando period more funds came gram with the Fundación la Gran Chorotega, a to the park. With the help of Earthwatch Europe local nongovernmental organization that pro- we obtained a grant from the Guinness Corpo- moted preservation of local cultural and biolog- ration’s “Water of Life” program to provide a ical resources of Guanacaste. building fund for the park. This grant allowed During the 1992/93 nesting season we moved the construction of a large dormitory building our research to Playa Grande to assist in con- for park volunteers. We also funded and men- servation efforts, population data collection, tored students from the Universidad Nacional and other research in the main portion of the in Heredia, who carried out tagging and nesting park. Earthwatch volunteers and students and ecology studies on Playa Langosta. They con- faculty from Drexel and Indiana-Purdue Univer- trolled poaching and when necessary brought sity at Fort Wayne tagged turtles and controlled park rangers to that beach to maintain order. ecotourists. Paladino taught guides in Spanish The park closed Playa Langosta to tourists so about turtles and control of ecotourists. In ad- that it was a refuge for leatherbacks. dition, we helped park rangers construct the In the 1998/99 season one of our Univer- first house for the park that served as the head- sidad Nacional students became the director of quarters and a home for the director and his the park after graduation. Rotney Piedra brought family. We also conducted botanical surveys to the position not only enthusiasm but also and produced guidebooks for the estuary and training in sea turtle biology. By 1999, residents dry forest. generally accepted the park. Tourists could only In the 1993/94 season we began to tag per- go on the beach with a guide, group size was manently all leatherbacks on the beach and officially limited to ten, and park rangers were continued studying their nesting ecology and more diligent about their duties. English- physiological ecology (Steyermark et al. 1996). language and conservation training at Drexel A cooperative agreement between MINAE and improved Piedra’s ability to manage the park. Drexel University provided English-language He increased controls on the beach and worked training to several ministry employees involved with local landowners to improve protection of in developing the park. The park, under director the turtles. By 2001 Piedra’s rangers were effec- José Quiros, continued to offer training courses tive, and guides cooperated with researchers. for local guides, and we continued our meetings As of 2002 there were still many problems. and education efforts, as well. We also met with The park law needed to be fixed to establish many of delegates of the National Assembly to boundaries properly along the land behind the educate them about turtles and the need for a beaches. Agreements were needed with land- national law to formalize the park. owners behind Playa Grande to support the park.

SEA TURTLE CONSERVATION 201 frankie chap 15 8/20/03 8:02 PM Page 202

A lighting law was needed to protect adults and thus protecting the 48-km nesting beach. In 1973 hatchlings from lights behind the beaches and Costa Rica signed the Convention on Inter- from Tamarindo. The urbanization plan for the national Trade in Endangered Species (CITES), area needed to be completed, and park consoli- making international trade in sea turtles and dation was still ongoing. Development was the other endangered species illegal. In 1975 the dominant factor threatening the park. Comple- government expanded Tortuguero National Park tion of the hotel in Tamarindo overlooking Playa to include the nearby hills and lowlands. Langosta created a new threat to that beach. Throughout this time Carr continued his Conservation at Las Baulas is thus still a work in green turtle studies at Tortuguero and estab- progress. To assist in this effort we formed a lished an ethic in Tortuguero that respected and nongovernmental organization, the Leatherback protected turtles. By employing villagers at the Trust, and registered it in Costa Rica as Fideo- Casa Verde field station and doing turtle nest comiso Baulas. Funds from the Leatherback counts in the park, Carr had a significant impact Trust support a biologist, two rangers, develop- on the local economy, and this formed an eco- ment of a management plan, and other ini- nomic basis for the development of turtle con- tiatives toward park consolidation. In 2003 we servation in the village. He linked economic raised another $325,000 for this effort. progress of the local community with conservation long before anyone made up terms such as “grassroots community-based conservation, 45 YEARS OF SEA TURTLE parks for people, sustainable development and CONSERVATION IN COSTA RICA use, and conservation for development” (Bran- Conservation biologist Archie Carr began sea don et al. 1998: 1). This later laid the ground- turtle conservation in Costa Rica in 1954 when work for ecotourism there. Carr’s students car- he visited Tortuguero with Costa Rican business- ried out numerous pioneering studies on the man Guillermo “Billy” Cruz. In 1955 he started hearing and orientation of hatchlings. Scientific the green turtle tagging project that continues to research was always a key part of the conserva- this day (Bjorndal et al. 1999). In 1959 U.S. busi- tion effort at Tortuguero (Spotila 1988). There nessman Joshua Powers formed the Brotherhood was little interest in research and conservation of the Turtle after reading The Windward Road at Tortuguero among Costa Rican universities (Carr 1955). This later became the Caribbean during this period. The Escuela de Biología was Conservation Corporation (CCC), which ensured just beginning at the University of Costa Rica. In continued funding for the green turtle project at addition, Limón Province was far removed from Tortuguero. The original goal of the CCC was “to the Central Valley in travel time and mind-set. save the green turtle from destruction, to give it In 1978 a major scientific expedition sailed to a chance to renew its numbers, and to redistrib- Tortuguero on the Alpha Helix, and several key ute it to all beaches where it was once common” physiological studies took place. It was here that (Godfrey 1999: 1). The approach was one of sci- Standora et al. (1982) accomplished the first suc- entific research, conservation, and restoration. cessful radio and sonic telemetry of sea turtles. Along the way Archie Carr worked with lead- In 1980 Morreale et al. (1982) demonstrated the ing citizens of Costa Rica to convince them to importance of temperature-dependent sex de- protect turtles. In 1963 President Francisco Or- termination to the conservation of sea turtles. lich signed the first executive decree regulating Throughout this period there was a continued hunting of sea turtles and collection of their harvest of 500 adult green turtles a year for eggs. Also in 1963 the first guards came to Tor- slaughter in Limón. In 1982 local pressure led tuguero to protect turtles. In 1970 Costa Rica to an increase in the regulated harvest to 1,800 prohibited all turtle hunting and egg collection green turtles per year. By the mid-1980s an and established Tortuguero as a national park, ecotourism boom started in Tortuguero, and the

202 SPOTILA AND PALADINO frankie chap 15 8/20/03 8:02 PM Page 203

nighttime population of the village jumped from tical encounter with a green turtle in the dark about 100 to 500 with the addition of several ho- and silence of the night. They did so under the tels and cabinas. The turtle ethic began to break watchful eye of a local conservationist/guide down as more people came to Tortuguero from trained and certified by the CCC. The guides other areas and money flowed into the commu- were well informed, friendly, and staunch de- nity from illegal arms traffic to Nicaragua during fenders of turtles. Protection of the beach was the Contra war. The paving of the runway, de- increasing. velopment of the area inland from Tortuguero, Meanwhile, on the Pacific coast, olive ridleys, and finally construction of an illegal road to Tor- along with leatherbacks at Las Baulas, dominate tuguero in 1996 ended its isolation. the nesting beaches of Pacific Costa Rica. Hughes Throughout this period the CCC increased and Richard (1974) were the first to document its conservation efforts. It erected an informa- the large number of sea turtles that nested on tion kiosk in 1985, initiated a comprehensive many beaches there and the mass nesting of conservation and development plan in 1989, and olive ridleys at Playas Nancite and Ostional. Steve obtained approval of a maritime terrestrial zone Cornelius, a Peace Corps volunteer, documented for Tortuguero in 1993. Despite the untimely the nesting of turtles on Playa Naranjo. Through death of Archie Carr in 1987, his family, students, the 1970s and 1980s he and Douglas Robinson and the CCC staff, energized by Cindy Taft, re- of the University of Costa Rica carried out a doubled their efforts to preserve green turtles and series of studies on sea turtles along the Pacific expand conservation efforts in Nicaragua and coast (Cornelius 1976, 1986). Robinson and his into the leatherback nesting season. We began university students began serious study of sea physiological and nesting ecology studies on turtles on most of these beaches. leatherbacks at Tortuguero in 1989 (Paladino et Robinson came to the University of Costa al. 1990) and included education and conserva- Rica in 1966 from Texas A&M University for a tion as a central part of our program. brief stay and spent the rest of his life there. He Talks in the schools, visits with community established the Museum of Zoology at the Uni- leaders, employment of local residents, and en- versity and the Programa de Tortugas Marinas, couragement of the park guards brought visibil- which served as a focal point for education of ity to the leatherbacks and gave them a status Costa Rican and foreign students in sea turtle bi- and level of protection approaching those of ology and conservation. He directed the thesis green turtles. Our rule of thumb became “no sci- studies of several students and established the entists, no protection,” and we expanded our first computer database for flipper tag returns project to the entire leatherback nesting season. from turtles along the Pacific coast of Central That extended protection on the beach to in- America. He also played a key advisory role in clude both the leatherback and green turtle sea- the development of conservation policies and sons, March through September. This developed laws in Costa Rica. The program continued at into an annual tagging program (Campbell et al. the university after his death in 1991 but, lack- 1996) and nicely complemented the CCC con- ing a director with his mature leadership and servation efforts in Tortuguero and Nicaragua. scientific training, was floundering by the mid- In 1994 the CCC completed its new field station 1990s, and the university ended it. Fortunately, and visitor center in Tortuguero, and in 1998 a vigorous program directed by Claudette Mo Costa Rica and Panama signed an agreement for remained at nearby Universidad Nacional. In the collaborative management of sea turtles in addition, a cadre of former students remained the Caribbean. from the Robinson group at the University of More green turtles arrived to lay eggs in 1998 Costa Rica, and Mario Alvarado, Randall Arauz, than ever before. The nesting beach was pro- Jorge Ballestero, Juan Carlos Castro, Anny tected. Thousands of people had their first mys- Chaves, and Isabel Naranjo continued to be

SEA TURTLE CONSERVATION 203 frankie chap 15 8/20/03 8:02 PM Page 204

active in sea turtle biology and conservation. trade, which still occurs on both coasts of Costa Unfortunately only Roldan Valverde continued Rica and invades national parks whenever rangers on to receive a Ph.D. or scientists let their guard down. Away from the Perhaps the most ambitious project started beaches enforcement is severely lacking or non- by Robinson was the experiment with a con- existent. In addition, since the loss of Robinson, trolled harvest of olive ridley eggs at Ostional. In efforts of biologists have lacked the guidance 1980 he saw that the little village situated on one needed to maintain the scientific integrity of the of the most important sea turtle nesting beaches project. The arribadas have decreased in num- in the world would play a critical role in the fu- bers of turtles during the late 1990s. It is not ture of sea turtle conservation. Even though har- clear how much of this decline is due to the long- vesting of sea turtle eggs had been illegal in term egg harvest and how much is due to the Costa Rica since 1966, it still took place at night mortality of olive ridleys in the net and longline at Ostional and elsewhere. Robinson established fisheries in the Pacific. Development in Nosara, the Ostional turtle station of the University of 12 km south, has forced land values out of the Costa Rica to study the arribada phenomenon reach of local people, just as at Las Baulas. De- and began to involve villagers in the protection velopers are buying land surrounding Ostional of the arribada. The concept evolved that vil- and planning to build tourist facilities. It would lagers could take the eggs from the first 24 hours be premature to term this experiment a success of the arribada, since they would be destroyed and a major mistake to take it as a model for sus- anyway, and protect them from the later portion tainable development of other communities of the event. This led to the current controlled near sea turtle nesting beaches. This is an un- harvest. From 1977, when Robinson first pro- finished experiment, one in great need of more posed the idea, until 1987, when the Costa Rican vigorous management by MINAE, in concert Congress reformed the Wildlife Conservation with more vigorous enforcement of laws for sea Law, which prohibited egg harvesting, to allow a turtle protection throughout Costa Rica. controlled harvest at Ostional, a vigorous debate within Costa Rica and worldwide among sea LESSONS AND RECOMMENDATIONS turtle biologists considered all aspects of the FOR THE FUTURE plan. The final arrangement saw the Ostional community form an economic development Many lessons can be learned from the story of association, Asociación de Desarrollo Integral sea turtles in Costa Rica. In 1964 Archie Carr de Ostional (ADIO), to manage the harvest and wrote the following about conservation in Africa: the University of Costa Rica entrusted with legal “But the saving of wild beings from obliteration responsibility for carrying out scientific studies cannot be expected to pay for itself in more than needed to sustain the population. a sprinkling of special cases. For most of the The Ostional experiment is a qualified suc- wild things on earth, the future must depend cess and has been the subject of numerous ar- upon the conscience of mankind. . . . The wel- ticles worldwide (e.g., Baker 1994). The village fare of the wildlife will have to be reckoned has a new school, a new clinic, a new Guardia against the rights of multiplying African man” Rural office, and a new sense of civic pride in (172–73). In 1993 Bonner echoed these senti- what the members of the community have ac- ments when he stated that if Africa’s wildlife is complished. Local wardens patrol the beach, to be saved it will require radical policies and villagers help to count the turtles, and the egg changes in attitudes. Certainly both of these sen- harvest is conducted in a regular and fairly well timents apply to the saving of sea turtles in Costa managed way (Ballestero et al. 1996). Rica. The central lesson of the past 45 years is Many problems remain, however. The Os- that there has to be a radical change in the ap- tional project has not diminished the illegal egg proach of Costa Ricans and their government if

204 SPOTILA AND PALADINO frankie chap 15 8/20/03 8:02 PM Page 205

they are to succeed in preserving sea turtles on and donated modest amounts of money (from their beaches and in the oceans. $5 to $10,000) to support sea turtle and other conservation projects, wealthy Costa Ricans, in SUCCESS IN SOME ASPECTS general, have not provided substantial amounts A dedicated cadre of conservationists have suc- to support the park system. In addition, there is ceeded in establishing and maintaining a won- little support in the government for the park derful park system in Costa Rica. Mario Boza, system. This is most obvious in the lack of Alvaro Ugalde, Pedro León, Sigfredo Marín, and commitment of funds to the parks. Despite eco- their colleagues have worked vigorously and ef- tourism being the number one source of foreign fectively for the parks. At the same time they revenue, inadequate funding hinders develop- have been helped by many North American bi- ment of park infrastructure, training of park ologists who have provided a scientific basis for personnel, and protection. Even more damag- management of the parks. From the beginning ing is the subservience of the park system and foreigners have played an essential role in the environmental protection to development and national park system (Wallace 1992). In general, tourism interests. When there is a choice be- parks have had the support of most of the people tween demands of protection of turtles on the of Costa Rica. Sea turtles are viewed as charis- beach at Las Baulas and access to turtles by matic creatures, and certainly people of the Cen- guides and tourists, the latter wins because of the tral Valley believe that Costa Rica is defending power of the tourism agency, Instituto Costarri- these animals. cense de Turismo (ICT), to overrule MINAE. There is a tax on tourists entering and leaving THE FUNDAMENTAL PROBLEM the country, but this money goes to support ICT, The consensus is that it is on the beaches and not the parks. Although Costa Rica is seen as a among local people where problems arise. This leader in ecotourism and parks (Terborgh 1999), is a convenient alibi that masks a more funda- it is essentially mining its parks for tourist dol- mental problem in the Costa Rican approach to lars and putting few resources back into parks to sea turtle protection and wildlife conservation sustain them. Finally, laws to prevent poaching, in general: the “power elite”—the ruling class protect beaches, regulate fishing, and defend that controls the political and economic power wildlife are weakly enforced (see chapters 22 in a country—of Costa Rica sees economic de- and 23). velopment and profit as more important than There is a lack of leadership at the highest conservation of natural resources, development levels in MINAE and the government and an of a modern park system, and protection of entrenched, inefficient bureaucracy in MINAE wildlife, including sea turtles. This is a familiar and the Park Service. The low priority the gov- problem in all countries. It is always a matter of ernment gives to protection leads to institu- balancing budget priorities in government and tional weakness. The political leadership fails to of raising private funds for conservation organ- demand successful conservation and fails to pro- izations. However, in Costa Rica conservation is vide adequate training and motivation to MINAE losing, and whereas in the United States and Eu- and Park Service staffs. Most personnel of the rope the “civil society” is committed to conser- park system and MINAE lack formal training in vation, this is not true in Costa Rica. There is a how to develop a park, build infrastructure, in- fundamental lack of philanthropy on the part teract with tourists, and defend the parks. They of wealthy Costa Ricans. They expect foreigners spend their time ensuring their own welfare in and their conservation organizations to con- the bureaucracy rather than the welfare of the tinue to provide monetary support for the park natural resources they are supposed to protect. system. Although many young Costa Ricans and Most rangers are poorly educated, poorly paid, middle-class adults have volunteered their time and little motivated to do their jobs. In fact, they

SEA TURTLE CONSERVATION 205 frankie chap 15 8/20/03 8:02 PM Page 206

do not know exactly what a park ranger is sup- use some of those resources to fund their con- posed to do. As in other countries (Terborgh servation infrastructure adequately. Personnel 1999), most guards are not empowered, and so need to be adequately trained, led, and paid in they occupy themselves with controlling tourists order for them to do their jobs effectively. Pro- and scientists while closing their eyes to the fessionals with university and advanced degrees more serious challenge of evicting squatters and need to be hired. Perhaps an exchange program controlling developers. Although it is true that with park systems in nations such as the United Costa Rica has a much more organized park sys- States and Canada, not based on a welfare pro- tem than most other tropical countries (Terborgh gram but on a work program, would be effective. 1999), it is also true that its parks sustain a Rangers from those nations could be given re- much greater number of ecotourists and a much lease time to come to Costa Rica to train rangers greater rate of development along their borders on-site and to learn about the biodiversity in Costa than parks in those countries. Rica firsthand, and rangers from Costa Rica could Costa Rica is caught in the same trap as many take their place in the home country and learn other nations. With an expanding population, a how an effective park system functions by work- declining resource base, and a large foreign debt, ing in it. there seems to be little hope for securing the If more progress is not made, it may be nec- resources needed to ensure protection of the essary to internationalize the parks in order to established national parks, let alone new ones save them (Boza 1993). Sea turtles are inter- such as Las Baulas. Without sustainable devel- national resources. Leatherbacks at Las Baulas are opment the future will be grim, and our play an international treasure, and the beaches there about sea turtles will indeed end as a tragedy. must be vigorously protected or leatherbacks However, no nation has captured or is close to will become extinct in the Pacific Ocean. If the reaching the golden egg of sustainable develop- people of Costa Rica cannot protect the nesting ment, and there is no guarantee that sustainable beaches, pressure will increase for the inter- development will lead to harmony between hu- national community to step in and carry out that mans and nature (Frazier 1997). function. Certainly the precedent has been established that the international community has NEED FOR RADICAL CHANGE the right to intervene in a country to protect hu- Costa Ricans will have to outgrow their laissez- man rights, as in Bosnia and Kosovo. It will not faire attitude toward environmental protection be too long in the future before the international and establish strong, clear laws to control devel- community establishes the precedent that it opment and protect the parks and natural re- has the right to intervene in a country to protect sources contained in them on land and sea. They nature and biodiversity vital to the future well- also will have to reform the legal system to pro- being of the global community. This is one basis vide vigorous enforcement of these laws. We of the evolving theory of environmental security. agree with Terborgh (1999) that there is no sub- This theory arose from the concept of preventive stitute for enforcement. Without it all is lost. defense (Carter and Perry 1999), which seeks to Next the park system will have to be revital- prevent wars before they occur. Environmental ized. It can no longer operate as a biodiversity security goes a step further and states that envi- welfare system dependent on donations from ronmental problems such as overpopulation, foreign governments and conservation organi- lack of resources, and environmental degradation zations. Plenty of resources are available in the can be the causes of conflict and will become the form of ecotourism dollars and biodiversity predominant causes of conflict in the decades to royalties. If Costa Ricans want to preserve their come (Myers 1993). biodiversity, their parks, and, most important Because Costa Rica has so much biodiversity for this story, their sea turtles, they will have to and the most critical sea turtle nesting beaches

206 SPOTILA AND PALADINO frankie chap 15 8/20/03 8:02 PM Page 207

in the Pacific, it has the obligation to protect Bonner, R. 1993. At the hand of man, peril and hope them by making changes in its conservation for Africa’s wildlife. New York: Knopf. 322 pp. strategy so that its parks accomplish their stated Boza, M. A. 1993. Conservation in action: Past, present, and future of the National Park System purpose. If it does not, it can expect that other in Costa Rica. Conservation Biology 7:239–47. nations will become increasingly involved. Brandon, K., K. H. Redford, and S. E. Sanderson. This play does not have to end as a tragedy. 1998. Parks in peril: People, politics, and protected Costa Rica has the basis for success in sea turtle areas. Washington, D.C.: Island Press. conservation and in reforming its park system Campbell, C. L., C. J. Lagueux, and J. A. Mortimer. 1996. Leatherback turtle, Dermochelys coriacea, and environmental ministry. It has a democrat- nesting at Tortuguero, Costa Rica, in 1995. ically elected government that enjoys the sup- Chelonian Conservation and Biology 2:169–72. port of the people. Most of its population appre- Carr, A. 1955. The windward road: Adventures of a ciates the intrinsic value of nature, and most of naturalist on remote Caribbean shores. New York: its people are educated. There is already a large Knopf. 266 pp. area of land set aside in parks and a cadre of ———. 1964. The land and wildlife of Africa. New York: Time. 200 pp. senior conservationists who can lead a revital- Carter, A. B., and W. J. Perry. 1999. Preventive ized park system, and the youth of the nation are defense: A new security strategy for America. committed to the protection of plants and ani- Washington, D.C.: Brookings Institution Press. mals. There is no nation in the tropics better 243 pp. positioned to take the next step into real nature Chaves, A., G. Serrano, G. Marín, E. Arguedas, A. Jimenez, and J. R. Spotila. 1996. Biology conservation. All that is needed is the political and conservation of leatherback turtles, Der- will on the part of the power elite to change their mochelys coriacea, at Playa Langosta, Costa Rica. priorities and make conservation an ethical im- Chelonian Conservation and Biology 2:184–89. perative (Oates 1999) instead of an advertising Congdon, J. D., A. E. Dunham, and R. C. van Loben initiative. Sels. 1993. Delayed sexual maturity and demographics of Blanding’s turtle (Emydoidea blan- dingii): Implications for conservation and man- REFERENCES agement of long-lived organisms. Conservation Biology 7:826–33. Arauz, R. M., and I. Naranjo. 1994. Hatching suc- Congdon, J. D., A. E. Dunham, and R. C. van Loben cess of leatherback turtles (Dermochelys coriacea) Sels. 1994. Demographics of common snap- in the leatherbacks of Guanacaste Marine Na- ping turtles (Chelydra serpentina): Implica- tional Park, Costa Rica. Proceedings of the 13th tions for conservation and management of Annual Workshop on Sea Turtle Biology and long-lived organisms. American Zoologist 34: Conservation (NOAA technical memorandum 397–408. NMFS-SEFC-278), 11–14. Cornelius, S. E. 1976. Marine turtle nesting activity Baker, C. P. 1994. Hatching their eggs and eating at Playa Naranjo, Costa Rica. Brenesia 8:1–27. them, too. Pacific Discovery (Summer): 10–18. ———. 1986. The sea turtles of Santa Rosa National Ballestero, J., G. Ordonez, and J. Gomez. 1996. Po- Park. Madrid: Hermanos Ramos. 64 pp. tential threats for the survival of sea turtles in Crouse, D. T. 1999. The consequences of delayed the Ostional Wildlife Refuge, Santa Cruz, Gua- maturity in a human-dominated world. In Life nacaste, Costa Rica. Proceedings of the 15th in the slow lane: Ecology and conservation of long- Annual Workshop on Sea Turtle Biology and lived marine animals, ed. J. A. Musick, 195–202. Conservation (NOAA technical memorandum Bethesda, Md.: American Fisheries Society. NMFS-SEFC-278), 31–34. Drake, D. L. 1996. Marine turtle nesting, nest pre- Bjorndal, K. A. 1999. Conservation of hawksbill dation, hatch frequency, and nesting seasonality sea turtles: Perceptions and realities. Chelonian on the Osa Peninsula, Costa Rica. Chelonian Conservation and Biology 3:174–76. Conservation and Biology 2:89–92. Bjorndal, K. A., J. A. Wetherall, A. Bolten, and J. A. Eckert, S. A., and L. Sarti. 1997. Distant fisheries Mortimer. 1999. Twenty-six years of green turtle implicated in the loss of the world’s largest nesting at Tortuguero, Costa Rica: An encour- leatherback nesting population. Marine Turtle aging trend. Conservation Biology 13:126–34. Newsletter 78:2–7.

SEA TURTLE CONSERVATION 207 frankie chap 15 8/20/03 8:02 PM Page 208

Frazier, J. G. 1997. Sustainable development: Mod- Myers, N. 1993. Ultimate security: The environmen- ern elixir or sack dress? Environmental Conser- tal basis of political stability. New York: Norton. vation 24:182–93. 308 pp. Frazier, J. G., and J. L. Brito Montero. 1990. Inci- Naranjo, I., and R. Arauz. 1994. Local guides in the dental capture of marine turtles by the sword- Leatherbacks of Guanacaste Marine National fish fishery at San Antonia, Chile. Marine Turtle Park: Sustained development and sea turtle con- Newsletter 49:8–13. servation. Proceedings of the 13th Annual Work- Godfrey, D. 1999. CCC celebrates 40 years of sea shop on Sea Turtle Biology and Conservation turtle conservation. Veladon, Caribbean Conser- (NOAA technical memorandum NMFS-SEFC- vation Corporation Newsletter (Winter 1999): 1–2. 278), 124–26. Heppell, L., B. Crowder, and T. R. Menzel. 1999. Nishimura, W., and S. Nakahigashi. 1990. Inciden- Life table analysis of long-lived marine species tal capture of sea turtles by Japanese research with implications for conservation and man- and training vessels: Results of a questionnaire. agement. In Life in the slow lane: Ecology and con- Marine Turtle Newsletter 51:1–4. servation of long-lived marine animals, ed. J. A. Oates, J. F. 1999. Myth and reality in the rain forest: Musick, 137–48. Bethesda, Md.: American How conservation strategies are failing in West Fisheries Society. Africa. Berkeley: University of California Press. Herzog, P. 1992. An assessment of ecotourism and 310 pp. its impact on leatherback sea turtles at Playa Paladino, F. V., M. P. O’Connor, and J. R. Spotila. Grande, Costa Rica. Programa Regional en 1990. Metabolism of leatherback turtles, gigan- Manejo de Vida Silvestre, Universidad Nacional, tothermy, and thermoregulation of dinosaurs. Heredia, Costa Rica. Unpublished report. Nature 344:858–60. Honey, M. 1999. Ecotourism and sustainable devel- Plotkin, P. T., R. A. Byles, D. C. Rostal, and D. W. opment: Who owns paradise? Washington, D.C.: Owens. 1995. Independent versus socially fa- Island Press. 405 pp. cilitated oceanic migration of the olive ridley, Hughes, D. A., and J. D. Richard. 1974. The nest- Lepidochelys olivacea. Marine Biology 122:137–43. ing of the Pacific ridley turtle Lepidochelys oli- Plotkin, P. T., D. W. Owens, R. A. Byles, and R. Pat- vacea on Playa Nancite, Costa Rica. Marine terson. 1996. Departure of male olive ridley Biology 24:97–107. turtles (Lepidochelys olivacea) from a nearshore Karl, S. A., and B. W. Bowen. 1999. Evolutionary breeding area. Herpetologica 52:1–7. significant units versus geopolitical taxonomy: Pritchard, P. C. H. 1982. Nesting of the leather- Molecular systematics of an endangered sea back turtle, Dermochelys coriacea, in Pacific turtle (genus Chelonia). Conservation Biology 13: Mexico, with a new estimate of the world pop- 990–99. ulation status. Copeia 1982:741–47. Leslie, A. J., D. N. Penick, J. R. Spotila, and F. V. Pal- Pritchard, P. C. H., H. Elizondo, C. Rodríguez, adino. 1996. Leatherback turtle, Dermochelys co- N. Guadamuz, E. Rodríguez, G. Rosales, and riacea, nesting and nest success at Tortuguero, Q. Jiménez. 1990. Las Baulas de Guanacaste, Costa Rica, in 1990–1991. Chelonian Conserva- a new national park for Costa Rica. Report to tion and Biology 2:159–68. Ministerio de Recursos Naturales, Energía, y Meylan, A. B. 1999. Status of the hawksbill turtle Minas, Programa de Rescate de Tortugas Mari- (Eretmochelys imbricata) in the Caribbean renas, San José, Costa Rica. 82 pp. gion. Chelonian Conservation and Biology 3: Spotila, J. R. 1988. Archie Carr: To the edge of 177–84. hope, 1909–1987. Herpetologica 44:128–32. Morreale, S. J., G. J. Ruiz, J. R. Spotila, and E. A. Spotila, J. R., A. E. Dunham, A. J. Leslie, A. C. Stey- Standora. 1982. Temperature dependent sex ermark, P. T. Plotkin, and F. V. Paladino. 1996. determination: Current practices threaten con- Worldwide population decline of Dermochelys servation of sea turtles. Science 216:1245–47. coriacea: Are leatherback turtles going extinct? Morreale, S. J., E. A. Standora, J. R. Spotila, and Chelonian Conservation and Biology 2:209–22. F. V. Paladino. 1996. Migration corridor for sea Spotila, J. R., R. D. Reina, A. C. Steyermark, turtles. Nature 384:319–20. P. T. Plotkin, and F. V. Paladino. 2000. Pacific Musick, J. A. 1999. Ecology and conservation of leatherback turtles face extinction. Nature 405: long-lived marine animals. In Life in the slow 529–30. lane: Ecology and conservation of long-lived marine Standora, E. A., J. R. Spotila, and R. E. Foley. 1982. animals, ed. J. A. Musick, 1–10. Bethesda, Md: Regional endothermy in the sea turtle, Chelonia American Fisheries Society. mydas. Journal of Thermal Biology 7:159–65.

208 SPOTILA AND PALADINO frankie chap 15 8/20/03 8:02 PM Page 209

Steyermark, A. C., K. Williams, J. R. Spotila, F. V. Wallace, D. R. 1992. The Quetzal and the macaw: Paladino, D. C. Rostal, S. J. Morreale, M. T. The story of Costa Rica’s national parks. San Fran- Koberg, and R. Arauz. 1996. Nesting leather- cisco: Sierra Club Books. 222 pp. back turtles at Las Baulas National Park, Costa Wetherall, J. A., G. H. Balazs, R. A. Tokunga, and Rica. Chelonian Conservation and Biology 2: M. Y.Y. Yong. 1993. Bycatch of marine turtles 173–83. in North Pacific high-seas driftnet fisheries Terborgh, J. 1999. Requiem for nature. Washington, and impacts on the stocks. North Pacific Fish- D.C.: Island Press. 234 pp. eries Commission Bulletin 53:519–38.

SEA TURTLE CONSERVATION 209 frankie chap 16 8/20/03 8:03 PM Page 210

chapter 16

Prospects for Circa Situm Tree Conservation in Mesoamerican Dry-Forest Agro-Ecosystems

David H. Boshier, James E. Gordon, and Adrian J. Barrance

ry forest once stretched almost con- primary, dry forest. Even in these areas it is rare Dtinuously along the Mesoamerican (Central to find a forest with no evidence of occasional American/Mexican) Pacific coast, from Sonora timber and firewood extraction, extensive live- in Mexico to Guanacaste in Costa Rica (area of stock browsing, or hunting. In Central America 550,000 km2). Conditions suitable for dry forest the dry-forest zone is reduced to a patchwork of also exist on the Yucatán Peninsula, the coast various types of agricultural land and small for- of the Mexican states of Puebla and Tamaulipas, ests of secondary origin. Unregulated selective with significant inland areas in dry valleys (e.g., felling of economically important tree species Motagua Valley, Guatemala) throughout the re- continues, resulting in the commercial extinction gion (see Graham and Dilcher 1995 and Murphy of timber species in some areas (e.g., Astronium and Lugo 1995 for the distribution and origins graveolens Jacq., Guaiacum sanctum L.). Reserve of this forest type). Human preference for the selection is often opportunistic, determined more seasonally dry tropical environment (Murphy by political, social, and economic constraints than and Lugo 1995) and the ease of clearing its veg- by optimal biological criteria. Consequently the etation have, however, led to the destruction location of reserves may bias floristic composi- and fragmentation of much of its forest. Janzen tion (at species and gene pool levels), limiting (1988: 130) estimated that less than 2 percent of their overall value for biodiversity conservation the original forest is in a state “sufficiently intact (Ledig 1988). to attract the attention of the traditional conser- Any conservation initiative requires consid- vationist,” with only 0.09 percent having official eration of the extent and value of and threat to reserve status. Only on the Pacific Mexican coast the resource to be targeted. Tropical dry forests are there significant areas of mature, possibly are species-rich in comparison with most of the

210 frankie chap 16 8/20/03 8:03 PM Page 211

world’s forest types (Gentry 1995). Biodiversity are left to conserve. The degree to which in situ can, however, be measured at a variety of levels, reserves are appropriate for conservation de- and species numbers alone are an insufficient pends on a variety of factors, such as forest size, and poor measure of the conservation impor- the extent of fragmentation, and the prevailing tance of any ecosystem. Consideration must socioeconomic context. Those forests that do also be given to the biota’s uniqueness, as well remain are usually small (e.g., 1–500 ha), below as the importance of interactions with other eco- the size that might be considered viable, and systems. At the species level, Mesoamerican dry highly intervened, such that the ideal of main- forests contain a woody flora that is distinct from taining a single large reserve is often irrelevant that of Mesoamerican wet forests (Gentry 1995). (as in the “single large or several small” debate; Even for tree species that occur in both wet and see Soulé 1987). Consideration must therefore dry forests, the dry-forest populations have typi- be given to the biological and social feasibility cally been shown to be genetically distinct from of managing networks of small patches within wet forest populations (e.g., Cordia alliodora [R. the current land-use mosaic. Conservation ini- et P.] Oken, Chase et al. 1995; Cedrela odorata L., tiatives for dry forest must examine approaches Gillies et al. 1997). Mesoamerican dry forests that depart from the traditional in situ conserva- are also host, depending on the season, to many tion paradigm, involving protected “wilderness migratory species, including species from the areas,” and focus as well on ways in which the wet tropics (chapters 7 and 8), such that its con- species of an already highly altered forest type servation is important in protecting more than can be conserved. just its own biota. Interest in the conservation of these species ECOLOGICAL RESTORATION and their habitat is not confined to biologists. Ecological restoration has been proposed as a Mesoamerican dry forests are the natural habitat tool to redress the inadequacies of current re- of a disproportionately high number of species serves (e.g., Janzen 1986a), and extensive areas commonly used in tropical agriculture and for- are currently under restoration management of estry for a variety of goods and services, such as various types in both the Guanacaste and Temp- Gliricidia sepium (Jacq.) Steudel, Leucaena spp., isque Conservation Areas in Costa Rica. Such Prosopis juliflora (Swartz) D.C., and Samanea restoration is facilitated by the very presence of saman (Jacq.) Merrill. These species are now existing in situ reserves, but in practice it is rarely propagated over much broader climatic and geo- feasible elsewhere given the socioeconomic con- graphical ranges than their natural distributions ditions prevalent in large parts of Mesoamerica. and have considerable socioeconomic impor- Even, for example, the establishment of Gua- tance (e.g., Stewart et al. 1992). Continued use nacaste National Park required the displacement and improvement of such species depend to of a human population living in, or farming, the some extent on the maintenance of genetic area, with smaller farmers and squatters faring variation within and between natural and semi- relatively poorly in the purchase and compensa- natural populations. tion process (Utting 1993). Nonetheless, in this case, sufficient national and international inter- OPTIONS FOR CONSERVATION est has probably been generated for the long- term maintenance of such management. Guana- IN SITU CONSERVATION caste, along with parts of Pacific lowland Mexico Conservation faces a problem. Mesoamerican and the Yucatán Peninsula, however, represents dry forest is diverse and distinctive, containing the better end of the spectrum in terms of the many socioeconomically important species; how- size and quality of remnant dry-forest patches. ever, existing reserves cover a very small fraction Although, at a smaller scale, restoration ecology of the original forest, and few intact dry forests may have great relevance in the creation of

CIRCA SITUM TREE CONSERVATION 211 frankie chap 16 8/20/03 8:03 PM Page 212

biological corridors between more or less intact genetic diversity within these on-farm popula- areas of forest, it and traditional in situ conser- tions (Chamberlain et al. 1996). Other examples vation will be inappropriate for much of the rest of traditional management systems that main- of the Mesoamerican dry-forest life zone, given tain tree species are given for the dry-forest zone the highly degraded and patchy nature of the of Central America (Kass et al. 1993) and the forest resource and prevailing socioeconomic Mixtec region of Guerrero, Mexico (Casas and conditions. Caballero 1996). Only recently, however, have tree conservation strategies sought to capitalize EX SITU CONSERVATION on such practices (e.g., Halladay and Gilmour Similarly, ex situ methods (e.g., seed banks, seed 1995), and consequently there is much to learn stands, botanic gardens) offer limited long-term if we are to realize the potential of circa situm potential for most tropical tree species. This is strategies and appreciate their limitations. In clear from the small number of tree taxa (some essence, circa situm conservation requires iden- 100 worldwide, mainly of economic importance; tification of compatibility between natural re- National Research Council 1991) subject to ef- source management systems found in native fective ex situ conservation; the scientific, tech- ranges of priority species (in this case trees) nical, and resource limitations that constrain ex and conservation objectives. Human inhabitants situ programs (National Research Council 1991); within such areas are thus recognized as agents and inherent deficiencies of ex situ populations of conservation. Circa situm conservation aims as conservation gene pools (Brown et al. 1989). to integrate conservation with prevailing socioeconomic trends, whereas the in situ and ex situ CIRCA SITUM CONSERVATION types of conservation aim to remove areas and The limitations of both in situ and ex situ ap- species from the influence of those trends. proaches, and the recognition that farmers maintain a large array of plant biodiversity in some FOCUS OF THIS CHAPTER traditional farming systems, have led to a re- examination of the potential role of “conserva- The focus of this chapter is to question the as- tion through use” on farms (e.g., Maxted et al. sumption that trees in scattered tropical agro- 1997). The term circa situm (also “circa situ,” ecosystems have no long-term viability and “farmer-based conservation,” “conservation in therefore no role in conservation. We discuss hortus”; Hughes 1998) has been used to distin- the role circa situm approaches could play in con- guish the very different circumstances of conserving the tree species of the Mesoamerican dry servation within altered agricultural landscapes forest. We are concerned here with the prospects (e.g., agroforestry systems, home gardens) out- for such an approach rather than the precise side natural habitats but within a species’ native prescriptions that will need to be formulated in- geographical range. Tree species have long been dependently for each species and situation. We maintained, and hence conserved, in Meso- do not propose that circa situm conservation can american agro-ecosystems. Hughes (1998) at- replace in situ conservation. Rather, we argue that tributes the relative abundance of a wide range for highly threatened, disturbed, and fragmented of Leucaena species in Mesoamerica, despite the vegetation types, it could have an important loss of forest cover in most areas, to their man- complementary role. This may be through the agement for several centuries by local commu- conservation of particular species and genes not nities and farmers in various ways and for vari- conserved in situ and/or the provision of addi- ous products and services. Indeed, he suggests tional habitat and the facilitation of gene flow that L. salvadorensis has been saved from extinc- between existing reserves. The potential gain is tion only because of its maintenance on farms, great if the many tree species that exist in such and there is also evidence for maintenance of disturbed vegetation can be shown to be conserv-

212 BOSHIER ET AL. frankie chap 16 8/20/03 8:03 PM Page 213

able there through existing land-use practices, sugarcane). The foothills show a fluctuating thus freeing scarce resources for the conser- mosaic of land uses, with a predominance of ba- vation of more critically threatened species that sic grain production (milpas), alternating with require more conventional, resource-intensive fallows, some low-density cattle raising and, in approaches. Provision of effective circa situm more humid areas, low-intensity coffee produc- options requires multidisciplinary research to tion. The lack of mechanized tillage and the establish the general potential for the integra- control of livestock movement typically allow tion of conservation and development and more natural regeneration of a highly variable range specifically which species are, or could be, sus- of tree species from seed banks, stumps, and tainably conserved in such systems, from both newly arrived seed. From these species many biological and human management perspectives. farmers actively protect a subset of those most The utility of particular species will not guaran- valued for a range of products (e.g., firewood tee their long-term conservation if, for example, and timber), reflecting a desire to optimize the genetic diversity and regeneration within critical use of available resources. Thus some trees reach, populations are compromised. Conversely, main- and remain in, reproductive maturity in milpas tenance of genetic diversity and adaptive capac- or pasture. Another subset of species remains as ity is irrelevant if current management drasti- stumps and resprouts (up to 17,000 per hectare) cally alters population persistence. We examine simply because the removal cost is not perceived results from our ecological, genetic, and socio- to produce adequate benefits (Barrance et al. in economic research conducted within dry forests review). Many of these trees reach reproductive of Costa Rica, Honduras, and Mexico, which maturity during fallow periods. The area there- overlap principally within southern Honduras. fore presents suitable conditions in terms of for- In particular, we consider results for two tim- est patchiness, the surrounding forest-agriculture ber species, Bombacopsis quinata (Jacq.) Dugand, interface, and socioeconomic circumstances to Bombacaceae, and Swietenia humilis Zucc., Meli- allow the synthesis of results across research aceae, both of local socioeconomic importance, disciplines. with superficially similar ecology and considered of conservation concern (see the section THE CASE STUDY SPECIES “The Case Study Species” later in this chapter). Bombacopsis quinata is a medium- to large-size We consider general principles that may in- deciduous tree from Central America, Colom- fluence under what conditions and for what bia, and Venezuela. It is hermaphroditic, largely species circa situm conservation will prove effec- self-incompatible (Sandiford 1998), and polli- tive and how it might be implemented. Finally nated principally by the bat Glossophaga soricina. we highlight research and training needs to ex- Its seeds are loosely attached to kapok, which plore further the feasibility and implementation facilitates their wind dispersal. Its timber is of this approach. much in demand and has led to investment in large plantations, as well as planting by farmers within Central America. This species can be prop- RESEARCH IN SOUTHERN HONDURAS agated by stakes and is used for living fences in Our research has had a common focus in south- parts of its range. Because of selective felling, ern Honduras (taken here as the lowlands and deforestation, and destructive agricultural prac- foothills of the departments of Choluteca and tices, this once common species is largely lim- Valle), an area within the dry subtropical life ited to isolated forest remnants. It is considered zone (Holdridge 1987) with a distinct dry season as endangered at the level of some populations from October to May. The southern Honduran (FAO 1986; Oldfield et al. 1998). lowlands are mainly occupied by cattle pasture Swietenia humilis is a monoecious, medium- and export agriculture (cantaloupe, watermelon, sized deciduous tree found along the Pacific

CIRCA SITUM TREE CONSERVATION 213 frankie chap 16 8/20/03 8:03 PM Page 214

watershed of Central America and Mexico. It is showed that, at the degree of isolation studied, pollinated by small butterflies, bees, and other in- fragmentation did not impose a genetic barrier sects, and its fruits contain large wind-dispersed between remnants, direct measures of current seeds. Under controlled pollination it is self- mating patterns giving altered and proportion- incompatible (D. Boshier unpubl. data). Its sawn ately higher levels of interfragment pollen flow timber is highly valued, but commercial refor- over longer distances. Common to both con- estation within its native range is hindered by trols and fragments for both species was a pre- the high incidence of attack by the shoot borer dominance of near neighbor mating (<300 m of Hypsipyla grandella. Where populations of S. hu- the maternal tree). A large proportion of pollen milis and S. macrophylla are sympatric, hybridiza- donors were, however, from outside each frag- tion is known, and their distinctness as species ment across the sampled area, indicating an ex- has been questioned (Styles 1981). Populations tensive network of gene exchange at this spatial over much of the species’ range have been re- scale (16 km2). So in two fragments (22 and 44 duced and fragmented, leading to its listing S. humilis trees), 62 percent and 53 percent, in 1973 in appendix II of the Convention on In- respectively, of pollen donors were from within ternational Trade in Endangered Species of the fragment, whereas 24 percent and 34 per- Wild Fauna and Flora (CITES) and classification cent were from distances greater than 1.5 km as “vulnerable” by the International Union for and 3.6 km, respectively. There was no evidence Conservation of Nature and Natural Resources for increased inbreeding in fragments, with both (IUCN) (Oldfield et al. 1998). species continuing to show high levels of outcrossing even in the smallest fragments. Indeed, GENETIC DIVERSITY AND ITS MAINTENANCE an S. humilis tree, “isolated” by 1.2 km from the Evidence to date shows that a high proportion of nearest flowering trees, showed 100 percent ex- tropical tree species are naturally outcrossing ternal pollen sources, with more than 70 percent (e.g., Bawa et al. 1985), with associated risks from from trees in the main area of forest (>4.5 km), inbreeding such as reduced fertility, growth, and in accord with the species’ self-incompatibility susceptibility to pests or diseases (e.g., Griffin and in contrast to predictions that spatially iso- 1990). Maintenance of genetic diversity is there- lated trees are more likely to deviate from random fore vital for the long-term viability and adapt- mating and receive pollen from fewer donors ability of populations of many tree species. (Murawski and Hamrick 1991). Reproductive biology, levels of genetic diversity Such enhanced levels of long-distance gene and gene flow, and extent of localized adaptation flow into smaller fragments, also seen in Spon- have been studied within the native range of dias mombin L. (Nason and Hamrick 1997), will both B. quinata and S. humilis (see Sandiford potentially restore or maintain genetic variation 1998; White 1998; Billingham 1999; White in populations of these species within the modi- et al. 1999; Boshier and Billingham 2000; and fied environment. This contrasts with tradi- White and Boshier 2000 for methodologies tional views of the genetic effects of fragmenting and results). Remnant stands of secondary dry populations whereby increases in spatial isola- forest, confined principally to hillsides and tion and population size reduction have been remnant trees in pastures in the Punta Ratón re- considered to reduce gene flow between frag- gion of the Honduran Pacific alluvial coastal ments (e.g., Saunders et al. 1991). Although the plains, near Choluteca, were used, along with genetic effects of fragmentation are complex, for control plots in more continuous forest in both some tree species under fragmentation, polli- Honduras and Guanacaste, Costa Rica. The frag- nation may occur over much greater distances ments varied in size (7–150 trees) and degree of than previously considered. The distance given spatial isolation (1–4.5 km). For both species, here for S. humilis is more than ten times greater genetic markers (allozymes or microsatellites) than that considered as distant by Chase et al.

214 BOSHIER ET AL. frankie chap 16 8/20/03 8:03 PM Page 215

(1996) and more in accord with distances previ- mary dry forest apparently no longer exists in ously identified by zoologists (e.g., Janzen 1971; southern Honduras, although it is probable that Frankie et al. 1976). There will, however, be a some older trees are remnants of much older distance beyond which genetic isolation will oc- forests (White et al. 1999). The forest cover that cur, with associated problems for population does remain is composed of stands of trees of viability and adaptation (see Young et al. 1996). as little as 2 ha in area and rarely greater than Although determination is experimentally prob- 20 ha, areas usually considered too small to be lematic, thresholds will vary between species of value to conservation practitioners. depending on pollinator characteristics, avail- The two species differ in terms of where ability, the specificity of the tree-pollinator rela- they occur. S. humilis, one of the most com- tionship, and the presence and strength of any monly encountered species (only 7 of more than self-incompatibility mechanism. Self-compatible 250 woody species identified were found more species that normally show some level of out- often), was found as frequently in farmland as in crossing (e.g., Murawski and Hamrick 1992) are forests (table 16.1). In contrast, B. quinata was likely to show increased levels of inbreeding found less than S. humilis, but much more often at much shorter distances of separation (lower in forests than on farmland. Both species were thresholds) than self-incompatible species. more common in fallows, fields, and pastures (table 16.1), which are typified as naturally re- PATTERNS OF DISTRIBUTION AT THE generating management systems, than in home SPECIES LEVEL IN SOUTHERN HONDURAS gardens and orchards, where planting is more Two rapid botanical surveys (RBS) were under- typical. Both species also showed considerable taken to determine the tree and shrub species variation in occurrence between the four com- composition of the southern Honduras agro- munities (table 16.1), probably owing to a com- ecosystem. The RBS technique follows the plot- bination of biological and anthropogenic factors, less or “unmeasured” sampling described by although in southern Honduras a lack of Hawthorne and Abu-Juam (1995), with minor undisturbed dry forest where the two species are modifications. A first, village-level survey con- sympatric makes it difficult to distinguish the sisted of species inventories in four rural commu- principal factors. Two of the larger forests in nities selected nonrandomly to reflect regional Choluteca suggest that both species may show variation across socioeconomic and environmen- similar levels of occurrence in mature forest. In tal gradients. Within each community 20 house- relatively undisturbed semievergreen forest on holds were selected along a socioeconomic gra- Cerro Guanacaure, B. quinata is common, as is dient and a random subset of land units selected S. humilis (J. Gordon pers. obs.). On Cerro Las for inventory from those farmed by these fam- Tablas, one of the most mature secondary for- ilies. Land unit categories followed definitions ests in the area, many mature trees of B. quinata given by householders. In each land unit species and S. humilis occur (densities up to 17.0 and were scored as present or absent, a species be- 9.6 per ha, respectively). In Costa Rica both ing considered present if it could be identified as species occur at similar levels in more undis- a woody individual, such that live stumps were turbed areas of protected dry forest (e.g., Lomas included. A second, forest-level survey inven- Barbudal, Playa Nancite mature forest area; toried tree diversity in a subset of forest patches table 16.2). Thus, we have grounds to speculate in southern Honduras. Forests were selected that in southern Honduras the relatively low nonrandomly to maximize geographical spread occurrence of B. quinata on farms results, at and biased toward the few remaining relatively least in part, from unfavorable management large forest areas (i.e., >50 ha). All forests ap- practices (further discussed later in this chap- peared to be largely, if not completely, composed ter) rather than a natural tendency toward low of secondary regeneration of varying ages. Pri- densities.

CIRCA SITUM TREE CONSERVATION 215 frankie chap 16 8/20/03 8:03 PM Page 216

TABLE 16.1 Occurrence of Bombacopsis quinata and Swietenia humilis by Land Use and Community in Southern Honduras, 1999

total percentage percentage number of containing containing land use samples b. quinata s. humilis

Forest 48 27 52 Farm 105 11 56 Fallow, ﬁeld, and pasture 58 16 67 Home garden and orchard 47 4 36

community/department San Juan Arriba, Choluteca 25 12 4 Agua Zarca, Valle 38 5 74 San José de las Conchas, Choluteca 24 17 46 Los Coyotes, Choluteca 25 16 64

The common on-farm occurrence of S. hu- areas of abandoned degraded pasture (table 16.2) milis cannot be explained simply by farmer pref- is abundant and that of B. quinata nonexistent. erence (see the section “Patterns of Management In contrast to other predictions and some studies of B. quinata and S. humilis in Southern Hon- (e.g., Ghazoul et al. 1998), seed production of duras” later in this chapter), given that natural S. humilis (table 16.3) was also more reliable regeneration has to occur before farmers can and higher in disturbed environments than in aid its recruitment. S. humilis, like its close rela- closed forest, whereas that of B. quinata showed tive S. macrophylla King, undoubtedly thrives in no differences by forest type, except at Playa regimes of heavy disturbance (Snook 1996) as Nancite, where it was low in all four fruiting provided by traditional agriculture in southern seasons. During this period 13 percent of the Honduras. Once established, it can survive for B. quinata trees at Playa Nancite died from ap- decades in the closed forests that may form when parent old age/creepers/wind. Given the lack fallows are abandoned. Gerhardt (1994) showed of any regeneration, the future of B. quinata at that germination of S. macrophylla in Guana- this site, although it is within a protected area, caste was not affected by differences in light seems insecure. Indeed, fires and tree felling ev- levels and could establish equally well in pasture ident over four years (table 16.2) confirm threats and young secondary forest. A study of localized to both species, even within “protected” areas. adaptation in both Costa Rica and Honduras, with We can conclude that species are likely to be seedlings in pasture and secondary and mature scattered nonrandomly in the various land-use forest, showed zero survival of B. quinata at three types and communities of an agro-ecosystem years of age in any of these land uses, compared and that these distribution patterns must be with 15–60 percent for S. humilis (M. R. Billing- known for circa situm conservation options to ham and D. H. Boshier unpubl. data). This is be assessed and for strategies to be devised and borne out at Playa Nancite (Santa Rosa National implemented. The fact that 76 percent of the Park), where S. humilis regeneration in adjoining species identified in the surveys were found in

216 BOSHIER ET AL. frankie chap168/20/038:03PMPage217

TABLE 16.2 Mortality of Bombacopsis quinata and Swietenia humilis Trees over a Four-Year Period at Sites in Costa Rica and Honduras

area studied no. of mean dbh in percentage of species site country site type (ha) trees cm (s.d.) trees cut/dead

B. quinata Lomas Barbudal Costa Rica Protected area 25 61 69.2 (25.5) 0/6.6a Biological Reserve Playa Nancite, Santa Costa Rica Protected area/ 28 62 71.6 (42.8) 0/12.9 Rosa National Park mature Playa Nancite, Santa Costa Rica Protected area/ 3 0 Rosa National Park regeneration Punta Ratón, Honduras Private forest/ 240 172 54.2 (20.9) 8.1/5.8a Choluteca farm S. humilis Lomas Barbudal Costa Rica Protected area 25 57 47.4 (17.0) 0/14.0a Biological Reserveb Playa Nancite, Santa Costa Rica Protected area/ 28 156 46.7 (16.0) 0/5.8 Rosa National Park mature Playa Nancite, Santa Costa Rica Protected area/ 3 23 20.3 (3.7) 0/0 Rosa National Park regeneration Cerro Las Tablas, Honduras Private forest/ 68 105 41.9 (14.0) 25.7/1.9 Choluteca farm Punta Ratón, Honduras Private forest/ 240 75 37.9 (14.1) 5.3/5.3 Choluteca farm

Note: Mortality occurred through either natural causes or human intervention. dbh = Diameter at breast height. aKilled by 1994 ﬁre. bPossibly S. macrophylla or S. macrophylla × S. humilis. frankie chap 16 8/20/03 8:03 PM Page 218

TABLE 16.3 Percentage of Bombacopsis quinata and Swietenia humilis Trees per Year with Moderate to Heavy Seed Production at Various Sites

sample 1994 1995 1996 1997 1998 species sitea sizeb (%) (%) (%) (%) (%)

B. quinata Lomas Barbudal >45 70.6 66.7 68.9 Playa Nancite >30 61.5 56.4 55.2 31.5 Punta Ratón >105 76.4 61.9 73.3 90.1 S. humilis Lomas Barbudal >46 12.1 0.0 16.3 20.8 Playa Nancite >143 32.9 11.0 18.4 17.2 Cerro Las Tablas >61 48.7 12.6 29.5 44.6 Punta Ratón >45 28.0 36.1 42.2 49.1 33.3 Comayagua 30 66.7 60.0 83.3 76.7

Note: Moderate to heavy seed production for B. quinata is more than 20 seed capsules; for S. humilis, more than 10 seed capsules. aSites are ranked for increasing degree of disturbance: Lomas Barbudal, relatively undisturbed area within a reserve; Playa Nancite, area of the Santa Rosa National Park with disturbance; Cerro Las Tablas, forest with remnant trees and secondary regeneration; Punta Ratón, remnant trees and secondary regeneration in small fragments and pasture; Comayagua, planted trees by roadside. bSample size varies from year to year owing to mortality.

agro-ecosystems (i.e., not limited to forest) em- “Genetic Diversity and Its Maintenance” earlier phasizes the potential for on-farm trees to make in this chapter). Evidently conservation strategies a contribution to conservation strategies. This is for these two species need to be very different. borne out by Harvey and Haber (1999), who found 190 tree species in 237 ha of upland pas- PATTERNS OF MANAGEMENT OF B. QUINATA ture near Monteverde Reserve in Costa Rica. As- AND S. HUMILIS IN SOUTHERN HONDURAS suming that greater occurrence means greater A study in the same small farmer communities conservation potential, the potential for different where village-level species inventories were taken sites to be the foci of successful circa situm con- (see previous section) shows that, in addition to servation will vary. For example, taking the dis- differences in reproductive biology, ecology, tribution evidence in isolation, the Agua Zarca and local distribution, B. quinata and S. humilis community has a high potential for conserva- differ in management by local people. In gen- tion of S. humilis but very little for B. quinata eral, within these communities, people men- (table 16.1). S. humilis appears ecologically suited tioned that many of the same tree species under to circa situm conservation “through use” on active protection are also used for timber, al- these farms, where tillage is minimal, livestock though there was a marked contrast between the movement controlled, and fallows sporadically frequency of active protection for S. humilis allowed. B. quinata, however, appears less suited, (second only to Cordia alliodora) and that of being much more susceptible to reductions in B. quinata (table 16.4). forest cover, although even for such species that A focal group meeting in Los Coyotes, where appear to be predominantly “forest species,” the species were rated for present and past use, also relatively few individuals on farms may be impor- revealed marked differences between the two tant for gene ﬂow between forests (see the section species in terms of their past and present socio-

218 BOSHIER ET AL. frankie chap 16 8/20/03 8:03 PM Page 219

TABLE 16.4 Principal Tree Species Mentioned as Used for Timber or Actively Protected within Agricultural Areas as Sources of Timber or Posts in Four Communities in Southern Honduras in 1998

percentage of percentage of species used intervieweesa species protected intervieweesa

Cordia alliodora 84.8 Cordia alliodora 38.0 Bombacopsis quinata 20.3 Swietenia humilis 22.8 Enterolobium cyclocarpum 19.0 Lysiloma spp. 20.3 Albizia saman 16.5 Enterolobium cyclocarpum 10.1 Lysiloma spp. 16.5 Albizia saman 10.1 Swietenia humilis 16.5 42 other tree species, 1.3–6.3 each including Bombacopsis quinata Calycophyllum candidissimum 15.2 Cedrela odorata 15.2 Conocarpus/Rhizopora spp. 15.2 (mangrove) Simarouba glauca 11.4

Note: The four communities were San Juan Arriba, Agua Zarca, San José de las Conchas, and Los Coyotes. aFigures given as a percentage of 79 interviewees.

economic niches. Previously (“in the time of their fields to meet their subsistence needs and as a grandfathers”) B. quinata was almost the only naturally regenerated cash crop. Furthermore, species used for timber; however, its exploitation stumps and seeds survive regardless of felling, to supply furniture workshops in the nearby town as stump removal is difficult and mechanical of El Triunfo led to a large decrease in its num- tillage is rarely practiced owing to topographical bers in the area. Farmers now use a broad range constraints. Farmers are also adept at control- of previously little-used species, including S. hu- ling grazing, a prerequisite of successful maize milis, others of which (e.g., Guazuma ulmifolia cultivation, so that livestock does not necessarily Lam) have far inferior timber value. Farmers also prevent regeneration and indeed may facilitate contrasted previous management during field it (Janzen 1986a). Patches of fallow and forest clearance, whereby many valuable timber trees within the mosaic continue, despite their small were felled and burned on-site or used for fire- size, to be propagule sources of many of the wood, with current practices, whereby valued species found in these fields. Up to certain lim- trees are protected and only timber offcuts used its the benefits of maintaining trees outweigh as firewood. negative effects (e.g., shade) that farmers per- Apparently because of a scarcity of off-field ceive the trees to have on crops (Barrance et tree resources, resulting from high human pop- al. in review). This form of management repre- ulation densities and growth rates, many farm- sents a rational response to resource scarcity ers, by favoring certain species during clearance, rather than a desire to conserve biological diver- protect and manage valued trees within their sity per se.

CIRCA SITUM TREE CONSERVATION 219 frankie chap 16 8/20/03 8:03 PM Page 220

The relatively low frequency of protection of CONCLUSIONS B. quinata in fields (tables 16.1, 16.4) does not appear to result from its being less valued than In situ conservation of the remaining areas of S. humilis or from differences in farmers’ per- mature Mesoamerican dry forest is likely to be ceptions of tree/crop interactions (no important the most effective and efficient means of safe- negative interactions were reported by farmers guarding a substantial part of the threatened between either species and crops). A more prob- specific and subspecific tree diversity of this able explanation is the relative scarcity and patchy forest type. However, many species and popula- distribution of B. quinata natural regeneration, tions of concern are not well or not at all rep- particularly in fields. Farmers protect it where resented in such areas and require a different they find it but find it much less frequently than approach. Habitat restoration may also have a S. humilis. The current scarcity of B. quinata role, but the socioeconomic conditions that make appears to stem, at least in part, from its past this a possibility are uncommon, and costs may overexploitation, prior to the self-imposition of be prohibitive. Circa situm conservation has much tree management controls by farmers as a re- to offer in these situations but also significant sponse to scarcity and its ecological characteris- limitations; for it to be effective, its application tics. S. humilis differs in that it appears to have must be guided by information from multidis- been rather less valued than B. quinata and as a ciplinary research of the type discussed in this result largely escaped overexploitation. It now chapter. As the contrasting examples of B. quin- benefits from the current practice of active pro- ata and S. humilis show, we cannot permit our- tection of the remaining valued species and its selves the luxury of assuming that all species are ease of regeneration. However, despite farmers’ likely to persist in an agricultural mosaic until clear preference for both species and their pro- this is proved otherwise. However, on the basis motion by extension agencies, neither species is of such evidence, neither should we underesti- commonly planted. This probably reflects the mate the capability of many species to persist in cost of planted trees, compared with “free” nat- large numbers in these agro-ecosystems under ural regeneration and possibly the greater risks current practices, as this could lead to the mis- to planted trees from cattle that are periodically direction of limited conservation resources to- introduced into milpas to eat crop residues. ward species not under threat. Circa situm con- Both B. quinata and S. humilis also occur in servation of tree diversity is already occurring, forest patches and pastures within larger estates albeit as a side effect of resource-poor farmer (e.g., Cerro Las Tablas), although the manage- practices. Given apparently similar conditions ment systems and socioeconomic conditions are across large parts of the dry zone foothills of El very different from those found in the small Salvador, Guatemala, and Nicaragua, circa situm farmer communities. In both pasture and for- populations may represent a considerable con- est remnants on these estates there appears to servation resource. If conservation planners were be a continued decline due to piecemeal felling to take this resource into consideration, how (table 16.2), possibly as the landowners do not many species currently assumed to be threat- face the same conditions of overall resource ened by habitat loss might prove to be thriving scarcity that motivate small farmers to nurture (Vandermeer and Perfecto 1997)? natural regeneration. In addition, conditions in An enhanced future role for circa situm con- many pastures (they are normally destumped servation in the dry tropics of Mesoamerica will on clearing and then periodically burned to en- depend principally on the existence of species of courage grass development) are less propitious conservation importance in the agro-ecosystems to tree regeneration than those on steeper land and forests in question. Species of widespread under subsistence production systems. occurrence whose status does not appear about

220 BOSHIER ET AL. frankie chap 16 8/20/03 8:03 PM Page 221

to change and species well represented in ade- little understood, and their compatibility with quately protected areas should not attract more conservation objectives are unclear. of the scarce resources available for conserva- For the species studied here and others else- tion. Are agro-ecosystems likely to contain the where (e.g., chapter 3), it appears that descrip- types of species we wish to conserve? To answer tions of remnant trees and forest patches as “iso- this question we offer the following generaliza- lated” or “living dead” (Janzen 1986b), with little tions. Agro-ecosystems are highly disturbed in or no conservation value, may be misleading comparison with the original vegetation, and the and more human perception than a true reflec- dominant species are likely to be either “pre- tion of actual gene flow and any biological real- adapted” to disturbance (i.e., weedy, pioneer, and/ ity. It is more realistic to view remnant forests or coppicing) or positively selected by farmers not as islands but rather as existing within a (or both). The long-distance dispersal propagules mosaic of land uses that differ in their capacity of these typically pioneer type species, coupled to provide habitat or permit movement for any with ever increasing disturbance, ensure that organism. In this sense it is important to recog- they will generally not be conservation priorities. nize the complementary habitat role that main- One priority must therefore be the relatively few tenance of trees on farms is already playing to species with much narrower distributions, en- in situ dry-forest conservation. With adequate demic to areas now completely converted to agro- gene flow and seed production, some remnant ecosystems (e.g., Leucaena salvadorensis; Hellin forest patches and trees may be important con- and Hughes 1993). For these species in situ is de tributors to connectivity and conservation, both facto, not a conservation option. A second prior- in situ and circa situm, more generally. The ca- ity would be those species and species assem- pacity of some pollinators to move long dis- blages whose continued in situ conservation tances has been shown, although the potential would be enhanced by their conservation circa to move between patches will depend on their situm. Here conservation in agro-ecosystems behavioral response to such a mosaic. Some bat between reserves increases habitat area, con- species move preferentially down forest tracks nectivity, and population sizes. and pathways (e.g., Estrada et al. 1993). Whereas Knowledge of which tree species occur where some bee species are generally restricted in their says nothing about the species’ viability within movements (Powell and Powell 1987), most an agro-ecosystem. The effects of fragmentation show some long-distance movements, and some and the regeneration of secondary forest patches will move more than 4 km across agro-ecosys- are likely to be dynamic, and to ensure that we tems between forest patches (e.g., Frankie et al. deal with populations with a long-term future, 1976; Raw 1989). With a range of nonspecialist we must carefully consider the reproductive and pollinators, pollen flow dynamics in S. humilis regenerative capacities of priority species and are probably far less susceptible to habitat dis- the perpetuation of management practices that turbance than tree species with more specialist allow natural or artificial regeneration. Yet knowl- pollinators, although Ficus species, with species- edge of the effects of human interventions (e.g., specific wasp pollinators, may form extensive logging, fragmentation) on tree gene pools is metapopulations in fragmented landscapes (Na- relatively poor. We do not know at what stage son et al. 1998). Changes in pollinator assem- forest fragments become genetically isolated; nor blages in fragmented landscapes may strongly do we understand the consequences of gene affect patterns of gene flow and reproduction flow between managed and remnant natural pop- in remnant tree populations. Concerns that de- ulations. Selection pressures exerted by farmers, clines in pollinator populations in such agro- intentionally or unintentionally, at species, pop- ecosystems may eventually limit tree reproduc- ulation, and within-population levels are also tion require monitoring of numbers, as well as

CIRCA SITUM TREE CONSERVATION 221 frankie chap 16 8/20/03 8:03 PM Page 222

evidence for pollinator limitation (Allen-Wardell intensification leading to massive extinctions et al. 1998). (cf. Janzen 1986b). Circa situm conservation strat- Active tree protection by farmers is strongly egies are more likely to be effective when planned dependent on their perceptions of the trees’ po- on a species-by-species basis, within the farm- tential value to them and on other community ing systems context of a particular area, with re- members’ recognition of their ownership rights sources focused on target species so as to mini- to the trees. However, the B. quinata case clearly mize negative socioeconomic effects. The “area” demonstrates that even if a species is valued its or “management unit” over which such strategies conservation is not guaranteed (Gordon et al. would operate would be measured in numbers 2003). In fact, use value may lead to a species’ of participating households or land units in decline until a threshold is reached whereby which target species conservation is successfully farmers are obliged to modify their practices if incorporated into existing, or acceptable, farm they wish to ensure a continued supply. It ap- management practices. pears that at this point the rate of decline of The prospects for circa situm conservation for many species tends to level out as mature indi- S. humilis appear good, but less so for B. quinata, viduals are managed on a more rational basis and it is ironic that of these two species the for- and removals replaced by natural regeneration mer is the one afforded more protection by in- that receives active protection by farmers. How- ternational convention (appendix II of CITES). ever, if regeneration is poor in agro-ecosystems, However, for it or similar species, some moni- as appears to be so for B. quinata, it may continue toring may be advisable to ensure that its niche to decline and disappear. Alternatively farmers in the farming system is maintained in the face may switch much of their demand for a particu- of changing socioeconomic trends. Current lar product to another species. trends, at least in some areas, involve an itera- Hughes (1998) suggests that circa situm con- tive process of property subdivision and elimi- servation is probably limited to species that tol- nation of forest and fallow areas, with a corre- erate disturbance and are actively used and pre- sponding increase in numbers of trees in grain ferred by farmers. However, although it indeed fields and pastures. It is possible that farm size favors disturbance-tolerant species, the examples will eventually reach a threshold beyond which from southern Honduras and those of Kass et al. farmers are unwilling to divide their property (1993) suggest its applicability to a broader range further among their heirs, and increased de- of species than just the actively preferred ones. pendence on off-farm income sources will divert However, survival of nonpreferred species may pressure from land and tree resources, leading to a large extent depend on the technology level to a stable situation. However, it is unclear at of land management practices; for example, a what point a threshold will be reached, to what larger number of stumps of nonvalued coppic- degree this will be influenced by available off- ing species is allowed to persist in low-intensity farm income, and whether farm size reductions swidden agriculture of southern Honduran hill- will lead farmers to increase tree densities on sides than in ploughed fields of the central valleys farms to satisfy tree product needs or reduce of Oaxaca, Mexico. It would be unfair to promote them to increase basic grain yields. The chal- conservation initiatives at a landscape scale across lenge offered by circa situm conservation is to the Mesoamerican dry zone that prevent poor adapt to these changes over a relatively wide farmers from adopting more profitable practices, geographical area rather than seek to halt such although in many parts there are significant bio- processes within a confined area such as a re- physical constraints (e.g., topography, resource serve. To this end there is a need to raise aware- availability) to those seeking to intensify produc- ness among development professionals of the tion. At least in southern Honduras there ap- value of natural regeneration as both conser- pears to be little immediate risk of agricultural vation and socioeconomic resources. Efforts to

222 BOSHIER ET AL. frankie chap 16 8/20/03 8:03 PM Page 223

promote planting and the use of multipurpose . Targeted tree populations need to be de- or exotic tree species (or both) at the expense of scribed in terms of size class distributions natural regeneration may have deleterious effects with respect to reproductive maturity. We not only on conservation but also on farmer wel- cannot assume that a circa situm population fare. This would be evident through a reduction contains the same proportion of repro- in the range of forest products normally avail- ductively active individuals as an in situ able to farmers from natural regeneration and population. an increased cost of tree establishment. . Fragmentation thresholds for gene flow need to be determined and the possible selection pressures exerted by farmers RECOMMENDATIONS FOR further elucidated. RESEARCH AND TRAINING . Species of conservation concern within The data requirements to answer the questions agro-ecosystems need to be understood and raised in this chapter are considerable, and real- classified in terms of farmers’ perceptions ization of the potential of circa situm conserva- of their value. Species guilds need to be tion requires a range of actions. For circa situm defined, combining biological and socio- conservation to have anything but a minor role economic variables to reduce, by extrapola- in the conservation of tropical forest resources, tion, the information demands of circa situm priorities among species and forest types will conservation. need to be set and shortcuts found. We end this . Education and training curricula need to be chapter by listing recommendations for research broadened so that more conservation biolo- and dissemination priorities for the further de- gists recognize the potential role of on-farm velopment of circa situm conservation. The list is conservation and more rural development neither prescriptive nor exhaustive and reflects practitioners recognize their role in bio- primarily the situation as we understand it in diversity conservation. southern Honduras. It is presented to illustrate . the types of information demands that circa si- The complementary habitat benefits of dif- tum prescriptions will make. However, overrid- ferent agro-ecosystems for conservation need ing such specificity is the need for conservation to be evaluated, recognized, and promoted. planners, more accustomed to in situ methods, . Rapid rural appraisal techniques (see Cham- to consider the possibility that populations of bers and Guijt 1995) need adapting and test- trees found outside protected areas have a role ing for their ability to provide adequate and in the conservation of biodiversity. This in turn cost-effective information for circa situm will require the direct involvement of develop- management. ment organizations in biodiversity conserva- . Strategies need to be explored on a species- tion and an effective two-way communication by-species basis to ensure effective use of between them and “traditional” conservation resources and minimum disruption to farm- organizations to ensure both conservation and ing practices. development benefits. . The costs/benefits to farmers of actively . Species distribution patterns need to be es- promoting conservation through a circa tablished not only at a broad level, to identify situm approach need to be carefully balanced species that are rare and therefore a conser- and, if necessary, forms of compensation vation priority, but also at a finer scale, to iden- considered. tify which species are not conserved in situ . Effective systems for long-term monitoring and which land uses within an agro-ecosys- of the biological (e.g., pollinator levels, re- tem are most appropriate for each species. generation levels) and socioeconomic

CIRCA SITUM TREE CONSERVATION 223 frankie chap 16 8/20/03 8:03 PM Page 224

dynamics (e.g., management practice Chambers, R., and I. Guijt. 1995. PRA—five years changes) of conservation need to be later: Where are we now? Forests, Trees and People established. Newsletter 26/27:4–14. Chase, M. R., D. H. Boshier, and K. S. Bawa. 1995. Population genetics of Cordia alliodora (Bor- ACKNOWLEDGMENTS aginaceae), a neotropical tree: 1. Genetic variation in natural populations. American Journal of This chapter is an output from projects funded Botany 82:468–75. through the Forestry Research Programme of the Chase, M. R., C. Moller, R. Kesseli, and K. S. Bawa. United Kingdom Department for International 1996. Distant gene flow in tropical trees. Nature Development (DFID) for the benefit of develop- 383:398–99. Estrada, A., R. Coates-Estrada, D. Meritt Jr., S. Mon- ing countries. The views expressed are not nec- tiel, and D. Curiel. 1993. Patterns of frugivore essarily those of DFID. Projects R5729, R6516, species richness and abundance in forest is- and R6913 of the Forestry Research Programme. lands and in agricultural habitats at Los Tuxtlas, Mexico. Vegetatio 107/108:245–57. REFERENCES FAO (Food and Agriculture Organization of the United Nations). 1986. Databook on endangered Allen-Wardell, G., et al. 1998. The potential conse- tree and shrub species and provenances. Rome: quences of pollinator declines on the conserva- FAO. 524 pp. tion of biodiversity and stability of food crop Frankie, G. W., P. A. Opler, and K. S. Bawa. 1976. yields. Conservation Biology 12:8–17. Foraging behaviour of solitary bees: Implica- Barrance, A. J., L. Flores, E. Padilla, J. E. Gordon, tions for outcrossing of a neotropical forest tree and K. Schreckenberg. In review. Trees and species. Journal of Ecology 64:1049–57. farming in the Honduran dry zone of south- Gentry, A. H. 1995. Diversity and floristic compo- ern Honduras 1: Campesino tree husbandry sition of neotropical dry forests. In Seasonally practices. dry tropical forests, ed. S. H. Bullock, H. A. Bawa, K. S., D. R. Perry, and J. H. Beach. 1985. Mooney, and E. Medina, 146–94. Cambridge: Reproductive biology of tropical lowland rain- Cambridge University Press. forest trees 1. Sexual systems and incompati- Gerhardt, K. 1994. Seedling development of four bility mechanisms. American Journal of Botany tree species in secondary tropical dry forest in 72:331–45. Gunanacaste, Costa Rica. Ph.D. diss., University Billingham, M. R. 1999. Genetic structure, lo- of Uppsala, Sweden. calised adaptation and optimal outcrossing dis- Ghazoul, J., K. A. Liston, and T. J. B. Boyle. 1998. tance in two neotropical tree species. D.Phil. Disturbance induced density-dependent seed set thesis, University of Oxford. 177 pp. in Shorea siamensis (Dipterocapaceae), a tropi- Boshier, D. H., and M. R. Billingham. 2000. Ge- cal forest tree. Journal of Ecology 86:462–73. netic variation and adaptation in tree popula- Gillies, A. C. M., et al. 1997. Genetic variation in tions. In Ecological consequences of habitat het- Costa Rican populations of the tropical timber erogeneity, ed. M. J. Hutchings, E. A. John, and species Cedrela odorata L., assessed using A. J. A. Stewart, 267–89. Oxford: Blackwell RAPDs. Molecular Ecology 6:1133–45. Science. Gordon, J. E., A. J. Barrance, and K. Schrecken- Brown, A. H. D., O. H. Frankel, D. R. Marshall, berg. 2003. Are rare species useful species? and J. T. Williams, eds. 1989. The use of plant ge- Obstacles to the conservation of tree diversity netic resources. Cambridge: Cambridge Univer- in the dry forest zone agro-ecosystems of sity Press. 382 pp. Mesoamerica. Global Ecology and Biogeography Casas, A., and J. Caballero. 1996. Traditional man- 12:13–19. agement and morphological variation in Leu- Graham, A., and D. Dilcher. 1995. The Cenozoic caena esculenta (Fabaceae: Mimosoideae) in the record of tropical dry forest in northern Latin Mixtec region of Guerrero, Mexico. Economic America and the southern United States. In Botany 50:167–81. Seasonally dry tropical forests, ed. S. H. Bullock, Chamberlain, J. R., C. E. Hughes, and N. W. Gal- H. A. Mooney, and E. Medina, 124–45. Cam- wey. 1996. Patterns of variation in the Leucaena bridge: Cambridge University Press. shannonii alliance (Leguminosae: Mimosoideae). Griffin, A. R. 1990. Effects of inbreeding on growth Silvae Genetica 45:1–7. of forest trees and implications for management

224 BOSHIER ET AL. frankie chap 16 8/20/03 8:03 PM Page 225

of seed supplies for plantation programmes. In Mooney, and E. Medina, 9–34. Cambridge: Reproductive ecology of tropical forest plants, ed. Cambridge University Press. K. S. Bawa and M. Hadley, 355–74. Man and the Murray, M. G., M. J. Green, G. C. Bunting, and Biosphere Series, vol. 7. Carnforth, England: J. R. Paine. 1995. Biodiversity conservation Parthenon Publishing Group. in the tropics: Gaps in habitat protection and Halladay, P., and D. A. Gilmour, eds. 1995. Con- funding priorities. Unpublished report. World serving biodiversity outside protected areas: The role Conservation Monitoring Centre, Cambridge, of traditional agro-ecosystems. Gland, Switzer- England. land: IUCN. 229 pp. Nason, J. D., and J. L. Hamrick. 1997. Reproduc- Harvey, C. A., and W. A. Haber. 1999. Remnant tive and genetic consequences of forest frag- trees and the conservation of biodiversity in mentation: Two case studies of neotropical Costa Rican pastures. Agroforestry Systems 44: canopy trees. Journal of Heredity 88:264–76. 37–68. Nason, J. D., E. A. Herre, and J. L. Hamrick. 1998. Hawthorne, W. D., and M. Abu-Juam. 1995. Forest The breeding structure of a tropical keystone protection in Ghana. Gland, Switzerland: IUCN. plant resource. Nature 391:685–87. 202 pp. National Research Council. 1991. Managing global Hellin, J. J., and C. E. Hughes. 1993. Leucaena genetic resources—forest trees. Washington, D.C.: salvadorensis: Conservation and utilization in Cen- National Academy Press. 228 pp. tral America. Serie Miscelanea CONSEFORH, Oldfield, S., C. Lusty, and A. MacKinven. 1998. The No. 39-21/93. Comayagua, Honduras: CONSE- world list of threatened trees. Cambridge: World FORH. 41 pp. Conservation Press. 650 pp. Holdridge, L. R. 1987. Ecología basada en zonas de Powell, A. H., and G. V. N. Powell. 1987. Popula- vida. San José: IICA. 216 pp. tion dynamics of male euglossine bees in Ama- Hughes, C. E. 1998. Leucaena: A genetic resources zonian forest fragments. Biotropica 19:176–79. handbook. Tropical Forestry Paper 37, Oxford Raw, A. 1989. The dispersal of euglossine bees be- Forestry Institute. 274 pp. tween isolated patches of eastern Brazilian wet Janzen, D. H. 1971. Euglossine bees as long dis- forest (Hymenoptera, Apidae). Revista Brasileira tance pollinators of tropical plants. Science 171: de Entomologia 33:103–7. 204–5. Sandiford, M. 1998. A study of the reproductive ———. 1986a. Guanacaste National Park: Tropical biology of Bombacopsis quinata. D.Phil. thesis, ecological and cultural restoration. San José: University of Oxford. 200 pp. UNED. 103 pp. Saunders, D. A., R. J. Hobbs, and C. R. Margules. ———. 1986b. Blurry catastrophes. Oikos 47:1–2. 1991. Biological consequences of ecosystem ———. 1988. Tropical dry forests: The most en- fragmentation: A review. Biological Conservation dangered major tropical ecosystem. In Biodiver- 5:18–32. sity, ed. E. O. Wilson, 130–37. Washington, D.C.: Snook, L. K. 1996. Catastrophic disturbance, log- National Academy Press. ging and the ecology of mahogany (Swietenia Kass, D. C. L., C. Foletti, L. T. Szott, R. Landaverde, macrophylla King): Grounds for listing a major and R. Nolasco. 1993. Traditional fallow systems tropical timber species in CITES. Botanical of the Americas. Agroforestry Systems 23:207–18. Journal of the Linnean Society 122:35–46. Ledig, F. T. 1988. The conservation of genetic di- Soulé, M. E. 1987. Viable populations for conserva- versity in forest trees. BioScience 38:471–79. tion. Cambridge: Cambridge University Press. Maxted, N., B. V. Ford-Lloyd, and J. G. Hawkes, eds. 189 pp. 1997. Plant conservation: The in situ approach. Stewart, J. L., J. J. Hellin, and C. E. Hughes. 1992. London: Chapman and Hall. 446 pp. Biomass production of Central American semi- Murawski, D. A., and J. L. Hamrick. 1991. The ef- arid zone species. Tropical Forestry Paper 26, fects of the density of flowering individuals on Oxford Forestry Institute. 83 pp. the mating systems of nine tropical tree species. Styles, B. T. 1981. Swietenioideae. In Meliaceae, Heredity 67:167–74. Flora Neotropica Monograph No. 28, ed. T. D. ———. 1992. The mating system of Cavanillesia Pennington, B. T. Styles, and D. A. H. Taylor, platanifolia under extremes of flowering tree den- 359–418. New York: New York Botanical Garden. sity: A test of predictions. Biotropica 24:99–101. Utting, P. 1993. Trees, people and power: Social di- Murphy, P. G., and A. E. Lugo. 1995. Dry forests of mensions of deforestation and forest protection in Central America and the Caribbean. In Season- Central America. London: Earthscan Publica- ally dry tropical forests, ed. S. H. Bullock, H. A. tions. 206 pp.

CIRCA SITUM TREE CONSERVATION 225 frankie chap 16 8/20/03 8:03 PM Page 226

Vandermeer, J., and I. Perfecto. 1997. The agro- tions, ed. A. G. Young and G. Clarke, 293–311. ecosystem: A need for the conservation biolo- Cambridge: Cambridge University Press. gist’s lens. Conservation Biology 11:591–92. White, G. M., D. H. Boshier, and W. Powell. 1999. White, G. M. 1998. A study of the population ge- Genetic variation within a fragmented popula- netics of Swietenia humilis Zucc. in fragmented tion of Swietenia humilis Zucc. Molecular Ecol- forest. Ph.D. thesis, University of Dundee, ogy 8:1899–1910. Scotland. 154 pp. Young, A. G., T. Boyle, and T. Brown. 1996. The White, G. M., and D. H. Boshier. 2000. Fragmen- population genetic consequences of habitat tation in Central American dry forests: Genetic fragmentation for plants. Trends in Ecology and impacts on Swietenia humilis. In Genetics, de- Evolution 11:413–18. mography and the viability of fragmented popula-

226 BOSHIER ET AL. frankie chap 17 8/20/03 8:04 PM Page 227

PART TWO

Transferring Biodiversity Knowledge into Action

The Record frankie chap 17 8/20/03 8:04 PM Page 228 frankie chap 17 8/20/03 8:04 PM Page 229

chapter 17

Biodiversity Inventories in Costa Rica and Their Application to Conservation

Paul Hanson

iodiversity inventorying and moni- vertebrates are affected. However, basic taxo- Btoring provide essential information used by nomic research is needed before these poorly many basic scientific disciplines as well as many known groups of organisms can be used in applied sciences such as biotechnology, agricul- monitoring and other conservation activities. ture, fisheries, and conservation. Most invento- The problem is that there are too few taxon- rying and monitoring have involved organisms omists working on the most species-rich groups that are relatively well known taxonomically— of organisms (Gaston and May 1992; Hawks- for example, vertebrates and vascular plants. Yet worth and Ritchie 1993; Hammond 1995). Worse the poorly known groups of organisms, such as yet, taxonomists who have a lifetime’s experience invertebrates and fungi, constitute the majority are retiring, and the knowledge they possess of the species. Conservation decisions based on (much of it unpublished) is not being recovered data for a limited range of organisms having rel- through firsthand transfer to young recruits (Cot- atively few species can be misleading (Prender- terill 1995). In recent years developed countries gast et al. 1993). Moreover, poorly known groups have begun reducing government expenditures, of organisms tend to be smaller in size and to and thus funding of national museums has have shorter generation times, which means diminished markedly in real terms. Increasingly that they respond more rapidly to environmen- such institutions are expected to become more tal changes (Brown 1991). Thus, taxonomically self-sufficient. This means that traditional taxo- difficult organisms are often good indicators of nomic research now has to compete with more environmental change and can help us to real- attractive proposals in evolutionary biology, the ize that a problem exists before the plants and latter often addressing species/population level

229 frankie chap 17 8/20/03 8:04 PM Page 230

questions in taxa that are already well known. Not COSTA RICAN INVENTORIES: surprisingly, job opportunities for taxonomists THE FIELD OPERATION have fallen drastically, and graduate students choose financially more promising fields of As in most countries, the early inventories in research. Costa Rica were quite sporadic (see Gómez and Because the majority of species do not yet Savage 1983 for a brief history). Subsequent have scientific names and the process of de- inventories, primarily by staff of the National scribing new species is generally slow, an attrac- Museum and the University of Costa Rica, have tive alternative is a rapid biodiversity assessment been primarily museum-based and driven by of taxonomically well-known groups (Oliver and the motivation of a few individuals. With the Beattie 1996). When such rapid assessments do creation of the National Biodiversity Institute include poorly known taxa, they must rely on the (INBio) in 1989 there was for the first time a co- separation of morphospecies. Although time is ordinated and institutionalized effort to survey saved by not having to obtain names for the the species present in the country. INBio is a pri- species, if the sorting of morphospecies is done vate, nonprofit organization whose mission is to by a nonspecialist, it is prone to serious errors (for promote a greater consciousness of the value of example, many insects can be separated only by biodiversity in order to conserve it and hence dissecting male genitalia). Although rapid bio- improve the quality of human life. INBio is not diversity assessments are relatively inexpensive, intended to be an institution dedicated to taxon- they are no substitute for the more costly system- omy as an end in itself, and the term museum has atic inventories. Moreover, species names are been carefully shunned. From the beginning it required for determining which sites harbor phy- was felt that neither the time nor the economic logenetically isolated species and for serving a resources were available to train Costa Rican tax- wider community of users interested in environ- onomists in foreign universities (Gámez 1999). mental education, bioprospecting, and the like. Instead, foreign taxonomists have been encour- In recent years a considerable amount of tax- aged to visit INBio in order to help identify spec- onomic inventory work has been carried out in imens collected and curated by INBio staff and Costa Rica in general and in dry forests in par- to train the collectors and curators. The primary ticular. Because the work done in the dry forests emphasis has been on vascular plants and in- has occurred in a national context, and because sects, although fungi, mollusks, and nematodes the problems facing inventories do not differ have been added. fundamentally between terrestrial ecosystems, At the front end of INBio’s inventory are the this chapter treats the broader national process. “parataxonomists”—rural people who grew up It thus differs from other chapters in this book in areas surrounding protected wild lands and by not focusing specifically on dry forests. This are trained to collect and mount specimens. chapter briefly evaluates inventory work in Costa After being mounted at the field stations, spec- Rica by considering what some of the problems imens are brought to INBio, where they are la- have been and how future inventories might be beled, bar-coded, and classified by technicians improved. Three aspects of the inventory pro- and curators (Janzen et al. 1993). How well have cess are discussed here: (1) the field operation of the parataxonomists performed with respect to collecting specimens, (2) the resource base for the biodiversity inventory? On the positive side biodiversity assessment, especially the biologi- they have amassed the largest collection of in- cal collections and human resources available sects in Central America in a very short period in the country, and (3) the information that is de- of time. Because insects were collected prima- rived from the inventory, especially as it applies rily by hand-netting and at lights, the inventory to conservation. has been most successful in those taxa that are

230 HANSON frankie chap 17 8/20/03 8:04 PM Page 231

readily collected by these methods. Many spe- a small fraction of the total time. Parataxonomists cies are represented by long series from various can, of course, be refocused in their collecting localities, thus providing information on geo- methods—for example, to obtain more biologi- graphical distribution and intraspecific variation cal information (which is being done), but this within the country. In general the parataxono- still does not address the shortage of human mists who have been most motivated have been resources for separating bulk samples. the most successful. For example, the team of To decide on a set of methods for a particular parataxonomists in the Guanacaste Conserva- taxon, we need to establish criteria for measur- tion Area have produced more specimen-based ing the performance of different collecting strate- information on host ranges of parasitoids than gies. The number of specimens collected and was previously known from all of tropical Amer- processed, although widely cited in annual re- ica (I. Gauld pers. comm.). ports, is a poor indicator of performance because Although the use of parataxonomists has it reveals nothing about how many species were proved to be a very effective means to jump-start collected or the quality of the processed speci- a collection, in the long run it becomes ever mens. In cases in which a specialist is actively more challenging to employ them efficiently. working with the collection, a much better in- Without sufficient feedback from curators or dicator of performance is the number of new taxonomists, overzealous collecting eventually species collected and how much it cost to obtain yields vast quantities of the common species, a them. problem that INBio is now beginning to address. Meanwhile, the minute insects are generally THE RESOURCE BASE FOR missed by hand-collecting, and the few that are BIODIVERSITY ASSESSMENT encountered are often poorly mounted, since the mounting is done in the field. To be fair, no Most inventories result in a collection of speci- human being could possibly do a first-rate job mens, which usually consist of both identified of collecting and mounting all types of insects; (voucher) specimens and unidentified specimens that is one of the reasons that entomologists of taxa for which no specialists are currently specialize in particular taxa. Recently INBio has available. Voucher collections are essential for trained parataxonomists who specialize in just the verification of field data and to provide a per- one order of insects, but there are still certain manent historical record. Biological reference taxa that are most efficiently collected by the tax- collections in general, with their associated li- onomists themselves. braries and staff, are a vital source of baseline In order to improve the collection of smaller- data for future inventorying and monitoring (Cot- sized taxa, INBio has begun placing more em- terill 1995). phasis on trapping methods. However, there are Biological collections are not mere ware- still not enough laboratory technicians to handle houses for specimens. At the very minimum these mass samples, perhaps because of the they require constant vigilance to prevent the institute’s legacy of being predominantly field- ravages of fungal and arthropod pests, which oriented. For example, a parataxonomist needs can rapidly destroy specimens. However, a col- only about an hour per month to maintain a lection whose staff does nothing more than Malaise trap, but a laboratory technician will re- ensure its physical well-being is effectively a quire a minimum of 20 to 30 hours to separate dead collection. An active and useful collection the focal taxa from this same sample. Thus, when requires a constant input of labor for sending it comes to processing bulk trap samples, INBio and receiving loans of specimens, organizing continues to invest a disproportionate amount (curating) specimens so that individual taxa are of human resources in a task that requires only readily accessible to taxonomists, and similar

BIODIVERSITY INVENTORIES 231 frankie chap 17 8/20/03 8:04 PM Page 232

tasks. Because the continuing cost of maintain- cess to scientific literature and people with ex- ing such collections is frequently viewed with perience in identifying particular taxa. Libraries suspicion by auditors, it is essential to publicize in Costa Rica cannot afford subscriptions to the importance of biological collections as re- most of the necessary scientific journals, which search facilities, every bit as valuable as their is not surprising because even libraries in de- more dazzling counterparts in nuclear physics veloped countries are reducing their subscrip- facilities. tions to these journals. Moreover, despite much In Costa Rica the oldest biological collection recent pontificating, the nuts and bolts of taxo- is that of the National Museum. This collection nomic literature are still unavailable on the In- was mismanaged in the late 1940s and did not ternet and show no sign of becoming available recover until the 1970s. Despite some unex- anytime soon. Thus, budding taxonomists in plained disappearances of type specimens in the Costa Rica, like those in other developing coun- past, the National Museum currently houses tries, have to obtain the necessary literature one of the best vascular plant collections in Cen- through their own efforts and those of foreign tral America. Its insect collection is primarily collaborators. restricted to the butterflies. At the University With respect to human resources, most of of Costa Rica the Herbarium was initiated in the the curators at INBio have received firsthand 1940s and the Zoology Museum and Insect Mu- training under the guidance of foreign taxono- seum in the 1960s. Unlike other large herbaria mists, and some are utilizing this training to in Central America, that of the university in- publish descriptions of new species. On the other cludes a considerable number of nonvascular hand, only a few of the curators have had grad- plants and fungi. The Zoology Museum has uate-level courses in systematics, genetics, or substantial collections of marine invertebrates, evolution, which would greatly facilitate sepa- arachnids, aquatic insects, and especially verte- rating the species, understanding species distri- brates. The Insect Museum has focused on bees, butions, and providing advice on conservation- certain families of beetles, and parasitic wasps. related issues. This deficiency is especially ironic The university’s biological collections would be because courses in these subjects are taught at impressive if their contents were better docu- the University of Costa Rica. mented. The preservation and growth of these biological collections have been assured primarily DATA AND INFORMATION MANAGEMENT by motivated individuals working from shoe- AND COMMUNICATION string budgets, although recently the university has taken a greater interest in them. A general problem among many institutions The collections at INBio have been an insti- worldwide is launching a database without clearly tutional priority since the inception of the in- defining its intended use, thus forgetting that stitute. The combination of a large, well-curated data management is a tool rather than an end collection and enthusiastic staff has acted as a in itself. Although developing one’s own soft- magnet for many foreign taxonomists. Perhaps ware for data management may be superficially for the first time in the history of Costa Rica, at attractive, it is usually difficult for a local insti- least some insect taxonomists are coming not tution do this in a cost-effective way (Olivieri et primarily to collect but rather to work with the al. 1995). INBio has invested heavily in develop- existing collection in INBio. As with the other ing its own software, and it would be interesting collections in the country, the staff of INBio has to analyze whether this effort has been cost- been quite enlightened about loaning specimens effective. Nonetheless, there has been an insti- to the international taxonomic community. tutional commitment to data management (see Besides collections, two other vital resources Zeledón 2000: 100), which allows upgrading for carrying out biodiversity assessment are ac- as new technology arrives. In contrast, data

232 HANSON frankie chap 17 8/20/03 8:04 PM Page 233

management in other collections in Costa Rica species Web pages, which has the potential to has generally relied on individual commitment, disseminate natural history information much which often results in disconnected and poten- more widely than traditional publications. tially incompatible databases. Finally, how is the inventory information be- Very little of the inventory effort in Costa Rica ing used in making decisions about conserving has entailed quantitative sampling, which can biodiversity? In general the staff of the conser- provide valuable data on the relative abundance vation areas has been concerned with how to of different species and what proportion of the protect rather than how to manage. The most biota remains to be collected. The notable excep- frequent requests coming from both the conser- tion is the Arthropods of La Selva project (Lon- vation areas and the private reserves are for lists gino and Colwell 1997). Yet it is difficult to fault of species and field guides, which are useful for the rest of the inventory work for being non- environmental education and ecotourism, but quantitative because most tropical organisms do less so for management and decision making. not even have names and merely addressing this Information on rare and endangered plant spe- essential first step is an enormous task. Differ- cies was used in planning the Osa and “Paso ent taxa require different sampling methods de la Danta” corridors and in planning a nature (New 1998), and for a particular taxon there is trail in Corcovado National Park, but such ex- often uncertainty as to how efficient one method amples are rare (R. García pers. comm.). Con- is in comparison with others (Hammond 1994; servation decisions often rely more on ecologi- Samways 1994). Not surprisingly, we must often cal information (e.g., mammal migrations) than rely on the taxonomists’ intuition. on raw taxonomic information, yet the latter Although further biological data are usually provides the foundation for the former. There- beyond the scope of most inventories, various fore, the sequence of information transfer that field biologists have been carrying out research needs to occur is systematics → ecology → con- in Costa Rica for many years, and this informa- servation, although systematics can occasionally tion is being compiled by the Organization for provide information directly to conservation (e.g., Tropical Studies (BINABITROP, or National Bib- phylogenetically isolated taxa). liography on Tropical Biology, currently comprising 12,000 references). In the future more CONCLUSIONS emphasis needs to be placed on gathering ecological information during the inventory, as is We obviously cannot afford to delay conserva- currently being done with macrolepidopteran tion efforts until all taxa have been fully inven- caterpillars in the Guanacaste Conservation Area toried. The urgency of the current situation re- (http://Janzen.sas.upenn.edu:591/rearingdb.htm) quires us to select areas that need protection and the ants of Costa Rica project (www.ever- on the basis of existing information and rapid green.edu/ants). biodiversity assessments of taxonomically well- The information obtained from inventories known organisms. At the same time, however, it has to be made available for a variety of uses, is prudent to continue the inventories of lesser- notably scientific advancement, environmental known taxa such as microorganisms, fungi, and education, and conservation management. To- invertebrates. These taxa constitute the majority ward these ends the National Museum, the pub- of the world’s species, and many of them are crit- lic universities, and INBio have all been actively ical to ecosystem functioning. Moreover, certain producing a variety of publications (both scien- species-rich areas would probably be overlooked tific and popular), although a perennial problem in conservation planning if we totally ignored in Latin America in general is the distribution of the species-rich taxa. these publications. Both INBio and the Arthro- Inventories in Costa Rica have been very pods of La Selva project are actively producing patchy in their coverage of the country. Inventory

BIODIVERSITY INVENTORIES 233 frankie chap 17 8/20/03 8:04 PM Page 234

work by the universities has been done prima- ciding what type of information is needed but rily outside the national parks, whereas INBio’s rather in providing it in an accessible format, inventory has been restricted to protected areas. disseminating it, and utilizing it. A great deal of Some areas (e.g., Guanacaste and the Osa Penin- the information already exists, some of it readily sula) have received more emphasis than others available but much of it requiring synthesis and (e.g., Braulio Carillo National Park). Very little dissemination. Perhaps the most serious prob- inventory work has been carried out in the pri- lem is the final step—applying the information. vate reserves, which are potentially quite impor- Too often inventory information is viewed as an tant in conservation (Herzog and Vaughan 1998). end in itself, and even when the information A national inventory that includes both protected reaches the decision makers, it is often exiled to and nonprotected areas would provide informa- the file cabinet. tion on the identity and geographical distribu- Based on the biodiversity inventory work that tion of the species occurring within the country has been done in Costa Rica, we can draw on cer- as a whole. This is essential for determining what tain facts to make suggestions for future biolog- proportion of the species are not in protected ar- ical inventories: eas and consequently which nonprotected areas are in greatest need of protection. . Fact: Although collecting is obviously an es- Inventories are also needed after lands have sential first step in the taxonomic process been acquired for conservation. These lands need and parataxonmists have been useful in this to be managed and monitored, but typically only process, the rate-limiting step in inventory vascular plants and vertebrates have been in- work is the description of new species. This ventoried. In fact, Costa Rican law requires only can be done only by taxonomic experts, but that trees and vertebrates be included in envi- they should not have to undertake all the ronmental impact studies. Yet inventory work in tasks themselves as they have traditionally the country has advanced to the point where done. Suggestion: Develop more and varied other taxa could and should be included. The human resources to support monographic laws are especially outdated in the case of fresh- taxonomy. For example, the concept of a water ecosystems, where environmental impact parataxonomist could be broadened to in- studies are required only to measure physical and clude people who assist taxonomists in de- chemical parameters (M. Springer pers. comm.), scribing new species. After the taxonomist which recover from dumping events (e.g., pulp has separated the species and written a pre- from coffee berries) more quickly than do popu- liminary key, these individuals could do lations of aquatic invertebrates. some of the more routine tasks (e.g., record- National biodiversity inventories have several ing locality data from labels and measuring functions, three of which have been highlighted specimens). Quality control should be main- here: (1) building local taxonomic expertise and tained by the taxonomist, and the usual peer biological collections, which are the basis for fu- review processes should be continued. ture scientific studies and conservation efforts, . Fact: There are not enough international tax- (2) providing basic information that can be used onomic resources to support an INBio clone in environmental education, and (3) supplying in the majority of tropical countries. INBio the information needed in developing manage- has benefited greatly from being one of the ment plans for conservation. In general, inven- first such institutions to appear. For example, tories in Costa Rica have succeeded much better British taxpayers have invested approxi- in the first two than in the third. There are many mately a million dollars in the Costa Rican good references treating the types of informa- inventory through salaries paid to staff of the tion needed in conservation (e.g., Noss and Coop- Natural History Museum who have partici- errider 1994), and so the problem is not in de- pated in this inventory (I. Gauld pers. comm.).

234 HANSON frankie chap 17 8/20/03 8:04 PM Page 235

Global taxonomic resources are simply not several important ideas for this chapter. I also large enough to duplicate such support in thank the following persons for providing use- most other tropical countries. Suggestion: ful comments on previous versions of the chap- Promote greater regional interinstitutional ter: Zaidett Barrantes, Wills Flowers, Gordon collaboration and increase employment op- Frankie, Randall Garcia, Jorge Gómez Laurito, portunities for taxonomists. It is worth re- Luis Diego Gómez, Barry Hammel, Jack Lon- membering that no country will ever be gino, Julian Monge, and Rodrigo Zeledón. self-sufficient with respect to taxonomic expertise. Even the United States and Canada had to import taxonomic expertise in order REFERENCES to produce keys to the genera of Chalcidoidea Brown, K. S. 1991. Conservation of neotropical en- of North America (Gibson et al. 1997). vironments: Insects as indicators. In The con- . Fact: Globally, taxonomy is primarily funded servation of insects and their habitats, ed. N. M. through science research budgets and is sub- Collins and J. A. Thomas, 350–404. London: Academic Press. ject to changes in science priority. The user Colwell, R. K., and J. A. Coddington. 1994. Esti- community wants taxonomists to produce mating terrestrial biodiversity through extrap- the tools by which organisms can be iden- olation. Philosophical Transactions of the Royal tified, but the research required to produce Society of London B 345:101–18. such tools is not currently viewed by the re- Cotterill, F. P. D. 1995. Systematics, biological search community as a priority undertaking. knowledge and environmental conservation. Biodiversity and Conservation 4:183–205. Suggestion: Potential user communities Gámez, R. 1999. De biodiversidad, gentes y utopías: should identify specific taxonomic needs Reflexiones en los 10 años del INBio. Santo (e.g., an identification manual for the Domingo de Heredia: Instituto Nacional de Neotropical genera of Chalcidoidea), and Biodiversidad. 144 pp. funding agencies should invite tenders to Gaston, K. J., and R. M. May. 1992. The taxonomy of taxonomists. Nature 356:281–82. produce these products rather than invest in Gibson, G. A. P., J. T. Huber, and J. B. Woolley, eds. more nebulous “capacity building.” Funding 1997. Annotated keys to the genera of Nearctic opportunities for specific products would Chalcidoidea (Hymenoptera). Ottawa: National stimulate organizations to collaborate in for- Research Council Research Press. 794 pp. mulating competitive bids, which in turn Gómez, L. D., and J. M. Savage. 1983. Searchers would create jobs for young taxonomists, on that rich coast: Costa Rican field biology, 1400–1980. In Costa Rican natural history, ed. particularly in developing countries with D. H. Janzen, 1–11. Chicago: University of Chi- comparatively low overheads. cago Press. Costa Rica has received international acco- Hammond, P. M. 1994. Practical approaches to the estimation of the extent of biodiversity in spe- lades for its efforts in conservation and inven- ciose groups. Philosophical Transactions of the tory. How well it lives up to this reputation in the Royal Society London B 345:119–36. future depends on a continued commitment to ———. 1995. The current magnitude of biodiver- these interrelated activities. The often repeated sity. In Global biodiversity assessment, ed. V. H. phrase “learning occurs through doing” is ap- Heywood and R. T. Watson, 113–38. Cambridge: propriate, and for other tropical countries that Cambridge University Press. Hawksworth, D. L., and J. M. Ritchie, eds. 1993. have barely begun doing it, the Costa Rican ex- Biodiversity and biosystematic priorities: Micro- perience can provide several take-home lessons. organisms and invertebrates. Wallingford, England: CAB International. ACKNOWLEDGMENTS Herzog, P., and C. Vaughan. 1998. Conserving biological diversity in the tropics: The role of I especially thank Ian Gauld for numerous private nature reserves in Costa Rica. Revista de thought-provoking discussions that provided Biología Tropical 46:183–90.

BIODIVERSITY INVENTORIES 235 frankie chap 17 8/20/03 8:04 PM Page 236

Janzen, D. H., W. Hallwachs, J. Jiménez, and Olivieri, S. T., J. Harrison, and J. R. Busby. 1995. R. Gámez. 1993. The role of the parataxono- Data and information management and com- mists, inventory managers, and taxonomists munication. In Global biodiversity assessment, in Costa Rica’s national biodiversity inventory. ed. V. H. Heywood and R. T. Watson, 607–70. In Biodiversity prospecting: Using genetic resources Cambridge: Cambridge University Press. for sustainable development, ed. W. V. Reid et al., Prendergast, J. R., R. M. Quinn, J. H. Lawton, B. C. 223–54. Baltimore: World Resources Institute. Eversham, and D. W. Gibbons. 1993. Rare spe- Longino, J. T., and R. K. Colwell. 1997. Biodiversity cies, the coincidence of diversity hotspots and assessment using structured inventory: Cap- conservation stategies. Nature 365:335–37. turing the ant fauna of a tropical rain forest. Rossman, A. Y., R. E. Tulloss, T. E. O’Dell, and R. G. Ecological Applications 7:1263–77. Thorn. 1998. Protocols for an all taxa biodiversity New, T. R. 1998. Invertebrate surveys for conservation. inventory of fungi in a Costa Rican conservation Oxford: Oxford University Press. 240 pp. area. Boone, N.C.: Parkway Publishers. 195 pp. Noss, R. F., and A. Y. Cooperrider. 1994. Saving na- Samways, M. J. 1994. Insect conservation biology. ture’s legacy: Protecting and restoring biodiversity. London: Chapman and Hall. 358 pp. Washington, D.C.: Island Press. 416 pp. Zeledón, R. 2000. 10 años del INBio: De una utopia Oliver, I., and A. J. Beattie. 1996. Designing a cost- a una realidad. Santo Domingo de Heredia: In- effective invertebrate survey: A test of methods stituto Nacional de Biodiversidad. 144 pp. for rapid assessment of biodiversity. Ecological Applications 6:594–607.

236 HANSON frankie chap 18 8/20/03 8:05 PM Page 237

chapter 18

Conﬂict Resolution

RECOGNIZING AND MANAGING DISCORD IN RESOURCE PROTECTION

Gregory A. Giusti

Editors’ note: This chapter represents the only viduals participating in such a discussion will contribution from an author who has not worked undoubtedly bring with them biased points of directly in Costa Rica. The topic presented here per- view to suit their particular needs and may have tains to collaboration, which is of great importance little or no regard for opposing perspectives. In in implementing any conservation project. As edi- order for a resource professional to assist in tors, we wanted to include a model for collaboration resolving resource management conflicts he or that could be used universally in cases in which there she must (1) recognize that such myopic view- was sufficient societal infrastructure. Greg Giusti’s points exist and (2) develop a process that enables model fits our need perfectly. It is based on years of a wide-ranging discussion allowing inclusion of work as a practicing cooperative extension specialist multiple viewpoints (Lee 1991). Schindler and and collaboration moderator for the University of Cheek (1999) expand on Lee’s approach by iden- California. tifying six criteria important to the success of citizen-agency interactions. They suggest that here can be no greater accomplishment the process is most effective when (1) it is open Tfor a resource professional than to bring and inclusive; (2) it is built on skilled leadership together a divergent group of individuals and in- and interactive forums; (3) it includes innovative terests to forge a comprehensive plan for bio- and flexible methods; (4) involvement is early logical conservation. Inherent in this proclama- and continuous; (5) efforts result in action; and tion is the recognition that multiple variables of (6) the process seeks to build trust among par- economic, environmental, and social prejudices ticipants. Simply stated, resource professionals will play a role in the dynamics of any group must recognize and accept that natural resource brought together to develop such a plan. Indi- conflict resolution is an exercise in social tutelage,

237 frankie chap 18 8/20/03 8:05 PM Page 238

directing and motivating participants in a pro- consider, and discuss complex biological con- cess of progressive instruction with the ultimate cepts and ecological relationships (Giusti et al. goal of actions that lead to resource conservation. 1991). Even then, conflict between divergent This chapter provides guidance for those who viewpoints may remain poignant and perhaps have been trained in the art and science of biol- bitter. ogy (or some other scientific discipline) but have The realization that broad public participa- never been taught how to deliver their discipli- tion and community involvement are needed nary expertise to a generally naive citizenry. This in conservation planning is becoming more ac- new style of resource management (Wondolleck cepted even by institutions that have historically and Yaffee 2000) has important implications been reluctant to involve external participation for both resource sustainability and professional in planning exercises, such as the United States credibility (see also chapter 20). Forest Service. It is important to remember that many of these institutions were not necessarily designed to consider community issues or co- PEOPLE AND NATURE: operative landowner programs when develop- MAKING THE CONNECTION ing regional conservation plans. To help address Most practicing biologists are aware that the some of the institutional limitations to commu- primary cause for the loss of biological diversity nity resource conflict resolution, the 1993 report throughout the world is the alteration of habi- of the President’s Council on Environmental tats through a variety of human-induced land-use Quality (PCEQ) recommended four broad goals practices (Soulé 1995; USGS 2001). Unfortu- aimed at directing participants toward conserva- nately, in today’s tumultuous world, few people tion strategies: (1) maintain the viability of native have the luxury to contemplate abstract and plants and animals; (2) encourage restoration complex theories of biological diversity, ecosys- of viable plant and animal populations; (3) com- tem integrity, or the relationship between Homo plement regional and global biodiversity con- sapiens and their natural environment. Yet there servation efforts; and (4) educate employees, are those who have repeatedly warned of the con- community leaders, and the public about bio- sequences of ignoring the protection of the nat- diversity conservation. ural systems and the potential impact on human Clearly, this report recognizes the impor- societies (Erlich 1986). Fortunately, if people (and tance of providing information to people in an communities) are provided with the informa- inclusive manner if they are expected to make tion, and if they make the effort, they can begin decisions that will ultimately protect biological to understand the basic tenets of evolution, eco- diversity. This approach inherently recognizes logical structures, and biological relationships— that both individuals and communities are inte- ecological principles that are the result of mil- gral components to any conservation strategy. lions of years of environmental factors. In order However, the report falls short in helping re- to direct efforts aimed at developing programs source managers understand the mechanisms that promote sustainable environments, people necessary to teach involved individuals how to must be provided with an opportunity to under- accomplish these goals. In simple terms, it is stand how these environments function and relatively easy to tell resource managers to edu- recognize how humanity plays an integral part cate employees, community leaders, and the pub- in altering the natural workings of ecological lic about biodiversity conservation. It is much systems. A necessary and important step in es- more difficult to deliver effectively a compre- tablishing a process of helping people assimilate hensive program aimed at people who may feel the connection between environmental degra- threatened by the whole process. dation and human effects is creating a forum A major obstacle to the conservation of bio- that provides the time people need to receive, logical diversity in today’s world is that envi-

238 GIUSTI frankie chap 18 8/20/03 8:05 PM Page 239

ronmental protection is often at odds with many flicts are caused and energized by the rapid deeply entrenched social practices, a situation population growth of California” (p. 112). He fur- that frequently results in protracted conflict, such ther lists the ultimate factors in this dispute: as that concerning the protection of salmonids “1)... population growth and settlement pat- and timber harvest practices in the Pacific North- terns; 2) absence of sustainable ecosystem plan- west. In many instances, these entrenched prac- ning; 3) new technologies; 4) transformation tices limit people’s ability to visualize and accept of rural economies; 5) popularity of outdoor new and innovative approaches necessary to recreational activities and, 6) point and diffuse address economic viability while providing a sources of anthropogenic pollution” (p. 112). In greater degree of environmental protection. An his discussion he uses these factors to defend example of this type of ingrained narrow view is his position that “explosive population growth a scenario in which someone who is confronted and land development have been accompanied with an ecological problem provides an economic by a plethora of environmental problems from answer. Such a response reveals that the person air pollution to hazardous wastes and—most of may not fully understand the ecological impli- all—large scale landscape transformation and cations of the problem being presented. It is degradation that have reduced some ecosystem then important for the resource professional to types such as wetlands and riparian systems to determine whether the person failed to, or chose less than 10% of their spatial expanse in less than not to, understand the problem being presented 100 years” (p. 112). and is merely reacting to the situation out of fear. The legacy of these past land-use practices, in In order to manage resource conflicts effectively combination with continuing practices that fur- and ensure an atmosphere for productive com- ther degrade existing habitats, is responsible for munication, all participants in a discussion must creating a political and social environment that be given the opportunity to comprehend fully the is often highly charged and emotional as both complexities of the argument. In the absence individuals and communities clash over rem- of this type of clear, concise communication, the nants of these highly modified habitats. dialogue most likely will not advance. The economic and social costs associated with the litigious actions of divergent interests are enormous and often result in conflicts that LAND USE AND go beyond the warring individuals, ultimately ENVIRONMENTAL CONFLICT spreading and encompassing entire commu- Several authors have cited land-use practices as nities. In many cases, though contentious, the having significant effects on both habitats and competing interests may both be publicly stat- dependent species across broad geographical re- ing that they share similar goals, such as pro- gions, including the Pacific northwest of North viding a sustainable economic base for the com- America (Franklin 1988; Moyle 1994) and the munity, promoting tourism, or promoting the tropical dry forests of the Western Hemisphere, natural beauty of the area. However, though Australia, Southeast Asia, and Africa (Janzen competing interests may use similar words to 1988). Increasingly, local and regional land-use describe their views, they often may have very practices and their effects on biological diversity different viewpoints and meanings. For ex- are becoming more prevalent in the scientific ample, to a timber company the statement “pro- literature (Brown et al. 1994; Noss and Cooper- viding a sustainable economic base for a com- rider 1994; Walter 1998). munity” could mean maintaining a sustainable Walter (1998) provides an overview of the timber supply, but to a tourism-based entity it factors he has identified that are driving the could mean maintaining a sustainable fishing land-use conflicts now widespread throughout industry. The two are not necessarily compatible California: “Most of the current land use con- activities and may reflect conflicting goals and

MANAGING DISCORD IN RESOURCE PROTECTION 239 frankie chap 18 8/20/03 8:05 PM Page 240

most likely differing options on land-use prac- 2. No individual party can impose its will on tices. It is becoming more apparent that resource the other. Participants must not be “forced” professionals must be capable of dissecting the to accept any decision. meaning behind the statements of competing 3. Representatives of stakeholder groups must interests if scientific-based approaches are to be in a position to commit themselves and be offered as a solution and ultimately provide a their constituencies to support any decision resolution to an existing conflict. that is agreed to by the group. 4. All parties must share a sense of urgency to ADDRESSING SOCIAL AND resolve the conflict. CULTURAL VALUES

Resource professionals, then, need to provide Although this list represents a simplified view information in an ongoing argument; merely of the complex nature of the conflict resolution providing information, however, usually will not process, it does guide the resource professional be sufficient to overcome the underlying value on how to proceed. Ensuring adequate process systems that may be the basis for the conflict is not an easy task, particularly when the dis- (Keller 1994). Often, these values are based on cussion may be between individuals with strong social, cultural, and economic systems (or per- emotional attachments to a subject or topic. sonal needs) having little to do with the biological criteria being addressed. Unfortunately, these anthropocentric assertions can serve as major DEVELOPING A PROCESS FOR impediments to conservation programs, dis- RESOURCE CONFLICT RESOLUTION tracting and directing the discussion. Further It is not uncommon for a resource professional complicating the matter is that most resource to be called on as the resident expert to assist professionals are not properly trained to facili- in developing conservation plans (Chess et al. tate what are often highly charged, emotional 1990; Magill 1991; Geisler et al. 1994). How- arguments and debates (Wondolleck et al. 1994). ever, conveying that expertise articulately to a One approach to managing social and cultural wide spectrum of lay people is a formidable task, difference in resource conflict resolution is rec- and it becomes more difficult in an antagonistic ognizing the importance of providing a process setting. Resource professionals directing a pro- in which these inherent disparities can be ac- gressive discussion in an attempt to develop a commodated. Paying attention to the need for conservation strategy should be aware of the providing a structured process is vital to the suc- following: cess of any effort aimed at developing a conservation strategy when working with divergent . Interest groups involved in the development interests. Failure to address adequately the im- of a conservation plan will be vying and lob- portance of process can lead to participants feel- bying for language that suits their particular ing that they have no ownership in the decision, need(s). thereby limiting their support of the outcome. . It is important for all parties to understand Wondolleck et al. (1994) provide some basic steps that when developing a plan that may even- that should be considered when developing an tually become policy, the group may not be inclusive process aimed at conservation: the final authority in accepting and imple- 1. Participants must view one another as menting the conservation strategy. equals. This does not imply that all parties . A well-articulated, science-based conserva- must always agree, rather, it recognizes the tion plan may be further refined through a need for all parties to work together in political process prior to becoming a conser- order to address the problem successfully. vation policy.

240 GIUSTI frankie chap 18 8/20/03 8:05 PM Page 241

FIGURE 18.1. An illustrative model to help guide conservation discussions.

I have developed a conceptual model (Giusti perience of the Quincy Library Group (QLG) 1994) to help guide resource professionals when (1996), which had to address the many nuances developing a process to assist various interest of identifying authority within a divergent group groups trying to develop a conservation plan or of people charged with a regional planning pro- strategy (fig. 18.1). cess. This group charted a five-year management plan on federal lands in the northern Sierra Nevada of California. In this case, the ultimate IDENTIFYING AUTHORITY: authority was the U.S. Congress, which would WHO’S IN CHARGE? have the jurisdictional powers to implement the A number of factors must be identified before a plan. However, at the local level, the discussion resolution can be attained through an inclusive needed to include three distinct national forests process. The first question involves identifying encompassing eight counties. If the group was the authority to which the plan will be directed. to develop and advance a successful plan, it was Addressing this question early in the process not enough for the QLG to view all three forests will ensure that all parties understand who will as one entity under the title of “Forest Service” have the final authority in implementing any or each county as a single jurisdictional author- strategy or plan. This step is crucial even if the ity as “local government.” In this case, for the process is intended to develop voluntary strate- QLG to develop an inclusive process, it had to gies (i.e., cost-share programs) requiring a will- recognize the autonomous authority of the indi- ingness on the part of the landowner to work vidual forests and counties. with the jurisdictional authority that will ulti- In most cases, identifying authority will mately be administering the program. not be as complex as the QLG example. Often, Identifying the proper authority can some- conservation-planning efforts are conducted at times be complicated, as exemplified by the ex- the local level involving a single jurisdiction, such

MANAGING DISCORD IN RESOURCE PROTECTION 241 frankie chap 18 8/20/03 8:05 PM Page 242

as the local board of supervisors or commis- managers are not thoroughly discussed. Exam- sioners, resource conservation districts, regional ples of such conflicts are (1) road construction boards, or commissions. In larger, regional-scale engineering standards that may not be sensitive planning efforts the responsible authority may to providing upstream migration corridors for be a resource agency at the state or federal level. migratory fish, (2) even-age forest management Regardless of what the authority may be, it is and the inherent limitations of providing suit- fundamentally important to identify and include able habitats to sustain viable populations of late the responsible jurisdiction early in the discus- seral dependent species, or (3) agricultural pest sion in order to provide an opportunity for it to management practices that may disrupt habitat assume some ownership in the process and, ul- quality for nontarget species. In many cases, the timately, the final outcome. Often, this can re- lack of attention to these conflicts has resulted sult in an invaluable allegiance and institutional in negative effects from modern technological support once a final proposal is submitted for advances on biological diversity (Walter 1998). consideration. It is imperative that resource professionals take the time to evaluate and then articulate how the various land-use management disciplines ADDRESSING THE SCOPE OF A PROJECT can have counterproductive results at the land- Most “-ologists” are trained to articulate the in- scape or ecosystem levels. It is not uncommon herent attributes of spatial analysis at the genetic, to have fierce disagreements between resource species, community, landscape, or ecosystem professionals who may feel adversarial toward levels when engaging in ecological deliberations. each other (e.g., biologists and foresters; engi- Although a number of academic questions deal- neers and ecologists); in these cases, personal ing with genetic integrity are of the highest emotions can impede fruitful discussions. importance in a global sense, most field biologists are dealing with conservation issues at the GOALS AND OBJECTIVES species level and beyond. Increasingly, conservation strategies are focusing on protecting bio- Undoubtedly, most people will agree on the logical communities at the landscape and eco- importance of establishing goals and objectives system levels as a way to ensure population in order to conserve natural resources. The chal- viability. The reality of this approach is that the lenge is to motivate a divergent group to agree discussion often involves multiple ownerships on a common set of goals and objectives that can and jurisdictions, making for complex group be addressed in realistic time frames and budg- discussions. It is imperative that people involved ets. Reaching such an agreement first requires in the discussion are provided with the neces- taking the time to explore the group’s needs sary information to understand the scope of the and desires; one way to begin this discussion is effects that are being proposed. Do not assume a brainstorming session to get a sense of the that everyone knows what you know. Biologists group’s interests. The next step should be to try are generally a professional minority in modern to capture some of the points or concepts that society. Few people will have had similar train- seem to have broad acceptance. Next, an attempt ing and experiences from which to make their should be made to begin the process of develop- decisions (chapters 20 and 24). ing a draft goal(s) statement. Points to consider when engaged in an exercise to develop a goal statement should include: MANAGEMENT OPTIONS (1) Goals should be realistic yet comprehensive. AND ASSESSMENTS (2) Goals are often described in lofty terms to mo- All too often, inherent conflicts that exist be- tivate people in a particular direction. (3) Goals tween the scientific disciplines used by land-use should help people visualize what the group is

242 GIUSTI frankie chap 18 8/20/03 8:05 PM Page 243

trying to accomplish. (4) Synonyms for goal are protection. Once such a forum is convened, a aim, intent, purpose, and end. These terms should fundamental issue to be addressed is the pro- be considered when developing a stated set of cess by which the group will decide how best to goals. choose its actions or recommendations. Examples of a goal statement: (1) The goal Paramount to the success of any decision- of the (name of the organization) is to provide a making process is the selection of the method sustainable timber-based economy while ensur- for deciding on actions—that is, by consensus or ing the protection of other, nontimber resources. majority rule. There are inherent pitfalls in ei- (2) Our goal is to develop a comprehensive ap- ther procedure. Obviously, the risk associated proach to the protection of aquatic habitats while with majority rule is that any decision will have recognizing human needs associated with those winners and losers. The risk associated with the habitats. consensus-driven process is the recognition that Although different than a vision statement, a every individual involved in the discussion is af- goal statement should help a reader understand forded veto power over the group. Facilitating the purpose of the conservation effort under way. the group process when participants are con- The purpose of establishing a goal statement is tentious should sometimes be left to someone to facilitate the development of objectives as a who has formal training in facilitation skills. It means of achieving the goal. Objectives should may fall to the resource professional to recog- be viewed as a set of actions necessary to achieve nize when it would be timely to involve an ex- the goal. Unlike goals, objectives (1) are described ternal facilitator. in simple, straightforward terms; (2) should The crux of most natural resources conflicts help people understand the tasks necessary to is the anxiety and mistrust that exist between achieve the goal; (3) can be described using short- competing interests. It is these elements that term, moderate-term, and long-term qualifying make it necessary to improve communications language; and (4) should be articulated in a way between stakeholders as a means of addressing that allows their success to be evaluated fully. the conflict. It is important to try to achieve some Examples of short-, moderate-, and long- level of balance when attempting to develop an term objectives: (1) This year, our organiza- inclusive process involving competing interests. tion will convene an educational workshop on I have provided a basic list of potential stake- managing habitat for endangered species (short- holders (Giusti 1994) that should be considered term objective). (2) In the next five years, our or- when trying to identify competing interests ganization will secure financial support neces- (fig. 18.1). Although not complete, the list is sary to implement riparian habitat restoration meant to serve as a guide to identifying special activities along five miles of stream (moderate interest groups that may want to be actively in- term). (3) In ten years, our organization will volved in developing a solution to a particular have a fully funded conservation center, staffed environmental problem. by one paid administrator and a cadre of docents who will maintain a continuous educational and RECOGNIZING THE TOOLS AVAILABLE restorative program aimed at riparian ecology TO ACHIEVE POSITIVE CHANGE (long term). The science of understanding and facilitat- What motivates people to action when it comes ing group process takes many years to master to conserving natural resources? The answer to (Jacobson 1999). Although it is not the respon- this question is highly variable and transitory. sibility of resource professionals to mediate all Initially, motivation can be caused by either the conflicts, it is incumbent on them to recognize real or perceived threat of financial loss to an in- when mediation can create a forum that will re- dividual (or individuals) or a community, per- sult in a positive progression toward resource ceived effects from land-use practices (erosion

MANAGING DISCORD IN RESOURCE PROTECTION 243 frankie chap 18 8/20/03 8:05 PM Page 244

from agricultural operations) that affect a per- Education programs should be used as a fo- son’s or group’s sense of stability, or any land- rum to increase awareness and understanding use practice or policy that creates economic, of natural resource management complexity. The environmental, or social uncertainty. purpose of an educational approach should be to Often the first response in a resource conflict provide objective, science-based information in is for one group or individual to lay blame on order to allow all participants to understand the the perceived perpetrator of the problem. The re- scope and complexity of the issues. sultant claims and counterclaims can often limit In order to educate, resource professionals the ability of the divergent views to focus on the must be willing to remain objective in their de- problem. Wondolleck and Yaffee (2000) stress livery of information. Objectivity should not be the need to get opposing viewpoints to focus on confused with neutrality. Neutrality implies that finding successful solutions to the problem but a professional does not care what happens to not be distracted by prolonged arguments that the resource. Objectivity suggests that we take focus on personalities or past events. the time to provide the information necessary for Once everyone’s views have been heard, an audience to understand the positive and neg- the next step is to guide participants toward re- ative consequences of a management decision. alizing what choices they have to address the One should consider the scope and the level problem. Generally, there are three broad cate- of complexity of the issue that is being addressed gories (menus) from which to choose when try- and develop the educational program according to identify solutions. These menus form the ingly. In some instances, it may be necessary to basis from which all decisions are ultimately develop a conservation strategy over an extended derived. series of educational meetings in order to allow participants to gather, digest, and understand the MENU 1: INCENTIVES issue. Incentives are those programs or actions that Education programs should not be consid- promote a positive response. Often referred to ered the end product of any conservation scheme. as the “carrot,” they are used to entice a positive They should be viewed as a vehicle by which action. Incentives can take on many forms. Their people are provided with the information neces- principal purpose is to stimulate positive motiva- sary to begin building the scheme. Different tion to achieve a desired outcome. Most incentive types of education forums include workshops, programs provide one or more of the following field trips, publications and mass media, and benefits: (1) financial gain, (2) improved self- working groups (chapter 20). esteem or position in the community, (3) public recognition, (4) financial support. Some common MENU 3: REGULATIONS forms of incentives include cost-share programs, Regulations are often considered the “stick” com- cost-deferred tax/zoning laws, public recognition ponent of the “carrot and stick” approach to programs, and public purchase programs. motivational mechanisms. Although necessary to curb human behaviors, regulations alone have MENU 2: EDUCATION not been proved to be fully successful in re- All too often special interest groups misuse the source protection. concept of education in order to promote their One recurring flaw in many regulatory ap- particular point of view. We commonly hear proaches is the lack of monitoring necessary to groups state the need “to educate the public” as evaluate effectiveness. This problem is most a means of promoting understanding. Usually, evident in many forest-related environments this approach simply implies that a campaign where the regulatory focus is often at the stand is needed to persuade a particular political con- level and may not be effective in evaluating the stituency to support an advocacy position. cumulative effects of stand management across

244 GIUSTI frankie chap 18 8/20/03 8:05 PM Page 245

a watershed or landscapes. Even stringent, pre- product of a political and social process and not scriptive regulations often fail to address ade- necessarily a scientific one. That said, it is in the quately the cumulative effects of past and present best interest of those resources that these pro- activities across the landscape, not to mention fessionals work in an environment that recog- the effects of these activities on large geograph- nizes the political and social realities of our ical areas, such as ecosystems. world and that they engage in open and straight- Another shortcoming of relying purely on a forward dialogues aimed at directing people in a regulatory approach to resource protection is that positive move toward resource protection. regulations are the product of a political process, not a scientific process. The inherent unpre- REFERENCES dictability in this approach often results in a Brown, L. R., P. B. Moyle, and R. M. Yoshiyama. compromised position that may not always be in 1994. Historical decline and current status of the best interests of the resource and may favor coho salmon in California. North American one special interest group over another. Journal of Fisheries Management 14:237–61. Regulatory approaches can take many forms Chess, C., B. Hance, and P. Sandman. 1990. Im- and can involve punitive actions against a viola- proving dialogue with communities: A short guide for government risk communication. New Bruns- tor. Some of the most common types associated wick, N.J.: New Jersey Agricultural Experiment with land use include zoning requirements, Station. 30 pp. prescriptive land-use regulations, environmen- Erlich, P. R. 1986. The machinery of nature: The liv- tal review programs, and performance stan- ing world around us and how it works. New York: dards. Although often necessary to address the Simon and Schuster. 320 pp. Franklin, J. F. 1988. Structural and functional di- most egregious situations or to establish a min- versity in temperate forests. In Biodiversity, ed. imal threshold of expected performance, regu- E. O. Wilson, 166–75. Washington, D.C.: Na- lations should be viewed in the context of a tional Academy Press. much larger, comprehensive approach to re- Geisler, M., P. Glover, E. Zieroth, and G. Payton. source management. 1994. Citizen participation in natural resource management. In Expanding horizons of forest ecosystem management: Proceedings of the Third PROFESSIONAL CHALLENGE IN THE FACE Habitat Futures Workshop, Oct. 1992, ed. M. Huff OF MOUNTING RESOURCE CONFLICT et al., 87–100. USDA General Technical Report PNW-GTR-336. Portland, Ore.: U.S. Department Resource scientists, educators, and managers of Agriculture, Forest Service, Pacific Northwest are faced with tremendous social challenges in Station, 100 pp. Giusti, G. A. 1994. Partnerships across owner- today’s fast-paced world. They are charged with ships. In Foresters together: Meeting tomorrow’s protecting the world’s natural wealth in the face challenges, 42–52. Indianapolis, Ind.: Society of of ever increasing demands placed on these re- American Foresters. sources by the very people who entrust them with Giusti, G. A., K. R. Churches, and R. H. Schmidt. this responsibility. Just as these professionals 1991. Sustainability: A challenge through public view the various biological components of the education and outreach programs. In Proceed- ings of the Symposia on Oak Woodlands and systems in which they work as intricate and com- Hardwood Rangeland Management; Oct. 31– plex, they must recognize the intricate and Nov, 2, 1990; Davis, Calif., R. Standiford, tech- complex human element in their managerial nical coordinator, 246–249. General Technical approaches. One of those approaches must Report, PSW 126. Berkeley, Calif.: PSW Re- consider providing a forum for open and honest search Station, USDA-FS. Jacobson, S. K.1999. Communication skills for con- dialogue that affords all interested groups access servation professionals. Washington, D.C.: Island to sound, science-based information. Resource Press. 351 pp. professionals must further recognize that many Janzen, D. H. 1988. Tropical dry forests: The decisions that affect natural resources are the most endangered major tropical ecosystem. In

MANAGING DISCORD IN RESOURCE PROTECTION 245 frankie chap 18 8/20/03 8:05 PM Page 246

Biodiversity, ed. E. O. Wilson, 130–37. Washing- propositional analysis. Conservation Ecology ton, D.C.: National Academy Press. 3(1):9. [On-line] URL: http://www.consecol.org/ Keller, S. R. 1994. A sociological perspective: Valu- vol3/iss1/art9 ation, socioeconomic and organizational fac- Soulé, M. E. 1995. Biodiversity indicators in Cali- tors. In Endangered species recovery: Finding the fornia: Taking nature’s temperature. California lessons, improving the process, eds. T. W. Clark, Agriculture 49(6):40–44. R. P. Reading, and A. L. Clark, 371–90. Wash- United States Geological Survey (USGS): National ington, D.C.: Island Press. Biological Information Infrastructure. 2001. Lee, R. L. 1991. Four myths of interface commu- http://www.nbii.gov/issues/biodiversity/ nities. Journal of Forestry 89(6):35–38. Walter, H. S. 1998. Land use conflicts in California. Magill, A. W. 1991. Barriers to effective public in- Ecological Studies 136:107–26. teraction. Journal of Forestry 89 (10):16–18. Wondolleck, J. M., and S. L. Yaffee. 2000. Making Moyle, P. B. 1994. The decline of anadromous fishes collaboration work: Lessons from innovation in in California. Conservation Biology 8:869–70. natural resource management. Washington, D.C.: Noss, R. F., and A. Cooperrider. 1994. Saving na- Island Press. 277 pp. ture’s legacy: Protecting and restoring biodiversity. Wondolleck, J. M., S. L. Yaffee, and J. E. Crowfoot. Washington, D.C.: Defenders of Wildlife and 1994. A conflict management perspective: Island Press. Applying the principles of alternate dispute res- Quincy Library Group [on-line]. 1996. http:// olution. In Endangered species recovery: Finding www.r5.pswfs.gov/robriefings/briefingssept96/ the lessons, improving the process, ed. Clark et al., 9609qlg.html 305–26. Washington, D.C.: Island Press. Schindler, B., and K. Aldred Cheek. 1999. Inte- grating citizens in adaptive management: A

246 GIUSTI frankie chap 19 8/20/03 8:06 PM Page 247

chapter 19

Conservation and Environmental Education in Rural Northwestern Costa Rica

LEARNING THE LESSONS OF A NONGOVERNMENTAL ORGANIZATION

Gordon W. Frankie and S. Bradleigh Vinson

t all started with fire! In the late 1970s damaging fires. Moreover, all wildfires were Imembers of our dry-forest research team human-caused, and the human population in were aware that wildfires were becoming com- the region was increasing. Fire was commonly mon in our general study area between Cañas used to “clean” fields and roadsides by local and Liberia and southwesterly to the Tempisque people, and “cleaning” fires often escaped and River in Guanacaste Province, Costa Rica (see burned their way into forests, including protected maps 1.1 and 1.2 in chapter 1). We also had rea- areas. Indifference to escaped fires by most local son to believe that the risk of fire would increase people and authorities was the rule. through time and endanger the remaining plants We were also aware that these fires could and wildlife throughout the area. Reasons for probably be managed with traditional, time- our concern included the following. Fire was not proven methods of prevention and suppres- a natural phenomenon in the dry forest, and the sion. This view was based on fire experience by biota was not adapted to cope with it. Thus, the one of us (GF), who had previously worked on flora and fauna were extremely vulnerable to fire fire crews for the U.S. Forest Service in several damage (Frankie et al. 1997) (Our research group pine forests of northern California. had worked in the Tempisque region since 1968, To deal with these fire concerns, a group of when fires were infrequent and small). Wildfires five biologists/naturalists (G. Frankie, J. Frankie, received their greatest combustible fuel from a L. D. Gomez, W. Haber, and S. B. Vinson) and fire-adapted exotic African grass, Hyparrhenia the National Parks Foundation of Costa Rica rufa, and repeated fires in an area usually led began working together in late 1979 to develop to substantially increased fuel loads of H. rufa, a proposal to protect a seasonal dry-forest thereby increasing the risk of future and more site, Lomas Barbudal, having unique biological

247 frankie chap 19 8/20/03 8:06 PM Page 248

characteristics, which was being threatened by DEFINITION OF EE these fires (see map 1.3 in chapter 1). After six years of work, the 2,400-ha site was decreed a In this chapter we use the following definition biological reserve of the National Park System of EE, which evolved over the period of years on 23 January 1986. At that time we had come to from 1969 to 1993 (EPA 1993: 1): “Environmen- believe that our work was finished and that the tal education is . . . the interdisciplinary process National Park Service would assume full respon- of developing a citizenry that is knowledgeable sibility for the new addition to the park system. about the total environment—including both its Such was not the case. In fact, local park service natural and built aspects—and that has the ca- officials informed all who had worked on the pacity and the commitment to engage in inquiry, project that there was neither “interest nor re- problem-solving, decision-making, and action sources” to care for Lomas Barbudal. that will assure environmental quality.” This defi- Based on this surprising response, three of nition reflects the kind of broad knowledge that the reserve founders (GF, JF, and SV) consulted we had as professional biologists; the desire to with the National Parks Foundation (private) outreach to a citizenry that consisted of several and quickly reached an agreement to assume different audiences; the fact that humans were responsibility for managing Lomas Barbudal for considered an integral part of the dry-forest en- an indefinite period (not an uncommon practice vironment; and the fact that some of our pro- at that time for individuals or small groups to grams were designed for proactive involvement. manage protected government land). Thus, by 1 February 1986 we began our role as land stew- OUTREACH PROGRAMS ards, biodiversity conservationists, promoters, and fund-raisers for the Lomas Barbudal Bio- Our center of operation was the town of Bagaces logical Reserve of the National Park System. We and vicinity in Guanacaste Province, where we also quickly became environmental educators on lived and worked (and still do). Bagaces is an behalf of the reserve. old town that has been in existence for more In this chapter we focus on environmental than 300 years. It has approximately 4,000 in- education (EE) associated with our work at the habitants and an economic base in agriculture, reserve and vicinity, which lasted from 1986 mostly rice and cattle currently. Town leaders are through 1997. During this 12-year period we de- protective of excessive development and modern- veloped a nongovernmental organization (NGO) ization, and thus the town has remained small and a wide variety of outreach programs for sev- and somewhat provincial in character. With re- eral different audiences. The goals of this chap- gard to wildlands, the location of the town, which ter are to is adjacent to the Pan American Highway, serves as a “gateway” to several nearby protected areas 1. describe the main outreach programs and of Guanacaste. The most relevant one for our intended audiences; work was Palo Verde National Park, of which the 2. describe major lessons learned from inter- Lomas Barbudal Reserve is a part (see map 1.3 in actions with different audiences and from chapter 1). environmental educators; At the beginning of our EE work in 1986 we 3. use selected historical events of our organi- established a small conservation organization, zation and guiding principles of EE from a Amigos de Lomas Barbudal, in Bagaces. Shortly 1977 UNESCO conference to evaluate the thereafter we legally registered the Friends of education program we developed; and Lomas Barbudal in California as a nonprofit con- 4. provide recommendations for future EE servation organization for the purpose of raising programs in rural communities. tax-deductible funds for projects of Amigos de

248 FRANKIE AND VINSON frankie chap 19 8/20/03 8:06 PM Page 249

Lomas Barbudal in Costa Rica. Finally, in the emerging in the Bagaces community to help early 1990s we registered Amigos de Lomas Bar- with fire suppression. These groups currently budal in Costa Rica as a nonprofit NGO. With work with the park service as volunteer fire- this action we also established a small board of fighters and have become an important part of directors of 12 local professionals. the fire suppression program (Loaiza 2000).

FIRE PROGRAM HIGH SCHOOL PROGRAM Our EE work began in February 1986 as one of During Easter week in 1988 we enlisted the vol- us (GF) trained a group of ten local residents unteer help of 54 Bagaces High School students in selected modern fire-fighting techniques that to work on a variety of projects at Lomas Barbu- could be applied locally. Working closely with dal for eight consecutive days. Results of this work this group produced positive results over a four- were so positive that we decided to develop a year period from 1986 to 1989. During the first year-round EE program with the high schoolers. start-up year, Lomas Barbudal had almost 50 per- We began by hiring a teacher (with bachelor’s cent of its 2,400-ha area burned (light to mod- degree in biology) from San José to coordinate erate burns). As resources were increased from the program. She, in turn, helped to organize a private donors and as everyone became more high school club with a focus on assisting Lomas familiar with the landscape, access routes, sea- Barbudal. sonal wind patterns, high fire risk areas, and An important lesson learned about this au- neighbors from 1987 through 1989, the percent- dience of students was their high energy and age of the burned area decreased progressively need to do hands-on work. As we had done dur- each year to a low of 15 percent in 1989. As an ing Easter 1988, we engaged them in field proj- NGO, we removed ourselves as land stewards ects that ranged from trail building and sign of the reserve in early 1990, but we continued painting to patrolling the reserve and fire fight- with EE work at the reserve and in Bagaces. The ing in and around Lomas Barbudal. Students park service had by then agreed to assume stew- were also instrumental in raising funds, plant- ardship responsibilities after our departure. ing trees at the reserve, and building benches The successful training and development of and developing a picnic area for visitors at Lo- local firefighters at Lomas Barbudal led to a se- mas Barbudal. Further, they helped to develop ries of one- to three-day workshops on regional interpretive displays for the reserve visitor cen- fire management from 1989 to the early 1990s. ter by collecting a variety of natural forest mate- The workshops, which were organized by Ami- rials such as fruits and seeds (see later in this gos de Lomas Barbudal, involved larger audi- chapter). Several students also did oil paintings ences that included many local residents and of animals on large waste disposal cans at the park service personnel. Professional firefighters reserve and in Bagaces. The program for this (volunteers) from the U.S. Forest Service in Cal- audience lasted for two years and was phased ifornia, Idaho, and Oregon regularly assisted in out when most students graduated from high these workshops, and they were well received. school. Toward the program’s end our focus be- During the early years, our organization rec- gan to shift toward another audience—primary ognized the need for education and develop- school children. ment of preventative measures as part of the overall fire management program. Thus, an OUTREACH TO OUTLYING SCHOOLS increasing amount of our time was invested in In the early 1990s we established an EE out- educational activities and projects that promoted reach program in 23 primary schools through- fire prevention and alternatives to burning. At out the Bagaces vicinity—some of which were the same time local volunteer organizations were in remote villages accessible only by a four-wheel-

CONSERVATION AND ENVIRONMENTAL EDUCATION 249 frankie chap 19 8/20/03 8:06 PM Page 250

drive vehicle. With the permission of teachers, Many other functions and activities were car- our education coordinator visited half of the ried out at the center. For example, once local schools every month (the other half were visited residents, teachers, and students from surround- less frequently) and presented two-hour talks and ing schools (including the University of Costa activities associated with some topic on conser- Rica in Liberia) learned about the library and vation or the environment (or both). This educa- other offerings, it became a source of new in- tional visit was always well received, as indicated formation for the community. It was also policy by the children’s responses and the teachers’ to secure requested new information if it could regular invitations to return. Once the program be located. At times, the requests challenged us was fully operative, we calculated that slightly to make site visits to gather new field data di- more than 1,000 children were receiving our rectly. Finally, the center occasionally served as a presentations. We also gave weekly talks to a meeting place for adult groups wanting to dis- group of local children at our Center for the cuss local environmental problems and issues Conservation of Nature in Bagaces (see the next such as the Mula Corridor conservation cam- section). paign (see the section “Evaluating the EE Pro- gram” later in this chapter). CENTER FOR THE CONSERVATION OF NATURE IN BAGACES VISITOR CENTER AT LOMAS BARBUDAL From 1988 to 1990 we conducted meetings and In 1989, we constructed a relatively large visitor gave presentations to a wide variety of audiences and interpretive conservation center at Lomas in several people’s homes in Bagaces. Realizing Barbudal’s northwestern boundary. The center, the need for better-quality space, we rented a rel- which is actually on the property of the Institute atively large house in Bagaces in the early 1990s of Agrarian Development (IDA, a large govern- that became our official headquarters. The house, mental agency), was designed for several func- or “Centro para la Conservación de la Natu- tions, two of which were to make ecological and raleza,” served many EE needs that were previ- environmental information available to reserve ously impossible to accommodate. For example, visitors and to provide employment for local res- the center had space for a full-time secretary, idents in the directly adjacent community of a library of almost 900 books and documents, San Ramon. Other functions included provid- shelves for nature displays, space for a large ing space for EE projects ranging from fire work- freshwater fish tank, and ample storage space shops to natural history classes for local schools. for equipment and supplies. It also had room for One unexpected function was to create a sense work tables and regular meetings with the chil- of new civic order and structure for the small dren of Bagaces. San Ramon community and IDA. Volunteer do- Once the Bagaces center was established, a nations were solicited from visitors who used formal program of weekly presentations was the center and nearby trail, picnic area, and river developed for Bagaces children of ages 9–12. for recreation. We employed local residents to Topics for presentations were highly variable operate the center on a daily basis during the dry and often pertained to relevant conservation and season and part of the wet season. We also em- environmental issues. Presenters ranged from ployed local people to maintain the nearby trail our immediate staff of professional biologists, to and picnic area. Volunteers from Bagaces also local experts, to professionals from San José. In helped in this work. addition, the center was open every weekday, all In 1994, we donated the center and its re- day, for children in our program to check out sources to a small women’s association of the library books and to work on special nature proj- community of San Ramon. Despite offers by our ects, including field trips, using the organiza- organization to assist it, the association attempted tion’s resources. to operate the center independently, which did

250 FRANKIE AND VINSON frankie chap 19 8/20/03 8:06 PM Page 251

not work very well. Part of the problem could be son 1999). This often requires extra time to be- traced to lack of transportation, electricity, and come familiar with the audience and to develop telecommunications in the community and the a sense of good will and trust. It may also require fact that many women in the association were taking a few calculated risks such as experiment- illiterate. Another part could be related to ex- ing with new approaches. Take audiences into pectations of collecting sizeable donations from the field to observe conservation/environmental visitors (which has never happened). Eventually, issues directly. Be aware of the wide variety of the association began accepting a small amount teaching tools for communicating information of assistance from students at the University (Ham 1992) and environmental teaching phi- of Costa Rica and the park service. As of this losophies such as Project Wild and Project writing, however, the future of the center is un- Learning Tree (PW 2000; PLT 1994). certain. One fact is clear from this comparative experience. Small, community-based ecotourism AVOID POTENTIAL PROBLEMS projects need regular outside assistance in order Despite our constant vigilance and awareness to maintain even a modest visitor center. of potential problem areas, there were always surprises and new challenges. One of our big- LESSONS LEARNED gest challenges was trying to keep current on frequent replacements of local park service di- We learned numerous lessons during our 12 rectors and their ever changing agendas. These years of work. These lessons fall into four main directors almost always promoted monopoliza- categories, which are described in the following tion of any matter that pertained to conservation sections. or the environment, whether or not they had the ABOUT THE NGO: expertise to do so. A second problem area was lo- “FRIENDS OF LOMAS BARBUDAL” cal gossip. Most was benign and harmless; often it was humorous. On occasion, however, it was It is necessary to legally and physically establish clearly divisive, mean spirited, and designed to and maintain an organization that is highly vis- attack the credibility and integrity of individuals ible, interactive with the community, and highly or, more commonly, organizations such as our professional in conducting its activities and busi- NGO. The easiest way for an individual or an ness affairs. It is critical to develop short (yearly) organization to cause harm was to claim that and long-term goals (five to ten years) and even- there was mismanagement of funds and per- tually to expect institutionalized status (and its sonal pocketing of money. It made no difference implications) for the organization. whether an organization such as ours had a KNOW LOCAL CULTURE certified public accountant who kept records on It is important to establish and maintain a vari- all incoming and outgoing money. The gossip ety of working and friendly relationships with came regardless. Fortunately, most of it could cooperators and local leaders. It is equally im- be dispelled if the source was known and could portant to be aware of agendas of groups involved be vigorously pursued. directly or indirectly with conservation. Get in- The small size of the “Friends” organization volved in community activities related to the worked against us in the long term as logical environment, but be selective and avoid over- extensions of ongoing work required more re- extension of time and resources (see chapter 18). sources and support. Despite an active and suc- Learn patience. cessful program in the early and mid-1990s, our organization lacked the basic ingredient OUTREACH AND KNOWING YOUR AUDIENCE of long-term support from a major consistent It is critical to know about the needs, expecta- donor(s) or an endowed fund. Eventually, the lack tions, and limitations of each audience (Jacob- of sufficient funds and depletion of personal

CONSERVATION AND ENVIRONMENTAL EDUCATION 251 frankie chap 19 8/20/03 8:06 PM Page 252

energy led to the closure of our EE program in among this audience. Further, we always had the 1997. volunteer support of local mothers for field ex- cursions. From 1990 to 1996, when our center was fully operative, we estimated that about 350 EVALUATING THE EE PROGRAM children took advantage of our program. Group During our 12 years of EE work in Bagaces and size at weekly presentations at the center was vicinity, we had ample opportunity to make 15–35 children. The length of time that a young- assessments as biologists and teachers as to ster spent in our program varied greatly, from whether our work was locally accepted and ef- about 2 months to 2 years. fective in this rural community. We also had In the early 1990s a local civic association in an understanding among the founders that we Bagaces solicited help from our organization to would develop and pursue EE projects only if stop a planned clear-cut in a forest corridor, La we deemed them successful and if we continued Mula, between Palo Verde National Park and Lo- to receive positive feedback from local residents. mas Barbudal. The corridor was considered to be We present here a qualitative evaluation of the the last effective biological connection between EE program in the sense of Jacobson (1999). See these two protected areas. The IDA had proposed also Jacobson and McDuff (1998). cutting the Mula on the basis of needed devel- Our best initial indication of acceptance was opment for local people (note: timber sales of realized through help and encouragement that the cut were estimated to be around U.S. $2 mil- we received in 1986 from the mayor of Bagaces, lion). The challenge was to quickly develop a who volunteered municipality assistance for us document demonstrating why this action was to begin our work at Lomas Barbudal and vicin- not in the best long-term interests of the region ity. Help from the mayor came primarily in the and country. Our organization gathered together form of building materials for construction of 35 biologists and over a three-month period pro- the visitor center at Lomas Barbudal. He also as- duced a document of 126 pages that spoke sisted by arranging for introductions to leaders strongly against the cut (Frankie et al. 1994). in Bagaces. During Easter week in 1988, we re- We were also aided by environmental journalists, ceived volunteer assistance from the Bagaces especially Gilda Aburto, and by leaders at that High School students through a collaboration time from the Institute of Biodiversity (INBio) with the mayor and school principal. Fifty-four in Heredia. These efforts, which are described students helped with a variety of field projects in more detail by Aburto (chapter 20), were suc- during this holiday period. cessful in canceling the government deforesta- From 1989 to 1994 we designed and offered tion plan. The Mula Corridor was subsequently several workshops on fire management, which protected through its annexation to Palo Verde were collaborative efforts between our profes- National Park. sional volunteers, park service personnel, and In addition to our positive historical experi- local townspeople. They were always well at- ences as an NGO, we assessed our program in tended, and the feedback was positive. Eventually relation to 12 guiding principles of EE, devel- some of our local trainees formed or joined local oped and agreed on at a UNESCO conference volunteer fire crews. Two of these individuals in 1977 at Tbilisi, Georgia (in the former USSR) were recently singled out as fire crew leaders in (UNESCO 1978, 1980), which follow. We were Bagaces (Loaiza 2000). able to relate our overall programs in positive Once the center for EE was established in ways to all principles except number 2. In this Bagaces, primary school children were readily case, the lifelong process was beyond our scope. attracted to the variety of programs offered. There was never a need to solicit the town’s children; 1. Consider the environment in its totality— word of our educational offerings spread rapidly natural and built, technological and social

252 FRANKIE AND VINSON frankie chap 19 8/20/03 8:06 PM Page 253

(economic, political, cultural-historical, DISCUSSION moral, aesthetic); 2. Be a continuous lifelong process, begin- To the best of our knowledge, NGOs dedicated ning at the preschool level and continuing specifically to the conservation of biodiversity through all formal and non-formal stages; began to emerge in Costa Rica around the mid- 1980s. Our organization also began its first 3. Be interdisciplinary in its approach, draw- outreach efforts at that time. Each of these or- ing on the specific content of each disci- ganizations has remained characteristically small pline in making possible a holistic and and localized. A very few of the organizations, balanced perspective; such as the National Museum in San José, de- 4. Examine major environmental issues veloped a mobile EE program that outreached from local, national, regional, and inter- to primary schools throughout Costa Rica. One national points of view so that students small conservation NGO in San José, FECON, receive insights into environmental developed a consortium of some 20 small NGOs, conditions in other geographic areas; which regularly sent representatives to make site 5. Focus on current and potential environ- visits to conservation projects or environmental mental situations, while taking into problem areas. FECON, in turn, communicated account the historical perspective; its findings and interpretations to its member 6. Promote the value and necessity of local, organizations and the media. FECON also rep- national, and international co-operation in resented conservation interests at the national the prevention and solution of environ- level through the Legislative Assembly. Despite mental problems; its small size, FECON has been one of the most prominent NGO voices for biodiversity conser- 7. Explicitly consider environmental aspects vation in the country. in plans for development and growth; In our Bagaces work, we used a variety of 8. Enable learners to have a role in planning approaches and teaching tools in our various their learning experiences and provide an outreach activities/projects and conservation opportunity for making decisions and campaigns (see examples in Ham 1992; Jacob- accepting their consequences; son 1999). Most of these worked well, especially 9. Relate environmental sensitivity, knowl- with regard to primary school children and high edge, problem-solving skills, and values schoolers. We regularly used the following topic clarification to every age, but with special avenues to reach these two audiences: familiar emphasis on environmental sensitivity to wildlife, common plant species, clean water, the learner’s own community in early healthy watersheds, art, music, public health, and years; recreation. We had a more cautious and guarded response from adult groups. We assessed audi- 10. Help learners discover the symptoms and ences and their possible responses, as far as real causes of environmental problems; possible, prior to any of the implemented activ- 11. Emphasize the complexity of environmen- ities. These early assessments greatly aided the tal problems and thus the need to develop execution of our programs (Jacobson 1999). critical thinking and problem-solving During the course of 12 years of EE work in skills; the canton of Bagaces, we developed numerous 12. Utilize diverse learning environments and contacts with a wide variety of individuals. Some a broad array of educational approaches contacts were made through direct cooperation to teaching/learning about and from the and collaboration as partners in projects. In many environment, with due stress on practical cases, neighbors (including cattle ranchers and activities and first-hand experience. rice growers) and other local residents, who took

CONSERVATION AND ENVIRONMENTAL EDUCATION 253 frankie chap 19 8/20/03 8:06 PM Page 254

the time to learn about our activities, served as and work structures. This antifire campaign spokespersons on our behalf. Our board of was in turn promoted through billboard adver- directors (12 individuals) provided support and tisements and pamphlets by several large gov- council, and others, such as environmental jour- ernment agencies in the country. Volunteer fire nalists and lawyers, also helped. It was clear that crews of private citizens have become more as a small NGO we regularly needed and de- commonplace in the Guanacaste region (Loaiza pended on cooperators, neighbors, friends, and 2000). All these developments in fire awareness other professionals to conduct our various proj- and management have occurred since the late ects (see also chapter 18). 1980s, and EE programs implemented by a wide There was only one small local group with variety of individuals have greatly contributed to whom we continually had problems. This group the success of the antiwildfire campaign. Rais- consisted of middle-management park service ing awareness, however, is only the first part of officials. At first we all seemed to be speaking the the education process. Continued vigilance to “same conservation and education language.” reinforce awareness is also necessary. After about five years of trying to cooperate, however, it became clear that cooperation and RECOMMENDATIONS partnerships were usually not possible. The difficulties were not with the projects, regular park Many recommendations can be offered based on service personnel, or most of the higher-ranking our 12 years of experience. Some of these rec- park officials in the capital city, San José. Rather, ommendations are implied in the “Lessons it came down to a strong sense of territory and Learned” section (earlier in this chapter), espe- monopolization on the part of middle-manage- cially with regard to establishing and maintain- ment individuals. ing a conservation NGO. It is important to Based on 20 continual years of observations mention again the need to develop a financially on wildfires in the canton of Bagaces, there are solvent NGO. High energy, numerous opportu- at least three important conclusions one can nities, and soaring expectations for transferring reach, and all of these are related to EE. First, meaningful environmental information must be forest wildfires are a relatively new phenome- matched with corresponding support funds. non in this region, and all are human-caused. Is EE needed to conserve biodiversity in Costa Damaging effects of fire increase in a forest area Rica? Based on our work and the experiences of if little or no responsible prevention and sup- colleagues in other parts of the country, EE is a pression are undertaken (Frankie et al. 1994, necessity if the system of natural areas is to be 1997; G. Frankie pers. obs.). Second, once forest conserved and protected for the future. There is habitats are destroyed and the exotic H. rufa both ignorance to dispel and the need for new grass invades, the risk of fire increases dramati- information by technically competent profes- cally for adjacent forestlands. Third, many fires sionals. EE should also be focused on protected can be prevented or suppressed using time- natural areas where rural people live, and they proven, traditional fire management techniques. should be the main target audience, as they are With regard to prevention, EE is one important likely to have the greatest long-term impact on component. Our fire workshops demonstrated biodiversity through their traditional use of the ease and interest by which local residents natural resources. Rural people also have great could become knowledgeable about fire ecology potential to become allies and extenders of en- and trained in traditional fire suppression tech- vironmental information. The most appropriate niques. Eventually, journalists (newspaper and type of EE program for rural communities is one television) also helped greatly to spread the word that has at least a ten-year life span to ensure about the destruction that fires were causing to enough time to become institutionalized and to forests, agricultural lands, and even private living observe and work with a generation of new citi-

254 FRANKIE AND VINSON frankie chap 19 8/20/03 8:06 PM Page 255

zens from ages 9 to 18. In contrast, there is also Well-trained and experienced educators the need for short-term interpretive information should be encouraged to conduct workshops to for visitors to the protected areas. train local teachers to provide at least some EE EE is best done by well-educated and trained to their pupils. In our experience, the effective- professionals. The natural environment is a com- ness of these new extenders will be quite vari- plex issue to explain in an appropriate manner. able, but there will always be some teachers who Adding people to the environment makes it even will embrace environmental information, and more complex. We recommend that a master’s- these individuals should be targeted for more level person be the target type of individual for training and resources. leading an EE program. This person should have There is a need for more professional bi- a degree and some research experience in biol- ologists to become involved in the process of ogy, ecology, and/or natural resource manage- transferring their biodiversity and conservation ment. The person also needs training in rural knowledge to audiences other than their own sociology, economics, and policy. A course in colleagues. They have much information to leadership training is also recommended, as share, and oftentimes they are in the very best environmental educators must be leaders. position to influence policy and decision makers Qualified environmental educators should (see chapters 18 and 20). This has not been a work in conservation NGOs and also within the common role for professional biologists in the Ministry of Natural Resources, and both must be past, but times are changing globally. Applying better supported by the government. The situa- and extending knowledge are becoming attraction to date in Costa Rica is that most EE comes tive to some biologists who have a sense of from the private sector and a few institutions urgency to protect the environment and who such as the National Museum in San José and want to play a proactive role in conserving bio- INBio in Heredia. Very little government invest- diversity (see O’Brien 1993; Meffe 1998; Brown ment in EE has been made within the ministry 2000; Kaiser 2000). (G. Frankie pers. obs.). The government should To conclude, EE is in its infancy as it relates also hire more educated and better-trained field to conserving and protecting biodiversity in ru- directors in the park service. At the present time, ral areas. There is much to be done, and a large most field park service personnel have little or part of the effort will be private. However, the no college experience (see chapters 5 and 21). government needs to play a more active role in EE needs to extend beyond the usual audi- supporting the general concept and programs ences mentioned here. It also needs to reach of EE, be they private or governmental. Govern- large and small rural landowners, as there is still ment leaders must make substantial financial much biodiversity in rural areas (see chapter 17), investments in EE! and this will require new creative efforts by educators. In our biological work on private ranches ACKNOWLEDGMENTS and farms, it is clear that many landowners have a strong love for their land and many of the wild We thank the following individuals for their sup- organisms that live there. They also think about port during our 12 years of EE work in the can- their children who will one day inherit the land ton of Bagaces: Gilda Aburto, Ramon Aguilar, and its wildlife. This recognition provides fertile Max Aragón, David Boshier, Mario Boza, Julio ground for creative environmental educators. In Bustos, Jorge Camacho, Magda Campos, Bob this regard, conservation easements, which are Carlson, Alfredo and Paulina Chacón, Dale used to buy future development rights, could be- Cuyler, Marvin Duarte, Jim Duzak, Maria Flexer, come more commonly used in rural Costa Rica, Jack and Maribell Fraser, Luis Diego Gómez, as they are in several places in the United States George Hagnauer, Verena Hagnauer, Lynn Hart- (Gustanski and Squires 2000). shorn, Tony Hartshorn, Stephen Hopkins, Tony

CONSERVATION AND ENVIRONMENTAL EDUCATION 255 frankie chap 19 8/20/03 8:06 PM Page 256

Leigh, Linda Newstrom-Lloyd, Lilliam Lopez, Frankie, G. W., S. B. Vinson, M. A. Rizzardi, T. L. Alfonso Mata, Emily Miggins, Beatriz Mosquiera, Griswold, S. O’Keefe, and R. R. Snelling. 1997. Pedro Ordoñez, Marianella Pastor, Susan Perry, Diversity and abundance of bees visiting a mass flowering tree species in disturbed seasonal Clara Padilla, Peter Ronchi, Susana Salas, Hilda dry forest, Costa Rica. Journal of the Kansas En- Sandoval, Valeria Solano, the Stewart family tomological Society 70:281–96. of Comelco, Jennifer Weber, Lewis Wilson, and Gustanski, J. A., and R. H. Squires, eds. 2000. Brian Wissenden. Don Dahlsten, Stephen Hop- Protecting the land: Conservation easements past, kins, Genesis Humphrey, Alfonso Mata, and present, and future. Covelo, Calif.: Island Press. 566 pp. Mary Schindler kindly read a first draft of this Ham, S. H. 1992. Environmental interpretation: A chapter. practical guide for people with big ideas and small We thank the following institutions and or- budgets. Golden, Colo.: North American Press. ganizations for their support: Consultoria Agro- 456 pp. economica, Cañas, Guanacaste; World Wildlife Jacobson, S. K. 1999. Communication skills for Fund, Washington, D.C.; Municipality of Ba- conservation professionals. Covelo, Calif.: Island Press. 351 pp. gaces; California Agricultural Experiment Sta- Jacobson, S. K., and M. D. McDuff. 1998. Conser- tion, Berkeley; Texas Agricultural Experiment vation education. In Conservation science and Station, College Station; Partners of the Ameri- action, ed. W. J. Sutherland, 237–55. Oxford, cas, Oregon chapter; Finca La Pacifica, Cañas; England: Blackwell Science. U.S. Forest Service; Fundación de Parques Na- Kaiser, J. 2000. Ecologists on a mission to save the world. Science 287:1188–92. cionales, San José; Museo National, San José; Loaiza, V. 2000. Ceden los incendios forestales. Rincon Corobici, Cañas; Finca Las Pumas, La Nación (national Costa Rican newspaper), Cañas; Cederena, San José; FECON, San José; 27 February. Dutch Embassy, San José; Canadian Embassy, Meffe, G. K. 1998. Conservation biology: Into the San José; British Embassy, San José; Del Monte millennium. In Conservation Biology 12:1–3. Corp., San José; BICSA bank, San José; and O’Brien, M. H. 1993. The professional biologist: Being a scientist means taking sides. BioScience Cheeseman’s Safaris, California. 43:706–8. PLT (Project Learning Tree). 1994. Project Learn- ing Tree: Environmental education activity guide. REFERENCES Washington, D.C.: American Forestry Founda- Brown, K. S. 2000. A new breed of scientist- tion. 402 pp. advocates emerges. Science 287:1192–95. PW (Project Wild). 2000. Project Wild: Activity guide. EPA. 1993. First report of the National Advisory Coun- Gaithersburg, Md.: Project Wild. 526 pp. cil on Environmental Education. Washington, UNESCO. 1978. Twelve guiding principles to define D.C.: U.S. Environmental Protection Agency. and guide the development of environmental edu- Frankie, G. W., V. Solano, and M. McCoy. 1994. cation. Final report, intergovernmental confer- Una evaluación técnica del Corredor La Mula ence on environmental education, organized entre La Reseva Biológica Lomas Barbudal y El and published by UNESCO in cooperation with Parque Nacional Palo Verde. Bagaces: Amigos UNEP, Tbilisi (Georgia), USSR, 14–26 October de Lomas Barbudal; Berkeley: University of 1977. California; and Heredia, Costa Rica: Universi- ———. 1980. Environmental education in the light dad Nacional de Heredia. 126 pp. Contact of the 1977 Tbilisi Conference. Paris: UNESCO. [email protected] for copies. 25 pp.

256 FRANKIE AND VINSON frankie chap 20 8/20/03 8:07 PM Page 257

chapter 20

The Media and Biodiversity Conservation

Gilda Aburto

or hundreds of years, humans have growth that demanded more and more resources. Fbeen in constant battle over natural resources, In his attempt to protect these wildlands, Muir some for their exploitation and others for their also found that there was a tremendous differ- protection and preservation for the future. One ence between delivering a seminar to a few of the ﬁrst people in North America—if not the people and publishing an article in a journal. ﬁrst—to raise their voice in favor of defending Published articles could transmit the impressive wildlife territories was the legendary John Muir beauty of these wildlife areas, the forests that he (Tolan 1990), known as the “Father of the Na- called the “temples of God,” to a much wider tional Parks.” For years he traveled through the audience. It was his passionate articles, published impressive valleys and mountains of the United in newspapers and magazines, that transported States, the forests and glaciers of Alaska, identi- the urban populations to these wild areas, where fying species and marveling at the natural beauty they would admire them so much that they were of these wildlands. He also observed how these moved to protect them. territories were being threatened by the growing Biologist Rachel Carson strengthened the en- population that was rapidly expanding westward, vironmentalist movement with the publication tearing down forests to establish settlements and of her book Silent Spring (1962), in which she raise livestock. warned of the dangers of agricultural chemicals. A passionate lover of nature, Muir found that She was especially concerned with the “miracu- it was not enough to be a passive scholar of na- lous insecticide” DDT, which was gravely con- ture, classifying species of plants and trees that taminating the environment, killing wildlife, and were rapidly being lost to the logging industry, poisoning humans. Once again it was the use of sheep ranchers, and the accelerated population a major media resource that alerted the public

257 frankie chap 20 8/20/03 8:07 PM Page 258

and raised the voices of protest; until then peo- if not impossible, to visit the numerous govern- ple had been unaware of the dangers to which ment offices, lawmakers, politicians, citizen as- they were being exposed. sociations, and international entities required to The struggle continues today. Whereas yes- effect policy, initiate campaigns, or raise funds terday pen and paper were sufficient to battle for environmentally oriented projects. Such a the ax, today chainsaws and tractors have given dedicated biologist would have to forgo much of an advantage to the destroyers. Entire forests his or her research to pursue this path of action fall overnight—before we can even begin to call because large amounts of time and economic attention to problems (Soulé 1986; Noss and resources would have to be invested. Moreover, Cooperrider 1994). Enormous stretches of trop- positive results would rarely be obtained. This ical rain forest disappear before we can discover path is too slow for our quickly changing world. what animal and plant species they held. It is It often happens that, by the time the goal is continuously more difficult for scientists to reached, the threats to the environment on which obtain funding for investigations into diverse the campaign was based are already resolved. natural habitats, yet large transnational corpora- Biologists who have traveled this road know tions easily acquire logging and mining conces- that it is unpaved, steep, and rough, like the sions, with outrageous economic advantages, forests they traverse. However, the often closed in the impoverished countries of Latin America. doors of the decision makers open miraculously Biologists alone cannot stop this destruction. as a result of an effective media campaign. Expe- They need the general public to donate money rience has repeatedly shown that journalists can necessary to carry out conservation projects, to be very effective and, above all, swift when com- write letters to politicians, and to join the voices municating important information presented to of protest against the imminent danger to na- them by biologists. Publications in the press can ture. Thus, although the problems involve bio- be very useful to the scientist requesting funds logical themes, it is the people who not only from an international organization, since jour- often create these problems but who also hold nal clippings can be affixed to proposals, giving the solution in their hands. Only through com- them more weight. More important, this infor- munication can biologists transmit their much- mation, widely disseminated by the media, has needed knowledge to the public, empowering it the potential to recruit thousands of willing and to act. dedicated citizens who can successfully pres- Scientists have an ally that can help them sure the authorities in ways an isolated biologist reach those who will make crucial decisions for simply cannot do (Jacobson 1999). the environment. That ally is the media. By Contact with the media is an absolutely cru- means of radio and television, information can cial step because, although there are many in- literally circle the globe and reach millions in a formed citizens who are concerned about the very short time. damages being done, few have the detailed infor- This chapter explores a set of guidelines for mation needed to act on these concerns. Many establishing effective communication between more people know nothing at all about the envi- biologists and the media for raising environ- ronment, the damage and destruction it faces, or mental awareness of the general populace. The what consequences this damage might have on guidelines are based on cumulative experiences their own well-being. These people are so im- of journalists working with biologists in Costa mersed in their fast-paced urban world, hurry- Rica and elsewhere in Central America. ing to work, hypnotized with sports, concerned Help from the public in creating an effective with problems of everyday life, that they seldom conservation campaign is absolutely essential, stop and think about the natural resources that as campaigns initiated by an isolated biologist provide the water that runs their in faucets, toi- will have slim chances of success. It is difficult, lets, and showers. They rarely notice when trucks

258 ABURTO frankie chap 20 8/20/03 8:07 PM Page 259

stacked with huge logs drive by and thus do not information accessible to a diverse audience (Ja- even think about where they come from or if cobson 1999). their extraction will affect their lives. Whatever path they choose to reach the me- Likewise, journalists often do not realize dia, scientists must establish and maintain their what goes on in the field of biologists. They need credibility. A scientist that has studied a specific scientists to inform them and to provide the nec- habitat for years has the knowledge and under- essary information to draft a news report. Pa- standing to pinpoint the immediate and poten- tience, above all, is needed to educate journalists tial dangers to this ecosystem. He or she can so that they can see the need to include the en- therefore effectively illustrate and defend his or vironment in their media agenda. Even if biol- her arguments (O’Brien 1993). ogists are not, at first, fully understood or if However, writing for the media is not the published information does not contain all the same as writing a paper for a scientific journal. data desired, contact with the media is an essen- One tool required to reach the public ear lies in tial first step in informing the public (Jukofsky “humanizing the story” as much as possible. The and Wille 1993). media, as well as the public, will definitely be Unfortunately, many investigators often act more interested if an ecological problem affects like the animals they study: shy, fleeing, and un- not only wild species but humans, as well. willing to approach people. They fear that they Scientists must also be careful not to write will not be understood or perhaps will be criti- lengthy horror stories or send extensive and com- cized by their peers and prefer to remain hidden plicated bulletins to the media. Many biologists in the depths of the forest. Initiating press cam- often wait until the last minute, when disaster is paigns also threatens to take up valuable time imminent, to report their findings to the public. from their research. Yet biologists must alert the Although such reports are sometimes necessary, population to the damage being done to nature people cannot be continually bombarded with or risk losing the same forests in which they disheartening news; otherwise, they will feel work. desperate and helpless to act meaningfully. To One route for the biologist who does not wish motivate people to act, biologists should make to speak directly to the media is to use a politi- it a point to update the public periodically about cian as a go-between. Politicians love the press, their work so that people can begin to feel a con- benefit from being on television and newspapers nection with the biologist’s area of study. More- as often as possible, and rarely miss an oppor- over, these updates must include positive news, tunity to make an appearance if there is a good such as the discovery of new plant and animal reason to do so. Not only do they enjoy and species, the development of effective conser- benefit from press attention, but politicians of- vation projects, and the detection of a species ten have the charisma and power to reach the once thought to be extinct. Finally, these reports media. If scientists can gain their attention, they should be short, clearly and concisely written, can become valuable allies in environmental and objective, abandoning overly technical lan- media campaigns. guage. To learn and appreciate the biologist’s Another avenue for “shy” scientists, if their message, the public must first be able to read it organizations have the budget, is to hire a jour- (Jacobson 1999). nalist or a public relations representative to be in charge of establishing contact with the me- LA MULA CREEK dia and organizing press conferences. These representatives also draft “translated” press In 1993, the Costa Rican Institute of Agrarian bulletins—reports written in plain, simple, and Development (IDA) provoked a commotion entertaining language free of hard-to-under- about a forest known as La Mula Creek in the stand scientific terminology—making scientific Tempisque region of Guanacaste Province (see

THE MEDIA AND BIODIVERSITY CONSERVATION 259 frankie chap 20 8/20/03 8:07 PM Page 260

map 1.3 in chapter 1). Located between the Lomas ogy, traveled through the forest for several days de Barbudal Biological Reserve and Palo Verde and nights, observing and collecting data on di- National Park, La Mula Creek is largely a pochote urnal and nocturnal species. Their results were (Bombacopsis quinata) forest whose timber, valued surprising in that the species discovered were at $2 million, covers about 350 ha. Intending to far more numerous than what was expected from donate the lots to local farmers who presumably the area. Hundreds of species, ranging from owned no land, the IDA planned to clear-cut La bats, small rodents, and marsupials to felines Mula Creek and divide it into 33 parcels. The in- (including a jaguar), monkeys, and rabbits, were stitute would then turn it over to the individual found in the forest. In addition, 144 species of farmers, who were to plant rice on the “cleared” birds, both native and migratory, and 101 tree land. species were identified in the small forest of La The town of Bagatzí, which was adjacent to Mula—a considerable diversity for such a small the pochote forest, protested this decision. It area (Frankie et al. 1994). argued that the forest protected the townspeople After all the information was collected, it was from strong summer winds, that nearby rice presented in a report that included aerial and plantations would poison their water resources, ground-level photographs showing the existing and that the fumigation planes would spray them vegetation density. The report was sent to all with toxic chemicals. Local biologists and con- authorities who had an influence on the deci- servationists joined the residents of Bagatzí and sions concerning the future of the forest. These emphatically opposed the cutting of the forest. authorities included the executive president of Focusing more on the biological aspects of the the IDA, Defensoría de los Habitantes, the min- forest, they maintained that it was the last haven ister of natural resources, officials at the Inter- of higher genetic quality pochote in the coun- american Development Bank that were to finance try, perhaps in all of Central America (where a system of irrigation canals throughout the area, pochote forests have been mostly eliminated for and some journalists. cattle production), and that it was a refuge for Meanwhile, inhabitants of Bagatzí, as well as wildlife species that were ecologically connected those from Bagaces, a nearby town in the same to the neighboring Palo Verde National Park. region, joined forces with biologists and conser- Moreover, they argued, during the winter, large vationists in more direct action to protect the currents of water flowed through La Mula Creek threatened environmental resources. Some were and reached La Bocana Lagoon, located inside constantly watching over the forest to avoid the Palo Verde National Park. If rice was planted clandestine cutting, especially during weekends, where the pochote forest was, precipitation would while others established contacts with the me- wash chemicals used on these crops into the dia, inviting journalists to visit the problem area lagoon, poisoning its waters. They pointed out and even providing transportation to take jour- that the government is obligated to protect the nalists to the forest. There, they were led on lagoon, since it is included in the international guided tours by members of the local organiza- RAMSAR Treaty (for migratory waterfowl) signed tion, the Voice of the People. This organization by Costa Rica years earlier. also contributed to the campaign by writing let- Between January and April 1994, Gordon ters to authorities, making calls to the media to Frankie (a coeditor of this volume), who had keep them informed of developing events, and studied the ecology of the region for 30 years, traveling to San José to visit lawyers who could asked 35 national and international biologists to instruct them in the law and guide them to the carry out a rapid biodiversity assessment of the courts to stop the cutting of the forest. flora and fauna of La Mula Creek. Specialists in The efforts of these diverse organizations many different fields, from botany to ornithol- and individuals were dramatically successful.

260 ABURTO frankie chap 20 8/20/03 8:07 PM Page 261

The journalists who visited the area published interested journalists, as did the residents of several reports on the importance of the forest Bagatzí in their attempt to save La Mula Creek. and the desire of the inhabitants to preserve it. Some biologists do not address the media be- They spoke extensively of the damage that would cause they think journalists will not understand occur with the destruction of the forest and ex- them or care about their situation or, even worse, erted constant pressure on corresponding au- will modify their statements. The popular image thorities to revoke the decision to cut it down of journalists and reporters hustling to cover (in the latter phase, the Institute of Biodiversity criminal events, politics, and economic swings in Heredia also spoke out against the cutting). and consequently having little time to think about Finally, after more than a year of struggle, the ex- nature is especially discouraging to these scien- ecutive president of the IDA revoked the deci- tists. Yet, if it is true that journalists often seem sion to clear-cut the forest. Instead, he donated to have little time to devote to environmental the plots to the Ministry of Natural Resources, issues, it is not because they are not interested. under whose authority they would be annexed to It is certainly true that they live under constant the Palo Verde National Park and would receive pressure to find and publish the daily news. special and continuous protection. Bogged down with many interesting and news- The campaign to conserve the Mula repre- worthy stories, these journalists need biologists sents an excellent example of how biologists and to seek them out, to knock on the media doors local people kept in regular communication and show them what is happening. When biolo- until the decision to abort the cutting was made. gists encounter a situation that must be pre- It is also noteworthy that journalists had direct sented to the public eye, they must not hesitate contact with government officials in San José to pick up a phone, send a fax, and communicate and used these opportunities to influence their in any way possible. If their efforts are not fruit- decision making, which in turn stemmed from ful the first time, they must try again. If one the journalists’ firsthand knowledge of the field station or journal was unwilling to publish their situation through biologists and local people. story, then they must contact another one. In a word, persistence is the key to reaching the press (O’Brien 1993). HOW TO REACH THE PRESS Moreover, biologists should not limit their From the example of the Mula, it is clear that contacts to one or two enthusiastic journalists conservation depends on three principal factors: but should try to identify journalists in several the knowledge of biologists, awareness and ac- different media entities, such as radio, television, tion on the part of the public, and political sup- and newspapers, who write about the environ- port. The connecting link between these three ment. The biologist should then formulate a list factors, which coordinates and makes possible of these journalists for future reference. This list the actions of biologists, the public, and politi- must be updated constantly because of the mo- cians, is the media. Use of the media is espe- bility of journalists and also because there are cially crucial in these times of economic glob- always new and enthusiastic reporters in each alization, powerful transnational companies, field. Scientists should call these journalists to and movement of large capital throughout the establish an initial contact and should try to world, in which the pressure to exploit natural learn about their work specialty to see if biolo- resources currently protected has increased. gist and journalist have interests in common. If However, if the media are the crucial link, the so, biologists should make an effort to build a scientists are responsible for the initial step in relationship with this journalist through con- connecting this link—by aggressively pursuing sistent communication. Favorable contacts can the media and providing their information to easily be maintained by periodically sending the

THE MEDIA AND BIODIVERSITY CONSERVATION 261 frankie chap 20 8/20/03 8:07 PM Page 262

journalist a bulletin he or she can publish and If the area is threatened by the actions of a even inviting him or her on a field trip to observe private firm, biologists should also provide phone the biologist’s work (Jukofsky and Wille 1993). numbers and names of people in charge, if avail- There are a number of media entities through able, so that the journalist’s work can be facili- which the concerned scientist can communicate tated with respect to the “balance,” or fairness, with the public. Scientists should take advantage of the report. They should also include informa- of as many of these channels as possible: tion on how they can be reached, such as fax and cellular phone numbers or E-mail. Even a 1. Bulletins home phone number is necessary, since it is 2. Interviews with the press often during non–office hours when journalists 3. Press kits have time to check these documents and search 4. Advertisements for newsworthy material. Scientists should also prepare themselves 5. Web pages beforehand for interviews with the media. Pre- 6. Press conferences paratory materials, such as maps, photographs, 7. E-mail lists (Jacobson 1999) scientific data, statistics, charts, results from studies, and any other documents considered Although preparation for each of these channels important to clarify or reaffirm concepts are of communication varies slightly, some general absolutely necessary to support the scientist’s guidelines can be pinpointed in the use of inter- words. Scientists planning for an interview views and press bulletins. should also anticipate questions, both negative and positive, that might be asked regarding the event in question, and they should prepare pos- HOW TO CONDUCT AN INTERVIEW itive, precise, and direct answers. Time, for both Those scientists who worry about misrepresen- journalist and biologist, is very valuable, and sci- tation of their evidence by the media have good entists should strive to say as much as possible reason. A good opportunity to report a threat to in the amount of time available. In terms of the environment to the media is often wasted be- content, there are two basic rules for every in- cause of the ignorance or lack of preparation terview with a journalist: (1) Scientists should of the reporter when interviewing the biologist. always tell the truth. Credibility is the most valu- However, miscommunications can also easily able possession the scientist has. (2) Scientists be avoided by following a few simple guidelines. should avoid saying anything they do not want First, the biologist must understand that the published. There is no such thing as off-the- journalist probably has no solid knowledge of record comments. When they are published, it the scientist’s area of interest because the range will be too late for retraction. Finally, after the in- of specialties is so wide. Taking this into ac- terview has taken place, the biologist must make count, the biologist can easily facilitate an ef- sure that the journalist understood the concepts, fective and fluid interview, minus the dreaded magnitude of the problem, and terminology distortion, by providing the journalist with pre- employed. paratory material in anticipation of the interview. Such information could include location HOW TO DRAFT A PRESS BULLETIN of the threatened area, type of habitat, affected animal and plant species, and who or what Another important mode of communication is threatens them. It is also extremely important to the press bulletin. As previously mentioned, these “humanize” the report by indicating if people or bulletins are an effective way of developing a re- settlements are affected and by making the value lationship with the press and educating people of the endangered land explicit. about their natural environment. The following

262 ABURTO frankie chap 20 8/20/03 8:07 PM Page 263

are some key factors in the preparation of an problem touches the human aspect. effective press bulletin: Whenever possible, write your message as a story rather than as a scientific paper. 1. Know what you have to say. Make a rough draft of the points to be discussed and enu- 5. Send your story expeditiously. If there is an merate them from most important to least urgent event to communicate, do not wait important. Jacobson (1999) has suggested to include it in your next bulletin; by then, that materials should be organized with the the event may not be of interest anymore. most significant information first, followed Draft a special bulletin and distribute it by key details added throughout the paper immediately. and ending with data that could be useful 6. Make sure the bulletin reaches the right for the comprehension of the subject contact person. Do not send your story to discussed. just anyone who will receive it in the edit- 2. Make your story newsworthy. Pinpoint why ing office. Always verify that your contacts your story would interest the public. Avoid still work in their organizations, and warn propaganda for yourself or your organiza- them of your incoming story. tion. If your information seems more like 7. Pay attention to the presentation of your an advertisement, it will be ignored by the bulletin. A document with inadequate editing office. diagrams, deficient composition, and poor handwriting will have difficulty 3. Write in a simple, direct, and objective way. attracting attention in the editing room. Present the situation in the least amount Make sure the bulletin is “clean,” or not of words possible without losing coherence completely covered with text; leave some or making it incomprehensible. Any spaces blank so that the reader will have bulletin should not be longer than two a chance to rest. If possible, use the double-spaced pages and should have stationery of the organization for the short sentences and paragraphs that are bulletin. Write double-spaced, only on easy to read. Set aside overly technical lan- one side of the paper, using wide margins guage that is incomprehensible. Extensive and an appropriate letter size, not exces- documents with difficult language scare sively small or large. away readers and journalists, who often do not have time to read them. More impor- 8. Include visual aids. Whenever possible, in- tant, if the reporter cannot understand clude colorful photographs and maps that what the scientist is saying, then there will attract attention to your story. If the will be no interest in publishing the story. bulletin is for television, videos can also be These documents often end up in the used. Whenever possible, invite the media trashcan. Even if the reporter does run the on a trip to the research area and offer story, there is the risk of publishing incor- room and board to the journalists, in rect data that will infuriate the environ- addition to a guided tour of the area. mental expert and create indifference from 9. Follow up. When you wish to establish a the public. Do not forget to include phone relationship with a media organization, and fax numbers and other ways to contact remember to follow up on your story even you or your organization. if your case was only published once. Keep 4. Emphasize the human side of the story. the journalist regularly informed on the Always include the preoccupations, inter- developments concerning your case, espe- ests, or desires of the majority of people. cially if he or she is far away from the case The public will be interested when the under study.

THE MEDIA AND BIODIVERSITY CONSERVATION 263 frankie chap 20 8/20/03 8:07 PM Page 264

nels of communication open (O’Brien 1993). NEW DIRECTIONS This will, of course, require biologists to invest Aside from newspapers, radio, and television, some of their time and energy in the art of in- scientists now have a new tool with which to formation transfer and perhaps to make other reach large audiences: the Internet. Web pages investments to enable their voices and points of are a new and effective channel for the diffusion view to be known and understood (Meffe 1998; of their campaigns. Not only is this form of Jacobson 1999). Many field biologists know a communication now the fastest and most eco- great deal about the biodiversity of their study nomical mode of distributing information of areas and some of the consequences of losing it. any used to date, but there are more than 67 mil- There is a great need for more of them to come lion people worldwide navigating daily through forward and share their knowledge and passion the Internet. The Internet is also extremely ver- with others (Primack 1998). satile. It allows the mixing of text, photographs, audio, and video. It also allows for a special connection with the audience. “Visitors” to the site can simply obtain information from the com- REFERENCES puter screen or “download” documents of in- Carson, R. 1962. Silent spring. Boston: Houghton terest and save them. Direct communication with Mifflin. 368 pp. these visitors, who can respond immediately with Frankie, G. W., V. Solano, and M. McCoy. 1994. feedback, is also possible through the use of Una evaluación técnica del Corredor La Mula entre La Reserva Biológica Lomas Barbudal y El E-mails listed on the Web site. The creation of Parque Nacional Palo Verde. Bagaces: Amigos Web pages is continuously becoming simpler; de Lomas Barbudal; Berkeley: University of Cal- one no longer needs to be a computer expert to ifornia; and Heredia, Costa Rica: Universidad be able to create a Web page. Even now, there are Nacional de Heredia. 126 pp. Contact frankie@ Internet sites that provide complimentary man- nature.berkeley.edu for copies. uals that can help interested scientists set up a Jacobson, S. 1999. Communication skills for conservation professionals. Covelo, Calif.: Island Press. more sophisticated and attractive page. 351 pp. E-mail lists are also becoming more and more Jukofsky, D., and C. Wille. 1993. Difundan su men- popular and can be targeted at specific groups of saje, guía para conservacionistas. San José and people. They allow one to communicate with New York: Centro de Periodismo Ambiental, one’s colleagues and with media organizations Proyecto de la Rainforest Alliance. 83 pp. Meffe, G. K. 1998. Conservation biology: Into the and influential personalities of the entire world. millennium. Conservation Biology 12:1–3. E-mail lists also have some advantages over the Nelson, P. 1994. Diez recomendaciones prácticas Web. The information proceeds directly to the para la investigación y redacción de reportajes so- subscribers, who do not have to search for it bre temas ambientales. Washington, D.C.: Inter- on the Web, and the people who have subscribed national Center for Journalists. 63 pp. Contact: are already interested in the information and are [email protected]. Noss, R. F., and A. Y. Cooperrider, eds. 1994. Sav- therefore much more receptive (see discussion ing nature’s legacy: Protecting and restoring biodi- in Jacobson 1999). versity. Covelo, Calif.: Island Press. 417 pp. In conclusion, the experiences to date between O’Brien, M. H. 1993. Being a scientist means tak- biologists, the media, the general public, and ing sides. BioScience 43:706–8. decision makers indicate that planned channels Primack, R. B. 1998. Essentials of conservation biology. 2d ed. Sunderland, Mass.: Sinauer Associ- of communication, perhaps coupled with a strate- ates. 660 pp. gic plan, can produce positive results for con- Soulé, M., ed. 1986. Conservation biology: The sci- servation projects. Biologists have an important ence of scarcity and diversity. Massachusetts: role to play in initiating and keeping the chan- Sinauer Associates. 584 pp.

264 ABURTO frankie chap 20 8/20/03 8:07 PM Page 265

Tolan, S. 1990. John Muir: Naturalist, writer, and ental en América Central. San José: Centro de guardian of the North American Wilderness. Har- Periodismo Ambiental, Rainforest Alliance. risburg, Pa.: Morehouse Publishing. 68 pp. 53 pp. Contact: [email protected]. Wille, C., and D. Jukofsky. 1993. Periodismo ambi-

THE MEDIA AND BIODIVERSITY CONSERVATION 265 frankie chap 21 8/20/03 8:08 PM Page 266

chapter 21

Threats to the Conservation of Tropical Dry Forest in Costa Rica

Mauricio Quesada and Kathryn E. Stoner

pproximately 550,000 km2 of tropical forest remains in Costa Rica and recognizes this Adry forest covered the Paciﬁc coast of habitat as the most endangered within the coun- Mesoamerica at the time that the Spaniards ar- try (Sánchez-Azofeifa 1997). rived. Today less than 2 percent of this forest Eleven different conservation areas are rec- remains (Janzen 1988), mostly in Mexico (Trejo ognized in Costa Rica today, and only two of and Dirzo 2000). It has been estimated that these contain the ten national parks and reserves the only protected sites of tropical dry forest in that protect tropical dry forest (Blanco and Mata Mesoamerica that are large enough to possibly 1994). Parque Nacional Guanacaste in the sustain dry-forest ecosystems are Parque Na- Guanacaste Conservation Area protects approx- cional Santa Rosa and Parque Nacional Palo Verde imately 50,000 ha, and Parque Nacional Palo in Guanacaste, Costa Rica, and the Chamela- Verde and Reserva Biológica Lomas Barbudal in Cuixmala Biosphere Reserve in Jalisco, Mexico the Tempisque Conservation Area protect an (Hartshorn 1988). additional 20,000 ha. Tropical dry forests in Costa Rica have al- In this chapter we discuss the major factors most disappeared. The total area of original dry that have affected tropical dry forests in Costa forest in Costa Rica estimated before 1940 was Rica and some of the effects that they have had approximately 400,000 ha (8% of the national on this ecosystem. We also discuss current man- territory), and by 1950 this area was reduced to agement practices used by the Tempisque Con- 40,200 ha (Sader and Joyce 1988). The most re- servation Area in Parque Nacional Palo Verde cent Landsat TM satellite image analysis esti- and Reserva Biológica Lomas Barbudal that may mates that less than 0.1 percent of tropical dry threaten the preservation of tropical dry forests.

266 frankie chap 21 8/20/03 8:08 PM Page 267

Finally, we present some recommendations for cattle began in 1954 and was soon replaced by the future successful conservation of tropical the exportation of refrigerated beef (Hall 1985). dry forest. The cattle business continued to grow as the government contributed with national and in- CAUSES OF DRY-FOREST ternational credits to support the industry finan- CONVERSION IN COSTA RICA cially. It has been estimated that approximately 50 percent of all agricultural credit went to CATTLE cattle in the early 1970s (Quesada-López-Calleja Deforestation because of the cattle industry has 1974; Leonard 1987). The Nicoya Peninsula and been the main cause of tropical forest destruc- the land surrounding Liberia (the capital of tion throughout Mexico and Central America Guanacaste) increased the area dedicated to pas- (Quesada-López-Calleja 1974; Toledo 1992; Maass tures 151 percent from 1979 to 1992 as a result 1995). Costa Rica is no exception, with the cattle of economic incentives given to cattle owners industry contributing more to deforestation in (Canet et al. 1996). the period before 1980 than all other economic In addition to forest destruction, another im- activities combined including commercial log- pact of the cattle industry on tropical dry forests ging (Lehmann 1992). On a countrywide basis, is caused by the use of fire to maintain open pas- by 1973 more than one-third of the country’s ter- tures. Fire has been one of the single most detri- ritory consisted of cattle pastures (Quesada- mental effects on tropical dry-forest ecosystems López-Calleja 1974; Lehmann 1992). The area of in Costa Rica within this century (Janzen 1986). the country dedicated to pasture increased from Natural fires in Neotropical ecosystems are ex- 630,000 ha in 1950 to more than 2 million ha tremely rare (but see Middleton et al. 1997); the in 1994 (IICA 1995). majority of fires are caused by humans (Janzen The first cattle ranch was founded in Gua- 1986). Fires became a problem in Guanacaste nacaste in the late 1500s (Janzen 1986), and ex- because of the introduction of African star grass tensive cattle ranches were established in this (Hyparrhenia rufa) in the 1920s (Parsons 1972). dry-forest area as early as 1800 (Boucher et al. This grass is an aggressive self-seeding species, 1983). The oldest post-Hispanic settlements in originally from the savannas of Africa, and is the dry forest were established along the Temp- adapted to survive fire (Parsons 1972). Despite isque River Basin in Guanacaste with subsequent the widespread use of H. rufa as a source of food development of pastures and agriculture in this for cattle in Guanacaste, it is not an effective cat- region. Data obtained from aerial photographs tle feed. It has been demonstrated that cattle and satellite images indicate that approximately rarely consume this grass during the dry season, 46 percent (115,000 ha) of this region consisted when it has a very low nutritional value (Dau- of pastures by the mid-1950s (Maldonado et al. benmire 1972a). 1995). To minimize the impact of the cattle indus- Several factors affected the continued growth try on the dry forest in Guanacaste, it will be nec- of the cattle industry in northern Costa Rica dur- essary to change the current grazing practices ing the twentieth century. The construction of and encourage alternative land uses for aban- the Pan American Highway in the 1950s allowed doned pastures. The Costa Rican government easy access to the entire country, facilitating both has already begun to encourage agricultural de- settlement and marketing in the rest of Gua- velopment in many degraded areas of formerly nacaste (Williams 1986). Furthermore, expan- dry forest (see the section “Agriculture” later sion of the cattle industry was a major goal of the in this chapter). Continued efforts to minimize Costa Rican government in the 1950s in order to burning of pastures should be combined with diversify export products. The exportation of live the implementation of strong fire management

THREATS TO DRY- FOREST CONSERVATION 267 frankie chap 21 8/20/03 8:08 PM Page 268

programs both within protected areas and in Most of the destruction of natural forests in adjacent areas. Fire management programs do Costa Rica has occurred because of a lack of gov- exist in many of the protected areas in the dry ernment control and a philosophy oriented to forest, but a lack of adequate infrastructure (e.g., provide economic incentives to transform forests fire-fighting equipment) and sufficient trained into tree plantations, pastures, and agricultural personnel limit their effectiveness. One excep- fields. It was not until 1969, with the establish- tion is the fire control program of the Guana- ment of a forestry law, that a government office caste Conservation Area (Janzen 1986). was created to enforce the protection of national parks and forest reserves. Despite this law, nat- TIMBER ural forests outside protected areas continued In 1943, Costa Rica still reported a forest cover to be exploited without control, and economic of 70–75 percent of the national territory (Sader incentives to maintain forest cover outside pro- and Joyce 1988). However, large-scale defor- tected areas did not appear until 1979. The gov- estation was initiated in the 1950s, and by 1980 ernment developed a tax exemption program for Costa Rica was experiencing 3.19 percent defor- landowners that adopted reforestation manage- estation annually, one of the highest rates in the ment within their property. This program has world (Leonard 1987). Forest destruction con- replanted only 0.8 percent of the national terri- tinued into the 1990s, and analysis of Landsat tory over the past 20 years (Canet et al. 1996). TM images revealed that by 1991 only 29 per- Other types of economic incentives were initi- cent of the national territory remained covered ated after 1988 to encourage reforestation and with some type of forest (Sánchez-Azofeifa 1997). include the following programs: Certificado de This same study estimated the rate of change of Abono Forestal, Certificado de Abono Forestal forest cover between 1986 and 1991 to be ap- por Adelanto, Certificado de Abono Forestal para proximately 45,000 ha per year, which is equiv- Manejo de Bosque, and Fondo de Desarrollo alent to a 21 percent loss of the forest cover Forestal. Although these programs represented originally present in 1986. In contrast to a creative initiative to protect tropical forests, a Sánchez-Azofeifa (1997), a second recent study series of problems associated with insufficient reported an increase of 10 percent forest cover monitoring, as well as changes in government between 1986 and 1997, which represents a re- administrations, have limited their success. To- forestation rate of 83,300 ha per year. The gov- day, these reforestation programs (tree planta- ernment attributed this change to the success tions) have replanted only 4 percent of the na- of reforestation programs and to the recovery tional territory and have received U.S. $98 million of abandoned pastures (Castro and Arias 1998). in economic incentives (Canet et al. 1996; Cas- Much debate has been generated over the intro and Arias 1998). terpretation of the Landsat TM images used in Although the main purpose of these programs these studies primarily because Castro and Arias has been to promote the development of tree (1998) classified the forest cover uniformly and plantations in open and degraded areas, as well did not distinguish between the different levels as the protection of natural forest in private of succession—abandoned pastures with some hands, the government does not have the insti- scattered trees and areas in reforestation pro- tutional capacity to monitor these agreements grams (i.e., tree plantations) were categorized as systematically. For example, a review of the na- forest. Even if the recent estimates provided by tional forestry policy indicated that several in- Castro and Arias (1998) were correct, the investment corporations sold or changed the use crease in forest cover is the result of secondary of land while continuing to receive economic regeneration in the fragments of land that dom- incentives for reforestation (Rodríguez and Var- inate the nation’s landscape. gas 1988). Today, logging permits are issued to

268 QUESADA AND STONER frankie chap 21 8/20/03 8:08 PM Page 269

extract timber from plantations or on the basis biological corridors (Aldrich and Hamrick 1998; of officially authorized management plans of Cascante et al. 2002; chapter 3). The sustainable natural forests; regulation is supposedly enforced use of tropical forests depends on the ability of through transportation controls (Kishor and Con- trees to reproduce and regenerate under natural stantino 1993). Transportation controls do not conditions; however, our current knowledge on monitor the impact of extraction on the natural these basic biological aspects is limited to only forest or the ecological status of the exploited a few dry-forest timber trees (Cascante et al. populations. In addition, it has been estimated 2002; Quesada et al. 2001). that 50 percent of logging in Costa Rica is ille- Although much commercial logging today gal, and logging trucks transport at night or on is accomplished within the legal framework of weekends when there are no controls (Kishor “forestry management plans,” there is great vari- and Constantino 1993). ation in the degree of government control on According to the current forestry policy (Gac- management plans authorized by different eta 91, Resolution No. 323, 13 May 1998; Gaceta conservation areas. For example, according to 100, Executive Decree No. 26977, 26 May 1998), the Sistema Nacional de Areas de Conservación, the programs mentioned earlier have merged between 1996 and 1997 the Tempisque Conser- into a single program to compensate private vation Area approved management plans to landowners for the environmental services pro- exploit timber from tropical dry forests from an vided by the forest or trees within their property. area of 607 ha (SINAC 1997). In contrast, the This program is designed to provide the follow- Guanacaste Conservation Area did not approve ing economic incentives: (1) land devoted to any forest areas for management plans in this reforestation with plantations will receive $576 same period (SINAC 1997). In Guanacaste to- per hectare over a five-year period (2) natural day there are approximately 14 active sawmills forests under management plans (i.e., selective with a capacity to process 47,000 m3 per day logging) will receive $352 per hectare over a five- (Canet et al. 1996); much of the timber pro- year period; and (3) natural forests under absolute cessed in these sawmills has been legally har- protection will receive $224 per hectare over a vested through management plans authorized five-year period. This policy is clearly oriented to by the respective conservation areas. It is ironic stimulate the exploitation of natural forests out- that dry-forest timber species continue to be har- side protected areas by investing the funds ob- vested today in Guanacaste, despite the fact that tained for conservation in economic incentives almost no natural forests are found outside pro- for deforestation. tected areas. Approximately 56 percent of timber harvested The development of forest management plans in Costa Rica in 1998 was obtained from forests; by the government has not been effective in re- the rest came from pastures, remnant forests, ducing the amount of deforestation but rather and agricultural areas (Estado de la Nación 1999). has provided a legal tool to facilitate the contin- Trees located in pastures and remnant forests ued exploitation of the few remaining natural accounted for 23 percent of the timber harvested forests. A clear example of the damage that can in 1998, the same amount that was harvested be done, even when logging is conducted under from pastures and remnant forests in 1992 and the auspices of a management plan, is found in 1993. This harvesting regime may have serious the area of the Osa Peninsula in southwestern long-term consequences on the remaining nat- Costa Rica. A recent study revealed that between ural forests in the country, since recent evidence 1997 and 1998, 164 forestry management plans has indicated the importance of isolated trees were approved in this area, authorizing the as “stepping-stones” for gene flow of tropical cutting of more than 14,000 trees resulting trees, especially in remnant forests that serve as in the deforestation of approximately 4,000 ha

THREATS TO DRY- FOREST CONSERVATION 269 frankie chap 21 8/20/03 8:08 PM Page 270

TABLE 21.1 Commercially Important Native Timber Species from Costa Rica’s Tropical Dry Forest That Are Legally Exploited Today, Market Value of the Wood, and Conservation Status of the Species

common name scientiﬁc name family valuea statusb

Ron ron Astronium graveolens Anacardiaceae 233 T Guayacan real Guaiacum sanctum Zygophyllaceae 233 E Cocobola Dalbergia retusa Fabaceae 233 T Guapinol Hymenea courbaril Caesalpinaceae 233 S Cortez amarillo Tabebuia ochraceae Bignoniaceae 88 S Cenizaro Samanea saman Mimosaceae 88 V Pochote Pachira quinata Bombacaceae 88 V Guanacaste Enterolobium cyclocarpum Mimosaceae 68 S Guayaquil Albizzia guachapele Mimosaceae 68 S Tempisque Sideroxylon capiri Sapotaceae 63 T Roble sabana Tabebuia rosea Bignoniaceae 63 S

Source: Figures in “Value” column are from the ofﬁcial newspaper of the National Congress of Costa Rica, which published the approved law legalizing the exploitation, Gaceta 16, 23 January 1988, Executive Decree No. 26612-MINAE. aValue in U.S. dollars for each cubic meter of wood of each tree species. bCurrent status of each tree species according to Jiménez (1993): S = stable; E = endangered; T = threatened; V = vulnerable.

(Barrantes et al. 1999). This study documented 1993). Table 21.1 shows a list of the most im- that much of the logging is being conducted on portant commercial tree species that the gov- the hills of watersheds of rivers and that many ernment allows for trade (Gaceta 16, 23 January biological corridors are being destroyed. Further- 1998, Executive Decree No. 26612-MINAE). more, an evaluation of the approved manage- One of these species has been recognized as in ment plans and field studies demonstrated that danger of extinction, three of these species are most of the requirements of the management recognized as threatened, and two species are rec- plans are not being carried out in the field. ognized as vulnerable (Jiménez 1993). Of these Deforestation to harvest trees for timber was 11 species, only Pachira quinata is currently be- one of the first activities that affected tropical dry ing used in plantations in reforestation programs; forests in Costa Rica. Historically, extensive log- the other species are theoretically harvested from ging of economically important timber species natural forests through management plans. Al- such as brazilwood (Caesalpinia sp.), ironwood though these native tree species are important (Guaicam sanctum), mahogany (Swietenia macro- for the construction of furniture, buildings, and phylla), rosewood (Dalbergia retusa), and Spanish artisan wood crafts, the economic incentives for cedar (Cedrela odorata) began in the early 1800s reforestation in plantations principally have been in northwestern Costa Rica as land was cleared awarded for the use of exotic species such as for the development of cattle ranches (Boucher melina (Gmelina arborea) and teak (Tectona gran- et al. 1983). The province of Guanacaste, owing dis ) (Canet et al. 1996). One exception to the pat- to its geographical location and high number of tern of using exotic species in plantations is the economically important tree species, has greatly reforestation and research program being devel- reduced most timber tree populations (Jiménez oped by the Guanacaste Conservation Area at

270 QUESADA AND STONER frankie chap 21 8/20/03 8:08 PM Page 271

Horizontes Biological Station. An innovative in the Tempisque River Basin of Guanacaste, program designed to evaluate the success of 12 whereas in the early 1990s approximately 30 per- native species in a variety of mixed plantations cent of this area consisted of agricultural fields is under way (Brenes 1994). In addition, the (Maldonado et al. 1995). The most common Guanacaste Conservation Area has developed crops today, sugarcane and rice, are produced the only active program of natural regeneration in the artificially irrigated flatlands replacing of the tropical dry forest, and between 1979 and pastures and dry forest. Approximately 17 per- 1985 this program successfully restored ap- cent of the land originally covered with dry for- proximately 2,100 ha of pasturelands into forest est in the Tempisque River Basin is cultivated (Kramer 1997). with these crops (Maldonado et al. 1995). Agricultural production in Guanacaste has AGRICULTURE been encouraged by the Servicio Nacional de Agricultural activities were conducted in Costa Aguas Subterraneas Riego y Avenamiento Rica in pre-Columbian times and perhaps were (SENARA), which has developed the largest ir- responsible for the first impact on tropical dry- rigation system in Costa Rica (SENARA 1998). forest ecosystems in Guanacaste (Quesada- The Proyecto de Riego de Arenal Tempisque López-Calleja 1980; Boucher et al. 1983). How- transports water from the Arenal Lake (located ever, indigenous populations were reduced by in north-central Costa Rica in the Guanacaste as much as 95 percent (Parsons 1975) during the Volcanic Mountain Range) in large cement canals conquest beginning in 1560. The elimination of to various areas in Guanacaste to provide irriga- the native inhabitants, combined with the fact tion water for agricultural crops. This program that Costa Rica remained sparsely populated un- has incorporated approximately 15,000 ha into til after independence (1821), allowed many for- this irrigation network, investing a total of U.S. merly cultivated lands to regenerate into tropical $38,460,000. Most of this project has been forest. It is therefore unlikely that cultivation by funded by loans from the Inter-American Bank indigenous groups had any long-lasting effect for Development, the Venezuelan Inversion on the tropical dry forest in Guanacaste (Boucher Fund, and local Costa Rican counterparts. It is et al. 1983). estimated that approximately 336 agricultural Today in Costa Rica, the traditional crops of producers have benefited from this irrigation coffee, bananas, cacao, and sugarcane are the system in Guanacaste (SENARA 1998). most important agricultural products, followed Although agriculture may be a more finan- by rice, beans, and corn. Seventy-four percent cially productive use of former pasturelands in of the land dedicated to agriculture is cultivated dry-forest habitats, efforts should be made to with these crops, and bananas and coffee remain minimize the effects of agriculture on protected as the most important sources of individual in- areas. One example of the agricultural area come (Estado de la Nación 1997, 1999). Most of covered by the Proyecto de Riego de Arenal Tem- these crops have not been produced in the trop- pisque irrigation system includes the rice fields ical dry forest, and only recently has agriculture contiguous with Parque Nacional Palo Verde and had an impact on natural forests in this area. Reserva Biológica Lomas Barbudal. This agri- Large-scale agricultural development in Guana- cultural development may affect the surround- caste did not occur until prices for beef dropped ing protected areas in several ways; for example, in the 1980s and many landowners began to waterfowl and many native terrestrial animals explore alternative uses for their land. Many for- from the protected areas nest or forage in the mer pastures were changed into agricultural rice fields with high levels of insecticides (see fields in an attempt at economic solvency. Data chapter 5), and an increase in fires in the agri- from aerial photographs indicate that in the mid- cultural zone likely will increase fires in the 1950s no agricultural crops were being produced adjacent protected areas. The development of

THREATS TO DRY- FOREST CONSERVATION 271 frankie chap 21 8/20/03 8:08 PM Page 272

buffer zones around protected lands should be Within the dry forest, approximately 13,000 enforced, and the use of these buffer zones international tourists visited the Tempisque Con- should be monitored. Finally, the use of fire in servation Area in 1996, and 26,643 visited the agricultural fields adjacent to protected areas Guanacaste Conservation Area (SINAC 1996). should be discouraged. Fire is often used during With the increasing number of tourists visiting the production of sugarcane to burn the stems this endangered ecosystem and the subsequent before harvesting. This practice is particularly income generated by this industry, it is impera- dangerous in Guanacaste because sugarcane is tive to assure that sufficient resources are in- harvested during the dry season, when the veg- vested in maintaining and improving protected etation is dry and high winds often cause the areas. fires to spread out of control. To determine if the increase in tourism has led to a comparable increase in money gener- TOURISM ated by protected areas, we compare income The uncontrolled development of tourism is the generated by the hotel industry and by the most recent industry in Costa Rica affecting trop- conservation areas in the tropical dry forest. To ical dry-forest ecosystems (Estado de la Nación estimate the income generated by the hotel sec- 1999). On a countrywide basis, international tor, we divided the hotels into three categories tourism is currently the main industry and has based on price range during peak tourist sea- experienced a more than fourfold increase in son: (1) luxury hotel, more than $80 per room; income generated between 1986 and 1996 (2) moderate hotel, from about $20–$80; and (Estado de la Nación 1997). In this same time (3) economy hotels, less than $20. We conserv- period, the number of international tourists atively assumed a 90 percent occupancy rate visiting the country increased from 261,000 (in for four months during the peak season and a 1986) to 779,000 (in 1996). Visitation of inter- 50 percent occupancy rate for the other eight national tourists to the Costa Rican National Park months. Average prices per room during the peak system similarly has increased from 70,056 to season for each category of hotel were $100, 268,774 in this same time period. $50, and $20, respectively, for luxury, moderate, In 1998, within the area recognized as tropi- and economy hotels, and these prices were esti- cal dry forest, more than 256 hotels and lodges mated as $80, $30, and $15 during the off sea- with more than 5,150 rooms were dedicated to son. Based on these data, a conservative estimate tourism (data summarized from Pariser 1998). is that the hotel sector of tourism (not including Approximately 63 percent of these businesses food and entertainment) in the tropical dry for- primarily accommodate international tourists; est generates approximately $73 million a year. the remaining facilities receive national and in- This accounts for more than 10 percent of the ternational visitors. The majority of these hotels total income generated by the entire tourist in- (123) are located on the beaches of Guanacaste dustry in all of Costa Rica during 1996 (Estado and Puntarenas (Pariser 1998), and consequently de la Nación 1997). most of the international tourists who visit the In contrast to the large income generated by dry forest of Costa Rica stay in beach hotels. tourism in the tropical dry forest by the private Many of the tourists go on day trips to national sector, in 1996 less than $165,000 was gener- parks or protected areas. Approximately one out ated from tourism for the two conservation ar- of every three international tourists who came eas within this habitat (SINAC 1996). Despite to Costa Rica in 1996 visited a national park or the increase in tourists visiting protected areas, protected area (Estado de la Nación 1997). There- a similar investment has not been made for fore, the national parks and protected areas are maintaining these areas in terms of infrastruc- an important attraction to many of the country’s ture and personnel (Brandon 1996). Since the international tourists. dry forest ecosystem is an important attraction

272 QUESADA AND STONER frankie chap 21 8/20/03 8:08 PM Page 273

for tourists, it should be compensated for this bi- a new settlement program. In 1977, 4,800 ha ological value. The hotel industry in Guanacaste of this land were declared as the Dr. Lucas Ro- could contribute by sharing the costs of protec- dríguez Caballero National Wildlife Refuge (Ex- tion and maintenance of protected areas within ecutive Decree No. 6942-A, 18 April 1977). An this area. Although theoretically the Instituto area of approximately 10,000 ha adjacent to Costarricense de Turismo is responsible for mon- the refuge was protected as Parque Nacional Palo itoring the development of tourist accommo- Verde in 1981 (Executive Decree No. 12765-A, dations, in 1999 a series of irregularities were 2 July 1981). The refuge and park were expanded discovered in the permits issued to this industry, several times during the next decade, and by especially in the coastal areas in Guanacaste 1990 Parque Nacional Palo Verde was expanded and Puntarenas (Estado de la Nación 1999). Thus to include the refuge, Reserva Biológica Lomas far in Costa Rica the development of tourism has Barbudal, and the connecting corridor for a total largely been a private enterprise with little or no of approximately 20,000 ha (Executive decree government regulations (Brandon 1996; Estado No. 20033-MIRENEM, 9 November 1990). de la Nación 1999). Although the long-term cu- Cattle have been present within this area mulative effects of tourism on tropical dry-forest under different densities for at least the past 80 habitats are unknown, some of the potential ef- years; however, the Stewart family removed fects include (1) damage to coastal and mangrove them from Palo Verde in 1979 shortly after the areas along the beach; (2) improper sewage and creation of the wildlife refuge (D. A. Stewart garbage disposal; (3) disturbance of nesting ar- interview). In 1987 some cattle were reintro- eas of endangered sea turtles; and (4) destruc- duced into the Palo Verde Lagoon as part of a tion of natural forests for tourist developments. management program for the wetlands, and ap- To assure the continued well-being of tropical proximately 4,000 head of cattle were released dry-forest ecosystems, it will be necessary for the into various lowland forested areas within the government to implement and enforce stricter park in 1991 when a technical committee of the regulations for the development of the tourist Tempisque Conservation Area decided to utilize industry. cattle as a management tool to control ﬁres and to manage the wetlands (Mozo 1995). In the WILDLIFE MANAGEMENT PLANS following section we analyze the management AND SUSTAINABLE DEVELOPMENT plan and discuss some of its effects on this en- IN THE TROPICAL DRY FOREST: dangered ecosystem. THEORY AND PRACTICE MANAGEMENT WITH CATTLE HISTORY The Tempisque Conservation Area, with sup- Parque Nacional Palo Verde and Reserva Biológ- port provided by the Inter-American Bank for ica Lomas Barbudal are located in the Temp- Development as part of the funding to develop isque River Basin in Guanacaste. These two the irrigation project of SENARA (see the sec- protected sites cover an area of approximately tion “Agriculture” earlier in this chapter), facili- 20,000 ha and encompass a variety of habitats tated the development of a wildlife manage- including dry deciduous forest, regenerating dry ment plan for Parque Nacional Palo Verde and deciduous forest of various ages, riparian and Reserva Biológica Lomas Barbudal (Vaughan et spring forests, savanna, mesic forest, mangrove al. 1995). This plan, in conjunction with a previ- forest, and wetlands (Frankie et al. 1993). ously established program of the Tempisque This area was historically a cattle ranch dat- Conservation Area, implemented the use of ing from 1926. In 1975 a portion of the land was cattle within these protected areas with the in- expropriated by the Costa Rican government tention of controlling ﬁres and restoring the (Instituto de Tierras y Colonización) to develop wetlands.

THREATS TO DRY- FOREST CONSERVATION 273 frankie chap 21 8/20/03 8:08 PM Page 274

The management plan is currently using While Palo Verde was part of the Stewart cattle in the Palo Verde wetlands as an attempt ranch, prior to 1979, five natural canals con- to maintain open waterbird habitat by eliminat- nected the Palo Verde Lagoon to the Tempisque ing cattails (Typha dominguensis) (Vaughan et al. River, allowing water to flood into the lagoon 1995). This area historically has been one of the during high tide. Cement gates or weirs were most important wetlands in Central America used at the end of these natural canals to help for more than 60 species of migratory and resi- regulate the water level in the lagoon. During dent waterbirds and is currently recognized as the dry season these gates were regularly closed a RAMSAR site (Vaughan et al. 1995). Since the after high tide by the ranch hands to help main- 1980s, this marsh has experienced a series of suc- tain the lagoon flooded with salty water (D. A. cessional changes that culminated in the reduc- Stewart interview). The weirs were not main- tion of open water and the establishment of large tained once the ranch was removed from Palo monospecific stands of cattails. Systematic re- Verde in 1979, and the natural canals silted in, introduction of cattle into the Palo Verde Lagoon contributing to the drying out of the lagoon and began in 1991 with the objective of eliminating eliminating the entrance of brackish waters from cattails and maintaining the waterbird habitat. the Tempisque River. These factors (and many According to McCoy (1994), cattle forage and others) should be evaluated because they un- trample on cattails, promoting open water. How- doubtedly contributed to the spread of cattails ever, since the introduction of more than 500 in the Palo Verde Lagoon during the following head of cattle (1 cow per ha) in 1991, these wet- decades. lands are still dominated by monospecific stands The decision of the management plan to use of cattails. This observation is consistent with ob- cattle to control fires and to promote regenera- servations made by D. A. Stewart from more than tion of the forest was based on a poorly designed 50 years of working with cattle in the Palo Verde study (supported by the Area de Conservación Lagoon, during which time he never saw the cat- Tempisque and the Organization for Tropical tle consuming cattails (D. A. Stewart interview). Studies) that concluded that cattle grazing re- The attempt to restore the Palo Verde Lagoon duces fires by controlling the expansion of pas- through the use of cattle has been implemented tures and favors the fast succession of dry forest without conducting a systematic experiment to (Barboza 1995, 1996). This study attempted to evaluate the impact and changes they may cause evaluate the effect of cattle grazing on the restora- in this ecosystem. Although the current man- tion of dry forest by comparing the following agement plan assumes that cattle are the key three treatments: (1) cattle grazing without fire; element in maintaining the open marsh and re- (2) no cattle grazing with fire; and (3) no cattle ducing the number of cattails, many factors grazing without fire (control). Three replicates contribute to the spread of cattails and thus may of each treatment were included in plots of 20 × be useful in controlling this species. The cattail 20 m. No standardized method of beginning is a fast-growing plant that successfully repro- each of the treatments at similar levels of regen- duces both sexually and asexually (McNaughton eration was utilized, and the treatments were not 1975). A single cattail head can produce up to randomly placed; the cattle without fire treat- 250,000 seeds, and the seeds may remain viable ments were all closer to the seed source of the for almost 100 years with an approximately 100 forest. In addition, the density of cattle within percent germination rate (Sojda 1993). Previous the treatments was not monitored. Since the research has demonstrated that the most effec- density of cattle was low and variable during the tive methods for controlling cattails include in- study, it is possible that no cows were ever within creasing the water level and salinity (Kelley et the treatments designated as cattle grazing and al. 1993; Magee 1993; Sojda 1993). that the proximity of these plots to the seed

274 QUESADA AND STONER frankie chap 21 8/20/03 8:08 PM Page 275

source (i.e., forest) was the principal contribut- grazed area was significantly greater than the ing factor to the greater number of stems and intermittently grazed sites (t = 3.14, d.f. = 743, species in this area. The conclusions of this p < 0.001). Janzen (1986) stated that the restora- study were based on qualitative interpretations tion of pastures to tropical dry forests is largely only without any quantitative comparisons or determined by the distance to the seed sources, statistical tests. the seed dispersal mode of the plant species, the The study in Palo Verde also considered proximity to the forest patches, and the number “controlled grazing” as a novel technique to gen- and species of animals involved. Once the seeds erate economic benefits for local communities are established in the pasture, they have to out- promoting sustainable development (Barboza compete the tall and dense grass community 1996). Between 1991 and 1994, 14 people were dominated by H. rufa. At this critical stage, live- given permission to have their cattle in the park. stock were originally used in certain areas of For each head of cattle, the participants were Parque Nacional Guanacaste, in low densities charged 100 colones (approximately 50 cents) and for short periods of time to reduce the per cow per month. In contrast, the approximate amount of grass. However, even under these charge for pasture rental for cattle in the Gua- controlled conditions, the resultant woody suc- nacaste area at this time was approximately cession may have a different species structure 600–900 colones ($3.00–$4.50) per cow per than does succession without livestock (Janzen month. Thus, it appears that the few people that 1986). It is likely that cattle grazing controls the were participating in this program found it more vertical expansion of H. rufa, but the horizontal economic to maintain their cows within the growth continues and outcompetes other woody national park than within privately owned pas- colonizers. This may be explained by the long tures. The cattle-grazing project was promoted evolutionary relationship between H. rufa and as sustainable development presumably in terms savanna grazers (Parsons 1972; McNaughton of a way for local people to generate a constant 1985; Milchunas et al. 1988). For the past several income, but few people benefited from this pro- years, cattle have been completely eliminated gram. In 1998 a similar number of people (15) from Parque Nacional Guanacaste and are not received permission to maintain their cattle considered to be a viable management tool for within the park for a rental fee of 200 colones controlling fires in this area. (approximately 67 cents) per cow per month. It appears that the wildlife management The economic benefit for the Tempisque Con- plan for Palo Verde and Lomas Barbudal was de- servation Area could not be estimated because veloped without any systematic research and of inadequate records documenting the collec- without reviewing the published information re- tion of payments and the use of money (Mozo garding the control of cattails (Kelley et al. 1993; 1995). In sum, this program, which obtained Magee 1993; Sojda 1993) or the effect of cattle on much support because of its supposed qualities vegetation (Daubenmire 1972a,b; Schulz and for sustainable development, has resulted in eco- Leininger 1990; Bock et al. 1993). It has not been nomic benefits for only a handful of local people demonstrated that cattle reduce the number of and has not generated income for the park. cattails in the lagoon or that the use of cattle We conducted a preliminary study to evalu- within the dry forest reduces the risk of fires ate the impact that the current management plan or promotes “ecological restoration” of the for- has on Palo Verde by comparing the floristic est. We suggest that systematic research be composition of an area within the cattle-grazing conducted to evaluate the effectiveness of cattle program with an area that had not been grazed as a management tool, as well as the evaluation over the past 15 years (Stern et al. 2003; chap- of other alternatives such as more and larger ter 5). A Shannon diversity index of the un- firebreaks or green breaks (Frere et al. 1992).

THREATS TO DRY- FOREST CONSERVATION 275 frankie chap 21 8/20/03 8:08 PM Page 276

RECOMMENDATIONS FOR THE should assure the Tempisque Conservation Area CONSERVATION OF TROPICAL adequate resources to achieve this type of inte- DRY FOREST IN COSTA RICA grative program.

PROTECTION AGRICULTURAL AREAS AND BUFFER ZONES The importance of complete protection of the Given the increase of land destined to be used few remaining tropical dry-forest habitats in as sugarcane and rice fields in the Guanacaste Guanacaste cannot be over emphasized. Efforts lowlands in the coming years, it is imperative must be made to purchase tropical dry-forest that adequate buffer zones be created in areas areas and to connect fragments of forest as well surrounding parks and reserves. Buffer zones as protected areas. Deforestation affects many should reduce the impact on the protected areas species of flora and fauna not only by eliminat- they surround. Since previous studies have in- ing their habitat but also by the inevitable frag- dicated that agricultural contamination, espe- mentation of habitats, which negatively affects cially that from rice production, may have severe population densities of many vertebrates and detrimental effects on the fauna within a region insects (Bierregaard and Stouffer 1997; Malcolm (see chapter 5), we do not recommend the con- 1997). Furthermore, protection of elevational tinued development of agriculture in buffer gradients is necessary for many species that zones. Land use that will be less harmful in annually migrate from tropical dry forests to al- buffer zones includes natural forest, forest plan- ternative habitats (Janzen 1986; Stoner 2001; tations (preferably with native species), or re- chapter 5). One example of a program connect- generating forest. Monitoring of buffer zones ing fragments over an elevational gradient is the should be conducted to assure that activities that Guanacaste Conservation Area. Parque Nacional may negatively affect the adjacent protected area Guanacaste now covers an elevational gradient are not initiated. Biological research should be from sea level up to the Cacao and Orosi volca- encouraged in these areas to provide more in- noes (1,100 meters above sea level). This inno- formation about the role of buffer zones and vative design should be taken as a model for edge effects in the protection of parks. other dry-forest areas in the Neotropics. RESTORATION ECOLOGY CATTLE AND FIRE If Costa Rica is to maintain and possibly in- To minimize the impact of the cattle industry crease forest cover within the dry forest, res- on the dry forest it will be necessary to change toration and natural regeneration programs, the current grazing practices that rely heavily on equivalent to the projects developed by the Gua- H. rufa and a regime of regular burning. Alter- nacaste Conservation Area, need to be imple- native feeds should be evaluated as potential mented immediately, within both protected sources of food for cattle. Continued efforts areas and privately owned lands. Furthermore, should be made to reduce fires in tropical dry the largest economic incentives given for the forests. To achieve this goal it is essential that the environmental services provided by private conservation areas have the following: (1) a com- landowners should be for total protection of ex- prehensive system of firebreaks surrounding and isting natural forests, not for the development of within protected areas; (2) adequate personnel forestry plantations. Finally, reforestation pro- and equipment for fire fighting; and (3) a programs should concentrate on utilizing native gram of environmental education within the re- timber species, not exotics. Today, the native tree gion. The effectiveness of this type of program species most wanted by loggers are mainly may be seen in the success of Parque Nacional found in protected areas. Flores (1985) predicted Guanacaste in recent years. The government that severe shortages in timber will result in a (Ministry of Environment and Energy [MINAE]) $350 million wood import bill by the beginning

276 QUESADA AND STONER frankie chap 21 8/20/03 8:08 PM Page 277

of the twenty-first century. If Costa Rica does not promote research within the respective conser- develop a program of plantations with its own vation areas. timber species, the last logging frontier will begin and end in the national parks and protected areas of this small country. CONCLUSIONS To accomplish any of these recommended goals, TOURISM it will be necessary to have both sufficient eco- The government should control the development nomic resources and adequately trained person- of the tourism industry, and regulations should nel. Terborgh (1999: 161) stated that “tropical be implemented and monitored to assure that parks are failing in country after country be- this industry does not destroy the dry-forest cause the institutions created to protect them ecosystem in Guanacaste. A hotel tax should be are weak and ineffectual. Institutional weakness imposed on hotels dedicated to the tourism in- derives from many sources, but it arises ulti- dustry. This tax should contribute to the main- mately from the low priority governments give tenance and development of protected areas to protection of parks.” If tropical dry forests near the hotels (i.e., the areas most frequently (and other habitats, as well) are to be success- visited by the hotel’s clients). fully conserved within Costa Rica, the government must recognize the importance of this task CATTLE AS “MANAGEMENT TOOLS” and see it as one of its most important priorities The use of cattle in Parque Nacional Palo Verde, within its administration. Many laws regarding Reserva Biológica Lomas Barbudal, and other the protection and use of biological resources protected areas of dry forest should be suspended have already been created (e.g., the new organic until further research has evaluated the impact law for the environment), but the challenge lies of this practice on the regeneration of the forest in designating sufficient resources (both eco- and the maintenance of wetlands. Alternative nomic and personnel) to enforce them. programs should be evaluated for controlling fires within the forest and for reducing the amount of Typha in the wetlands and restoring REFERENCES the marsh. Aldrich, P. R., and J. L. Hamrick. 1998. Reproduc- tive dominance of pasture trees in a fragmented SCIENTIFIC PANELS TO EVALUATE tropical forest mosaic. Science 281:103–5. WILDLIFE MANAGEMENT PLANS Barboza, G. 1995. Pastoreo controlado para control de incendios y restauración de bosque tropical seco, The Costa Rican government and international en el Parque Nacional Palo Verde, Guanacaste, organizations that provide funding for wildlife Costa Rica. Informe de avance de resultados de management plans should utilize scientific monitoreo del proyecto, 12 de junio de 1995: panels to evaluate the quality and potential ef- San José: Organización para Estudios Tropicales. fectiveness of these plans before implementa- 7 pp. tion within protected areas. This review process ———. 1996. Pastoreo controlado evita incendios y ayuda a ganaderos. OET al Día 1:4. is essential to evaluate and identify innovative Barrantes, G., Q. Jiménez, J. Lobo, T. Maldonado, ideas that may contribute to the protection of M. Quesada, and R. Quesada. 1999. Evaluación endangered habitats. Each conservation area de los planes de manejo forestal autorizados en el within the country should have an independent periodo 1997–1998 en la Peninsula de Osa. scientific panel composed of Costa Rican and Cumplimiento de normas técnicas, ambientales e impacto sobre el bosque natural. San José: Fun- international biologists. The participants of this dación Cecropia. 96 pp. panel should be academic individuals who are Bierregaard, R. O., Jr., and P. C. Stouffer. 1997. independent of the government and the national Understory birds and dynamic habitat mosaics park system. This panel can also identify and in Amazonian rainforests. In Tropical forest

THREATS TO DRY- FOREST CONSERVATION 277 frankie chap 21 8/20/03 8:08 PM Page 278

remnants: Ecology, management, and conservation Flores, J. G. 1985. Diagnóstico del sector industrial of fragmented communities, ed. W. H. Laurance forestal. San José: Editorial Universidad Estatal and R. O. Bierregaard Jr., 138–55. Chicago: Uni- a Distancia. 72 pp. versity of Chicago Press. Frankie, G. W., L. Newstrom, S. B. Vinson, and Blanco, O., and A. Mata. 1994. La cuenca del Golfo J. F. Barthell. 1993. Nesting-habit preferences de Nicoya: Un reto al desarrollo sostenible. San of selected Centris bee species in Costan Rican José: Editorial de la Universidad de Costa Rica. dry forest. Biotropica 25:322–33. 235 pp. Frere, E., P. Gandini, C. Hayslip, J. C. Martinez- Bock, C. E., J. H. Bock, and H. M. Smith. 1993. Sanchez, and D. Milum. 1992. Aspects of Proposal for a system of federal livestock ex- biological management in Guanacaste National closures on public rangelands in the western Park. Institute for Environmental Studies 12:1–29. United States. Conservation Biology 7:731–33. Garita, D. 1989. Mapa de cobertura boscosa. Re- Boucher, D. H., M. Hansen, S. Risch, and J. H. port, Unidad de Cartografía y Topografía. San Vandermeer. 1983. Agriculture: Introduction. José: DGF and MIRENEM. In Costa Rican natural history, ed. D. H. Janzen, Hall, C. 1985. Costa Rica: A geographical interpreta- 66–73. Chicago: University of Chicago Press. tion in historical perspective. Boulder, Colo.: West- Brandon, K. 1996. Ecotourism and conservation: A view Press. 348 pp. review of key issues. Environmental Department Hartshorn, G. S. 1988. Tropical and subtropical Papers, Global Environmental Division. Wash- vegetation of Meso-America. In North American ington, D.C.: World Bank. 69 pp. terrestrial vegetation, ed. M. G. Barbour and W. D. Brenes, G. 1994. Descripción general del programa de Billings, 365–90. New York: Cambridge Uni- restauración y silvicultura del bosque seco del Area versity Press. de Conservación Guanacaste. San José: Ministe- IICA (Instituto Interamericano de Ciencias Agri- rio del Ambiente y Energia. 14 pp. colas). 1995. Area de concentración I (Componente Canet, G., M. Chavarria, O. Gamboa, D. Garita, de política socioeconómica, comercio e inversiones). M. Jiménez, S. Lobo, P. Marín, L. Sevilla, Z. Tre- San José: Ministereo de Agricultura. jos, and M. Valerio. 1996. Informacíon estadís- Janzen, D. H. 1986. Guanacaste National Park: Trop- tica relevante sobre el sector forestal 1972–1995. ical ecological cultural restoration. San José: Edi- San José: Ministerio del Ambiente y Energía, torial Estatal a Distancia. 103 pp. Sistmea Nacional de Areas de Conservación, ———. 1988. Tropical dry forests: The most en- Area de Fomento. 54 pp. dangered major tropical ecosystem. In Biodiver- Cascante, A., M. Quesada, J. A. Lobo, and E. Fuchs. sity, ed. E. O. Wilson, 130–37. Washington, D.C.: 2002. Effects of dry tropical forest fragmenta- National Academy Press. tion on the reproductive success and genetic Jiménez, Q. 1993. Arboles maderables en peligro de ﬂow of the tree, Samanea saman. Conservation extinción en Costa Rica. 1st ed. San José: INCAFO Biology 16:1–11. S.A. 121 pp. Castro, R., and G. Arias. 1998. Costa Rica: Hacia la Kelley, J. R., Jr., M. K. Laubhan, and F. A. Reid. 1993. sostenibilidad de sus recursos forestales. San José: Options for water-level control in developed Ministerio del Ambiente y Energía y el Fondo wetlands. Fish and Wildlife Leaﬂet 13(4):1–7. Nacional de Finaciamiento Forestal. 23 pp. Kishor, N. M., and L. F. Constantino. 1993. Forest Daubenmire, R. 1972a. Ecology of Hyparrhenia management and competing land uses: An eco- rufa (Nees) in derived savanna in north-western nomic analysis for Costa Rica. The World Bank, Costa Rica. Journal of Applied Ecology 9:11–23. Latin America and Caribbean Region’s Envi- ———. 1972b. Some ecological consequences of ronmental Division (LATEN), Dissemination converting forest to savanna in northwestern Note 7:1–30. Costa Rica. Tropical Ecology 13:31–51. Kramer, E. 1997. Measuring landscape changes in Estado de la Nación en Desarrollo Humano Sos- remnant tropical dry forests. In Tropical forest tenible. 1997. Un analisis amplio y objectivo so- remnants: Ecology, management, and conservation bre la Costa Rica que tenemos a partir de los indi- of fragmented communities, ed. W. H. Laurance cadores mas actuales (1996). 1st ed. San José: and R. O. Bierregaard Jr., 386–99. Chicago: Editorama S.A. 46 pp. University of Chicago Press. ———. 1999. Un analisis amplio y objectivo sobre la Lehmann, M. P. 1992. Deforestation and changing Costa Rica que tenemos a partir de los indicadores land-use patterns in Costa Rica. In Changing mas actuales (1998). 1st ed. San José: Editorama tropical forests: Historical perspectives on today’s S.A. 37 pp. challenges in Central and South America, ed. H. K.

278 QUESADA AND STONER frankie chap 21 8/20/03 8:08 PM Page 279

Steen and R. P. Tucker, 59–76. Durham, N.C.: Pariser, H. S. 1998. Adventure guide to Costa Rica. Forest History Society. 3d ed. Edison, N.J.: Hunter Publishing. 546 pp. Leonard, H. J. 1987. Natural resources and economic Parsons, J. J. 1972. The spread of African pasture development in Central America. New Brunswick, grasses into the New World tropics. Journal of N.J.: International Institute for Environment Range Management 20:13–17. and Development, Rutgers University. ———. 1975. The changing nature of New World Maass, J. M. 1995. Conversion of tropical dry forest tropical forests since European colonization. In pasture and agriculture. In Seasonally dry tropi- The use of ecological guidelines for development in cal forests, ed. S. H. Bullock, H. A. Mooney, and the American humid tropics, 28–38. IUCN Pub- E. Medina, 399–422. Cambridge: Cambridge lications. Morges, Switzerland: IUCN. University Press. Quesada, M., E. Fuchs, and J. Lobo. 2001. Pollen Magee, P. A. 1993. Detrital accumulation and pro- load size, reproductive success and progeny kin- cessing in wetlands. Fish and Wildlife Leaflet ship of natural pollinated flowers of the tropical 13(3):1–7. dry forest tree, Pachira quinta. American Journal Malcolm, J. R. 1997. Biomass and diversity of small of Botany 88:2113–18. mammals in Amazonian forest fragments. In Quesada-López-Calleja, R. 1974. La forja de una Tropical forest remnants: Ecology, management, nación, tomo II. Tesis de Licenciatura en Dere- and conservation of fragmented communities, ed. cho, Universidad de Costa Rica. W. H. Laurance and R. O. Bierregaard Jr., 207– ———. 1980. Costa Rica: La frontera sur de Meso- 21. Chicago: University of Chicago Press. américa. 2d ed. San José: Instituto Costarricense Maldonado, T., J. Bravo, G. Castro, Q. Jiménez, O. de Turismo. 228 pp. Saborio, and L. Paniagua. 1995. Evaluación eco- Rodríguez, R., and E. Vargas. 1988. El recurso fore- logica rapida región del Tempisque Guanacaste, stal en Costa Rica: Políticas públicas y sociedad. Costa Rica. San José: Fundación Neotropica, Heredia, Costa Rica: Editorial de la Universidad Centro de Estudios Ambientales y Politicas. Nacional. 61 pp. 104 pp. Sader, S. A., and A. T. Joyce. 1988. Deforestation McCoy, M. B. 1994. Seasonal, freshwater marshes rates and trends in Costa Rica, 1940 to 1983. in the tropics: A case in which cattle grazing is Biotropica 20:11–19. not detrimental. In Principles of conservation Sánchez-Azofeifa, G. A. 1997. Assessing land use/ biology, ed. G. K. Meffe and C. R. Carroll, 352–53. cover change in Costa Rica. San José: Centro de Sunderland, Mass.: Sinauer Associates. Investigaciones en Desorrollo Sostenible, Uni- McNaughton, S. J. 1975. “r” and “k” selection in versidad de Costa Rica. 181 pp. Typha. American Naturalist 109:251–61. Schulz, T. T., and W. C. Leininger. 1990. Differ- ———. 1985. Ecology of a grazing ecosystem: The ences in riparian vegetation structure between Serengeti. Ecological Monographs 55:259–94. grazed areas and exclosures. Journal of Range Méndez, G. 1997. Centro de investigación del Management 43:295–99. bosque tropical seco y estaciones biológicos: SENARA (Servicio Nacional de Aguas Subter- Una opción para el estudio, la investigación y el raneas Riego y Avenamiento). 1998. Gestión in- ecotourismo. ACG Rothschildia Revista informa- stitucional, 1994–1998. Informe de las acciones tiva 4:9–10. implementadas en los programas y proyectos Middleton, B. A., E. Sánchez-Rojas, B. Suedmeyer, estrategicos del SENARA durante la admin- and A. Michels. 1997. Fire in a tropical dry for- stración Figueres Olsen. San José: SENARA. est of Central America: A natural part of the dis- 19 pp. turbance regime? Biotropica 29:515–17. SINAC (Sistema Nacional de Areas de Conserva- Milchunas, D. G., O. E. Sala, and W. K. Lauenroth. ción). 1996. Informe de ingresos. San José: SINAC. 1988. A generalized model of the effects of 10 pp. grazing by large herbivores on grassland com- ———. 1997. Hectareas aprobadas para planes de munity structure. American Naturalist 132:87– manejo de bosque. San José: SINAC. 15 pp. 106. Sojda, R. S. 1993. Management and control of cat- Mozo, E. T. 1995. Pastoreo con ganado vacuno, una tails. Fish and Wildlife Leaflet 13(41):1–7. alternativa del ACT para prevención de incendios Stern, M., M. Quesada, and K. E. Stoner. 2003. forestales, recuperación de humedales y restaru- Changes in composition and structure of a ración del bosque tropical seco. Bagaces, Costa tropical dry forest following intermittent cattle Rica: Convenio MIRENEM–Opción Colombia, growing. Revista de Biología Tropical 50(3/4): Universidad Sergio Arboleda. 66 pp. 1021–34.

THREATS TO DRY- FOREST CONSERVATION 279 frankie chap 21 8/20/03 8:08 PM Page 280

Stoner, K. E. 2001. Differential habitat use and re- Trejo, I., and R. Dirzo. 2000. Deforestation of sea- productive patterns of tropical dry forest frugiv- sonally dry tropical forest: A national and local orous and nectarivorous bats in northwestern analysis in Mexico. Biological Conservation 94: Costa Rica. Canadian Journal of Zoology 79: 133–42. 1626–33. Vaughan, C., M. McCoy, J. Fallas, H. Chaves, Terborgh, J. 1999. Why conservation in the tropics G. Barboza, G. Wong, M. Carbonell, J. Rau, and is failing: The need for a new paradigm. In Re- M. Carranza. 1995. Plan de manejo y desarrollo quiem for nature, 161–86. Washington, D.C.: del Parque Nacional Palo Verde y Reserva Biolog- Island Press/Shearwater Books. ica Lomas Barbudal. Contrato SENARA-BID- Toledo, V. M. 1992. Bio-economic cost. In Develop- MIRENEM-UNA. Heredia, Costa Rica: Univer- ment or destruction? The conversion of tropical for- sidad Nacional. 110 pp. est to pasture in Latin America, ed. T. Downing, Williams, R. G. 1986. Export agriculture and the cri- S. Hecht, and H. Pearson, 63–71. New York: sis in Central America. Chapel Hill: University Westview Press. of North Carolina Press. 52 pp.

280 QUESADA AND STONER frankie chap 22 8/20/03 8:09 PM Page 281

chapter 22

Environmental Law of Costa Rica

DEVELOPMENT AND ENFORCEMENT

Roxana Salazar

ver the past 20 years, Costa Rica has creased help to developing countries, must go Okept pace with the evolution of environ- beyond what has been done until now.” Further, mental policy in the Central American region it was pointed out that the main threats against and the Caribbean, developing a fairly complete the environment, which were discussed in Rio legal frame and, in general, a good set of poli- de Janeiro in 1992, are now global and have cies. However, Costa Rica has also followed the been sharpened by extreme consumer habits. general global trend, despite the importance that In Costa Rica, environmental destruction is, environmental issues have gained, by allowing at least in part, the product of poor interpreta- the destruction of biodiversity to continue at an tion and lack of enforcement of the laws, as well accelerated pace (see chapters 7, 12, 15, 21, and as of shortcomings in the laws and public poli- 23). Multiple declarations and agreements at cies themselves. On the one hand, we must con- the international level have not succeeded in sider that attitude change begins with education stopping this worldwide destruction, as has been and that sanctions and ﬁnes should be used as recognized by world authorities who attended secondary measures. However, this change in the Ministers of the Environment World Fo- attitude has yet to be systematically encouraged. rum (Ministers of the Environment World Forum The policies coming out of public institutions 2000): “There is an alarming discrepancy be- have been more of control and short-term re- tween commitments and action. The goals and sponses than of prevention, education, and par- priorities agreed upon by the international com- ticipation. The policies for biodiversity protec- munity regarding sustainable development, like tion are also vertically controlled, which means the adoption of national strategies and the in- that all respective duties are granted to only one

281 frankie chap 22 8/20/03 8:09 PM Page 282

ministry. As a result, laws and regulations are ticle 89, containing the cultural goals of the re- created that are inconsistent with social, eco- public, which establishes the protection of nat- nomic, and/or cultural customs. ural beauty, conservation, and development of On the other hand, the judicial branch, which the historic and artistic heritage of the nation. is responsible for exacting compliance with en- Aside from protection of the environment for vironmental law, lacks a basic understanding cultural purposes, the constitution also guaran- of ecology as well as of pertinent environmen- tees the right of each individual to “a healthy and tal issues. It thus has trouble identifying oppor- ecologically balanced environment,” which ap- tunities for legal actions. Another problem is pears in article 50, amended in recent years to that environmental offenses, and corresponding contain the following sentences: “The State will penalties, have not been classified. Without es- provide for the greater well-being of the inhabi- tablished categories of offenses and respective tants of the country, organizing and stimulating penalties, judges have difficulty assigning ap- production and the most appropriate sharing propriate sanctions for the offender and com- of wealth” and “ Every person has the right to a pensation for damages committed against the healthy and ecologically balanced environment. environment. As a result, the judicial branch is Therefore, he or she is justified in denouncing often unable to carry out the intent of the law. any act that infringes upon that right and claim- Enforcing environmental law is further hin- ing reparations for the damage caused.” dered by a deficiency of clear policies, inade- In pursuit of these goals, the state has cre- quate budgets and human resources, and lack ated several general guidelines for the environ- of follow-up evaluation and verification mech- ment’s defense, including the “precautionary anisms. An especially pernicious problem is the principle,” or in dubio pro natura, and “the pol- lack of coordinated actions between institutions luter pays.” The first of these principles has been of the public sector, as is the case for water re- a persistent theme in the most important inter- sources management (see chapter 9), which national environmental gatherings in the past means inefficiency in addressing ongoing envi- decade, including the summit at Rio de Janeiro ronmental issues. This problem is evident in in 1992, the Biological Diversity Convention several well-known cases in which squatters (UNEP 1992), and the Convention on Climate who had invaded national wildlife refuges were Change (Law 7414). Emphasizing preventive already receiving utility services from national rather than reactive measures, the precaution- companies before anything was done to expel ary principle was formally adopted in Costa them. Although Costa Rica’s environmental law Rica’s Biodiversity Law (Zeledón 1999c) in the frames are innovative and far-reaching in intent 11th article: “2.—Precaution criterion or in du- and have had an important presence in the coun- bio pro natura: Whenever there exists a danger try, there are many shortcomings that hinder of serious or irrevocable damage to biodiversity their effectiveness. or to its associated knowledge, the absence of This chapter presents the general legal frame scientific certainty must not be used as a reason for environmental and biodiversity protection to postpone the adoption of effective measures in Costa Rica, identifying its weaknesses as well for its protection.” as important areas for action that must be con- The second principle, which demands that the sidered for improvement of these conditions. polluter bear the cost of damage to the environment, also has roots in the declarations of internationally recognized conventions. Both the 16th THE SPIRIT OF THE LAW principles of the Río Declaration and the Cli- The Political Constitution of Costa Rica (GOCR matic Change Convention insisted that polluters 1984) grants a series of fundamental guarantees internalize environmental costs and consider related to the environment. Among these is ar- the public interest without distorting commerce

282 SALAZAR frankie chap 22 8/20/03 8:09 PM Page 283

and international investments. Costa Rica’s EIAs. This law forces economic parties to con- variation is article 103d of the Biodiversity Law sider environmental costs in the design and ex- (Zeledón 1999c), which states that it is the obli- ecution of their projects and to take appropri- gation of the Ministry of Environment and En- ate precautionary measures by requiring EIAs. ergy (MINAE) to eliminate negative incentives These documents are then made public, allow- that encourage damage to the environment and ing neighbors and interested parties to know the to create disincentives that will promote envi- details of the project and to present their opin- ronmental conservation. ions for consideration before the project is One measure that puts this principle into authorized. When existing projects or activities practice is article 69 of the Forestry Law (Zele- do not comply with the terms under which they dón 1999b). Combining both a disincentive and were approved, any interested party can dean incentive to promote conservation, this law nounce them before the National Environmen- places a tax on fuels and compensates forestry tal Technical Bureau, which is responsible for plantation owners for the amount they spend on reviewing, approving, and monitoring these services that mitigate fuel emissions. However, projects. However, despite this law’s potential as there are many laws that do not follow the pre- an environmental defense mechanism, the suc- viously mentioned guidelines. For example, there cess of EIAs in protecting the environment is is tax exemption for synthetic chemicals in cer- highly questionable. The most serious problems tain percentages. This measure constitutes a lie in enforcement. The National Environmental “negative incentive” by encouraging the use of Technical Bureau has too often approved EIAs harmful synthetic chemicals without taking into for projects that clearly have strong negative consideration the high environmental or social environmental effects. In many areas, commu- costs that such use entails. Thus, even though the nity participation in public hearings and con- principles mentioned earlier are far-reaching in sultations has become one more requisite with their intent, the laws themselves often distort which to comply, and the challenges facing neg- the spirit of these guidelines, an issue that is atively affected communities are rarely taken into discussed further in the following section. account. Territorial Ordering Plans. The Organic Environ- mental Law encourages active participation from THE LETTER OF THE LAW the inhabitants and organized society in the The principles discussed earlier are the basis creation and application of territorial ordering for some of the most important environmental plans and urban regulatory plans. These plans laws produced by Costa Rica’s Legislative As- determine what type of industries, commercial sembly. Establishing a series of general regula- establishments, or projects can be allowed in tions for the environment’s preservation, these what areas and with what procedures. By partic- are the Organic Environmental Law, the Forestry ipating in the creation of these plans, farmers, Law, the Biodiversity Law, the Wild Life Protection entrepreneurs, conservationist groups, and res- Law, and the Water Law (Zeledón 1999a–e). idents all have clear knowledge of the “rules of the game,” and problems caused by an activity ORGANIC ENVIRONMENTAL LAW inappropriately placed can be avoided. Above all, The Organic Environmental Law strives to pro- this law requires that various administrative and vide Costa Ricans and the state with the neces- local government entities cooperate to deﬁne sary instruments for preserving a healthy and special protection areas and work together for ecologically balanced environment. It is made their true protection. up of three parts: environmental impact assess- However, territorial ordering plans suffer ments (EIAs), territorial ordering plans, and re- from some of the same problems that affect en- strictions on land use. vironmental assessment plans: lack of proper

ENVIRONMENTAL LAW 283 frankie chap 22 8/20/03 8:09 PM Page 284

enforcement. As in the case of the EIAs, experi- sible for creating wildlife refuges (Zeledón ence shows that many county officials in charge 1999d: art. 82, p. 110) as well as classifying them of these plans do not exact compliance with their as (1) state property, (2) mixed property, or (3) pri- conditions. On the contrary, their refusal to en- vate property. No commercial activities are theo- force the plans is often tied to a deceitful agree- retically permitted in areas classified as state ment between the private owner and corrupt mu- property; only authorized activities such as re- nicipal representatives for their own private gain. search and tourism are allowed, if properly con- Limitations for Land and Forestry Property. The trolled and with low impact. Private property is Forestry Law defines as an essential and primary allowed in other management categories, but this function of the state the conservation, protec- is limited for environmental reasons. In fact, tion, and administration of natural forests. It many wildlife refuges are mixed (article 82), such consists of a set of general regulations concern- as those on the coast. In these refuges, the state ing the use and improvement of the nation’s owns the maritime-terrestrial zone, and private forest resources in accordance with the princi- landowners may have properties in the rest of ple of correct and sustainable use of renewable the refuge. Controlled development of private natural resources. Put simply, the Forestry Law activities such as tourism and commerce is per- places restrictions on property use—even private mitted in these private holdings. These activities property—and does not permit change of land must also be low impact and are subject to spe- use inside national forests. cial regulations that guarantee compatibility be- Protecting the environment through restric- tween protection of the environment and right tions on land use, however, has problems not of property. only in its enforcement. It also conflicts with The institution in charge of these protected several important laws, especially those that areas is the MINAE, in cooperation with the Na- protect private property. In response to these re- tional System of Protected Areas, which clusters strictions, some private landowners have in- the National Park Service, the State Forestry Ad- sisted that the Forestry Law is contrary to the ministration, the General Wildlife Bureau, and inviolability principle of property rights, granted the Bureau of Geology and Mines. This agency, by article 45 of the Constitution (GOCR 1984). along with its associated organizations, is re- Meanwhile, the Supreme Constitutional Court sponsible for creating technical measures for has maintained the position that forestry re- the appropriate management and conservation sources are protected in order to guarantee the of wild flora and fauna, recommending the es- integrity of the natural environment. Supreme tablishment of national refuges, and requesting Constitutional Court Vote 5893-1995 (Supreme from respective competent authorities “the ar- Constitutional Court 1995) reemphasizes that the rest of persons who invade property under the preservation of “forestry resources is not only, by national wildlife refuge regime” (article 36 of itself, reasonable, but is constitutionally viable, its the Forestry Law). Along with these responsibil- protection and rational exploitation being an ob- ities, the ministry has the power to authorize ligation of the State.” Although this interpretation certain activities in refuges of mixed and private has supported the importance of forest protec- categories if their EIA is approved and as long tion, it goes against long-established notions of as they continue to abide by strict sustainability private property and is met with much resistance. criteria. According to article 81 of the bylaws, activities such as tourist development may be au- NATIONAL PARKS AND WILDLIFE REFUGES thorized if they are “in agreement with the prin- Conflict between laws, at least in the case of wild- ciples of sustainable development established in life refuges, is really a management problem. the management plans.” The laws themselves are fairly straightforward. It is the ministry’s abuse of this power to The government’s executive branch is respon- authorize activities that creates conflict and ulti-

284 SALAZAR frankie chap 22 8/20/03 8:09 PM Page 285

mately threatens the national wildlife refuges. Despite these laws, the condition of the It is logical—and legal—for the ministry to ap- rivers that cross the Greater Metropolitan Area, prove permits in privately owned lands inside an although showing some improvements, contin- officially declared refuge. However, even though ues to be deplorable (see chapter 9). The actions the limits are not clearly defined and there are for the control and adequate emission of do- some contradictions in the bylaws, it is very clear mestic wastes have not received proper atten- that the state should not grant permission for tion, even though this is a serious problem that developments in protected areas meant to con- requires an urgent solution. Although citizens serve wildlife. Yet the ministry has allowed the continue to denounce the condition of these pub- invasion of protected state lands in several places, lic waters and laws are in place to enable actions has even granted permits for construction in against the perpetrators, the state has rarely re- these protected areas, and has ignored, to a sponded. Part of the problem is that many dif- greater or lesser degree, the permission limits ferent activities use water as a receptor of waste. and agreements for activities in privately owned In order for the state to enforce regulations it lands. The ministry has created bylaws that con- has to create new rules and controls for these tradict state environmental law, introducing activities—a politically risky business. hotels into wildlife preserves, which were never intended to be tourist areas (Gandoca-Manzanillo PROTECTION ZONES National Wildlife Refuge and nearby coastal ar- Another special case is the protection zones eas are an example of this contradictory policy. or protection areas. The Forestry Law estab- See Rossi 1992). As a result, Costa Rica’s pre- lishes that all Costa Rican lands with forests and cious refuges and the biodiversity of wild species especially those with the potential for forestry, they contain have begun to be transformed, little whether they are state-owned or subject to pri- by little, into tourist sites. vate control, will fall under the coverage of the law. Of particular importance are forested areas adjacent to rivers, lagoons, or lakes. As with wa- WATER PROTECTION ter, not only is conservation of these areas im- The conservation and sustainable use of water portant for environmental reasons, but it also is are a special case because protection of water is necessary for preventing natural threats to civil not only of environmental importance but also works and building settlements. These areas of social interest. Again, as with the Organic En- have a propensity for landslides and floods, which vironmental Law and wildlife refuges, the legal have, every year, destroyed homes, bridges, high- frame concerning water conservation is abun- ways, and other infrastructure, with serious con- dantly clear. Water is the property of the state, sequences for the population. The slides have which is in charge of granting concessions for reached truly frightening dimensions, causing its use. The administration and management of severe damages. To prevent these disasters, the water are the responsibility of the Water Depart- state has declared forested areas along rivers, ment of the MINAE, which operates in the Na- lagoons, and lakes “protection zones,” entitled tional Meteorological Institute. With respect to to special protections. water pollution, article 132 clearly prohibits pour- This law prohibits tree cutting in protection ing wastewaters, solid wastes, or any other pol- zones, and article 19 of the Forestry Law forbids luting substance in national waters. The Wildlife land-use change. However, there are several ex- Conservation Law includes norms about water ceptions to the law. The second paragraph of pollution (article 132, for example, prohibits article 19 states that permits for logging and water contamination) that have enabled citizens construction can be granted in forested areas to file a number of complaints against industries with the objective of “carrying out infrastructure that lack adequate waste treatment systems. projects, state or private, that are of national

ENVIRONMENTAL LAW 285 frankie chap 22 8/20/03 8:09 PM Page 286

convenience.” This exception to the law is reit- Moreover, it clearly reflects the spirit of Costa erated in article 34, which prohibits the cutting Rica’s environmental principles. All three meas- of trees in protection areas, except in those proj- ures of the Organic Environmental Law as well ects designated of “national convenience” by the as the establishment of protection zones seek to executive branch. prevent damage to the environment by demar- The bylaws in the Forestry Law clarified what cating areas of protection, restricting land use, was meant by national convenience: “The activ- and requiring those who do intend to use their ities of national convenience are those related to land to file EIAs. EIAs and water pollution laws the study and execution of projects or activities have the added impact of forcing those planning of public interest carried out by the centralized projects to internalize costs to the environment State entities, independent institutions, or the by making them find alternatives to potentially private sector, which bring about benefits to all damaging procedures. or a large part of society, such as: capture, trans- However, effectiveness of these measures is portation, and supply of water; pipelines; road dramatically diminished by the lack of proper construction; generation, transmission, and enforcement by the institutions in charge of car- distribution of electricity; transportation; min- rying out the laws. Conflicts between these laws ing activities; irrigation and drainage channels; and social policy only increase the opportunities recovery of forestry vocation areas; conservation for corrupt officials to neglect the conservation and sustainable use of forests; and others of the effort. Finally, problems exist within the laws same nature to be determined by the Ministry of themselves, as evidenced by laws that regulate Energy and the Environment according to the land use in protection zones, allowing broad needs of the country.” interpretations, which do not promote environ- This definition largely refers to projects under- mental preservation. taken by the public sector. However, the state subsequently incorporated an article into the TWO CASES OF PARTIAL SUCCESS Biodiversity Law in which the activities of national convenience were defined as “activities It is useful to analyze two cases that reflect prob- carried out by centralized State entities, in- lems with the law and the institutions appointed dependent institutions, or the private sector, to enforce them. The first of these concerns the which have social benefits greater than the Geest Caribbean (a British company) banana socio-environmental costs.” plantation, which began to expand its territories The threat to protection zones, as evidenced at the beginning of the 1990s. In doing so, the by the cited definitions, lies within the laws them- company failed to comply with various legal en- selves. As various civilian groups, such as the vironmental requirements on some of its farms, AMBIO Foundation, have pointed out, the dis- clear-cutting trees near stream corridors that cretional power of the executive branch to inter- were officially protected and deviating streams pret the law is far too wide. These groups have in Tortuguero and the Santa Clara lowlands, in openly expressed their concern with this ruling, Costa Rica’s northeastern Caribbean slope. The noting that almost any activity could be consid- damages also affected the Tortuguero National ered of “national convenience.” Yet despite the Park and the Barra del Colorado Wildlife Refuge. danger to both the environment and civilian Although these areas were officially desig- population living and working in these areas, nated protected areas, the government at that logging and construction continue to occur in time openly permitted the cutting of primary protection zones. forest trees, considering agricultural expansion In summary, a broad and comprehensive a national priority. The public was aware of these constitutional and legal frame for the protection violations, however, and the state was sued by of the environment does exist in Costa Rica. the general procurator (procuraduria general),

286 SALAZAR frankie chap 22 8/20/03 8:09 PM Page 287

legitimizing the claim presented by the non- tion of some of the already finished works—as governmental organization Justicia para la Nat- always, after the damages against the environ- uraleza. In 1995 the penal process entered the ment were already done. Despite strong charges judgment phase, and the Tropical Science Cen- of habitat destruction and the obvious damages ter, a nonprofit scientific organization based in that were taking place, the constitutional procu- San José, Costa Rica, was officially requested to rator could not avoid the legal actions proposed conduct an EIA. Most of the economic evalua- by the interested companies, and the process tion (nonmarketing damages) was based on the was prolonged and continues to this very day. United States’ Clean Water Act, which permits This case showed that MINAE action was weak an evaluation equal to the replacement costs and did not enforce rules until the damage was (restoration of damaged resources), and the Oil done. Part of the problem was that the EIA did Pollution Act of 1990, as well as on the cumu- not become compulsory until 1995, with the new lated experiences of the Tropical Science Center Organic Environmental Law. on EIAs and natural forest science. In order to avoid a long process and a very CONCLUSIONS AND possible larger penalty, the process ended in RECOMMENDATIONS conciliation between parties. Although the penalty did not compensate all the damages, it For several decades, Costa Rica has encouraged established an important legal precedent in the the protection of natural areas, bringing the ex- history of Costa Rican legislation. The fine was isting environmental legal frame up to date with to be employed in environmental restoration the most current international law. Its commit- projects in the affected areas, as well as for legal ment to preserving the environment and rich and expertise expenses needed to complete them. biodiversity of species it contains is reflected in This case shows the importance of civil par- a national system of conservation areas that ticipation, but it also reveals, once again, the enjoys international prestige. However, there political influence of economic and development are serious problems in the interpretation and interests that prevent proper establishment of enforcement of laws as well as in the structure environmental and social frames. It also shows of the laws themselves. In some instances, there that enforcement is difficult to achieve for au- is a clear lack of government interest in exacting thorities because there are many interconnected compliance with the regulations. interests associated with the activities. Finally, Problems within institutions responsible for these authorities are often weak. protecting the environment, such as the MINAE, In the Gandoca-Manzanillo case, a group of are reflected in the lack of long-term policies and investors began in 1992 to develop a tourist adequate budgets and human resources. Despite center on the Caribbean coast on a concession its mandate to monitor carefully all activities in the maritime zone granted by the Talamanca within conservation areas, the ministry lacks Municipality (province of Limón). Construction the follow-up, evaluation, and verification mech- of the hotel complex was carried out with dra- anisms that would prevent disasters such as matic modifications to coastal resources. These that of the Gandoca case and others (see also damages included draining and filling wet- chapter 23). Finally, the absence of coordinated lands for tennis courts, coral rock extraction for actions between institutions of the public sector landfills, raw sewage dumped directly into the creates conflicts between laws that are not easily sea, and displacement and burying of official resolved. landmarks. All these problems have a direct effect on the Local residents denounced the situation to the health of the environment. There is also a very MINAE. After several years, the ministry was serious and continuing consequence to the forced to investigate and finally to order demoli- quality of life of every Costa Rican citizen when

ENVIRONMENTAL LAW 287 frankie chap 22 8/20/03 8:09 PM Page 288

the environment is seriously damaged or de- 6. Citizens and citizens’ organizations must stroyed. Several mechanisms have been pro- be allowed to exercise their rights to a posed to remedy these serious problems, but healthy environment by participating in many have not offered even a glimmer of hope conservation efforts, including the preven- for the environment’s protection. Thus, rather tion, mitigation, and control of damages to than adding new measures, it is urgently im- the environment as well as its restoration portant that the current legal frame be reviewed after these damages have occurred. and modified according to the following general 7. Environmental education for the general recommendations: public is absolutely necessary if they are to 1. Clear, long-term national environmental be involved in protecting the environment. policies that focus on prevention rather Officials responsible for the environment than control of damages to the environ- need higher levels of education as well as ment are urgently needed. It is necessary specialized training to complete their obli- to classify environmental violations so that gations successfully. penalties and sanctions can be issued to violators. The loopholes that now exist in REFERENCES the legal frame, as evidenced in the Gan- GOCR (Government of Costa Rica). 1984. Political doca example, make enforcement of envi- Constitution of the Republic of Costa Rica. San ronmentally oriented laws an impossible José: National Printing Office. 100 pp. task. ———. 1994. Amendment to the constitution, by 2. Once environmental laws have been cre- Law 7412 of May 24, 1994. La Gazeta (San José), 10 June 1994. ated, they must be monitored and their Ministers of the Environment World Forum. 2000. efficiency and effectiveness evaluated. Malmoe Encounter, Sweden, 2000. These laws must continue to be revised Rossi, A. 1992. La loca de Gandoca. San José: Edi- until they can guarantee the environment’s torial Universitaria Centroamericana. 140 pp. protection. Supreme Constitutional Court. 1995. Vote of 1995. Supreme Constitutional Court Archives, San 3. It is important that social law and environ- José. mental law are compatible. Contradictions UNEP (United Nations Environmental Program). between private property rights and envi- 1992. Biological Diversity Convention. Rio de ronmental law must be eliminated. Janeiro. Zeledón, R., ed. 1999a. The Organic Environmen- 4. The judicial branch must have a basic un- tal Law N°7554(10/4/95). In Codigo ambiental, derstanding of both ecological processes 4. San José: Editorial Porvenir. and the environmental legal frame so that it ———. 1999b. The Forestry Law N°7575 (2/5/96), can accurately interpret these environmen- amended by laws N°7609 (6/11/96) and ° tal laws and can issue appropriate penalties N 7788 (4/30/98). In Codigo ambiental, 31. San José: Editorial Porvenir. (see also chapter 23). ———. 1999c. The Biodiversity Law N°7788 5. It is necessary that all levels of action be (4/30/98). In Codigo ambiental, 61. San José: horizontally integrated so that environmen- Editorial Porvenir. tal law can be put into action. One institu- ———. 1999d. The Wildlife Conservation Law N°7788 (4/30/98). In Codigo ambiental, 93. San tion (the MINAE) is not enough to regulate José: Editorial Porvenir. all activities in all conservation areas. Other ———. 1999e. The Water Law N°276 (8/27/42), institutions must be given the power to amended by laws N°2332 (4/9/59), N°5046 authorize activities and enforce legislation, (8/16/72) and N°5516 (5/2/74). In Codigo am- and all institutions must work together to biental, 141. San José: Editorial Porvenir. efficiently accomplish these tasks.

288 SALAZAR frankie chap 23 8/20/03 8:10 PM Page 289

chapter 23

Dispute over the Protection of the Environment in Costa Rica

Julio Alberto Bustos

lthough there are many laws for the laws. Such a high ﬁgure suggests that, in Costa Aprotection of the environment, there is also Rica, the solution for every problem is “passing incongruity among them. a law” or increasing the sanctions provided in the existing laws. Whether these laws are in agree- LEGAL PROTECTION OF BIODIVERSITY ment with previous legislation is rarely considered, much less the requirements to put these INCONGRUITY OF ENVIRONMENTAL LAWS laws into action, such as the economic and tech- The existence of a law for practically every en- nical resources for their application, the estab- vironmental problem (forestry law, water law, lishment of specialized tribunals, and the in- biodiversity law, and so on) has led to an en- struction of judicial representatives in charge of tanglement of laws that are often redundant, administering justice in environmental matters. contradictory, and ambiguous and thus hinder Laws in general stumble into frequent legisla- cooperation between institutions and limit effective errors that leave loopholes through which tive action. From its independence in 1821, Costa perpetrators of environmental damages escape. Rica has passed more than 19,000 laws for a For example, the legal norm in forestry law in- country of fewer than 4 million inhabitants. More dicates that, independent of its vegetative cover, than 8,000 laws are currently in force, taking a plot is not a forest if it is smaller than 2 ha.1 into consideration the abolition of some of these This allows for deforestation of small but vital

1. Forestry Law #7575 of 5 February 1996, art. 3, sec. d: “Forest: native or autochthonous ecosystem, intervened or non-regenerated by natural succession or other forestry techniques, occupying a surface of two more hectares.”

289 frankie chap 23 8/20/03 8:10 PM Page 290

forested areas, even if they protect a headwater or creek or are part of a larger biological corridor ECONOMIC LIMITATIONS OF THE and their conservation is of utmost importance. JUDICIAL POWER AND THE MINISTRY Despite laws that prohibit the activity, there is OF ENVIRONMENT AND ENERGY (MINAE) no judicial sanction for dumping wastewaters, There are many complaints concerning eco- most of which are untreated, into rivers unless nomic limitations for protecting the environment they contain heavy metals, trash, fuels, and sed- effectively. The environmental comptroller, in iments (see chapters 9 and 10). Thus, waste- charge of the fiscalization of all laws related to waters from the Greater Metropolitan Area the environment, has only one assistant and (San José and neighboring counties) are poured does not even have a vehicle for the appropriate into the Tárcoles River at an annual volume of duties of the post. The Environmental Prosecu- 35.2 million m3, producing a yearlong index tor of the Public Ministry, the specialized office of critical contamination (see chapter 9). The that tends to all cases pertaining to public action Tempisque and Barranca Rivers are subject to crimes that involve pollution, phytosanitary pro- similar contamination. A joint study by the Na- tection (e.g., quarantine and pest eradication), tional University (province of Heredia) and the wildlife abuses, unregulated urban development, Utrecht University (the Netherlands) regarding and the destruction or theft of archaeological pollution of the Gulf of Nicoya established that heritage, has but three investigators to cover the the principal contaminator of the basin is the entire country. The claims are actually received human population. Out of 41 municipalities in by assistant prosecutors around the country, the area, only 22 dump their wastes in sanitary who have shown an almost complete lack of landfills and the rest in open dump sites.2 knowledge regarding the environment’s legal protections. JUDGES’ LACK OF Costa Rica has theoretically protected 25 per- ENVIRONMENTAL KNOWLEDGE cent of its territory under the categories National Even though there is a judicial school in Costa Park, Forestry Reserve, or Wildlife Refuge and Rica that provides instruction to judges in the has achieved international prestige by elevating fields they oversee, there are always those who to a constitutional rank the right of its citizens show a lack of consciousness and interest re- to a healthy and ecologically balanced environ- garding protection of natural resources and their ment. This constitutional right allows every citi- importance for public health. As a result, seri- zen to denounce acts that violate environmental ous claims of destruction of specific natural re- law and to demand compensation for damages sources are disregarded in the justice tribunals, caused.3 However, reality is different. Original starting with prosecutors who conduct perti- private owners have not been paid in full for nent investigations and ending with the final ab- many lands expropriated by the government. solution from judges who perhaps do not under- These landowners exhibit a range of responses stand the damage caused by cutting down trees from patience to bitter frustration. Many, tired or pouring toxic residues into a river (see chap- of waiting for promised compensation (some- ters 9, 10, and 22). times for decades), exploit these lands as they

2. Economic-ecological appraisal of water deterioration have the right to denounce actions that threaten this right in the Gulf of Nicoya, National University–Utrecht Univer- and demand restoration of the damage caused. sity, December 2000. The State will guarantee, defend, and preserve said 3. Political Constitution of the Republic of Costa Rica, right. The law will determine responsibilities and corre- art. 50, paras. 2, 3, 4: Every person has the right to a healthy sponding sanctions. (Article reformed by Law #7412 of and ecologically balanced environment. Therefore, they all 24 May 1994.)

290 BUSTOS frankie chap 23 8/20/03 8:10 PM Page 291

see fit. Moreover, as private forests diminish, follow-up of the dismissed cases and those that loggers are surreptitiously moving in on “pro- reached trial and ended in sentencing or abso- tected” areas. There is a daily parade of logging lution (J. A. Bustos pers. obs.). There is also no trucks loaded with huge specimens from ances- registry of repeat offenders—data that could tral primary forests, especially during weekends determine the existence of professional delin- or evenings when forest rangers are not on quents in environmental crimes. Although there watch. MINAE estimates that 25 percent of the are no figures on environmental crimes that commercial wood in the country is extracted are punished with a sentence, ecologists, prose- illegally. cutors, and the environmental comptroller agree MINAE, the institution responsible for stop- that they are very few. In statements to the press, ping forest clear-cutting under a scheme of one well-known environmental prosecutor, José absolute protection, claims to have insufficient Pablo González, has suggested that the country funds to hire enough park rangers to patrol the is on the verge of total impunity with respect to forests, much less supply them with the neces- environmental crimes.4 sary equipment, such as vehicles and radios, for González and other prosecutors have blamed implementing their surveillance duties. At least the lack of convictions and sentencing for envi- that was its explanation when 3,000 ha from ronmental crimes on MINAE. This claim is only Tortuguero National Park were recently cut down partially correct, as a lengthy legal process must right under its nose, without any inspector issu- be followed in order to prosecute a named party ing a warning of the damages, as discussed later. for an environmental crime. Investigation of an unlawful environmental act begins with MINAE IMPUNITY OF VIOLATORS BEFORE and its initial collection of field evidence for pos- ENVIRONMENTAL LAWS sible prosecution of the case. As prosecutors Weakness of action, owing in part to lack of com- have pointed out, the record shows that MINAE munication between controlling entities, strongly often does not conduct a thorough collection contributes to violators’ evident impunity regard- of field evidence in the vicinity of the alleged ing environmental crimes. These crimes are crime. Further, it does not present the evidence currently denounced at the regional offices of at a trial and commonly has little follow-up con- MINAE, the Office of the Environmental Comp- tact with the Public Ministry. These deficiencies troller, the Environmental Tribunal, and any often lead to poorly prepared cases of the pros- Prosecutor of the Republic, especially the Envi- ecutors, resulting in acquittals. Clearly, the two ronmental Prosecutor, which is the specialized ministries should be required to work more entity of the Public Ministry. Incredibly, these closely together. institutions do not communicate with one an- The Public Ministry is the entity of the judi- other, and there is no state entity that consoli- cial power in Costa Rica that exercises the duties dates information on environmental damages of plaintiff in the penal justice field. The Public at a national level. Ministry must conduct the preparatory investi- The Environmental Prosecutor has records gation in class action lawsuits and has com- only of the number of submitted claims (137 in plete functional independence in the exercise of 1999 and 178 in 2000) but has no statistical its faculties and legal and regulatory credits.5

4. In Ernesto Rivera, “Weak Control of Environmental ber 1994, modiﬁed by Judicial Reorganization Law #7728, Crimes,” La Nación, 31 December 2000, 4A, 5A. 16 December 1997, arts. 1, 2, and 3. 5. Organic Law of the Public Ministry, #7442, 25 Octo-

DISPUTE OVER PROTECTION 291 frankie chap 23 8/20/03 8:10 PM Page 292

Furthermore, Penal Procedural Code indicates ward the environment. This failure is the result that the Public Ministry will carry out penal ac- not so much of flaws in the management plans tion as established by law, exercise the pertinent themselves as of their lack of enforcement. and useful duties for determining the existence In response to numerous complaints, MINAE of an unlawful act, and will see to effective contracted and paid for forestry audits (mid- compliance with guarantees recognized by the 1990s) to evaluate “management plans” in the Constitution and the existing international and regions of Tortuguero, Osa, and Sarapiquí. The community rights in the country and the law.6 audits demonstrated that mechanisms and tech- Therefore, determination of the existence of a nical guidelines supporting forestry management crime and its investigation are the responsibility have, for the most part, been entirely ignored by of the Public Ministry and not MINAE. those companies that submitted plans and by MINAE itself. The following figures provide MINAE’S LACK OF ENFORCEMENT OF examples: 40 percent of the forestry files in the FORESTRY MANAGEMENT PLANS Sarapiquí region and 78 percent in the Tor- Costa Rica has suffered a drastic loss in forested tuguero area (both in the Caribbean lowlands) areas. Whereas 53 percent of the national terri- were incomplete for one reason or another. tory was forested in 1961, by 1977 that figure had Further, of those plans that had been approved, decreased to 31.1 percent. By late 1980, suppos- 40 percent in the Sarapiquí and 36 percent in edly as a result of the Forestry Law of 1969, Tortuguero were found to be incomplete. Other deforestation dropped from 50,000 ha per year problems were also discovered. Forestry officials to 22,000 ha. The figures are evidence that the are supposed to adhere to the “60/40 logging country has suffered an accelerated disappear- criterion,” whereby 60 percent or fewer of the ance of forests, with resulting ecological dam- inventoried trees in a given management plan ages. Soils appropriate for forest production may be cut but at least 40 percent must be left. were converted to agricultural and cattle areas, In the Sarapiquí, this criterion was not observed segmenting the land into small farms and for- in 60 percent of the cases; in Tortuguero, this est fragments. At the root of this destruction is figure was 44 percent. Cutting had exceeded the Costa Rica’s “forestry management plans,” which 60 percent limit at both locations. Finally, the have been strongly criticized and denounced forest inventories in both areas had statistical by environmentalists, communities, and non- errors in about 40 percent of the cases. governmental organizations. Management plans must also include maps The Forestry Law defines a forestry manage- showing where trees scheduled for cutting and ment plan as “a group of technical norms that residual trees are located. Further, these maps will regulate activities in a forest or forestry plan- are supposed to indicate the locations of promi- tation, a farm or part of it with the purpose of nent landmarks, boundaries, creeks, and roads exploiting, preserving, and developing the exist- to be used for accessing trees to be harvested. ing or intended tree vegetation, in accordance The auditors found that maps accompanying with the principle of rational use of renewable most management plans often did not agree natural resources that guarantees sustainability with the field reality. Common problems in- of the resource.”7 Despite its mandate to protect cluded extra roads cut into forested areas, trails the sustainability of renewable resources, these poorly sketched, and eroded soils due to exces- plans have completely failed in their duties to- sive access routes, which were often wider than

6. Penal Procedural Code, Law #7594 of 26 March 1996. In force since 1 January 1998, arts. 62 and 63. 7. Forestry Law #7575, art. 3, sec. e.

292 BUSTOS frankie chap 23 8/20/03 8:10 PM Page 293

permitted. Some trees were cut too low to the law, according to Jiménez, is thus the result of ground, which did not permit checking or MINAE’s authorization of incomplete manage- marked locations on maps. “Mother or seed” ment plans as well as its refusal to enforce them trees were often cut. Yet, despite these problems, once they have been authorized. all examined plans had been approved, which indicates little effort had been made to follow up ABSENCE OF EDUCATIONAL CAMPAIGNS on proposed plans and enforce environmental TO RAISE CONSCIOUSNESS laws when infractions were discovered.8 There is no better defender of the environment Once a management plan is approved, the than a citizen who is conscious of the impor- Forestry Division allocates special tags to be at- tance of preserving and protecting natural re- tached to the cut logs for identification during sources. Citizens who know and understand transport to and receiving at the sawmill. The laws do not pollute and do not allow others to question that has arisen with regard to these damage the environment. Thus, one of the main tags is, Do cut logs correspond to specific tags setbacks to an effective system of environmen- that were allocated according to the submitted tal protection in Costa Rica is the lack of educa- (and approved) plan? There is evidence that some tional campaigns in schools and for the general logs come from other sites, including protected population (see also chapters 19 and 20). Al- natural areas such as forest reserves and national though there is a law requiring environmental parks. The current tree-tagging stage leaves wide education in primary and secondary schools, the open yet another avenue for avoiding full com- Education Ministry has made only marginal ef- pliance with forestry law. forts to promote and advance environmental One reason that MINAE accepts and approves education. The same is true for MINAE’s em- incomplete and illegal management plans is barrassing program of environmental education that its representatives, like the environmental (see also chapter 19). Environmental issues rarely judges, are ignorant of environmental processes appear in the media’s daily agenda, being rele- and the laws that protect them. The following gated to the lowest order of news, after sports, excerpts are taken from dendrologist Quírico entertainment, events, and political news (see Jiménez’s article published in the national news- also chapter 20). In addition, weak environ- paper La Nación in 2000: “The lack of biological mental conscientiousness among Costa Rica’s knowledge of the forest as an ecosystem of those leaders does not lead to watchdog duties or edu- in charge of executing the field management cational media campaigns for the general popu- plans and those with approval power in the En- lation. Consequently, people commonly throw vironmental and Energy Ministry (MINAE) has trash on the roads, dump waste in rivers, burn made way for legalized deforestation.” He ar- trash in neighborhoods without a care for the gues further that “the political will to do some- impact on surrounding homes, hunt in protected thing about it has been missing. Therefore, areas, and generally disrespect, out of their own about 75,000 trees in protection areas have been ignorance, environmental laws. sacrificed in two years of logging as a result of unclear management plans that do not feature THE CASE OF ALTOS DE LA BONITA even the minimum requirements. Corrections Altos de la Bonita is a farm located on the south- and sanctions have not been carried out, and ern Pacific coast, facing Ballena Marine National impunity is prevalent.”9 Impunity before the Park in Puntarenas Province. The farm has steep,

8. Alpízar V. Edwin [forestry engineer, member of the 9. Quírico M. Jiménez, “Legalized Deforestation,” La Work Group on Forests and Forestry Products of FECON], Nación, 14 December 2000, 14/A. “Forestry Regencies,” Environmental Dialogues (trimester bulletin, June–August 2000).

DISPUTE OVER PROTECTION 293 frankie chap 23 8/20/03 8:10 PM Page 294

clay soils, unsuitable for agriculture, and was town the farm is a part of], and was granted completely covered with primary and secondary one.” Although the rural police has never had forests; the latter had not been disturbed for the authority to grant land-use permits, the ju- about 20 years. Because of its rugged slopes, the dicial authority carried out no actions to end the farm constituted an amphitheater for the park, destruction, despite the proprietor’s charges. with impressive scenic beauty provided by the Environmental deterioration and the consequent abundance of trees and the mist that often rose ecological damage continued to be sheltered by slowly from the sea. It was the wish of the owner judicial inertia. to preserve the farm in its natural state, without The investigation, taking place in March 1998, exploiting the trees. He had even requested a was done a year and a half after the claim was Certificate of Forest Protection, an incentive of- submitted and showed that the accused had re- fered by MINAE to proprietors of private forests. mained on the farm and had already converted In 1996, the owner found evidence of log- several hectares of forest into areas of perennial ging that had taken place on the edge of a trail agricultural products. Further, there were burned and immediately requested an inspection of the areas, pasturelands, trails cut with heavy ma- area from the ministry’s regional office, which chinery, a storage yard with piles of wood that then submitted its claim to the Environmental were being actively exploited, and land displace- Prosecutor. The claim, in brief, was as follows: ments that changed the property’s nature. All “The purpose is and has been . . . to maintain the of these infractions are detailed in the Forestry forest cover of the farm, especially to preserve Law and punishable with a prison sentence. the scenic value of said forests in front of Ballena The first trial took place on 6 November 1998. Beach, declared Maritime Reserve, and to pro- In the decision, the judge ignored the crimes tect the Ballena, Mercedes, and Piñuela creeks, verified through sight inspection, the report, the latter a source of drinking water for the small and the photographic evidence supplied with settlement of Las Brisas. The first two creeks the claim. Further, the judge demonstrated an empty into Ballena Beach and the latter into inability to establish a chronological sequence Piñuelas Beach, which is part of the Reserve. of verified facts despite several reiterations of The farm also protects Tortuguita Creek; it events. Although the usurper was sentenced for empties into the Tortuga River, itself a tributary illegal logging, the previous Forestry Law, #7174, near the mouth of the Balso River in a mangrove a law that had already been abolished but that area. Conservation of the farm’s forest cover is the judge believed to be in force at the time of essential for future tourist developments and the events, was applied. survival of the existing ecosystem.” The usurper got away with a sanction of 500 Without protection of the trees, the head- colones per day for 200 days (slightly more than waters would be in danger of drying up and $300), even when the evaluation of 1998, with torrential rains would drag clay sediments and the new Forestry Law in full force, proved that deposit them in the coral reefs of the marine the value of the extracted wood (to be used as park, destroying them. The motion was presented firewood) exceeded the imposed fine by more before the Penal Tribunals of Pérez Zeledón, than 11 times. Amazingly, the judge did not file #96-00430-195 PE, against Asdrúbal Cruz, order the accused to vacate the farm in question, who was charged by the owner of the farm with even though the usurper was not the legitimate usurpation and illegal logging. The accused owner. The accused was absolved of the usur- testified before the prosecutor’s office in Novem- pation charge by the in dubio pro reo principle ber 1996 and claimed that he was carrying out (innocent until proven guilty), even when he “agricultural ownership actions” (coffee crops himself admitted to having “worked” the farm and pastures for cattle) because “he requested without demonstrating any right of ownership. a permit from the Platanares Rural Police [the Finally, the Court of Final Appeal (cassation court)

294 BUSTOS frankie chap 23 8/20/03 8:10 PM Page 295

completely nullified the sentence and ordered a The request for discontinuance in favor of new trial. the accused was treated as a new case, despite Because of the grievous violations of due the efforts from the owner to attach the claim of process and the errors when handing down the the forestry inspector to the already existing pro- sentence, the judge was accused by the farm cess. The statements from the owner were reg- owner of the crimes of prevarication (issuing de- istered by the prosecutor in a format file known cisions contrary to law or based on false events), as “Interview with a Witness,” and the prosecu- bias, and noncompliance of duties. However, tor himself, once the statement was signed and the charge was dismissed, and a full discontinu- in the absence of the owner, committed the se- ance of the trial followed based on the decision vere infraction of adding text to the statement that “the fact that the judge ignored logical rea- to adjust it in accordance with the request for soning does not translate into a crime as long as discontinuance. intent is not shown.”10 The falsified document was presented be- When it was noted that damages continued fore the Public Ministry’s Prosecutor Investiga- on the farm, the owner again appeared before tion Unit, which absolved the prosecutor of all MINAE to request its involvement and to put an fault and responsibility. The investigators ar- end to the systematic destruction of the ecosys- gued that the Document Analysis Section of tem. The forestry inspector who visited the site the Judicial Investigation Bureau of Costa Rica reiterated the charge before the prosecutor of lacked the technical means necessary to deter- Pérez Zeledón. The prosecutor did not investi- mine in what sequence the plaintiff’s signature gate the facts or attach the claim to the process and the added writing occurred, although it was already in progress but limited his involvement possible through simple observation to see the to requesting a final discontinuance in favor of discontinuity between the last added part and the usurper and destroyer of the environment. the rest of the document.11 They even excluded In the analysis of events, the prosecutor main- the evidence obtained through the use of a foren- tained that “there is insufficient convincing sic study that left no doubt that part of the text evidence to require a trial.” When requesting had been added afterward.12 the absolution, the prosecutor declared that the In distress, the owner of the farm personally forestry inspector “did not witness the accused appeared before Elizabeth Odio, the environ- logging” (hard to come by in an evidence in- mental minister, and asked for her cooperation spection) and emphasized that no kind of out- in helping him to find justice and avoid further side investigation was carried out to determine ecological destruction. After claiming in exasper- that someone had executed the alleged logging, ation that “it was the one hundredth claim she even when the accused constantly admitted to en- received,” the minister harshly stated that “she gaging in these activities. In response, the prose- knew what had to be done,” which she failed to cutor announced that “the statements from the demonstrate by doing absolutely nothing. accused to the administrative authorities lack The second trial was held on 13 July 1999. proof value.” Again the judge absolved the defendant of the

10. The ﬁnal discontinuance in favor of Judge Franz of the General Prosecutor of the Republic at 8 A.M.on Paniagua Mejía was issued at 4 P.M. on 26 May 2000 by 22 September 1999, signed by the general prosecutor, Car- Judge Hanz Roberto Leandro Carranza of the Penal Court los Arias Núñez, and by the assistant general prosecutor, of Pérez Zeledón. Prosecutor José Efraín Sanders Quesada Jorge Segura Román. made the request for discontinuance. 12. Discontinuance issued at 7:47 A.M. of 2 November 11. The accusation against Prosecutor Ronald Carmona 1999 by Judge Emilia Ureña Solis of the Penal Court of González was ﬁled as Administrative Cause #207-98 of the Pérez Zeledón. Prosecutor Investigation Unit and resolved by Vote #177-99

DISPUTE OVER PROTECTION 295 frankie chap 23 8/20/03 8:10 PM Page 296

charges of usurpation and Forestry Law infrac- are more a case of “weak conduct” of the prose- tions, based on a supposed “agricultural owner- cutors, who commit crimes of omission and duty ship” over which a penal judge has no decision noncompliance, than of “weak claims” of the power because such ownership must be estab- plaintiffs. lished by agricultural judges. The Court of Final Appeal (Cassation) once again nullified sentenc- USURPATION AND LOGGING IN ing of the farm invader. Finally, in a third trial, LANDS DONATED TO THE STATE on 20 September 2000, more than four years af- The second case illustrates the Costa Rican state’s ter the claim was initially presented, the usurper inability to protect areas guaranteed absolute was sentenced to one year in prison. However, protection by the national park system. The in- the old abolished Forestry Law was applied, and capacity of judicial organizations allows for un- a fine of 500 colones per day for 200 days was challenged destruction, but the government is levied for the illegal logging, for a grand total of also guilty of negligence in its obligations to 100,000 colones (a little more than $300).13 donations from other countries of tremendously Final restitution of the property to its rightful valuable ecosystems. owners was ordered on 12 February 2001. How- The case is more disheartening considering ever, there is a sad side note: As a result of the that part of the funds used to finance the pur- ignorance and total negligence of the authorities chase of lands donated to the Costa Rican state in charge of serving justice, there is probably ir- were collected by concerned students who reversible destruction in a high-risk area that is wished to contribute to forest conservation. The part of a biological corridor and serves as buffer Nepenthes Conservationist Group from Den- zone for a marine national park. After four years mark collected and donated more than $300,000 and three trials in court, during which the squat- so that the Neotropical Foundation could pur- ter continued to cut trees, plant coffee, and keep chase and afterward donate lands to the Costa livestock, the court finally recognized the prop- Rican state. These lands are part of the so-called erty’s legal owner and penalized the squatter with Tortuguero–Barra del Colorado Biological Cor- a jail sentence and a fine to cover the damages. ridor and were to be added legally to the Tor- However, in a bizarre twist of fate, the squatter tuguero National Park to be forever protected. died shortly after this last trial, leaving no one to Before the donation, the Neotropical Foun- pay for the damages to the owner and to the en- dation fought for five years before the Justice vironment, not only for the trees cut but also for Tribunals to expel illegal squatters who had in- the soil sediments that will now be transported vaded about 3,000 ha. Once the lands were de- to the coral reefs of Ballena Marine National Park. livered to the government of Costa Rica, MINAE The case in question clearly identified the was in charge of their administration and mon- possible culprit and provided substantial mate- itoring. Although the Nepenthes Foundation no rial evidence of environmental damage. The real longer had any power over these lands, some problem begins with the obvious lack of knowl- of its members visited the area in early February edge on the part of the judges, whose ignorance 2001 and were able to ascertain illegal actions of the environment and its protection and ele- such as wood extraction, use of land for produc- mentary knowledge of agricultural ownership, tion activities, and shifting of farm limits that and how it is exercised, are overwhelming. The encroach on the officially established limits. Rep- deficiencies in Costa Rica’s environmental law resentatives of MINAE were warned of the proven

13. Sentence #215-2000, passed by Judge Jaime Hernández Granillo, South Zone Court, Pérez Zeledón, at 12:30 hours of 20 September 2000.

296 BUSTOS frankie chap 23 8/20/03 8:10 PM Page 297

irregularities and inspected the area on 19 March commit these acts only when law enforcement 2001. They discovered people living in ranches, representatives do not carry out their duties and evidence of illegal wood extraction, and indica- when the government itself neglects the lands it tions of the existence of a sawmill in the area. In is called on to protect. All the laws in the world addition, there were planks tied together in the will be worthless unless there are effective means Tortuguero River channels, indicating that rivers to apply them. were exploited for wood transportation and there- Recommendations are offered as possible fore that the perpetrators were eluding possible solutions to this problem: monitoring control on the roads. The evidence further showed that logging was 1. Costa Rica must comply with the obligation systematically carried out for more than a year. it assumed through the signing of the MINAE denounced before the Public Ministry American Convention on Human Rights. those responsible for invading the area, and the In accordance with article 25 of said con- drawn-out process in the tribunals began once vention, Costa Rica must create “a quick again. MINAE’s own excuse for overlooking the and simple process or any other effective damages was a lack of personnel and transporta- process before competent judges or tri- tion means, excuses exploited with exasperating bunals that provides protection against regularity. actions that violate fundamental rights Representatives of the Nepenthes Conserva- recognized by the Constitution, the law, or tionist Group have stated that “the government the present Convention, even when such [of Costa Rica] seems incapable of protecting the violations are committed by people exercis- national resources of the Costa Rican people.”14 ing their official functions.” To date, there A question remains: What would have happened is no request for an injunction of constitu- if members of the Neotropical Foundation had tional violation against jurisdictional ac- not visited the donated lands? There may have tions. Norms of the Organic Law of the been no notice of the invasion and flagrant Judicial Power, which prohibit acts in viola- destruction of a national park that is supposed tion of the Constitution and the precedents to be constantly patrolled by park rangers. and jurisprudence of the Supreme Consti- tutional Court, are only “paper” norms without established sanctions. There must RECOMMENDATIONS be an awareness that the right to a healthy These examples demonstrate that the accelerated and ecologically balanced environment is a deterioration of the Costa Rican environment constitutional-level norm. and its biodiversity is a product of the impunity 2. Judges presiding over environmental cases with which those who abuse and destroy natural must have a more thorough knowledge of resources are allowed to act. The farmer who the environmental field. It is essential to takes over a plot and changes the land use, com- know the difference between a primary panies that contaminate aquifers, and the log- forest, a secondary forest, and a forestry ging entrepreneur who hides behind an alleged plantation, what levels of biodiversity each “management plan” and uncontrollably destroys contains, and how each affects its eco- forests are only part of the problem. They can system. Judges must be able to differentiate

14. Information given by Sonia González, administrative director of Neotrópica Foundation, and newspaper clip: Magaly Batista, “Danes Indignated with the Country,” Al Día, 2 April 2001.

DISPUTE OVER PROTECTION 297 frankie chap 23 8/20/03 8:10 PM Page 298

between a native, endangered, and com- coordinating them as an organic and con- mon tree species in order to know what gruent whole. forests are most in need of protection. 4. The Education Ministry and MINAE must Finally, they must have a basic understand- make greater investments and commit- ing of ecosystem theory, which describes ments to educate the people of the country the close relationship between soil, water, as to the importance of the environment air, and wildlife, to defend the environment and its relationship to their quality of life— properly. on an individual and community level. In 3. All institutions in charge of protecting the addition to preserving natural areas, the environment must be better integrated. The general topics of clean air, water, pollution, effective protection of biodiversity will not and improved public health must be be achieved by increasing the number of stressed in all areas throughout the laws and penalties but by integrating and country.

298 BUSTOS frankie chap 24 8/20/03 8:11 PM Page 299

chapter 24

The Policy Context for Conservation in Costa Rica

MODEL OR MUDDLE?

Katrina Brandon

orldwide, a debate is under way over from examining the cases in this volume about Wwhether we can protect biodiversity in the politics and policies surrounding biodiver- situ, whether we should bother to try, and sity conservation; they not only present the situ- whether these efforts should include any areas ation as it exists within Costa Rica but also facil- that limit human uses. Critics have claimed that itate an understanding of the degree to which protected areas are hard to manage and there- Costa Rica is, or is not, a model for other tropi- fore we should not even try, that they are too cal countries. small to make a difference, or that they are so- This chapter builds from the other chapters cially unfair and unjust (Brechin et al. 2002). in this book and examines Costa Rica’s leader- Although these criticisms are mostly voiced by ship as a “model” for conservation in other social scientists, alarmingly, some ecologists and countries, expanding outward from dry forests policy makers, often those who work at levels far to the broader issues of conservation in Costa from the ﬁeld, have echoed them (Sayer et al. Rica and how conservation in Costa Rica is 2000). linked to larger debates concerning biodiversity Within this debate, Costa Rica often features conservation worldwide. Although conservation prominently and ironically—as an example of a can, and must, take place in many ways and at country that has wholeheartedly embraced sus- different scales, this chapter focuses on pro- tainable development with protected areas as tected areas because they are the organizing the centerpiece or as a country with policies that nexus for biodiversity conservation efforts and fail to support conservation and parks too small offer the most concrete examples of how con- to be ecologically viable. Much can be learned servation actions occur.

299 frankie chap 24 8/20/03 8:11 PM Page 300

entered a crisis phase. Although many parks COSTA RICA: A BRIEF REVIEW OF had been created legally, there was no money to INNOVATION WITHIN CONSERVATION compensate landowners for land expropriated to In this chapter I identify the relevant contextual create them and little money left to manage ex- factors of scale, history, and social forces that isting parks, and the high inﬂation rate meant have affected the development of both Costa that the buying power of colones was substan- Rican conservation policy and other policies that tially reduced. Finally, International Monetary have affected Costa Rican conservation. Costa Fund requirements led to hiring freezes, leaving Rica is held up as a leader in three areas: many parks with few managers or staff members. At the same time, threats to parks increased . Creating strong parks with a clear mission dramatically from multiple sources. Certain gov- . Linking protected areas with surrounding ernment ministries wanted to open parks to lands logging to generate cash, and the land titling . Politics, policies, and management affecting and settlement agency saw parks as good areas parks for settlement of landless peasants. Landless peasants and unemployed banana workers also CREATING STRONG PARKS saw parks as commons areas available for colo- WITH A CLEAR MISSION nization, and Corcovado National Park suffered Unlike remote areas in many parts of the world a serious invasion by gold miners (Christen that have been protected, few, if any, areas of 1994). Costa Rica could be considered remote—even The pressures on Costa Rica’s tropical dry within the past 50 years. By the mid-1900s, most forest in the northwestern Paciﬁc region of the of the land suitable for agriculture in the central country were intense, and forestland was rapidly valley had been cleared, and settlers had begun being converted. By the mid-1980s, many con- to expand up the mountain slopes. In response servationists were ready to give up on this area. to this expansion of the agricultural frontier, Between 1979 and 1992, in the Liberia region, laws were passed as early as 1913 to protect 119,712 ha of primary and secondary forests were wildlife and limit expansion into certain areas transformed into pastures. During that time, such as those around Poás Volcano. Other pro- Dan Janzen and Winnie Hallwachs acquired a tected areas created prior to the 1950s generally 101,200-ha ranch and farm in the Guanacaste emphasized scenic, historic, and recreational region and began the process of restoring it to value—not biodiversity. wilderness (Janzen 1988). Costa Rica’s modern system of parks started It is vital to understand the policy context in in 1969, when a major goal of the government which the parks were created. Deforestation was was to establish new parks and protected areas promoted both directly and indirectly by a set rapidly, increase the size of existing areas, and of government policies that favored conversion increase support to the National Parks Service of forests to export-oriented commodities such (Janzen 1983). A 1970 law set aside slightly more as coffee, cattle, cacao, sugar, and bananas. The than 10 percent of the land in strictly protected following statistics on deforestation are com- areas and an additional 17 percent in “buffer monly cited:1 zones” or forest reserves. Park creation continued during the 1980s, but a severe economic . Deforestation levels in Costa Rica were the crisis struck the country, and the park system highest in Latin America.

1. See World Bank (2000: 4–6) for a more complete discussion of the numbers, sources, and discrepancies.

300 BRANDON frankie chap 24 8/20/03 8:11 PM Page 301

. Between 1979 and 1992, primary forests (15%). In 1987, the Arias administration pro- declined by 38 percent and secondary forests posed creating a national system of conservation by 8 percent. areas (the agency now known as Sistema Na- . Rates decreased from approximately cional de Areas de Conservación [SINAC]) to in- 46,500 ha per year in the 1950s, to less tegrate protected area management and decen- than 31,000 ha per year between 1979 and tralize much of the decision-making power to 1989, to about 16,000 ha per year in 1997 regional levels (Umaña and Brandon 1992). The (World Bank 2000). country was divided into nine Regional Con- servation Units now called ARCs (Regional Protected areas within Costa Rica were estab- Conservation Areas) and sometimes called lished amid levels of deforestation and popula- “megaparks.” The current number of adminis- tion growth that were among the highest in the trative units is now 11. Each ARC is composed world. Annual population growth rates reached of three land-use categories: 3.8 percent per year into the 1960s. Although . Core areas subject to absolute protection, the exact numbers are open to debate, it is clear such as national parks that tremendous changes in land use have taken . place in Costa Rica in the past 50 years. Buffer zones, or multiple use areas, often As the economic crisis improved in the 1990s, forest or indigenous reserves, although the number of officially protected areas contin- some are highly populated or under ued to increase: from 16.8 percent in 1990, to agricultural use 23.8 percent in 1997, to 24.8 percent in 1999 . Intensive extraction zones, such as agricultural (World Bank 2000: 12). Strictly protected areas lands (International Union for the Conservation of Nature [IUCN] categories I–V) amount to 14.2 Conceptually, the conservation objectives of the percent of the national territory in 85 sites pro- ARC system were to tecting approximately 723,000 ha (World Re- . improve intersectoral coordination for each sources Institute 2000). The mission of Costa region; Rica’s strictly protected areas overlapped with . increase local participation in resource plan- the IUCN’s conceptions of “strictly” protected ning and use; areas—recreational uses are allowed, extractive uses are not. There was an emphasis on con- . create vertical linkages (between local people serving biodiversity within parks, since most and government) within each region; and parks were established specifically to help stop . decentralize decision-making authority to the transformation of the land and everything regional levels. on it to other uses. All the ideas introduced under the ARC sys- LINKING PROTECTED AREAS tem were visionary at that time, and there was WITH SURROUNDING LANDS strong support from both international conser- Recognition is increasing that although protected vation organizations and the emerging eco- areas should remain the cornerstone for the pro- tourism market. These ideas sought to mini- tection of biodiversity, “requiring them to carry mize threats to parks from surrounding areas, the entire burden for biodiversity conservation is increase local support for conservation, and a recipe for ecological and social failure” (Bran- prevent large-scale threats by government agen- don et al. 1998: 2). Costa Rica was one of the cies while the system was in the early planning countries that first tried formally to link with stages. surrounding lands the absolutely protected land Costa Rica’s emergence as an ecotourism (14%) and land in private hands but protected destination supported these links in two ways.

POLICY CONTEXT FOR CONSERVATION 301 frankie chap 24 8/20/03 8:11 PM Page 302

First, the park system became the focal point for tion relative to other government sectors, such nature-based tourism, and “between 1990 and as agriculture, forestry, mining, or public works. 1997, 38% of tourists visited national parks.” At Recognizing this, the government consolidated the national level, revenue from coffee, cane, a set of functions and in 1986 created a new cattle, and bananas was replaced by revenue from ministry called the Ministerio de Recursos Nat- tourism. By 1999, tourism revenue reached urales, Energía y Minas (MIREMEM) by trans- $950 million, or 9 percent of gross domestic ferring the Energy and Mines sections from the product (World Bank 2000: 23). Second, these Ministry of Industry, Energy, and Mines and links made private investment in ecotourism at- the National Park Service (NPS) and the General tractive to landowners with forests, particularly Forestry Directorate from the Ministry of Agri- those with lands adjacent to parks. Private re- culture. Although it took nearly four years for serves adjacent to public areas extend ecosystem the National Assembly to approve the creation functions, expand habitats, and help attract of this ministry, MIRENEM operated in a de tourists to areas, generating revenue through facto manner during the Arias administration. multiplier effects (Brandon 1996; Langholz et It is now known as the Ministry of Environment al. 2000). Privately managed reserves now total and Energy (MINAE). about 5 percent of the country’s land area and Costa Rica has been viewed as a model also 25 percent of the forested lands (World Bank because of its innovation in financing arrange- 2000: 12–13). ments to support conservation. In 1978, the government created the National Parks Founda- POLITICS, POLICIES, AND tion, an independent, private nonprofit founda- MANAGEMENT AFFECTING PARKS tion dedicated to the planning, management, In the past 20 years, Costa Rica has also under- protection, and development of national parks taken a broad-based set of political actions that and reserves. It attracted donations from inter- were intended to support conservation. It has national conservation organizations and govern- been considered a leader in a wide array of poli- ments to support parks and played an important cies and actions that, broadly characterized, role in essentially privatizing the financing of supported natural resource management, con- the ARC system. Channeling financing through servation, and biodiversity. Key policy actions the foundation offered increased flexibility and have included responsiveness, and the financing was not subject to governmental constraints, such as the con- 1. consolidating power for conservation by ditions imposed by the International Monetary creating a new ministry; Fund. Costa Rica began an intensive interna- 2. creating new financing mechanisms to tional fund-raising effort, whose success relied support conservation; heavily on the participation of foundations, non- 3. establishing the in-country technical capac- governmental organizations, and private con- ity to catalog Costa Rica’s tremendous bio- servation groups. The country has also initiated diversity and identify potential uses; and a program of payment for environmental serv- 4. creating market-based incentives to support ices, which include carbon sequestration, sus- conservation. tainable logging, ecotourism, hydroelectric power, water supply, bioprospecting, and payment in In many Latin American countries, the great- recognition of option and existence values (World est threats to parks and protected areas are from Bank 1993). These actions are briefly described weak, inconsistent, poor, or conflictive govern- in the following paragraphs. ment policy (Rudel and Roper 1996, 1997; Bran- The country used debt-for-nature programs don et al. 1998). These weaknesses are largely (debt swap) to turn a serious national problem due to the low level of import given to conserva- into new opportunities for conservation. In mid-

302 BRANDON frankie chap 24 8/20/03 8:11 PM Page 303

1988 Costa Rica proposed a specific debt-for- 5. create incentives for improved forest man- nature swap with the Swedish government for agement (World Bank 2000). the completion and endowment of Guanacaste National Park, a major conservation project in Payments for these services are from taxes on northern Costa Rica. Swedish students and pri- fossil fuel use; from an implicit resource trans- vate conservation groups supported the project, fer from urban to rural sector users; and from a and a total of U.S. $3.5 million was allocated for sector that imposes high costs on the economy, debt conversion and targeted at Guanacaste in terms of foreign imports and pollution costs, National Park, out of a donated total of nearly to a sector that broadly provides services both $15 million, significant financing for the largest nationally and internationally. block of tropical dry forest in the country. Other debt swaps were channeled through the National RHETORIC TO REALITY: Park Foundation to support a specific ARC. IMPLEMENTATION OF POLICY REFORMS This is an example of how the ARC system was effectively privatized to facilitate fund-raising Most of the rhetoric about Costa Rica portrays it efforts and improve the regional disbursement as a progressive country, politically committed to of funds for conservation. sustainable development. The previous section Through the National Biodiversity Institute outlined some of the actions that have led to this (INBio), created in 1989, the country created the perception. This section briefly addresses how technical capacity to conduct an inventory of life well this set of actions has been implemented forms in Costa Rica and is analyzing the central in reality. and potential contributions of biodiversity to society. A deal between INBio and Merck Phar- THE CONTEXT AND MISSION OF maceutical Company publicized the value in ex- PARKS WITHIN COSTA RICA ploring the country’s biodiversity in pursuit of Understanding how best to conserve tropical dry marketable products, an area sometimes called forest in Costa Rica can be only partially in- ecoprospecting. formed by ecological analyses, which provide Costa Rica has also been an international information on what to conserve and where to leader in market-based mechanisms that sup- conserve it. What is omitted by such ecological port conservation, principally through payments analysis is how to conserve it, which is not an for environmental services. In 1996, Costa Rica ecological issue but a social and political one adopted a program for payment of environmen- (Redford and Sanderson 1992; Brandon et al. tal services to 1998). The historical forces and social actors that shape the local context influence parks and 1. eliminate subsidies that acted as perverse the management they require. Just as biodiver- incentives for resource depletion; sity is scale-dependent, so too are the social 2. collect revenue from users of environmen- forces that shape how biodiversity is used, man- tal services to pay for them; aged, or destroyed. In any country and at any scale of analysis, it means that one must look at 3. eliminate the dependence on the govern- geographical scales (site to regional to national) ment for forest sector subsidies and get and temporal scales (historical to present to the the forest sector to be viewed as providing future) to understand the array of policies that services beyond wood; affect biodiversity conservation. These policies 4. fix the unequal distribution of benefits include ones made at international levels (poli- from many environmental services, which cies such as the Kyoto protocol) that in turn accrue nationally and internationally, not shape those decisions made at regional or pro- to the owner of land and forests; and vincial levels. Understanding these contexts—of

POLICY CONTEXT FOR CONSERVATION 303 frankie chap 24 8/20/03 8:11 PM Page 304

scale, history, and social forces—is essential to based on a meaningful ecological framework— developing lasting management approaches for they were opportunistic expressions of what was a particular site. possible at a given moment in time. In looking Many of the protected areas established world- at Costa Rica, it is clear that most of the parks wide in the past 30 years have been established within the country were established with the re- in areas that are geographically remote or viewed alization that if any area was to be spared from as a political backwater with little power. Parks chainsaws, it would have to be protected quickly. were created in areas viewed as places with “noth- In such a context of rapid change, this was the ing there,” and the act of creating a park con- only way that biodiversity conservation could sisted of little more than drawing lines on maps have proceeded. Yet this context of creation has to satisfy the clamoring of international conser- left many problems for Costa Rican parks— vation organizations. Yet consolidating the man- from both political and biodiversity conservation agement of these types of areas is generally perspectives. It has meant that parks were not simple; the threats are locally based and often at established low levels. Management can often be initiated . with local support in a given area but to by compensating people for lost access to land counter quickly pressures exerted at local and resources, demarcating boundaries, educat- levels; ing people about conservation, and beginning . what is more conventionally thought of as park with sufficient area to protect the range of management—the array of actions undertaken species that would be left within as parks inside the park. The legacy of how a park is es- became islands; tablished influences the present-day manage- . as part of any regional land-use framework ment context (see Brandon 1998). or strategy; The social context in which Costa Rica’s . with consideration to corridors or elevational park system was created included the following gradients; or elements: . to represent the full spectrum of biodiversity . Rapid declaration of parks to protect areas of (representation) to the greatest degree 2 land from conversion to other uses possible. . Declaration of parks on the basis of what was All of this matters a great deal in understanding socially and politically feasible, not on the where Costa Rican parks, particularly those con- basis of ecological criteria serving tropical dry forests, fit into the inter- . High levels of social hostility toward parks, national debates on parks. Although many view resulting from expropriation without legally Costa Rica as an exemplar of park creation, in required compensation fact, in biodiversity terms, it represents a worst- . High levels of pressure on land proposed for case scenario. Costa Rican parks cannot serve as parks to be converted to other uses—both at a model for other countries—except those with national levels for foreign exchange and at a similar set of social pressures: rapid land-use local levels for agriculture conversion, strong demographic pressure, rapid fragmentation of habitat, and relative ease of ac- The parks that were established, then, were cess to most parts of the country. When viewed created amid great social, economic, and de- from the context in which they were created, mographic pressure. They were not developed Costa Rican parks are relatively free from threats,

2. For excellent discussions of this, see Palminteri et al. (1999) and Powell et al. (2000, 2002).

304 BRANDON frankie chap 24 8/20/03 8:11 PM Page 305

particularly large-scale threats. But it is also TABLE 24.1 impressive that in one generation, a relatively Deforestation Rates in Costa Rica’s National Parks short amount of time, most Costa Ricans have and outside Its Protected Areas come to value the park system. Costa Rican parks in outside demonstrate that even areas established under national protected difficult circumstances can function, if not parks areas flourish (Sanchez-Azofeifa et al. 1999). This (% per year) (% per year) is consistent with recent findings by a Conser- vation International team studying 93 parks (in 1976–86 0.56 3.6 22 countries) that are more than 5,000 ha each, 1986–91 0.21 2.8 more than five years old, and subject to human pressure (Bruner et al. 2001). The study found 1991–95 0.16 3.2 that despite high levels of pressure, more than 80 percent of the parks had as much natural veg- Source: Adapted from World Bank (2000). etative cover today as they did when they were first established, and a substantial percentage forest reserves (which have a less stringent pro- had more. There was also a notable difference tection status). What this demonstrates is that in deforestation levels between parks and sur- the parks have been relatively effective at halting rounding areas. The analysis of Costa Rica and land-use conversion, whereas the forested areas this recent Conservation International study al- around them have been converted at high rates. low room for cautious optimism on the effec- This marked difference in deforestation rates tiveness of parks. also reflects the difficulties in making the kinds of links desired at the regional level. LINKING PARKS AND SURROUNDING LANDS Overall, the lack of available forests left to log The model of how each of the regional conser- has been a greater limiting factor for the con- vation areas would operate is visionary. The idea version of forestland than any integrated plan of linking protected areas with their surround- under the SINAC. This volume contains nu- ing lands in order to improve intersectoral coor- merous examples of activities taking place in or dination, increase local participation in resource near a park that are clearly incompatible with the planning and use, create linkages between local conservation uses within. For example, the people and government, and decentralize deci- Arenal-Tempisque Irrigation Project (see chap- sion-making authority to regional levels is one ter 21) is clearly not a good neighbor for the park. that conceptually leads to better conservation. Other attempts to provide linkages between However, although the SINAC systems and each parks and surrounding areas have gone astray of the ARCs appear to be functional on paper, or been misinterpreted or have the potential to the system does not yet appear to have made become problematic. For example, chapter 21 substantial gains nationwide. in this volume addresses the use of “controlled Part of the difficulty lies with stopping the grazing” as a “novel technique to generate eco- existing forces leading to land-use change. As nomic benefits for local communities providing noted earlier, parks in Costa Rica were in large sustainable development.” Few have an image part created to be somewhat “separate” or “dif- of cattle as a discriminating tool for weed man- ferent” from the lands that surrounded them. It agement, fire control, and forest regeneration. is worth exploring just how significant these It may well be that cattle can perform some of differences are, as shown in table 24.1. these services. However, decisions that should Table 24.1 shows that there has been a marked be rooted in science and range management difference in deforestation rates between na- quickly and easily become clouded when scien- tional parks and surrounding areas, including tists and managers begin to focus on providing

POLICY CONTEXT FOR CONSERVATION 305 frankie chap 24 8/20/03 8:11 PM Page 306

tangible, local, direct benefits to nearby residents. development of a national-level technical capac- Since providing such benefits has become part ity for conservation, and changes in forest sector of the mission of each of the ARCs, making policies are all examples of actions taken to sup- decisions about park management quickly be- port conservation. At the other extreme are poli- comes entangled with making decisions that cies that undermine conservation efforts, such are popular at decentralized levels. Unfortu- as those promoting large-scale development ac- nately, this often leads to weak outcomes for tivities or road construction, both of which often conservation—and to inequitable distributions pose the most significant threats to biodiversity. of benefits, as well (see Brandon 1998, 2000). A review of what Costa Rica intended by the ac- This volume also contains other examples that tions described earlier and what actually took highlight the failure of conservation to be ac- place when they were implemented reveals a vast corded importance equal to that of other sectors, discrepancy. Although Costa Rica established even within the ARC system. what could reasonably be called a broad set of Increasingly, conservationists are identify- “model” policies to support conservation, there ing the kinds of uses that are likely to conserve have been substantial weaknesses in how these biodiversity outside protected areas and act as policies have been implemented in practical sound buffers. Examples of areas that encour- terms. age consumptive and extractive uses but that The creation of an environmental ministry can be “good neighbors” to parks include shade- has been an important action in attempting to grown coffee farms, organic cacao plantations, elevate and integrate many of the diverse agen- agro-ecosystem models, and extractive reserves. cies that affect conservation. Simply putting these Clearly, there will be a need for what Boshier et diverse agencies under one roof, however, has al. (chapter 16) describe as circa situm efforts— not led to better intergovernmental coordination even if those efforts focus on species conserva- within the ministry. It has also not given enough tion, as opposed to ecosystem processes. Such weight or political capital to MINAE for it to be efforts are vital within the mosaic of landscape a progressive advocate for change. uses, but they do not negate the need for pro- The World Bank concluded in a recent re- tected areas (Redford and Richter 1999). Yet view: “Sustainable development is still not an we lack examples of functional, well-funded sys- official national objective. The government does tems, such as the SINAC, in which decisions are not give support to forest management and con- made about landscape uses within regions in servation because it has not fully internalized ways that preserve biodiversity and allow devel- the importance of forests to production, exports, opment, at appropriate levels of intensity and at employment, and a sound environment” (World appropriate scales, to take place. The SINAC sys- Bank 2000: 61). This is largely due to politics tem and ARCs are an attempt at this, but they and power. For example, under President Rodri- are still a long way from demonstrating that ef- guez’s administration, MINAE lost significant fective integration of parks and surrounding power at the national level when the president areas is possible. reduced the overall number of ministers and assigned the second vice president dual respon- POLITICS, POLICIES, AND sibilities: that of second vice president and head MANAGEMENT AFFECTING PARKS of MINAE. This suggested that MINAE, and the Policies can affect parks directly in at least two functions under it, were viewed as less worthy ways. Policies designed to support conserva- of representation than other sectors of gov- tion affect them favorably. Costa Rica’s creation ernment. Furthermore, environmentalists had of an environmental ministry, implementation loudly complained that the second vice presi- of policies to support financing for conservation, dent lacked environmental credentials and was

306 BRANDON frankie chap 24 8/20/03 8:11 PM Page 307

antienvironment. For example, despite the rhet- resulting in the loss of capable people from oric given to ecotourism as a basis for supporting the public sector. local-level development, government policies in 2. It created an impression that conservation fact support large-scale projects and put numer- should not be a mainline governmental ous obstacles in the way of small-scale tourism responsibility, since it was not treated as development (Honey 1999). Such political dis- such. cussions have concrete ramifications that reach 3. It created significant funding discrepancies all the way to the site level, as evidenced in between the popular ARCs, which were chapter 15. well supported, and those ARCs with less Costa Rica has been viewed as a model in its cachet. innovative and creative financing arrangements, such as debt swaps, international fund-raising, 4. It created a rapid influx of funds that raised the use of the National Parks Foundation and local expectations and made people feel that special relationships between donors and each they should directly benefit from the con- ARC, and a system of payments for environ- servation sector. mental services. Yet each innovative action has had difficulties when applied. Debt swaps were The legacy of these problems, and the relatively a great idea, but the market for them proved weak political power that MINAE has, have fairly short lived. Most were earmarked for a meant that even when the policy framework for specific park or ARC. Costa Rica has benefited conservation has been great, implementation has from high levels of international fund-raising, been weak. which was largely channeled through the Na- The most glaring current example of this tional Parks Foundation. However, there were a discrepancy between policy and implementa- number of problems with channeling support tion is the 1996 law on payment for environ- for conservation in this manner. First, obtaining mental services from fossil fuel taxes. The law external financing for conservation is reason- allocates a third of the revenues from taxes on able when the premise is that most benefits are fossil fuels to conservation. For one 15-month enjoyed internationally. But such a premise does period, $25 million should have been generated little to instill national or local-level responsibil- for conservation; however, the government used ities or appreciation for the benefits that indeed the money for other things, and only $7 million are recognized at those levels. It also under- had been paid out to forest owner associations mines arguments that biodiversity has high in- (World Bank 2000). Costa Rica’s payments for trinsic value and that a range of localized con- environmental services, joint implementation servation services are performed by forests and for carbon sequestration, and the World Bank– wild areas. It sets up an expectation that external funded “Ecomarkets” project are all examples of support for conservation is the responsibility of innovation. However, only when the implemen- the international community. This line of think- tation of these three innovations matches the ing, along with the high levels of funding given rhetoric will Costa Rica truly be providing lead- to the foundation for conservation, had the fol- ership in these areas. lowing consequences: Costa Rica’s actions to develop a technical capacity for conservation are in fact exemplary. 1. It created a “shadow” civil service whereby However, the high technical capacity can be people hired by the foundation for conser- traced as much to the abolition of the army and vation were paid much more than “real” the investment in education as to other actions. civil servants, weakening the position of en- With the creation of INBio, a wide variety of tities such as the National Park Service and local people have been trained in parataxonomy;

POLICY CONTEXT FOR CONSERVATION 307 frankie chap 24 8/20/03 8:11 PM Page 308

helping to create local support for conservation others, combined with a serendipitous drop in within the ARCs. The country’s technical capac- cattle prices and out-migration to the central val- ity has also been bolstered by the huge number ley, have taken a huge amount of pressure off the of students and researchers from other coun- Guanacaste region for cattle ranching. This has tries who have come to Costa Rica annually un- made it possible to take a small park and expand der the auspices of the Organization for Tropical it upward and outward so that it includes eco- Studies. As a result of their presence and the logical gradients from the lowlands to the moun- technical expertise they have brought, much of tains and is a major wilderness area rather than the research on the tropics has been generated a fragment. This example, which is of extreme in Costa Rica.3 Other training that supports importance in tropical dry-forest conservation, conservation has taken place in the ecotourism highlights the need to start with parks as an- sector—this is one of the best ways in which chors and then think at different scales of both local people have become directly involved and space and time. Instead of giving up on conser- trained in areas that support conservation. vation, even on a small scale, we must look at the contributions each park can make within the scale it occupies. From there, we must look to CONCLUSION the future and think what might be possible if Costa Rica has won strong international acclaim significant changes occur in an area. for the fundamental action of creating a national Such an understanding of scale, on geo- system of parks. Yet, as this chapter demon- graphical and temporal levels, is important in strates, there is little in the creation of these both ecological and policy analysis. Many of the parks that serves as a model, other than the find- policies that Costa Rica has implemented look ing that boundary demarcation and enforcement, good in a particular place, at a particular level, in a context in which land tenure is respected, or within a particular time period (see Cuello et can stop the rapid transformation of sites. The al. 1998). As a result, innovative policies encour- design of Costa Rica’s parks was based on what aging decentralization, which may work within was politically feasible, not on what was ecolog- one ARC, are captured by local-level politics and ically best. What remains, then, is a series of distorted in another ARC. Creative financing islands in the midst of a landscape undergoing mechanisms defined for one park or ARC fail to transformation. Yet even those parks that are materialize for others, leaving them as the des- islands or fragments—and virtually all the un- titute sibling ARCs. Creative ideas in one area linked parks are small enough to suffer the prob- make for bad policy in others. So when Dan lems of fragmentation—have value. Although Janzen decided to allow dumping of orange by- we may have lost keystone species in some frag- products from a juice factory in the Guanacaste ments, we have learned that something valuable Conservation Area in exchange for payments for has been lost, and we are seeking to “rewild” and ecosystem services (the by-products would de- restore it (Soulé and Terborgh 1999). grade and create a well-fertilized area enhancing The impressive gains made in the past 20 forest regeneration), problems quickly arose. years in the Guanacaste ARC demonstrate this. How can a good idea such as this one be made The indomitable efforts of Dan Janzen and to work across the system? What if oil palm or

3. The Organization for Tropical Studies began its activities in Costa Rica in 1963, and these helped provide a research foundation for many of the conservation actions taken.

308 BRANDON frankie chap 24 8/20/03 8:11 PM Page 309

coffee-processing wastes are dumped in an ad- REFERENCES jacent park for payment? What about pesticide loads? Who will speak for the park and ensure Barrett, C. B., K. Brandon, C. Gibson, and H. Gjert- that what is dumped is sound? Policies can often sen. 2001. Conserving tropical biodiversity amid weak institutions. BioScience 51:497–502. ensure that damage is not done, but they are less Brandon, K. 1996. Ecotourism and conservation: A good at encouraging good ideas at local levels. To review of key issues. Environment Department discourage misuse, they must also discourage Paper 33. Washington, D.C.: World Bank. 70pp. what might, in some cases, be good ideas. ———. 1998. Perils to parks: The social context The difficulty of scale arises again in looking of threats. In Parks in peril: People, politics, and protected areas, ed. K. Brandon, K. Redford, at innovative national policies such as payments and S. Sanderson, 415–39. Covelo, Calif.: Island for ecosystem services. Despite the creation of Press. the ARC system and the intention of linking ———. 2000. Moving beyond integrated conser- parks with buffer areas, ARCs have had few vation and development projects (ICDPs) to concrete ways to do this. The law allowing fossil achieve biodiversity conservation. In Tradeoffs or fuel taxes to be directed toward conservation synergies? Agricultural intensification, economic development and the environment, ed. D. R. Lee could prove to be the best way of implementing and C. B. Barrett, 417–32.Wallingford, England: ARCs—it provides immediate financing for for- CAB International. est protection in recognition of the services they Brandon, K., K. H. Redford, and S. E. Sanderson, provide. This in turn helps the country by en- eds. 1998. Parks in peril: People, politics, and pro- suring a reasonable supply of future timber as tected areas. Covelo, Calif.: Island Press. 519 pp. Brechin, S., P. R. Wilshusen, C. L. Fortwangler, well as biodiversity values and retains the qual- and P. C. West. 2002. Beyond the square wheel: ity of hydrological systems, vital in a country Toward a more comprehensive understand- with an 83 percent dependence on energy from ing of biodiversity conservation as a social and hydropower generation. What has eluded most political process. Society and Natural Resources of the ARCs, concrete ways to offer incentives to 15:41–64. extend the ecosystem services of parks, would Bruner, A. G., R. E. Gullison, R. E. Rice, and G. A. da Fonseca. 2001. Effectiveness of parks in pro- be possible for perhaps the first time. But at the tecting tropical biodiversity. Science 291:125–28. national scale, MINAE enjoys limited political Christen, C. A. 1994. Development and conserva- power, and there is not yet sufficient popular tion on Costa Rica’s Osa Peninsula, 1937–1977: understanding of the law to demand that pay- A regional case study of historical land use pol- ments be made. This demonstrates the complex- icy and practice in a small Neotropical country. Ph.D. diss., Johns Hopkins University. ity and contradictions inherent in Costa Rican Cuello, C. 1995. Sustainable development in theory policies, a condition rampant in most countries and practice: Costa Rican thinking and experi- (Barrett et al. 2001). At the national level, in ence on sustainability. Dissertation, University terms of rhetoric and conceptual actions, Costa of Delaware. Rica has developed policies that are exemplary Cuello, C., K. Brandon, and R. Margoluis. 1998. and could, with appropriate adaptation, serve Costa Rica: Corcovado National Park. In Parks in peril: People, politics, and protected areas, ed. as models within other countries. But in the K. Brandon, K. Redford, and S. Sanderson, 143– implementation of these policies, Costa Rica’s 92. Covelo, Calif.: Island Press. progress has been uneven. Therein lies the Honey, M. 1999. Costa Rica: On the beaten path. “muddle”—the contradictions, lack of applica- In Ecotourism and sustainable development: Who tion, and politicization that undermine sustain- owns paradise?, chap. 5. Washington, D.C.: Is- land Press. able development objectives. Yet there is hope Janzen, D., ed. 1983. Costa Rican natural history. that with the right framework and political will Chicago: University of Chicago Press. 816 pp. in place policy changes will lead to improve- ———. 1988. Buy Costa Rican Beef. Oikos 51: ments and lasting conservation. 257–58.

POLICY CONTEXT FOR CONSERVATION 309 frankie chap 24 8/20/03 8:11 PM Page 310

Langholz, J., J. Lassoie, and J. Schelhas. 2000. In- Sanchez-Azofeifa, G. A., C. Quesada-Mateo, centives for biodiversity conservation: Lessons P. Gonzalez-Quesada, S. Dayanandan, and from Costa Rica’s private wildlife refuge pro- K. Bawa. 1999. Protected areas and conserva- gram. Conservation Biology 14:1735–43. tion of biodiversity in the tropics. Conservation Palminteri, S., G. V. N. Powell, A. Fernandez, and Biology 13:407–11. D. Tovar. 1999. Talamanca Montane-Isthmian Sayer, J., N. Ishwaran, J. Thorsell, and T. Sigaty. Paciﬁc Ecoregion-based conservation plan: Pre- 2000. Tropical forest biodiversity and the World liminary reconnaissance phase. Unpublished Heritage Convention. Ambio 29(6):302–13. WWF report. Soulé, M., and J. Terborgh, eds. 1999. Continental Powell, G. V. N., J. Barborak, and S. Rodriguez. conservation: Scientiﬁc foundations of regional 2000. Assessing representativeness of pro- reserve networks. Covelo, Calif.: Island Press. tected natural areas in Costa Rica or conserving 227 pp. biodiversity: A preliminary gap analysis. Biolog- Umaña, A., and K. Brandon. 1992. Inventing in- ical Conservation 93:9335–41. stitutions for conservation: Lessons from Costa Powell, G. V. N., S. Palminteri, B. Carlson, and Rica. In Poverty, natural resources, and public M. Boza. 2002. Successes and failings of the policy in Central America, ed. S. Annis, 85– Monteverde reserve complex and Costa Rica’s 107. Washington, D.C.: Overseas Develop- system of national protected areas. In Rescuing ment Council. tropical nature: Making parks work, ed. J. Ter- World Bank. 1993. Costa Rica: Forest sector review. borgh, C. P. van Schaik, L. Davenport, and Internal Division Reports, Agricultural Opera- M. Rao, 156–71. Covelo, Calif.: Island Press. tions Report 13850. Washington, D.C.: Environ- Redford, K. H., and B. Richter. 1999. Conservation ment Department. of biodiversity in a world of use. Conservation World Bank, Operations Evaluation Department. Biology 3:1246–56. 2000. Costa Rica: Forest strategy and the evolu- Redford, K. H., and S. E. Sanderson. 1992. The tion of land use. Internal report. Washington, brief barren marriage of biodiversity and sus- D.C.: World Bank. tainability. Bulletin of the Ecological Society of World Resources Institute, United Nations Devel- America 73:36–39. opment Programme, United Nations Environ- Rudel, T., and J. Roper. 1996. Regional patterns ment Programme, World Bank. 2000. World and historical trends in tropical deforestation, Resources: 2000–2001: People and ecosystems: The 1976–1990: A qualitative comparative analysis. fraying web of life. Washington, D.C.: World Re- Ambio 25(3):160–66. sources Institute. 389 pp. ———. 1997. The paths to rain forest destruction: Crossnational patterns of tropical deforestation, 1975–90. World Development 25:53–65.

310 BRANDON frankie chap 25 8/20/03 8:12 PM Page 311

chapter 25

Conclusion and Recommendations

Gordon W. Frankie, Alfonso Mata, and Katrina Brandon

n order to survive, human beings must ecological basis, of Costa Rica’s biodiversity and Ibegin to consider the deterioration and de- conservation focused in the tropical dry forest. struction of natural resources as a capital loss, The biological, socioeconomic, cultural, and particularly for development and management political knowledge presented in this volume options of future generations. Our choices are suggests several general lessons for future pro- simple. Either we take care of our planet and its jections on ecological restoration and biodiver- natural resources, or we continue to participate sity conservation. Interconnecting and under- not only in our natural world’s progressive dete- lying each of these lessons are the socioeconomic rioration but also in our self-destruction. The and political contexts that have developed around fundamental issue is that of appropriate envi- protected areas and the future of biodiversity ronmental management, or in other words, how conservation. In this ﬁnal chapter we summa- people can interact responsibly with nature. A rize the main biological and social lessons that friendly relation, nondestructive but sustainable, are stated or, we believe, implied by the con- is something that must be accepted and adopted tributing authors. Where appropriate, we also as a means of life and culture, and protecting include obvious recommendations for improve- and conserving nature is something that in- ment on conserving biodiversity. The summary volves the whole of society. Thus, society needs does not in any way suggest that individual les- to be informed about the potential and weak- sons and recommendations pointed out in the nesses of the environment in which humans various chapters are unimportant; they are all live and work. The general scope of this book is relevant and useful. We also looked for simi- to offer an interdisciplinary exploration, with an larities in lessons that were shared with other

311 frankie chap 25 8/20/03 8:12 PM Page 312

regions of Costa Rica. Thus, most lessons of- Lesson 2. Studies on several groups of animals and fered here apply to the country as a whole and to habitats clearly indicate that the dry forest is ecolog- other regions of Mesoamerica, as well. ically linked to other, adjacent life zones and in some cases to geographical areas outside the country. Many bird species (chapter 12) and sea turtles LESSONS (chapter 15) regularly migrate long distances each Lesson 1. A great amount of information already year between the dry forest and other, widely exists for many prominent groups of plants and an- separated geographical areas, and international imals of the dry forest, but much more information treaties theoretically protect some of these well- is needed. known organisms. What is not generally recog- A wealth of knowledge already exists—in nized is that large numbers of diverse taxa also published scientific studies, unpublished re- make regular migrations, most of which are sea- ports, on-line sources, and semitechnical pub- sonal, between dry and adjacent wetter forests of lications. Further, there are large collections of Costa Rica. The case for in-country migration of Costa Rican plants and animals in several mu- resident dry-forest animals is made for birds by seums. Most notable in Costa Rica are the Insti- Barrantes and Sanchez (chapter 15) and for bats tute of Biodiversity (INBio) in Heredia, the Na- by Stoner and Timm (chapter 5) and LaVal (chap- tional Museum in San José, and the University ter 13). Janzen (chapter 7) offers an introduction of Costa Rica in San Pedro. In the United States, to the large numbers of phytophagous and some large collections, especially of plants, are housed predator insect groups that travel seasonally be- at the Missouri Botanical Garden in St. Louis tween dry and wet forests. Haber and Stevenson and at the Chicago Field Museum. Additional (chapter 8) provide considerable survey informa- sources of information are the many dry-forest tion on three types of migrations that dry-forest researchers themselves and their ongoing stud- butterfly species exhibit traveling between major ies, which continually produce new and useful biotic regions. Their survey also reveals that about findings. Many of these workers also regularly half of all dry-forest butterfly species migrate! devote some of their time to educational out- Related to the organismal connectivity be- reach efforts. tween dry and wet forests is another type of eco- Despite the accumulated and continuously logical connection. This can be readily observed emerging knowledge, there will always be the in watersheds that traverse several life zones need and curiosity to obtain basic information. from high elevations to lower ones (chapter 9). There will also be an increasing demand to ap- These watersheds or corridors connecting diverse ply existing knowledge to solve problems (Meffe habitats have received little study to date. Yet 1998) and to obtain new information on how hu- they provide important aquatic and terrestrial mans affect biodiversity (see chapters 1, 4, 7, 12, environments for much biodiversity and, at the 15, 21, and 23). One important biological area in same time, important environmental services for need of more research is that of monitoring se- humans (Daily 1997; see also chapters 9–11). lected taxonomic organisms to assess environ- The important message of Lesson 2 is that mental health (chapters 4, 6, 9, 10, 12, and 15). the dry forest should not be considered as an is- Monitoring will also become increasingly nec- land of special habitats and organisms. Rather, essary for surveying rare and endangered species much of the dry-forest biota has evolved many (chapters 12 and 15). Another area that will re- different ecological relationships with wetter for- quire considerable new information and knowl- ests and in some cases with widely separated geo- edgeable professionals is restoration ecology. graphical areas. The relationships are complex However, before this knowledge can be applied, and must be considered when developing effec- other new actions must be put into motion, as tive legislation, local or regional management indicated in the lessons that follow. plans, and collaborations to protect them.

312 FRANKIE ET AL. frankie chap 25 8/20/03 8:12 PM Page 313

Lesson 3. Conserving biodiversity and the habitats many organizations jointly develop a work plan that support it consists of many interacting biological, is that they are each bound by their own policies socioeconomic, political, policy, and organizational and traditions (Clark et al. 1994). Multiple or- components, all of which are important and have a ganizations without a shared sense of mission place in conservation. or clear goals often had difficulties working to- In referring to programs for recovering en- gether. This study suggests that involvement of dangered species in the United States, Clark et all relevant groups is essential but that compli- al. (1994: 419) state, “Regardless of the biologi- cated organizational structures to implement cal status of the species and its habitat, the ulti- conservation projects may not lead to improved mate causes of most species’ endangerment lie conservation outcomes. Clark et al. (1994: 421) in human values that are manifest in varying came to a similar conclusion when considering social, economic, and political institutions and species restoration projects, recommending that activities.” They expand on this fundamental high-performance teams (i.e., lead groups) con- proposition by pointing out that there is a basic duct the actual conservation work. need for holistic understanding of the problem, which should then receive interdisciplinary at- Lesson 4. Gaps in transferring information from tention from the conservation community. biologists to databases (large and small) to policy The same case can be made for projects to makers and land managers/stewards must be recog- protect healthy species and habitats. Biologists nized and bridged. and nonbiologists need to work together in a truly Conserving biodiversity implies that there is integrated manner to achieve stated goals (chap- an orderly transfer process, which entails know- ters 4, 9, 15–18, 20, 22, and 23). Working together ing the organisms (naming), their biologies and has not been the rule in most conservation proj- ecologies, and how to get this information to ects throughout the country (chapters 15, 17, 19, the people who will decide on plans for manag- 21, and 23). The integrated approach requires ing the organisms and their habitats. Hanson more work through advanced planning, indi- (chapter 17) states that “the sequence of infor- vidual agreements among participants, frequent mation transfer that needs to occur is systematics meetings to review and critique progress, and re- [leading to] ecology [leading to] conservation”— ports reflecting progress from all parties involved. a logical concept. We know most of the promi- Bringing participants together also requires new nent organisms of the dry forest and something organizational structure, effective leadership, pos- of their ecologies, but numerous accounts in sibly additional funding, and group learning this volume indicate that we have a serious prob- (Clark et al. 1994: 420; chapters 16 and 18). lem in the transfer of knowledge to the practice There are a few caveats for increased collab- of conservation. Actually, the problem is a multi- oration, however. Although it is important for all faceted one. We have identified the most obvi- relevant organizations to be informed about ous gaps in information transfer: (a) between and consulted on what is happening, this does biologists and government officials (decision not mean that all should be involved in man- makers) charged with protecting biodiversity aging conservation programs. A recent study and the environment in the field as well as in the conducted by the Biodiversity Support Program Ministry of Environment and Energy (MINAE) in (Margoluis et al. 2000) found that involving all the capital, San José; (b) between biologists and relevant groups in conservation did not lead to the media; (c) between biologists and local peo- the best conservation outcomes. What was most ple adjacent to protected areas; (d) between local effective was having one strong lead organiza- people and government officials; and (e) between tion that had a clear sense of mission and that the laws and their effective implementation. was strategic in its collaboration with other or- These are extremely important gaps that con- ganizations. One of the difficulties of having front professionals who wish to see biological

CONCLUSION AND RECOMMENDATIONS 313 frankie chap 25 8/20/03 8:12 PM Page 314

knowledge invested in the most productive ways would have to be offered to attract good profes- in order to protect biodiversity. Each of these sional people. Last, if new people are hired, they gaps is discussed in the lessons that follow. would need to be effectively incorporated into the system. Lesson 5. Officials in the central government as well Hiring more qualified people would repre- as field directors charged with protecting biodiversity sent a big step in the right direction. In addition, and the environment must be better educated and this change would allow for more effective com- trained. Upgrading capacities and careful hiring of munication and collaboration between biologists committed personnel are a necessity. and directors of the large conservation areas As pointed out in Stoner and Timm (chap- (chapters 1 and 24). At the present time, with a ter 5), the vast majority (98%) of government few exceptions, many biologists shy away from officials charged with protecting biodiversity and working in parks and reserves because of the the environment in MINAE have never received difficulties of working with government offi- any university education or training, and many cials and their cumbersome regulations (see Les- of them are lacking basic ecoethical values, en- son 6). This is unfortunate because protected vironmental commitments, or a minimal moti- areas need biologists. In addition, it seems that vation for protecting the environment. Yet these the decentralization of MINAE into conserva- individuals are charged with the huge task of tion areas permitted laxity in leadership, poor protecting the country’s biodiversity and envi- enforcement of the law, and a lack of monitoring ronment—a complex and difficult proposition of violations (chapters 21–23), with corruption in by any measure. The deficiencies in under- the form of granting government permissions standing, much less protecting, biodiversity are for forest cutting and land-use permits (e.g., pointed out with clear examples in chapters 7, 9, Playa Grande and Ostional wildlife refuges). 12, 15, 17, 19, and 21–24. This scenario is further There is also a disconnection in academia be- complicated if corrupt situations are involved, tween biology and conservation. Many academic especially in relation to land-use permits in pro- biologists feel that if they pursue applied aspects tected areas (chapters 22 and 23). of conservation, they will not be taken seriously The obvious remedy seems clear enough. or granted tenure. This has been a subject of Hire a large number of university graduates over debate within the Society for Conservation Biol- a period of time and screen personnel for their ogy and also holds true throughout Latin Amer- commitment to conservation and understand- ica. Another key element is that the courses ing of its importance. These individuals should and training most biologists receive have little have bachelor’s and master’s degrees in a wide to do with applied conservation (Meffe 1998). variety of biological and social fields in order to deal with the many complex problems that oc- Lesson 6. More collaborative partnerships must be cur in conservation (chapters 15, 19, 21–24). A built between the public and private sectors in con- good case could easily be developed to justify serving biodiversity. The private sector, including this change. However, it would not be easy to biologists, nongovernmental organizations, and implement for many reasons. First, there is tre- private landowners, has a great role to play in bio- mendous inertia in MINAE, which is not un- diversity conservation. expected (Clark et al. 1994). Second, this change The need for more collaboration in conserva- must have leadership and support at the highest tion work has never been greater. Collaboration political level of the government. As Brandon is at the core of truly integrated, multidiscipli- points out in chapter 24, the central government nary conservation projects (Lesson 3), and it has not been recently inclined to support con- depends largely on effective information transfer servation in effective ways. Third, more money (Lesson 4). Collaborations can be simple agree- would be needed support the new salaries that ments or more complex contracts, but regard-

314 FRANKIE ET AL. frankie chap 25 8/20/03 8:12 PM Page 315

less of their form, it is highly desirable to view resent a good start. Private biologists and govern- these relationships as partnerships (see Won- ment land stewards should form special teams or dolleck and Yaffee 2000). committees that work together to produce man- There are many types of collaborations that agement plans, and biologists must be assured need to be developed or strengthened, including that their contributions will be appropriately rep- (a) government agency to government agency; resented (O’Brien 1993). Finally, professional (b) government agency to biologists; (c) gov- biologists must be invited periodically to review ernment agency to private organizations, espe- management plans critically (Lesson 9). cially nongovernmental organizations (NGOs); (d) government agency to private landowners; NONGOVERNMENTAL and (e) government to local associations and CONSERVATION ORGANIZATIONS other civic groups. NGOs that focus on conservation and environ- In this lesson, attention is focused on the mental issues have played a considerable role in government and its many relationships because protecting the country’s biodiversity. Their work it plays a major role in the protection and regu- is, however, not always recognized, mostly be- lation of biodiversity by virtue of its authority. cause of their small size and lack of media at- Yet its record of collaborations within its own tention. Collaborations between government and structure and with the private sector falls far NGOs are gradually increasing in the country, short of what it could be (chapters 7, 12, 15, 19, but many more need to be established. Further, and 21–23), and this deficiency in collaboration government incentives need to be offered, espe- is negatively affecting biodiversity in a variety cially to those organizations that become legally of ways (chapters 21–24). To emphasize the im- registered in the country. portance of the potential for more effective col- Small nonprofit organizations have several laborations between government and the private advantages over government agencies: they sector, we focus on biologists, NGOs, and pri- can organize and mobilize quickly, require little vate landowners. start-up money, are staffed with dedicated workers, often have leaders with bachelor’s or higher NONGOVERNMENTAL BIOLOGISTS degrees, are in contact with a concerned segment As mentioned in Lesson 1, professional biolo- of society, often have many available volunteers, gists have worked for years in Costa Rica to pro- and can serve as advocacy groups to offer alter- duce a vast amount of biological knowledge nate viewpoints on the health of the environ- (Janzen 1983). Yet most biologists are aware that ment. Their big disadvantage is that they are their best science is not being applied to protect always desperately short on funds (chapter 19). biodiversity in protected areas (chapters 7, 12, 15, Large international NGOs such as the World 17, 19, 21, 23, and 24). Further, only rarely are Wildlife Fund (WWF), World Wide Fund for Na- they being asked to contribute directly to prob- ture (WWF), Conservation International (CI), lem solving and overall management (see chap- and the Nature Conservancy (TNC) have occa- ter 17). sionally contributed directly to protecting bio- Collaborations between biologists and gov- diversity, but these organizations and their funds ernment land stewards need to increase quickly are often targeted early on toward specific proj- because species are being lost and more will be ects. They can be an excellent source of targeted lost in the future (chapter 7, 12, and 15; see also assistance, but they are often reluctant to speak Lesson 11). Several accounts in this volume sug- out forcefully and publicly on inappropriate gest that a greater effort be made by government government actions. Even though much of officials to attract and work more closely with conservation is in fact political, it has to happen biologists (Lessons 4 and 5), and government behind the scenes. In several countries, large incentives to encourage collaborations would rep- international NGOs seen to be publicly taking

CONCLUSION AND RECOMMENDATIONS 315 frankie chap 25 8/20/03 8:12 PM Page 316

strong stands on issues would likely be asked to stewards and adjacent landowners have been leave the country. For this reason, they often have mixed throughout the dry forest. Relationships networks of in-country NGOs that they work with vary from tolerance to outright distrust and dis- and support, which can take stronger stances on regard. This is unfortunate because many pri- issues of national importance. vate landowners still have much intact habitat The first Costa Rican association for the on their properties, and if they are approached conservation of nature (ASCONA), now FECON in a professional manner with proposals, many (Costa Rican Federation for Environmental Con- could be effective partners in biodiversity conservationists), was founded at the Tropical Sci- servation. Consider also, for example, that many ence Center (TSC), a private NGO in San José; landowners like wildlife, stream corridors, and almost all members of the TSC were members trees (see chapter 19) and that they discourage of ASCONA. In contrast, Fundación de Parques hunters, control fire on their properties, and in Nacionales (NPF) and Fundación Neotropica general love the land they work. Many also want (NF) have programs favoring conservation, but their hard work and family legacy passed on to they do not participate openly in defense of their offspring. conservation and in denouncing transgressions These landowners are mostly responsible or corruption cases. NGOs such as the Orga- stewards who could be doing more to protect nization for Tropical Studies (OTS), TSC, and their lands if there were effective collaborations Monteverde Conservation League (MCL); York with the government. These could range from University; and a few other private institutions simple agreements with landowners to more manage considerable areas with strict protection involved projects in which resources and per- and also are dedicated to research and ecotourism sonnel are shared for clearly stated goals. Fur- activities. Projects such as AMISCONDE (Wille ther, some landowners are in the process of 1994), in the southern part of Costa Rica, which considering conservation easements (Gustanski have shown to be very successful, have been and Squires 2000) to protect their properties sponsored by private companies such as Mc- from further development. Conservation ease- Donald’s and Nestle through the intermediacy ments are just beginning to play a role in biodi- and technical guidance of CI and the TSC. versity conservation in Costa Rica. These facts suggest a great potential for collaborations that PRIVATE LANDOWNERS would further protect biodiversity. However, there The flora and fauna of Costa Rica do not recog- must be a change in attitude on the part of gov- nize societal and political boundaries. Obviously, ernment personnel to be able to negotiate these only part of the country’s biodiversity is con- collaborations. We recommend that new govern- tained within the boundaries of national parks ment personnel, trained in public relations and and reserves. What needs to be taken into con- other socioeconomic skills, be directed to work sideration is the surrounding areas and their with landowners on new conservation projects. capacity and potential to conserve biodiversity Private biologists and NGOs may need to be in- (chapters 16 and 17) or their capacity to serve as vited to facilitate and implement agreements. a buffer for effects by human activities. Another important consideration is what the govern- Lesson 7. Conserving biodiversity requires environment is doing to encourage biodiversity protec- mental education of many different public and pri- tion on adjacent public and private lands. See vate audiences. Bustos (chapter 23) and a theoretical discussion Several authors expressed the need for in- of this topic in Brandon (chapter 24) and Bran- creased environmental education to expand the don et al. (1998). knowledge base concerning the natural and Collaborations between government land human-altered environments of Costa Rica (chap-

316 FRANKIE ET AL. frankie chap 25 8/20/03 8:12 PM Page 317

ters 5, 6, 9, 10, 13–16, 19, 21, and 23). The contri- warming and climate change, protected areas, butions on water by Mata (chapter 9) and Vargas and payments for environmental services. Palo and Mata (chapter 10) are extremely important Verde National Park in the dry forest is an ex- in this regard because they relate to environ- ceptional site to demonstrate biological links mental issues of human health and food supply. between the United States and other countries, They point out important relations between en- given its importance as a site for migratory vironmental impacts, human health, and environ- waterfowl (chapter 12). It also provides a window mental stress and stability on the one hand and on the complexities of park management amid economic activities, quality of life, and ecologi- government programs aimed at promoting cal restoration on the other and how education development. Courses of this type should be is necessary to improve the general scenario. greatly expanded to include more Costa Rican Knowing the characteristics of target audi- decision makers, an issue raised in the National ences is a fundamental component of environ- Strategy of Conservation of Costa Rica (Quesada- mental education (Jacobson 1999), and a few Mateo 1990). authors have focused on selected audiences The need for environmental education can (chapters 9, 17, and 19). Aburto (chapter 20) ad- be easily justified—with such education, all au- dresses another important audience in need of diences associated with biodiversity conservation special environmental education—the biologists will be better informed. The critical question is themselves. She stresses the need for biologists whether appropriate investments by government to learn how to communicate their important and private groups will be made to better inform findings to the media, with the goal of trans- the citizenry of the country. To date, individual ferring new information to the public. Bustos private groups have mostly recognized the need (chapter 23) points out the need for more envi- for environmental education and have taken the ronmental education of judges who preside over initiative to implement programs. The govern- biodiversity and environmental laws (chapter 22), ment, through MINAE, has yet to make signi- about which they know little. ficant investments in meaningful environmen- The media also need to be better informed tal education programs (G. Frankie pers. obs.), about environmental issues. For groups such as and much more should be expected from the the media and judges, what they need to know Ministry of Education. We strongly recommend consists of two components. The first is envi- that the government increase spending for environmental education. The second, although re- ronmental education to a variety of audiences. lated to the first, is a more complex understand- See also Holl et al. (1995). ing of economics, politics, and environmental issues. For example, an increase in oil prices Lesson 8. Connections between biodiversity, natu- in San José is not only an economic issue but ral resources and their use, and ecosystem services— also an environmental issue—from increased as well as economic expenditures and accounting— revenue for payment for ecosystem services, to need to be made explicit and clearly understood, from foreign exchange needs, to sprawl, congestion, the highest government level to the local level. and pollution in the Central Valley. Many issues This lesson has many components. The first in Costa Rica have environmental repercussions, is that biodiversity conservation is not necessar- but, as in the United States, there is often no in- ily synonymous with natural resource manage- terest or general understanding of how these ment or ecosystem services. One can have well- issues are related to one another. The OTS offers forested watersheds that provide stable sources courses on this topic for U.S. decision makers, of timber and abundant clean water but that are who visit Costa Rica for a week to learn about devoid of the multiplicity of species that once rain forests, primary forests, energy issues, global resided there. In many parts of Latin America,

CONCLUSION AND RECOMMENDATIONS 317 frankie chap 25 8/20/03 8:12 PM Page 318

this “empty forest” syndrome has already been GDP) and our national accounting systems recognized—forests that lack viable populations (green accounting), politicians have had little of characteristic fauna (Redford and Sanderson use for these numbers. Politically, it is to the 1992). benefit of most politicians to boost exports and Well-managed natural resources can exist generate foreign exchange—even if these come without biodiversity—they represent areas where from unsustainable resources practices. The the ultimate management objective is to ensure political benefits are immediate, and by the time a continuous stream of resources for human the damage begins to exact a toll on the econ- use. Ecosystem services can be thought of as a omy, voters rarely remember who did what, and broader set of processes that act as the foun- the politician is long gone. Circumventing this dation for life—they represent clean air, clean means incorporating these green measures into water, fertile soils, and the like (Daily 1997). standard ways of measuring productivity and ex- Both natural resource management and ecosys- penditures in order to understand how well we tem services can occur at reasonable levels with- are managing natural capital. Yet, as demon- out maintaining the biodiversity that was once strated by the case of Costa Rica with the pay- intact in an area (see Redford and Richter 1999). ment for environmental services from fossil fuel Sound natural resource management policies taxes, understanding is not equivalent to action and policies that ensure ecosystem services (chapter 24). If Costa Rica is exceptional in that represent the context within which biodiversity it has made the first strides at least to establish conservation must occur. this, it is all the more saddening and sobering Managing for biodiversity implies putting that actions that make good economic and envi- existence values ahead of use values. In our ronmental sense are still ignored. changing world, it may mean being active land and biodiversity managers in ways not thought Lesson 9. There is an urgent need to use ecoregional of a decade ago. For example, we now know that planning as the basis for land-use management Costa Rica’s fabled golden toad is probably ex- countrywide, to make parks the anchors for bio- tinct because the level of the cloud forests has diversity within each ecoregion, and to monitor these receded, and the toads were unable to change actions continuously. habitats quickly and move upslope to areas that Many lessons within this chapter have more retained the humidity they needed. Now, golden than one dimension to them. This lesson ad- toads are famous for being the first species dresses the need to (a) move toward ecoregional whose probable extinction is directly linked to actions; (b) reinforce the necessity for parks to climate change. Keeping viable populations of be the cornerstone of biodiversity conservation; amphibians in this setting means either carry- and (c) ensure that both planning and imple- ing frogs and toads up the steep slopes in Mon- mentation of development activities are compati- teverde or figuring out how to get moist clouds ble with conservation. Within the conservation to linger lower for longer periods of time. Are we community, there has been a dramatic shift to ready to do this? Is this a requirement of actively thinking at larger scales than parks, owing to a managing for biodiversity? variety of ecological problems that are raised in A second key lesson in this is that we can numerous chapters in this book and elsewhere attach values to the many ecosystem services (e.g., Soulé and Terborgh 1999). provided. These values do not appear in any- Costa Rica developed the Regional Conserva- one’s accounting. Although there have been tion Areas (ARC) system to be responsive to the huge advances in the past decade among eco- need to link protected areas with the surround- logical and environmental economists in how to ing lands. Unfortunately, the ARC system has measure the value of these services and include not led to better coordination or to actions that them in our measures of productivity (e.g., green are consistent with biodiversity conservation.

318 FRANKIE ET AL. frankie chap 25 8/20/03 8:12 PM Page 319

Some might question why biodiversity conser- development, that is, for ecotourism, and there- vation should serve as the basis for regional fore land-use permits can be granted to hotels in planning within the country. There are several protected refuges. answers to this, but the bottom line is money. It is only within the past few years that products Lesson 10. Biodiversity and environmental laws and of the Intel plant in the Central Valley have sur- policies need periodic critical review to identify those passed other exports. But the fact remains that, that work and those that do not work and why. Im- aside from Intel, Costa Rica’s economy depends plementation must be reviewed critically. on tourism and agriculture. Within the rural sec- Laws and policies need to represent the best tor, these two and biodiversity are inextricably science and highest certainty of implementa- linked. Costa Rica has developed a strong repu- tion. This often creates deep conflicts between tation as an ecotourism destination, a position science and policy. As Mata notes in chapter 9, that could easily be lost without strong manage- a recent Costa Rican forestry law (No. 75757, ment of protected areas. We are increasingly 1996) bans tree cutting along riparian corridors seeing shifts to forms of both tourism and agri- using categorically defined distance measures culture that are not sustainable. Negative effects that reflect urban versus rural and type of to- in one area—such as pesticide runoff and pography. He notes that “the appropriate meas- wildfires—affect not only biodiversity but also urement should be determined by practical and tourism and the overall health of the general scientific observations of the stream’s conditions: population (chapters 5, 9, and 19). its role in local ecological processes, the amount Many of Costa Rica’s parks are too small to of biodiversity it contains, stream dynamics, be ecologically viable for what they seek to pro- and the equilibrium [relationships] of related tect (e.g., see Cuello et al. 1998). Either these ecosystems.” areas should be enlarged, or they need to have Unfortunately, laws cannot be established adjacent lands in compatible management forms this way because they would quickly become (chapters 14 and 16). Ecologically, this is para- arbitrary. Neither can they be precise enough to mount, but it is socially important, as well. specify every eventuality. They must be blunt Smaller-scale entrepreneurial activities, such as instruments, established in ways that represent nature-based tourism lodges and shade-grown the “best science” (or “best guess”) and the best organic coffee, offer a chance for people to be chance that the laws will be implemented. So in owners and managers. This differs vastly from this case, 50 m represented the best science or employment in commercial tourism, which of- guess and the best hope that it would be fol- ten provides only seasonal work in low-paying, lowed. Finding the balance between science and menial categories. the reality of what people will do is tricky. In Strong ecoregional plans with participation general, because many environmental problems by different government agencies, plans that cannot be reversed, it is better to be more cau- include green accounting and provide a realistic tious on the “science” side and give a greater push understanding of the inherent economic and toward implementation (see also Soulé 1986: 6). ecological trade-offs, are likely to lead to good Like indicators in a good research design, planning—from social, economic, and envi- laws and policies must be precise, measurable, ronmental sides. The challenge is to see if the consistent, and sensitive. Precision is an ex- political will exists for their implementation. tremely important criterion—there must be a Scientifically based management plans are ab- way to tell if the law or policy applies to a partic- solutely necessary to place clear boundaries on ular issue or condition. Within this volume are economic interests related to tourist businesses. numerous examples in which laws and policies Here again, there is a paradox in the definition were formulated with insufficient clarity and con- of policies: that conservation is meant for future sequently were open to a variety of interpretations.

CONCLUSION AND RECOMMENDATIONS 319 frankie chap 25 8/20/03 8:12 PM Page 320

Precision is desirable unless one is guaranteed Biologists have certainly posted their warn- that the direction of interpretation will always ings, but the general public and decision makers be pro-environment, an unlikely case. Measura- probably have not heard and understood these bility is important because it determines whether warnings very well. Many scientists generally the law or policy is accomplishing what it was in- couch their findings in ways that are appropriate tended to accomplish. In general, simple meas- to academia, referring to limitations of the data ures work best. Consistency is important, as and what can and cannot be extrapolated from well—the policy or law should apply every time findings (Soulé 1986). However, what may be in an equal situation or for all cases. This brings appropriate within the scientific community is us to Dan Janzen’s attempt to generate revenue generally inaccessible to others because key for Guanacaste National Park by allowing or- conclusions are buried in jargon. Scientists also ange by-products to be dumped in it (see chap- often wait for the media to come to them, rather ter 24). Although the idea was a good one— than being proactive (chapter 20). economically and ecologically—how can a law Reed Noss (1995: 701–3) writes, “We should allowing the dumping of by-products be applied not be so assured that the degradation of nature consistently without harmful effects in other is general knowledge.” Evidence of this lack of cases? Great ideas often die because the policy knowledge dates from Leopold: “Much of the and legal context rejected them; but the trade-off damage inflicted on land is quite invisible to lay- for the lack of progress (and in this case revenue) men” (Leopold in Noss 1995: 701). If damage is protection from misuse of a law opening on land is invisible, how difficult must it be to parks to such uses. The last criterion of a law or detect very subtle changes in insects, birds, fish, policy is sensitivity—its ability to capture the and mammals? Biologists can, and must, become variety of changes or conditions that are relevant much more vocal with their findings (O’Brien to the issue at hand. 1993; chapters 7, 9, 20, 21, and 23). For those Laws and policies can quickly be evaluated to doing research in Costa Rica, it means taking determine if they meet these standards and if more dramatic action than publishing in the the standards capture the best available sci- journals TREE or Conservation Biology. It means ence. Implementation is the second part of the bringing the findings to a Costa Rican NGO, a equation—there must be a basis on which peo- television station, and the offices of La Nación ple want or need to act in the specified manner. (Costa Rica’s largest newspaper). It means ask- Wanting to act in a specified manner can be pro- ing to speak at a local school. It may even mean moted in a variety of ways, from church sermons buying a round of drinks in the local cantina near to television advertisements. Needing to act a a research site on a Saturday night and giving certain way is more likely to come about from an emotional speech (chapters 18–21 and 23). a deterrence or incentive point of view. Because There is good news here—evidence suggests implementation is malleable, laws that are a few that crisis leads to action (Brandon et al. 1998). degrees harsher than one might expect should Two of the strongest allies for conservation are be pushed. All of this necessitates reviewing pride and a sense of ownership. People react and revising laws as needed—there is nothing when something they care about may be lost; sacred in laws and policies. The tricky element, however, people need to be told what is being however, is to get good laws implemented, and lost and what they can do about it. One study that requires political will. within Costa Rica found that even when people had some idea about an environmental problem, Lesson 11. Biologists have warned that species have they seldom knew what action they should take, recently been lost from the country and that more if any. There was, even among different socio- will be lost unless new steps are taken to avert these economic groups, a very high (91%) willingness projected losses. to pay more for services such as water and elec-

320 FRANKIE ET AL. frankie chap 25 8/20/03 8:12 PM Page 321

tricity if the money was used to protect biodi- estry law has been frequently revised as needed versity (Holl et al. 1995). This suggests a set of to attempt to control deforestation. Early efforts findings: (a) we need to craft messages that are at decentralization have been made. INBio has clearer and louder about the impending species been a model for bioprospecting and training losses and how these affect every citizen; (b) bi- committed local naturalists (parataxonomists) ologists must take a more proactive role in this (see chapter 17). Debt-for-nature swaps and process (chapter 19); (c) there is generalized other financing mechanisms have been signi- support for environmental and biodiversity pro- ficant in what they have accomplished. Costa tection; (d) financial support for biodiversity Rica helped to define the potential for interna- conservation exits in Costa Rica but it is not tional ecotourism. Notable also has been the law allocated; and (e) greater public political aware- allowing payment for environmental services. ness and outcry are needed to relate environ- All of these actions have served as a model in mental issues to quality of life—clean air and some form and have influenced the overall tra- water, food, and recreation (chapter 20). jectory of international conservation. What has been missing, however, is consistent and effec- Lesson 12. National politics concerning larger so- tive implementation. cial issues and international conservation trends have played, and will continue to play, a great role CONCLUDING REMARKS in determining policies for conserving Costa Rica’s biodiversity. The defense and appropriate management of In the coming decade, many factors will in- protected areas are crucial for the future of bio- fluence how well biodiversity is conserved, pro- diversity. In the case of Costa Rica, which is a tected areas are maintained, and ecosystems politically and socially stable country, we can de- function. These factors are both direct and indi- rive from this chapter’s lessons four major prin- rect. The most important direct factor will be ciples that are indispensable in order to protect how well the laws and policies that are already ecosystems effectively: on the books are implemented (chapters 22 1. Clear (political) definitions of the national and 23). Indirect factors will shape conservation, concepts of “protection” and “natural pat- as well—in more subtle and hard-to-measure rimony” are needed to consolidate the ways. But the bottom line is that even proactive integrity of what is already gained, terri- measures designed to support conservation in- torially and institutionally. directly will have little impact if direct implementation is not improved. Equivalently, nega- 2. Strong, governmentally backed institutions tive actions will exert a negative impact. In the should be in charge of this protection, with universe of indirect effects, lack of implementa- clear goals and obligations toward effective tion of sound policies and practices generally is conservation. a lose-lose proposition. 3. Any corruption in these institutions must The role of international conservation organ- be controlled and environmental ethics izations has been significant in the development developed, especially for those in charge. of Costa Rica’s conservation policies and pro- 4. Active citizen participation is needed. grams; however, this has been a two-way street. Many policy reforms that Costa Rica has under- A clear definition of “protection” is lacking, taken have been initiatives that started within even in the bylaws. Despite several mishaps (e.g., Costa Rica and have in turn shaped the inter- Gandoca, Nosara, Tamarindo, Osa), “develop- national conservation community. ments” and resource extraction or land-use Costa Rica has exerted strong leadership in a permits are still officially granted in areas that large number of environmental arenas. The for- have been declared refuges, protected areas, or

CONCLUSION AND RECOMMENDATIONS 321 frankie chap 25 8/20/03 8:12 PM Page 322

reserves (see example in Rossi 1999). The legal Clark, T. W., R. P. Reading, and A. L. Clarke, eds. protection of an especially diverse area, by de- 1994. Endangered species recovery: Finding the les- cree or by law, is not a guarantee by itself. Al- sons, improving the process. Covelo, Calif.: Island Press. 450 pp. though creation of protected areas is a necessary Cuello, C., K. Brandon, and R. Margoluis. 1998. first step, appropriate management by strong Costa Rica: Corcovado National Park. In Parks institutions and dedicated functionaries are the in peril: People, politics, and protected areas, ed. most important factors. The critical opinion and K. Brandon, K. Redford, and S. Sanderson, monitoring eye of citizens are also important 142–91. Covelo, Calif.: Island Press. Daily, G. C., ed.. 1997. Nature’s services: Societal de- and may become a key element of supervision pendence on natural ecosystems. Covelo, Calif.: and protection. Island Press. 392 pp. Priorities in conservation efforts are also Frankie, G. W., V. Solano, and M. McCoy. 1994. necessary. For example, environmental protec- Una evaluación técnica del Corredor La Mula tion does not make sense unless ecosystems, entre La Reserva Biológica Lomas Barbudal y El whose processes are inextricably linked to one Parque Nacional Palo Verde. Bagaces: Amigos de Lomas Barbudal; Berkeley: University of another, can be rescued intact. Costa Rica is at a California; and Heredia, Costa Rica: Universi- critical juncture. The coming decade will largely dad Nacional de Heredia. 126 pp. Contact decide what wildlife and which wildlands will [email protected] for copies. endure in the future. Far from being a total suc- Gustanski, J. A., and R. H. Squires, eds. 2000. Pro- cess, conservation in Costa Rica still needs a tecting the land: Conservation easements past, present, and future. Covelo, Calif.: Island Press. final and effective push toward consolidation 566 pp. and even restoration. Many important and ef- Holl, K. D., G. C. Daily, and P. R. Ehrlich. 1995. fective laws already exist; but a large part of the Knowledge and perception in Costa Rica resolution to problems lies in strong implemen- garding environment, population, and biodiver- tation and enforcement, with the backing of sity issues. Conservation Biology 9:1548–57. wise political decisions at the highest level. Jacobson, S. K. 1999. Communication skills for conservation professionals. Covelo, Calif.: Island Biologists, conservationists, and informed Press. 351 pp. citizens have major roles to play in ensuring that Janzen, D. H. 1983. Costa Rican natural history. protected areas serve as the cornerstones of Chicago: University of Chicago Press. 816 pp. biodiversity conservation, that laws are strictly ———. 1986. Guanacaste National Park: Tropical enforced, that parks and lands around them are ecological and cultural restoration. San José: Edi- torial Universidad Estatal A Distancia. 103 pp. better managed, that the general public is well Margoluis, R., C. Margoluis, K. Brandon, and informed, and that conservation receives its due N. Salafsky. 2000. In good company: Effective within the political process. There is still the alliances for conservation. Washington, D.C.: Bio- potential to take action; however, doing so will diversity Support Program. require the coordination of many specialists and Meffe, G. K. 1998. Conservation biology: Into the their colleagues. Doing less than this will cer- millennium. Conservation Biology 12:1–3. Noss, R. 1995. The perils of Pollyannas. Conserva- tainly lead to the demise of tropical dry forests tion Biology 9:701–3. and their unique biodiversity. O’Brien, M. H. 1993. The professional biologist: Being a scientist means taking sides. BioScience REFERENCES 43:706–8. Quesada-Mateo, C., ed. 1990. Estrategia de conser- Brandon, K. 1996. Ecotourism and conservation: A vación para el desarrollo sostenible. San José: review of key issues. Environment Department Servicios Litograficos. 180 pp. Paper 33. Washington, D.C.: World Bank. 70 pp. Redford, K. H., and B. Richter. 1999. Conservation Brandon, K., K. H. Redford, and S. E. Sanderson, of biodiversity in a world of use. Conservation eds. 1998. Parks in peril: People, politics, and Biology 3:1246–56. protected areas. Covelo, Calif.: Island Press. Redford, K. H., and S. E. Sanderson. 1992. The 519 pp. brief barren marriage of biodiversity and sus-

322 FRANKIE ET AL. frankie chap 25 8/20/03 8:12 PM Page 323

tainability. Bulletin of the Ecological Society of reserve networks. Covelo, Calif.: Island Press. America 73:36–39. 238 pp. Rossi, A. 1999. La loca de Gandoca. San José: Edi- Terborgh, J. 1999. Requiem for nature. Covelo, Calif.: torial Universitaria Centroamericana—EDUCA. Island Press. 224 pp. 138 pp. Contact: [email protected]. Wille, C. 1994. Nature’s hothouse: Can we save the Soulé, M., ed. 1986. Conservation biology: The sci- Neotropics? American Birds 48:56. ence of scarcity and diversity. Sunderland, Mass.: Wondolleck, J. M., and S. L. Yaffee. 2000. Making Sinauer Associates. 584 pp. collaborations work: Lessons from Innovation in Soulé, M., and J. Terborgh, eds. 1999. Continental Natural Resource Management. Covelo, Calif.: conservation: Scientiﬁc foundations of regional Island Press. 277 pp.

CONCLUSION AND RECOMMENDATIONS 323 frankie chap 25 8/20/03 8:12 PM Page 324 frankie contributors 8/20/03 8:13 PM Page 325

CONTRIBUTORS

GILDA ABURTO San José, Costa Rica GORDON W. FRANKIE Division of Insect Biology— College of Natural Resources, University of California, VICTORIA J. APSIT Biology Department, Univer- Berkeley, California sity of Missouri, St. Louis, Missouri GREGORY A. GIUSTI University of California, ADRIAN J. BARRANCE Overseas Development Cooperative Extension, Ukiah, California Institute, London JAMES E. GORDON Department of Plant GILBERT BARRANTES Escuela de Biología, Univer- Sciences, University of Oxford, Oxford, England sidad de Costa Rica, San Pedro, Costa Rica WILLIAM A. HABER Monteverde, Puntarenas, KAMALJIT S. BAWA Department of Biology, Uni- Costa Rica, and Missouri Botanical Garden, versity of Massachusetts, Boston, Massachusetts St. Louis, Missouri

FEDERICO BOLAÑOS Escuela de Biología, Univer- JAMES L. HAMRICK Departments of Botany and sidad de Costa Rica, San Pedro, Costa Rica Genetics, University of Georgia, Athens, Georgia

DAVID H. BOSHIER Department of Plant Sciences, PAUL HANSON Escuela de Biología, Universidad Oxford Forestry Institute, University of Oxford, de Costa Rica, San Pedro, Costa Rica Oxford, England DANIEL H. JANZEN Department of Biology, Uni- KATRINA BRANDON Center for Applied Biodiversity versity of Pennsylvania, Philadelphia, Pennsylvania Science, Conservation International, Washington, D.C. JORGE A. JIMÉNEZ Organization for Tropical JULIO ALBERTO BUSTOS San José, Costa Rica Studies, San Pedro, Costa Rica

JAIME ECHEVERRÍA Centro Scientiﬁco Tropical, RICHARD K. LAVAL Monteverde, Puntarenas, San José, Costa Rica Costa Rica

325 frankie contributors 8/20/03 8:13 PM Page 326

ALFONSO MATA Centro Scientiﬁco Tropical, San JAMES R. SPOTILA School of Environmental José, Costa Rica Science, Engineering, and Policy, Drexel University, Philadelphia, Pennsylvania SEAN T. O’KEEFE Department of Biological and Environmental Sciences, Morehead State University, ROBERT D. STEVENSON Department of Morehead, Kentucky Biology, University of Massachusetts, Boston, Massachusetts FRANK V. PALADINO Department of Biology, Indiana-Purdue University, Fort Wayne, Indiana KATHRYN E. STONER Departamento de Ecología

MAURICIO QUESADA Departamento de Ecología de los Recursos Naturales, Instituto de Ecología, de los Recursos Naturales, Instituto de Ecología, Universidad Nacional Autónoma de México Universidad Nacional Autónoma de México (UNAM), Morelia, Michoacán, Mexico (UNAM), Morelia, Michoacán, Mexico ROBERT M. TIMM Department of Ecology and PETER S. RONCHI Amigos de Lomas Barbudal, Evolutionary Biology, University of Kansas Natural Bagaces, Guanacaste, Costa Rica History Museum, Lawrence, Kansas

ROXANA SALAZAR Fundación Ambio, San José, JOSÉ A. VARGAS Centro de Investigación en Costa Rica Ciencias del Mar y Limnología (CIMAR), Universidad de Costa Rica, San Pedro, Costa Rica JULIO E. SÁNCHEZ Museo Nacional de Costa Rica, San José, Costa Rica S. BRADLEIGH VINSON Department of Entomology, Texas A&M University, College MAHMOOD SASA Organization for Tropical Station, Texas Studies, San Pedro, Costa Rica, and Instituto Clodomiro Picado, Universidad de Costa Rica, NELSON ZAMORA Instituto de Biodiversidad San Pedro, Costa Rica (INBio), Santo Domingo, Heredia, Costa Rica

326 CONTRIBUTORS frankie index 8/20/03 8:14 PM Page 327

INDEX

Abangares River, 117 Agraulis vanillae, 107 Ammodramus savannarum, 150 Aburto, Gilda, 252 Agricultural waste, contamination Amphibians. See Reptiles and Acacia spp., 70 with, 122, 129, 141 amphibians Accipitridae, 156 Agriculture Anacardiaceae, 32, 270 ACG. See Area de Conservación buffer zones and, 63, 276, 301 Anartia fatima, 107 Guanacaste deforestation and, 271–72 Anatidae, 156 Acinote leucomelas, 106 land use and, 56 Ancyloscelis spp., 73 Acraeiniae, 106 pollution from, 122, 129, 141, 171 Andara tuberculosa, 139, 141 Acrocomia vinifera, 55 water use and, 7, 119 Andira inermis, 25, 67, 70, 90 Acromyrmex spp., 70 Aguacate group, 5 Animals, 9. See also individual ACT. See Area de Conservación Aguna asander, 88, 108 animal names Tempisque Aimophila botterii, 150 Ant-lion nests, 90 Actata spp., 89 Aimophila ruﬁcauda, 150 Ant-plant mutualism, 70 Adelpha iphiclus, 108 Albizia caribaea, 21 Anteater, 50, 51, 138, 142 ADIO. See Asociación de Desarrollo Albizia guachapele, 271 Anteos clorinde, 107 Integral de Ostiona Albizia saman, 219 Anthidium spp., 73 Aellopos titan, 89 Alouatta palliata, 49 Anthophorids, 21 African star grass, 42, 57–58, 67, 267 Altos de la Bonita, 293–96 Anthophorinae, 69 Africanized honeybee, 40, 68 Alvarado, Mario, 203 Ants Agalychnis callidryas, 184 Amazilia boucardii, 138, 139, 157 biodiversity of, 69, 74 Agave species, 165 Amazilia rutila, 150 as bioindicators, 71–75 Agelaius phoeniceus, 150 Amazona albifrons, 150 ecological impact of, 71 Agkistrodon bilineatus, 186, 187 Amazona auropalliata, 150, 157 needs of, 70–71 Agouti paca, 56 Amazona ochrocephala, 138, 140 sampling of, 72

327 frankie index 8/20/03 8:14 PM Page 328

Anurans, 179–80, 181 Arribadas, 195, 196, 204 Basilisk lizard, 183, 186 adaptations of to seasonal climate, Artibeus hirsutus, 161, 162 Bat Conservation International 184 Artibeus inopinatus, 161, 162 (BCI), 167 feeding ecology of, 183 Artibeus jamaicensis, 52, 53 “Bat houses,” 170 Mayan-Nahual tradition and, 186 Artibeus lituratus, 53, 162 Bat pollination systems, 163–64 microhabitat of, 179, 182 Artibeus phaeotis, 52, 53 in cloud forest, 23–25 reproduction of, 183–84 Artibeus spp., 167 in dry forest, 22, 23, 35, 40 Anypsipyla univitella, 90 Artibeus watsoni, 53 Bat roosts, 168–70 Aphelandra spp., 153 Artiodactyl, 49 Bats, 160–73 Aphrissa statira, 106, 107 Ascia monuste, 107 biodiversity of, 49, 50, 52–55, Apinae, 73 Asociación de Rescate y Conser- 160–61 Apis mellifera, 90 vación de Vida Silvestre biological control of insect pests Apis spp., 73 (ARCAS), 196 by, 165–66 Aquatic birds, 148, 154–55 Asociación de Desarrollo Integral climate change and, 171–72 Aquatic invertebrates, 118 de Ostional (ADIO), 204 conservation concerns, 161–62 Aquifers, 6 Astronium graveolens, 210, 270 deforestation and, 166–67 Ara macao, 138, 157 Asturina nitida plagiada, 150 endangered, 51, 52, 53–55, 162 Aratinga canicularis, 150, 157 Atalea butyracea, 55 environmental education for, Aratus pisonii, 139 Ateles geoffroyi, 50 172–73 Arauz, Randall, 200, 203 Atelopus varius, 187 global warming and, 172 ARC system, 301, 308 Atherinidae, 139 habitat alteration and, 166–67 ARCAS. See Asociación de Rescate y Atta spp., 70, 74, 90 habitat destruction and, 167–71 Conservación de Vida Silvestre Authority, identifying, in conflict in lowland rain forest, 49 Ardeidae, 156 resolution, 241–42 migration of, 162–63 Area de Conservación Arenal, 117 Avalanches, 7 mist netting of, 52 Area de Conservación Guanacaste Avicennia bicolor, 137, 138, 139 in Monteverde Cloud Forest, 161, (ACG), 8, 9, 81–82, 117, 119, Avicennia germinans, 137, 138, 140 164, 172 308 Avifauna. See Birds pollination by, 22, 23–25, 35, 40, carnivores in, 87 Azteca ant, 70, 74 163–64 climate of, 83–84 Azuero Peninsula, 178 pollution and, 171 environmental education for, 158 rabies and, 167–68 geology of, 83 seed dispersal by, 163 insect migration in, 87–91 Bactris major, 55 threats to bat roost sites, map of, 82 Bactris minor, 55 168–70 physical setting of, 82–85 Bagaces toxic chemicals and, 171 rainfall in, 83, 84 Centro para la Conservación de la Bauhinia ungulata, 88 rainy season in, 83, 85–87 Naturaleza, 250 Bay-winged Hawk, 149 temperature in, 83, 84 flowering phenology in, 18 BCI. See Barro Colorado Island; Bat tourism in, 272 volvanic aquifer in, 6 Conservation International Area de Conservación Tempisque Bagatzí, 121, 260 Beach. See under Playa (ACT), 8, 9, 117, 119 Baird’s tapir, 50 Bebedero River, 7, 127 tourism in, 272 Bait traps, for ants, 72 Bees, 20–21, 22, 40, 67–76 wildlife management plan for, Balantiopteryx plicata, 49, 54, 161 biodiversity of, 25, 69 273–74 Ballestero, Jorge, 203 as bioindicators, 71–75 Arenal-Corobici-Sandillal Hydro- Banded Wren, 154 fires and, 67–68 electric Project, 7 Bare-throated Tiger Heron, 149 honeybees, 40, 68, 72, 90 Arenal Lake, 117, 271 Barra del Colorado Biological nectar and, 70 Arenal River watershed, 7 Corridor, 296 needs of, 69–70 Arenal Tempisque Irrigation Barra del Colorado Wildlife Refuge, nest construction by, 70, 74 System, 7 286–87 pollination by, 20–21, 39 Ariidae, 139 Barranca River, 117 Beetles, 21, 23–25, 87, 90 Ario River, 7 Barro Colorado Island (BCI), polli- Belvosia spp., 89 Armadillo, 56 nation of dry-forest tree species Bignoniaceae, 22, 270 Army ant, 70 in, 32, 33 BINABITROP. See Organization for Arremonops rufivirgatus, 150 Basiliscus basiliscus, 183, 184, 186 Tropical Studies

328 INDEX frankie index 8/20/03 8:14 PM Page 329

Biodiversity Bombycoid species, 86 Cairina moschata, 156 of ants, 69, 74 Borgainaceae, 33, 34 Callinectes arcuatus, 128, 138 of bats, 49, 50, 52–55, 160–61 Botaurus pinnatus, 156 Callinectes toxotes, 138 of bees, 25, 69 Bothrops asper, 186 Calocitta formosa, 150 of birds, 9, 147–52 Boza, Mario, 200 Calycophyllum candidissimum, 219 of butterflies, 102–5 Brachiopoda, 128 Camponotus spp., 74 defined, 17 Braconidae, 86 Camptostoma imberbe, 150 of flowering plants, 18–29 Bradypus variegatus, 50 Campylorhynchus rufinucha, 150 of frugivorous bats, 49, 50, 52 Brahman cattle, 57 Cañas River, sugar industry and, 10 in Gulf of Nicoya, 127–39 Brazilwood, 270 Canids, 49 inventorying, 229–35 Breeding. See Reproduction Canis latrans, 49 Ley de Conservación de la Vida Breeding structures, of neotropical Cantil, 187 Silvestre, 111 dry-forest trees, 32 Caparis odoratissima, 137 management of, 317–18 Bright-rumped Attila, 155 Capuchins, 49, 50 of primates, 49, 50 Brosimum alicastrum, 58 Caracara, 156 of reptiles and amphibians, Brotherhood of the Turtle, 202 Caracara plancus, 150 178–79, 180–82 Brown-throated three-toed sloth, 50 Cardisoma crassum, 138, 140 of rodents, 49, 50, 58 Bruchid beetle, 87, 90 Caribbean Basin, climate of, 3 of sea turtles, 195–97 Buffer zones, 63, 276, 301 Caribbean Conservation Corpora- of trees, 9, 18, 19 Bufo marinus, 185, 186 tion (CCC), 196, 202 Biodiversity inventorying, 229–35 Bufo periglenes, 187 Caribbean tectonic plate, 5 Bioindicators Burhinus bistriatus, 150 Carnivores, Area de Conservación ants and bees as, 71–75 Burrowing frog, 183 Guanacaste (ACG), 87 threshold levels of, 74–75 Bursera simaruba, 58 Carollia brevicauda, 53 Biological Diversity Convention, Busarellus nigricollis, 156 Carollia castanea, 53 282 Buteo albicaudatus, 150, 156 Carollia perspicillata, 50, 53 Biotemperature, 8 Buteo albonotatus, 150 Carollia spp., 167 Birds, 147–58, 312 Buteogallus subtilis, 150 Carollia subrufa, 49, 54 biodiversity of, 9, 147–52 Buteogallus urubitinga, 156 Carr, Archie, 202, 204 conservation of, 155–58 Butterflies, 99–111 Carson, Rachel, 257 diet and habitat use of, 154 biodiversity of, 102–5 Cassia biflora, 70 endemism of, 152 census of in forest sites, 102 Cassia spp., 70 extinction of, 155–57 climate change and, 110–11 Castro, Juan Carlos, 203 in mangrove forests, 138, 142, cloud-forest pollination and, Caterpillars, 23, 89 148, 149 23–25 Cats, 49 monitoring of, 68 dry forest pollination and, 21 Cattails, 59, 274 pollination systems and, 25 dry season and, 104–6 Cattle, cattail control by, 274 reproductive activity of, 152–53 extinction of, 110–11 Cattle Egret, 148 in watershed, 118 migration behavior of, 99–102, Cattle grazing, 10, 58, 267, 275, 276 Bivalve mollusk, 139, 140 106–10 Cattle industry, fires and, 267, 274, Black-bellied Whistling Duck, 149 monitoring of, 109 276 Black-crowned Night Heron, 148 pesticide contamination and, 110 Cattle ranching Black-headed Trogon, 154 reproductive diapause and, 104–6 and deforestation, 267–68 Black-hooded Antshrike, 149 seasonality of fauna and, 104–6 grazing and, 10, 58, 267, 275, Black iguana, 185, 186 “Buzz-collecting,” 70 276 Black turtle, 195, 196, 197 Buzz pollination, 70 history of, 267 Blackbird, 157 Byrsonima crassifolia, 58, 70, 75 impact of, 267 Blepharipa spp., 89 mammals and, 56–57, 58 Blue crab, 128 wildlife management plans and,

Blue-throated Goldentail, 149 C4 grasses, 57 273–74, 277 Boa constrictor, 183, 188 Caceilians, 179–80, 181 Caves, as bat roosts, 169 Boat-billed Night Heron, 148 Cactus, 18, 19, 164 Cavity nesters, 153 Bolitoglossa mexicana, 187 Caesalpinia eriostachys, 67–68, 70 CCC. See Caribbean Conservation Bombacaceae, 22, 163, 270 Caesalpinia spp., 270 Corporation Bombacopsis quinata, 22, 213–23, 260 Caesalpiniaceae, 21, 270 Cebus capuchinus, 49, 138

INDEX 329 frankie index 8/20/03 8:14 PM Page 330

Cecidomyiidae, 138 CITES. See Convention on Inter- Conservación de Murciélagos de Cecropia peltata, 50 national Trade in Endangered Costa Rica, 173 Cecropia spp., 70, 167 Species Conservación de Murciélagos Cedrela odorata, 34, 219, 270 Clerodendrum pittieri, 137 Migratorios (PCMM), 172 Cenizero (tree), 58, 270 Climate, 3–5, 100. See also Conservation, 11–12 Central Valley, 130 Phenology biodiversity inventorying and, Centris bicornuta, 70 of Area de Conservación Guana- 229–35 Centris flavifrons, 70 caste (ACG), 83–84 of birds, 155–58 Centris flavofasciata, 70, 74 of Caribbean Basin, 3 circa situm conservation, 212 Centris spp., 21, 23, 25, 59, 73, 75 of dry forests, 48 conflict resolution and, 237–45 Centro para la Conservación de la of Pacific Basin, 2 ecological restoration for, 211–12 Naturaleza, 250 rainy season, 83, 85–87 education. See Environmental Centropomidae, 139 Climate change. See also Phenology education Centurio senex, 52, 53 bats and, 171–72 environmental law and, 281–88, Cerambycid beetles, 90 butterflies and, 110–11 289–97 Cerithidea montagnei, 139 changes in species composition ex situ conservation, 212 Cerithidea pulchra, 139, 140 and, 43–44 farmer-based conservation, 212 Cerithium stercusmuscarum, 139 flowering and, 41 in Gulf of Nicoya, 131 Cerro El Hacha, 6 gene flow and, 43 in situ conservation, 211, 220 Cerro Guanacaure, 215 genetic diversity and, 43 innovation in, 300 Cerro Las Tablas, 217–18 tree mating systems and, 42–43 of mangrove forests, 140–41 Cerro Vista al Mar, 7 Climbers, 18 media and, 257–65 Cerros La Carbonera, 7 Cloud forest, 9 nongovernmental organizations Cerros Las Tablas, 215 bats in, 161, 163, 164, 172 and, 315 Chaetognatha, 128 butterflies in, 102 of plants, 25 Chamela-Cuixamala Biosphere pollination systems in, 23–25 policy context of, 299–309 Reserve (Mexico), 266 Coastal-Maritime Law (1977), 140 private landowners and, 316–21 Chaves, Anny, 200, 203 Coatis, 49, 50 of reptiles and amphibians, Chelonia mydas, 186, 195 Cochliomyia hominivora, 167 187–90 Chetogena spp., 89 Cochlospermum vitifolium, 67–68, 70 restoration ecology and, 276–77 Chicon, Didier, 196 Cockles, 141 of sea turtles, 198–204 Chione subrugosa, 139 Cocobola (wood), 270 sustainable development and, Chiquito River, 117 Cocoon-forming frogs, 185 63–64 Chiroderma villosum, 54 Coco’s tectonic plate, 5 threats to, 266–77 Chironectes minimus, 50, 138 Cocytius anteus, 89 of trees, 210–24, 268–69 Chiroxiphia linearis, 150 Coelioxys spp., 73 of watershed, 122–23 Choeronycteris mexicana, 161, 162 Coendou mexicanus, 49 Conservation areas, 4, 8–9. See also Choloepus hoffmanni, 50 Coenipita spp., 86 Area de Conservación Arenal; Chondrohierax uncinatus, 156 Coffee mills, pollution by, 130 Area de Conservación Guana- Choreja, 59 Coleoptera, 166 caste; Area de Conservación Chorotega people, hunting habits Colinus cristatus leucopogon, 150 Tempisque of, 55–56 Collared anteater, 138 Conservation in hortus, 212 Chorotega region, 8, 118 Collared Forest Falcon, 149 Contamination. See Pollution biodiversity in, 118 Collared peccary, 56 Controlled grazing, 275 hydrogeology of, 6 Colletes spp., 73 Convention on Climate Change, land-use imbalances in, 6 Colletinae, 73 282 life zones in, 3, 8 Columbina inca, 150 Convention on International Trade tourism in, 11 Columnar cacti, bat pollination of, in Endangered Species Chrotopterus auritus, 53, 162 164, 165 (CITES), 202 Ciconiidae, 153, 156 Combretum farinosum, 153 Convict cichclids, 118 Circa situm conservation, 212, Conflict resolution, in conservation Cordia alliodora, 34, 218, 219 213–20 work, 237–45 Cordillera de Tilarán, 100–102 Circus cyaneus, 156 Conocarpus spp., 219 Core areas, 301 Cissia renata, 105 Conophis lineatus, 183 Cornelius, Steve, 203

330 INDEX frankie index 8/20/03 8:14 PM Page 331

Corobicí River, mammal biodiversity, Ctenosaura spp., 183 Drino spp., 89 50 Cuckoos, 149 Dry forests (Costa Rican) Cortez amarillo (wood), 270 Cultural values, 185–86, 240, 251 agricultural impact on. See Cossidae, 90 Curatella americana, 58, 70 Agriculture Costa Rica Curculionidae, 90 bats of, 160–73 agriculture in, 271–72 Cyclopes dydactilus, 142 bees of, 20–21, 22, 40, 67–76 biodiversity inventories in, 230–35 Cytarops alecto, 54 birds of, 9, 25, 68, 147–58, 312 biodiversity law in, 111 buffer areas and, 63, 276, 301 cattle industry in, 267–68 butterflies of, 99–111 climate of, 3–5, 100 Dalbergia retusa, 67–68, 70, 270 cattle ranching and, 56–57, conservation areas in, 4, 8–9 Danainae, 107 267–68, 273–74, 275, 276 environmental law. See Environ- Danaus eresimus, 107 climate of, 48 mental law Danaus gilippus, 107 deforestation of. See Deforestation geography of, 1–5 Danaus plexippus, 99, 107 ecological threats to, 39 geohazards and disasters in, 7–8 Dasypus novemcinctus, 56 flowering plant pollination in, geology of, 5, 6 DDT, toxicity of, 61 17–23, 25–27 hydrology of, 6–7, 115–23 Deadwood-nesting bees, 67 herpetofauna of, 177–91 land use capacity in, 5–6 Decapod crustaceans, 128 history of, 48 life zones in, 3, 8 Deciduous forests, birds in, 149 insects of, 80–92 logging in, 187, 267–71 Deforestation, 6, 10, 40–43 land conversion in, 56 park system in, 301–5 agriculture and, 271–72 mammals of, 48–64 poaching in, 56, 196 bats and, 166–67 mangrove forests of, 136–42 pollution in, 11 cattle ranching and, 267–68 regeneration of, 63 river basins in (map), 2 changes in species composition reptiles and amphibians of, soils of, 5 due to, 43–44 177–91 terrestrial-maritime zone in, 11 genetic diversity and, 43 sea turtles of, 68, 194–207, 312 timber industry in, 187, 267–71 logging and, 187, 267–71 stream corridor biota in, 118 tourism in, 11, 272–73, 277, 301–2 mammals and, 56, 57 threats to, 266–77 watersheds in, 115–23 pollinator population and, 41–42 tourism and, 11, 272–73, 277, wildlife of, 9 reptiles and amphibians and, 187 301–2 Costa Rican dry forest. See Dry timber industry and, 267–71 tree reproduction in, 38–45 forests (Costa Rican) Dendrobatidae, 184 trees in fragmented landscapes Cote Lake, 117 Dendrocolapidae, 142 in, 30–36 Coto Brus (Costa Rica), 34, 35 Dendrocygna viduata, 155–57 Dry season Cotton rat, 58 Dendroica erithacorides, 138, 140 bird reproduction during, Cottontail rabbit, 58 Dendroica petechia xanthotera, 150 152–53 Coyotes, 49, 50 Denham, John, 196 butterflies in, 104–6 Crabs, in mangrove forests, 138–39, Deppe’s squirrel, 51 insects in, 90 140 Dermatemys mawii, 186 Dryas iulia, 107 Crane, 153 Dermochelys coriacea, 195 Drymarchon corais, 183, 188 Crematogaster spp., 71, 74 Desmodus rotundus, 53, 167 Ducks, 149, 153, 155, 156 Crescentia alata, 22, 58 Detracia graminea, 139 Crevices, as bat roosts, 169 Diapause, 104–5 Crocodiles, 179–80, 181, 184, 186, Diclidurus albus, 54 Eagles, 156 188 Didelphis virginiana, 49 Earthquakes, 8 Crocodylus moreletii, 186 Diphaglossinae, 73 Echiura, 128 Crotalus durissus, 183, 186, 188 Diphylla ecaudata, 53 Eciton burchelli, 70–71 Crustaceans, 128, 138, 139, 140 Diptera, 166 Eciton spp., 74 Cruz, Guillermo “Billy,” 202 Diversity, of stream corridor biota, Eclosion, 85, 86 Cryptic soil-litter ant, 70 118 Ecological restoration, 211–12 Crypturellus cinnamomeus, 150 Dolichostelis spp., 73 Ecosystem monitoring, 68–69 Ctenosaura palearis, 187, 190 Dove, 149 Ecotourism, 301–2 Ctenosaura quinquecarinata, 187 Dr. Lucas Rodríguez Caballero Ectatomma spp., 74 Ctenosaura similis, 186 National Wildlife Refuge, 273 Edentates, 49

INDEX 331 frankie index 8/20/03 8:14 PM Page 332

Education. See Environmental Environment Law (1995), 140 Fabaceae, 25, 32, 270 education Forestry Law (1996), 120, Fagaceae, 22 Egret, 148, 156 283–86, 292 Falco peregrinus, 156 EIAs. See Environmental impact impunity of violators, 291 Falconidae, 156 assessments incongruity in, 289–90 Falconiformes, 152 Eichhornia crassipes, 59 judges’ lack of environmental Falcons, 149, 152, 156 Eichhornia heterosperma, 59 knowledge and, 290 Farmer-based conservation, 212 El Niño, 10, 109, 139 judicial power and, 290–91 Farms, circa situm conservation, 212 El Salvador Ley de Conservación de la Vida Fauna, 9. See also individual animal reptiles and amphibians in, 179 Silvestre (1992), 111 names sea turtles in, 196 national parks and wildlife FECON, 253 Elegant Trogon, 154 refuges, 284–85 Felids, 51–52 Eleutherodactylus rhodopis, 187 Organic Environmental Law, Felis jaguarundi, 138 Eleutherodactylus spp., 183–84 283–84 Fertilizers, pollution with, 121 Eleutherodactylus stejnegerianus, 187 protection zones, 285–86 Ficus spp., 164 Ellobium stagnalis, 139 Water Law No. 276 (1942), 120 Fiddler crab, 140 Encospilus spp., 86 for water protection, 285 Fig wasps, 21, 23–25 Enea, 57 Wildlife Conservation Law (1987), Finqueros, 57 Engraulidae, 139 204, 285 Fire conservation, environmental Enterolobium cyclocarpum, 21, 22, Wildlife Law (1992), 140 education for, 158, 169, 247–55, 32–34, 42, 58, 219, 270 Enyo ocypete, 89 268 Environment Law (1995), 140 Eolian erosion, 6 Fires, 267–68 Environmental change, 10–11 Epargyreus spp., 86 bees and, 67–68 bioindicators to monitor, 71–73 Epicharis spp., 21, 69, 73 cattle industry and, 267, 274, 276 mating systems and, 39–40 Epiphytes, 18, 19 environmental education to pre- phenology and, 39 Eptesicus fuscus, 170 vent, 158, 169, 247–55, 268 reproductive biology of dry-forest Eptesicus serotinus, 167 environmental impact of, 10, 42, trees, 38–45 Eretmochelys imbricata, 195 158 Environmental conflict, land use Erosion, 6 habitat destruction by, 187 and, 239–40 Estero de Tamarindo, 200 jaragua grass and, 42, 57–58, 67 Environmental education Estero San Francisco, 198 Fish absence of, 293 Estero Tamarindo, 198 freshwater fish, 118 bats and, 172–73 Estivation, 184 in Gulf of Nicoya, 128 Center for Conservation of Estivation cocoons, 185 in mangrove forests, 138, 139 Nature, 250 Ethmia spp., 86 Fishing resources, in Gulf of definition of, 248 Ethmiidae, 86 Nicoya, 128–29 for fire conservation, 158, 169, Euglossinae, 21, 69, 74 Flatworms, 128 247–55, 268 Eulepidotis spp., 86 Flies, 138 materials for, 75 Eumomota superciliosa, 150 Flooding, 6, 7 outreach programs for, 248–49 Eumops auripendulus, 54 Flora, 9. See also individual plant and to schools, 249–50 Eumops glaucinus, 162 tree names sea turtles and, 203 Eumops underwoodi, 55, 161 Floral visitation counts, 72 tree conservation and, 268–69 Euphonia affinis, 150 Flowering at visitor center at Lomas Euptoieta hegesia, 107 climate and, 41 Barbudal, 250–51 Eupyrrhoglossum sagra, 89 of mangroves, 139–40 Environmental impact assessments Eurema daira, 100, 105 Flowering plants (EIAs), 283 Eurema dina, 104, 105 biodiversity of, 18–29 Environmental impacts, 10–11. See Eurema spp., 108 phenology of, 17–20, 39 also individual impacts Eurybia elvina, 108 pollination systems of, 20–27 Environmental law, 281–88, Eurytides epidaus, 86 Flowers 289–97 Eurytides philolaus, 86 bat flowers, 22 Biodiversity Law (1992), 111 Euscirrhopterus poeyi, 86 climate change and, 41 case studies in, 286–87, 293–97 Ex situ conservation, 212 large-bee flowers, 20–21, 23 Coastal-Maritime Law (1977), 140 Exomalopsis spp., 73 moth flowers, 22, 23

332 INDEX frankie index 8/20/03 8:14 PM Page 333

phenology and, 41 Genetic diversity phenology in, 41 pollination systems and, 20–27 deforestation and, 43 soils of, 5–6 small-bee/generalist flowers, 21 phenology and, 43 Guanacaste Conservation Area. Flycatcher, 152, 154 of trees, 214 See Area de Conservación Foam nests, of frogs and toads, 184 Genetic drift, forest fragmentation Guanacaste Forest fragmentation, 30–36, 40–43, and, 31 Guanacaste Moutain Range, 3 58 Genetic variation, forest fragmen- Guanacaste National Park, 50, 51 changes in species composition tation and, 31, 35 Guapinol (wood), 270 and, 43–44 Geocarcinus quadratus, 140 Guarea glabra, 21 effects of, 30–32 Geology, 5 Guatemala pollination and, 32–36 Geranospiza caerulescens, 156 reptiles and amphibians in, 179 pollinator population and, 41–42 Gerreidae, 139 sea turtles in, 196 Forest regeneration, 63 Giant anteater, 50, 51 Guayacan real (wood), 270 Forestry Law (1996), 120, 283–86, Gill nets, 141 Guayaquil (wood), 270 292 Glaucidium brasilianum, 150 Guazuma ulmifolia, 58, 219 Forests. See also Dry forests (Costa Gliricidia sepium, 70, 211 Guiraca caerulea, 150 Rican); Forest fragmentation; Global warming, bats and, 172 Gulf of Nicoya, 3. See also Nicoya Monteverde Cloud Forest Glossophaga alticola, 161 Peninsula inbreeding, effects of, 42–43 Glossophaga commissarisi, 53, 165 biota of, 127–39 logging in, 187, 267–71 Glossophaga leachii, 49, 54 conservation in, 131 Formicidae, 69 Glossophaga morenoi, 161 as estuary, 126–32 Fragmentation Glossophaga soricina, 53 fishing resources of, 128–29 of forests. See Forest fragmentation Glossophaga spp., 50 geology of, 5 of riparian system, 120–21 Glossophagine bats, 162 mangrove forests of, 137, 140, 141 Fragmented habitats, 30–36, 110 GMA. See Great Metropolitan Area maps of, 2–4 Frankie, Gordon, 260 Gmelina arborea, 270 pollution in, 129–30 Freshwater fish, 118 Gnampogenys spp., 74 population growth and, 130 Freshwater habitats, birds in, 149 GNB. See Gulf of Nicoya Basin rivers flowing to, 117–18 Freshwater turtles, 186 Goacampa variabilis, 86 threats to, 122 Frogs and toads, 179–80, 181 Golden toad, 187 Gulf of Nicoya Basin (GNB), 3–4, 5 adaptations of to seasonal climate, Gopher, 51 Gulf of Nicoya Contingency Plan, 184 Grandiarca grandis, 139, 141 129 feeding ecology of, 183 Granivores, 149 Gymnopis multiplicata, 187 Mayan-Nahual tradition and, 186 Grass-cutter ant, 70 microhabitat of, 179 Gray sac-winged bat, 49 reproduction of, 183–84 Gray short-tailed bat, 49 Habitat, definition of, 81 Frugivorous bat, 49, 50, 52 Grazing, 10, 58, 267, 275, 276 Habitat fragments, 68, 74 Fruit bat, 167 Great false vampire bat, 51 Hacienda Monteverde, mammal Funnel-eared bat, 50 Great Metropolitan Area (GMA), 130 biodiversity in, 50 Green-backed Night Heron, 148 Haciendas River, 6 Green iguana, 183, 186 Halictids, 21 Galictis vittata, 50 Green turtle, 195, 202, 203 Halictinae, 73 Galls, in Avicennia, 138 Greta oto, 107 Hardwood trees, biodiversity of, 9 Gandoca-Manzanillo case, 287 Grison, 50, 51 Harlqeuin frog, 187 Gastropods, 139, 140 Ground-dweller mollusk, 140 Harvest mouse, 49, 50 Gecarcinus quadratus, 138 Groundwater, 6, 117–18 Hawkmoths, 22 Geest Caribbean, 286 Guacimal River, 117 Hawks, 149, 153, 156 Gene flow Guaiacum sanctum, 210, 270 Hawksbill turtles, 195 climate change and, 43 Guanacaste (tree), 32–34, 58, 270 Heavy metal contamination, 130 deforestation and, 43 Guanacaste (region) Heliconiidae, 107 forest fragmentation and, 31, 34 biodiversity in, 9 Heliconius sara, 109 Generalist insect pollination systems climate of, 41 Heliomaster constantii, 150 in cloud forests, 23–25 geology of, 5 Heliotropium curassavicum, 137 in dry forest, 21–22, 23 land-use capacity in, 5 Heloderma horridum, 184, 190

INDEX 333 frankie index 8/20/03 8:14 PM Page 334

Herbs, 18, 19 Insect Museum, University of Costa Laguncularia racemosa, 137, 138 Heron, 62, 148, 149, 153, 156 Rica, 232 Lake Arenal, 117 Herpetofauna. See Reptiles and Insecticides, toxicity of, 61 Land birds, 154–55 amphibians Insectivorous bats, 50, 52, 165–66, Land crabs, 138 Hesperiidae, 86, 89, 103, 104, 105, 167 Land use, 239 108 Insects, 80–92 Landslides, 6, 7 Heteromyidae, 90 bats as biological control of, Large-bee pollination systems Hirtella racemosa, 90 165–66 in cloud forest, 23–25 Hoffmann’s two-toed sloth, 50, 51 in dry season, 90 in dry forest, 20–21, 22, 40 Holochroa spp., 86 eclosion by, 85, 86 Large predatory ponerines, 70 Homoptera, 138 in mangrove forests, 138 Las Baulas Marine National Park, 7, Honduras phenology and, 85 198–202 circa situm conservation in, in rainy season, 83, 85–87, 90 Lasiurus blossevillii, 53 213–20 Institute of Agrarian Development Laterallus ruber, 150 patterns of distribution of trees (IDA), 250 Laurel (tree), 34 in, 215–16 Institute of Biodiversity (INBio), Leaf-chinned bat, 50 reptiles and amphibians in, 230–33, 252, 303 Leaf-cutter ant, 70, 90 179–80 Intensive extraction zones, 301 Leaf gall, in Avicennia, 138 Honey bee, 40, 68, 72, 90 Ironwood, 270 Leatherback turtles, 195 Hooded skunk, 49 Irrigation water, 7, 10 conservation of, 203 Howler monkey, 49, 51 Isla Capitan, 198 extinction of, 196, 197–98 Humenaea courabaril, 21 Isla Pajaros, 129 nesting sites of, 196 Hummingbirds, 157 Itaballia demophile, 105 Lechuga spp., 59 cloud-forest pollination by, 23–25 Ithomiinae, 105, 106, 107 Legal system, environmental law dry forest pollination by, 21, 35 Ixobrychus exilis, 156 and, 289–97 in mangrove forests, 138, 139 Leopardus pardalis, 51 reproduction of, 152, 153 Leopardus tigrinus, 51 Hunting, of mammals, 55–56 Jabiru, 149 Leopardus wiedii, 51 Hydrogeology, 6 Jabiru mycteria, 156 Lepidochelys olivacea, 195 Hydrology, 6–7, 119–22 Jacana, 149, 153 Lepidoptera, 86, 89, 100 Hyla ebraccata, 184 Jaguar, 51 Leptodactylus labialis, 184 Hylonycteris underwoodi, 49, 54, 162 Jaguarundi, 51, 138 Leptodactylus melanonotus, 184 Hymenaea courbaril, 90, 270 Jamaican fruit-eating bat, 52 Leptodon cayanensis, 156 Hymenoptera, 69, 71, 166 Janzen, Dan, 308 Leptonycteris curasoae, 162 Hyparrhenia rufa, 42, 57, 254, 267 Jaragua grass, 42, 57–58, 67, 267 Leptonycteris nivalis, 162 Hyphantrophaga spp., 89 Jobo (tree), 32 Leptonycteris spp., 163 Hypopachus variolosus, 183 Judicial power, and environmental Leptothorax spp., 74 Hypsipyla grandella, 214 law, 290–91 Leptotyphlops goudotti, 183 Junonia evarete, 138, 139 Lesser Ground Cuckoo, 149 Letis spp., 86 Ibis, 148, 156 “Let’s Save the Estuary” program, Ichneumonidae, 86 Kinorhyncha, 128 129, 131 Icteridae, 157 Kinosternon scorpioides, 187 Leucaena salvadorensis, 212, 221 Icterus pectoralis, 150, 157 Kinosternon spp., 186 Leucaena spp., 211, 212 Icterus pustulatus, 150 Koberg, Maria Teresa, 200 Ley de Conservación de la Vida IDA. See Institute of Agrarian Silvestre, 111 Development Lianas, 18, 19, 165 Iguana, 183, 186, 187 La Mula Creek, conservation in, Liberia River, sugar industry Iguana iguana, 183, 186 259–61 and, 10 In situ conservation, 211, 220 La Selva Licania arborea, 90 INBio. See National Biodiversity bat biodiversity in, 49, 161 Lichonycteris obscura, 54 Institute pollination systems of, 25, 164 Life zones, 3, 8 Inbreeding, in forests, 42–43 Labidus spp., 74 Limacodidae, 86 Inca Dove, 149 Lagartos River, 117 Limbopsylla lagunculariae, 138 Inga vera, 22, 90 Lagoons, birds in, 149 Liomys salvini, 49, 90

334 INDEX frankie index 8/20/03 8:14 PM Page 335

Lirio de agua, 59 Manataria maculata, 99, 100, 106, Megalopta spp., 73 Lithurgus spp., 73 107 Megalopygidae, 86 Little Blue Heron, 148 Manatee, 51 Melampus carolianus, 139, 140 Littoraria zebra, 139, 140 Manduca dilucida, 86 Melanerpes hoffmannii, 150 Lizards, 179–80, 181 Manduca lefeburii, 86 Meliaceae, 21, 34 adaptations of to seasonal climate, Manduca occulta, 89 Melina, 270 184 Manduca sexta, 89 Melipona spp., 73 feeding ecology of, 183 Mangled howler monkey, 49 Meliponinae, 21, 69 Mayan-Nahual tradition of, 186 Mangrove cockle, 141 Melipotis spp., 86 microhabitat of, 179, 182 Mangrove crab, 140 Melitoma spp., 73 reproduction of, 183 Mangrove forests, 136–42. See also Melongena patula, 139 Local culture, 240, 251 Mangroves Mena River, 6 Logging, 187, 267–71 birds in, 138, 142, 148, 149 Mephitis macoura, 49 Lomas Barbudal conservation of, 140–41 Mesoamerica mammal biodiversity in, 50 crustaceans in, 138, 139, 140 federal and state reserves in, 189 tree distribution in, 215, 217–18 fauna of, 137–41 lowland dry forest in, 177–78 visitor center at, 250–51 fish in, 138, 139 mammal biodiversity in, 48, 49 wildlife management plans for, in Gulf of Nicoya, 137, 140, 141 Mesocheira, 21 273–76 insects in, 138 Mesoplia spp., 21, 73 Lonchorhina aurita, 162 mammals in, 138 Metal contaminants, 130, 141 Long-tailed Manakin, 149, 154 mollusks in, 138, 139, 140 Metavoria spp., 86 Long-tongued bat, 49 phenology and, 139–40 Mexican porcupine, 49 Lucas Rodríguez Caballero National plant species in, 137 Mexico Wildlife Refuge, 273 pollution of, 141 reptiles and amphibians in, 179 Lutra longicaudis, 50 seasonal patterns of, 139–40 sea turtles in, 196 Lycaenidae, 103, 104 soil salinity and, 137 Micrastur semitorquatus, 156 Lygropia erythrobathrum, 138 structural attributes of, 137 Micronycteris brachyotis, 52, 53 Lysiloma auritum, 21, 22 Mangrove vireo, 138 Micronycteris hirsuta, 54 Lysiloma spp., 219 Mangroves, 128. See also Mangrove Micronycteris microtis, 53 forests Micronycteris minuta, 52, 53, 162 flowering of, 139–40 Micronycteris nicefori, 53 Macao, 138 in Gulf of Nicoya, 137, 140, 141 Micronycteris schmidtorum, 53, 161 Macrophyllum macrophyllum, 54 in Pacific Coast, 136–37 Micronycteris sylvestris, 54, 162 Mahogany, 270 timber use and, 219 Microplitis spp., 86 Mammals, 48–64 Manilkara chicle, 90 Micrurus nigroinctus, 183 biodiversity of, 9, 49 Manilkara zapota, 22 Migration cattle ranching and, 56–57, 58 Mantled howler monkey, 51 of butterflies, 99–102, 106–10 deforestation and, 56, 57 Margay, 51 definition of, 99 distribution of, 49–50 Marine turtles, 186 of insects, 87–91 endangered species of, 51–52 Marine worms, 128 of moths, 100 endemism, 49–50 Marinula concinna, 139 of spider monkey, 55 extirpated species, 50–51 Marpesia petreus, 106, 108 Mimon cozumelae, 54 hunting pressures on, 55–56 Marshes Mimon crenulatum, 54 in mangrove forests, 138 birds in, 149 Mimosaceae, 21, 22, 270 migratory species of, 52, 55 cattails in, 59 Mimosoid legumes, 32 monitoring of, 52, 68 Marsupials, 49 Mimosoidae, 32 pesticide contamination and, 59, Masked Tityra, 155 Ministerio de Recursos Naturales, 61–62 Mating systems, of dry-forest trees, Energía y Minas (MIRIMEM), poaching of, 56 39–40, 42–43 302 programs about for park personnel, Mayan/Nahual civilization, amphib- Ministry of Environment and 62 ians and reptiles in, 185–86 Energy (MINAE), 158, 302, 314 research priorities involving, 62 Mechanitis polymnia, 105 lack of enforcement by, 292–93 riparian mammals, 118 Media, conservation and, 257–65 mission of, 283 sustainable development and, Megachile spp., 73 problems of, 205, 291 63–64 Megachilinae, 21, 69, 73 Mist netting, 52, 167

INDEX 335 frankie index 8/20/03 8:14 PM Page 336

Mites, 138 Myrmecophaga tridactyla, 50 climate of, 4 Mo, Claudette, 203 Myrospermum frutescens, 70 hydrology of, 7 Mollusks, 128, 138, 139, 140 Mytella guyanensis, 139 Noctilio albiventris, 54 Molossids, 167 Noctilio leporinus, 53, 162 Molossus ater, 54 Noctuidae, 86, 89 Molossus molossus, 54 Nacardium excelsum, 50 Nomada spp., 73 Molossus pretiosus, 54 Naranjo, Isabel, 203 Nomadinae, 73 Molossus sinaloae, 54 Nasua narica, 49 Nomonyx dominicus, 156 Monarch butterfly, 99 Natalus stramineus, 50, 53 Nongovernmental conservation Monitoring Natica chemintzii, 139 organizations, 315 biodiversity inventorying, 229–35 National Bibliography on Tropical Norops cupreus, 183, 184 of butterflies, 109 Biology, 233 Norops pentaprion, 187 ecosystem health and, 68–69 National Biodiversity Institute Norops sagrei, 183 of mammals, 52, 68 (INBio), 230–33, 252, 303 Northern-barred Woodcreeper, 154 of pollinators, 26–27, 42, 44 National Meteorological Institute, Northern Jacana, 149 Monomorium spp., 74 285 Northern raccoon, 138 Monteverde Cloud Forest, 9 National Museum, 232, 253 Nosara River, 7 bats in, 161, 164, 172 National parks, 56, 284–85, 301–5 Notodontidae, 86 butterflies in, 102 Guanacaste National Park, 50, 51 Nyctinomops femorosaccus, 161 pollination systems in, 23–25 Las Baulas Marine National Park, Nyctinomops macrotis, 161 Monteverde Reserve Complex 7, 198–202 Nymphalidae, 89, 103, 104 butterfly migration in, 100 politics and policies affecting, Nymphalinae, 106–8 map of, 101 306–7 Mormoopid bats, 165 Santa Rosa National Park, 103, Mormoops megalophyla, 162 163, 217–18, 266 Ocelot, 51 Morococcyx erythropygius, 151 Tortuguero National Park, 286, Ocotea veraguensis, 90 Moth pollination systems 296 Odocoileus virginianus, 49, 138 in cloud forest, 23–25 National System of Conservation Oedipina taylori, 187 in dry forest, 22, 23, 39, 40 Areas, 4, 9. See also Area de Olive ridley turtles, 195, 196, 203, Moths Conservación Arenal; Area de 204 in mangrove forests, 138, 139 Conservación Guanacaste; Area Oncilla, 51 migration of, 100 de Conservación Tempisque Open-area butterflies, 105 phenology and, 86–87 National System of Protected Areas, Opossum, 49, 50, 56, 138 Mucuna urens, 165 284 Orange-throated bat, 52 Mud turtles, 187 Nectar feeders, 149 Organic Environmental Law, Muir, John, 257 Nectar plants, 70 283–84 Mula Creek, conservation in, Nectarivorous bats, 50 Organization for Tropical Studies 259–61 Neococytius cluentis, 89 (BINABITROP), 233 Multivoltinism, 86 Neotropical Foundation, 296 Organochlorine pesticides, 61, 129 Museums, biological collections of, Nepenthes Conservationist Group, Organophosphate pesticides, 61 232 296 Oriole, 149, 155, 157 Musonycteris harrisoni, 161, 162 Nests Ornate Hawk-Eagle, 149 Myiarchus nuttingi, 151 ant-lion nests, 90 Orosi Volcano, 6 Myiarchus tyrranulus brachyurus, 151 bees, nest construction by, 70, 74 Ortalis vetula, 151 Myiodynastes maculatus, 151 cavity nesters, 153 Orthogeomys underwoodi, 51 Myotis albescens, 54 foam nests, 184 Osmia spp., 73 Myotis carteri, 161 of frogs and toads, 184 Ostional National Wildlife Refuge, 7 Myotis elegans, 54, 162 sea turtle nesting sites, 196–97 Otters, 50–51 Myotis findleyi, 161, 162 wood-nesting bees, 75 Otus cooperi, 151 Myotis grisescens, 171 Nicaragua Ouratea lucens, 90 Myotis lucifugus, 170 reptiles and amphibians in, 180 Owl, 149 Myotis nigricans, 54 sea turtles in, 196 Myotis riparius, 54 Nicoya Peninsula, 3, 119. See also Myotis velifer, 162 Gulf of Nicoya Paca, 56 Myotis yumanensis, 170 biodiversity of, 9 Pachira quinata, 270

336 INDEX frankie index 8/20/03 8:14 PM Page 337

Pachycondyla spp., 74 butterflies and, 110 Plethodontid salamanders, 183–84 Pachylia ficus, 89 mammals and, 59, 61–62 Poaching, 56, 196 Pachylioides resumens, 89 Pet trade, 186 Pochota quinata, 163, 270 Pachyramphus aglaiae, 151 Peter’s tent-making bat, 55 Pocket mouse, 90 Pacific Basin, climate, 2 Pheidole spp., 70, 74 Point-source pollution, 11 Pacific lowlands, 49, 101–2, 103 Phenology, 38. See also Climate; Policy reforms, conservation and, Pacuarae Reserve, 196 Climate change 299–309 Padilla, Clara, 200 changes in species composition Polioptila albiloris, 151 Paelloa spp., 89 and, 43–44 Polistes instabilis, 107 Painted lady butterfly, 99 environmental global change Polistes variabilis, 88 Palo Verde and, 39 Political Constitution of Costa Rica, mammal biodiversity in, 51, 52–55 flowering plants and, 17–20, 41 282 marshes, cattails in, 59 genetic diversity and, 43 Pollen movement, dry forest trees photos of over time, 60–61 insects and, 85 and, 34, 40 Palo Verde Lagoon, 60–61, 274 managing climate change, 44–45 Pollination systems Palo Verde National Park, 56, 62, mangrove forests and, 139–40 additional information needed 157, 266, 273–76 plants and, 87 about, 26 Palustrine wetlands, 9 pollinator populations and, 41–42 declining pollinator populations, Panama reproductive diapause, 104–5 26–27 reptiles and amphibians in, 180 trees and, 30–35 in dry forest, 20–23, 25–26 sea turtles in, 196 Philoxerus vermicularis, 137 forest fragmentation and, 32–36 Pandemos godmanii, 103 Phoebis sennae, 104 in La Selva, 25 Pandion haliaetus, 156 Phragmosis, 184 in Malaysian forest, 25 Panurginae, 73 Phrynohyas venulosa, 185 in Monteverde Cloud Forest, Papilio homerus, 110 Phyllostomidae, 51, 164, 165, 169 23–25 Papilio thoas, 105 Phyllostominae, 51 Pollinators Papilionidae, 86, 89, 103, 104, 105, Phyllostomus discolor, 54 environmental change and, 40 110 Physalaemus pustulosus, 184 foraging ranges and, 40 Parabuteo unicinctus, 151, 156 Phytophagous insects, phenology, monitoring of, 26–27, 42, 44 Parasitic plants, 18, 19 85–87 Pollution Parataxonomist, 230–31 Pieridae, 103–8 bats and, 171 Paratetrapedia spp., 73 Pinae, 73 control of, 11 Parides iphidamus, 105 Piper spp., 164, 167 of Gulf of Nicoya, 129–30 Parque Marino Las Baulas, 7, Pistia stratiotes, 59 by heavy metal contamination, 198–202 Pit viper, 187 130 Parque Nacional Guanacaste, 50, 51 Pithecellobium elegans, 42 of mangrove forests, 141 Parque Nacional Palo Verde, 56, 62, Pithecellobium lanceolatum, 22 by metal contaminants, 141 157, 266, 273–76 Plant associations, 8 by pesticides, 59, 61–62, 110, Parque Nacional Santa Rosa, 103, Plant inventories, 26 121–22, 171 163, 217–18, 266 Plant reproduction, 17–27 point-source pollution, 11 Parrots, 138, 153, 157 phenology and, 17–20 population growth and, 121 Passerina ciris, 151 pollination systems, 20–27 from residual waters, 121 Passiflora auriculata, 109 Plants. See also individual plant names by sewage, 121 PCMM. See Conservación de conservation of, 25 by solid waste, 122, 129, 141 Murciélagos Migratorios overview of, 9 of Tempisque River, 61, 129 Pecari tajacu, 56 phenology and, 87 of Tempisque River Basin (TRB), Peccary, 50, 51, 55, 56 plant inventories, 26 11, 61 Pelliciera rhizophorae, 138, 139 Platyrinchus cancrominus, 151 Polymesoda inflata, 139, 140 Penaeus stylirostris, 141 Platyrrhinus helleri, 53 Ponerine ants, 71 Peponapis spp., 73 Playa Grande, 198, 200 Population growth Perching moth, 22 Playa Langosta, 198, 200 bats and, 171 Peropteryx kappleri, 54 Playa Nancite, 203, 215, 217–18 in Gulf of Nicoya, 130 Peropteryx macrotis, 54 Playa Naranjo, 203 pollution and, 121 Pesticide contamination, 121–22 Playa Ostional, 203, 204 Porcupines, 49 bats and, 171 Plegadis falcinellus, 151 Portunidae, 139

INDEX 337 frankie index 8/20/03 8:14 PM Page 338

Powers, Joshua, 202 Regional Conservation Areas, 301 River otter, 50 Predaceous bats, 51 Reithrodontomys gracilis, 49 Rivers Predatory ants, 70–71 Reithrodontomys paradoxus, 49, 50 ecological importance of, 116 Premontane rain forest, 102 Reproduction fragmentation of riparian systems, Premontane wet forest, 102 of birds, 152–53 120 Press bulletin, writing, 262–63 of plants, 17–27 human impact on, 119–22 Primates reproductive diapause, 104–5 hydrology, 6–7, 119–22 biodiversity of, 49, 50 of reptiles and amphibians, modiﬁcation of natural course of, endangered, 51 183–84, 186 10–11, 119–20 in mangrove forests, 138 of trees, 38–45 in northwestern coastal strip, Pritchard, Peter, 200 of waterbirds, 153 116–17 Private landowners, conservation Reproductive diapause, 104–5 Roadside Hawk, 149 and, 316–21 Reptiles and amphibians, 118, 177–91 Roaming army ants, 70 Procyon lotor, 49, 138 adaptations of to seasonal climate, Robinson, Douglas, 203 Prosopis juliﬂora, 211 184–85 Roble sabana (wood), 270 Protection zones, 285–86 biodiversity of, 178–79, 180–82 Rodents, biodiversity of, 49, 50, 58 Protothaca asperrima, 139 conservation of, 187–90 Rodríguez, Bernal, 172 Proyecto de Riego de Arenal deforestation and, 187 Rodríguez, Esperanza, 200 Tempisque, 271 diel activity in, 179 Rolling seasonal invasion, 88 Pseudomyrmex ants, 70, 74 feeding ecology of, 182–83 Ron ron (wood), 270 Psittacidae, 157 habitat destruction and, 187 Rose-throated Becard, 155 Psylloidea, 138 hunting of, 186 Rosewood, 270 Pternohyla fodiens, 185 microhabitat of, 179, 182 Rosthramus sociabilis, 151, 156 Pterocarpus michelianus, 70 pet trade in, 186 Rothschildia lebeau, 85 Pteronotus davyi, 53 reproduction of, 183–84, 186 Ruiz, Georgita, 196 Pteronotus gymnonotus, 53 reserves for, 188–89 Pteronotus parnellii, 53 threatened and extinct species of, Pteronotus personatus, 54 185–87 Sabalos River, 6 Pteronotus spp., 50 Reserva Biológica Lomas Barbudal, Saccopterys bilineata, 54 Ptiloglossa spp., 73 50, 215, 217–18, 250–51, 273 Saccopterys leptura, 54 Puma, 51 Resource protection. See Saimiri oerstedii, 51 Puma concolor, 51 Conservation Salamanders, 179–80, 181, 183–84 Punta Ratón, 217–18 Restoration ecology, 276–77 Salpinctes obsoletus, 151 Puntarenas estuary, 129–30 Rhinoclemmys areolata, 186 Salvin’s spiny pocket mouse, 49 Pygmy fruit-eating bat, 52 Rhinophrynus dorsalis, 186 Samanea saman, 22, 58, 211, 270 Pyralid moth, 90 Rhizophora mangle, 137 Santa Rosa National Park, 103, 163, Pyralidae, 86, 90, 138 Rhizophora racemosa, 137 217–18, 266 Rhizophora spp., 139, 140, 219 Sapoa River, 6 Rhogeesa alleni, 161, 162 Sapotaceae, 22, 270 Quercus oleoides, 22, 55 Rhogeesa genowaysi, 161 Sarcoramphus papa, 156 Quiscalus mexicanus, 151 Rhogeesa minutilla, 166 Saturniidae, 86 Rhogeesa mira, 161, 162 Satyrinae, 105, 106 Rhogeesa parvula, 162 Sauria, 179–80, 181 Rabbit, 49, 58 Rhogeesa spp., 165–66 Savanna, birds in, 149 Rabies, vampire bat and, 167–68 Rhogeesa tumida, 53 Scarlet macao, 138 Raccoons, 49, 50, 138 Rhynchonycteris naso, 54 Schausiella santarosesis, 86 Rails, 153 Rhynophrynus dorsalis, 183 Scheelea rostrata, 55 Rainy season, 83, 85–87, 90 Rice Sciurus deppei, 51 Ranching. See Cattle ranching land use and, 56 Scolecophis atrocinictus, 183 Raptors, 149, 152 pesticides and, 62 Screwworm, 168 Rattlesnake, 186 Riodinidae, 103, 104, 108 Sea turtles, 68, 194–207, 312 Refugio Nacional de Fauna Silvestre Riparian system, 10 biodiversity of, 195–97 Gandoca-Manzanillo, 196 birds in, 149 conservation of, 198–204 Refugio Nacional de Vida Silvestre, fragmentation of, 120–21 environmental education about, 200 River basins, map, 2 203

338 INDEX frankie index 8/20/03 8:14 PM Page 339

extinction of, 196, 197–98 Solanum umbellatum, 164 Tamarindo (Costa Rica), 198 nesting sites of, 196–97 Solenopsis spp., 70, 71, 74 Tamarindo Bay, 198 in Parque Marino Las Baulas, Solid waste, 122, 129, 141 Tamarindo National Wildlife 198–202 Solitary bees, 72 Refuge, 7, 200 Seacockle, 141 Southern river otter, 50 Tapirs, 49, 50, 51 Seasonal marshes, birds in, 149 Spanish cedar, 34, 270 Tapirus bairdii, 49, 50 Seba’s short-tailed fruit bat, 50, 52 Spectacle Owl, 149 Tárcoles River, 122, 126, 127, 141 Seed dispersal, 43, 163 Sphingidae, 86, 89 Tayassu pecari, 50 Seed-eating weevil, 90 Spider monkey Teak, 270 Senna spp., 70 biodiversity of, 50, 51 Tectona grandis, 270 Serpents, 179–80, 181 migration of, 55 Telmapsylla minuta, 138 Servicio Nacional de Aguas Subter- Spiroeta epaphus, 107 Tempisque (wood), 270 raneas Riego y Avenamiento Spizaetus ornatus, 156 Tempisque Conservation Area. (SENARA), 271, 273 Spondias mombin, 32, 33, 214 See Area de Conservación Sesarma spp., 140 Spondias radlkoferi, 32 Tempisque Sesarma sulcatum, 138 Spoonbill, 156 Tempisque River, 6, 119, 126 Sesuvium portulacastrum, 137 Sporophila torqueola, 151 colluvial-alluvial aquifer, 6 Sewage, pollution by 11, 121, 129 Squirrel monkey, 51 modification of natural course of, Shoot borer, 214 Squirrel, 51 10–11 Shrews, 50 Stennorhina freminivilli, 183 pesticide contamination of, 61 Shrimp, 128, 141 Stenoderminae, 167 pollution of, 61, 129 Shrubs, 18, 19 Stingray, 128 sugar industry and, 10 Sibon anthracops, 173 Stork, 149 Tempisque River Basin (TRB), 1, Sibon spp., 187 Streak-backed Oriole, 149, 155 4–5 Sideroxylon capiri, 270 Streak-headed Woodcreeper, 154 climate of, 4 Silent Spring (Carson), 257–58 Stream corridor biota, 118 fragmentation of riparian system Silk anteater, 142 Strigiformes, 152 by cane fields, 120 Simarouba glauca, 219 Sturnira lilium, 53, 162 geology of, 5 Singless bee, 21 Sturnira ludovici, 164 land-use capacity of, 6 Siphosturmia spp., 89 Sturnira spp., 167 mammal biodiversity of, 50 Siproeta stelenes, 107, 108 Sublegatus arenarum arenarum, 151 pollution of, 11 Sipuncula, 128 Successional stage, 8 rice production in, 56 Skipper butterfly, 86, 88, 103, 105 Sugar industry soils of, 5 Slender harvest mouse, 49 fragmentation of riparian system watershed, 117 Sloanea terniflora, 90 by cane fields, 120 wetlands, 9–10 Sloths, 50, 51 land use and, 56, 120 Teometridae, 86 Small-bee pollination systems pollution by sugar mills, 129, 130 Tepezcuintle, 56 in cloud forest, 23–25 water use by, 10, 119 Terrestrial-maritime zone, alteration in dry forest, 21–22, 23 Surface water, 7 of, 11 Smilisca baudinii, 185 Swartzia cubensis, 90 Territorial ordering plans, 283 Smyrna blomfildia, 106 Swietenia humilis, 213–23 Tetraodontidae, 139 Snakes, 179–80, 181, 188 Swietenia macrophylla, 90, 216, 270 Tetrapedia spp., 73 conservation of, 188–89 Sylepta belialis, 86 Theodoxus luteofasciatus, 139, 140 feeding ecology of, 183 Sylvilagus floridanus, 49 Thomas’ fruit-eating bat, 52 Mayan-Nahual tradition and, 186 Symphonia globulifera, 35 Three-toed sloth, 51 microhabitat of, 179, 182 Threskiornithidae, 153, 156 reproduction of, 183 Thygater spp., 73 reserves for, 188–89 Tabebuia ochracae, 270 Thyreodon spp., 86 Social values, 240 Tabebuia rosea, 270 Thyrothorus pleurostictus, 151 Social wasps, 88 Tabebuia spp., 67, 70 Thyrothorus rufalbus, 151 Soil-litter ants, 70, 71 Tachinidae, 86, 89 Tilapia, 129 Soil salinity, in mangrove forests, 137 Tadarida brasiliensis, 162, 165, 169, Timber industry, 187, 267–71 Soils, 5 171 Tiny big-eared bat, 52 Solanaceae, 164 Tadarida spp., 163 Tityra, 154, 155 Solanum spp., 167 Tamandua mexicana, 138, 142 Toads. See Frogs and toads

INDEX 339 frankie index 8/20/03 8:14 PM Page 340

Tonatia bidens, 54 Uca spp., 139, 140 human impact on, 119–22 Tonatia brasiliensis, 53 Ucides occidentalis, 140 hydrology and, 6–7, 119–22 Tonatia silvicola, 55 Underwood’s long-tongued bat, 49 pollution of, 121–22 Toro Island, 122 Underwood’s pocket gopher, 51 rivers ﬂowing to Gulf of Nicoya, Tortuguero-Barra del Colorado University of Costa Rica, biological 117–18 Biological Corridor, 296 collections of, 232 system interaction, 118–19 Tortuguero National Park, 286, Univoltine species, 85–86 Tempisque River Basin (TRB), 117 296 Unzela japix, 89 trees in, 118 Tourism, 11, 272–73, 277, 301–2 Urban areas, bats in, 167, 170 Weasels, 49 Trachemys scripta, 186 Urban planning, 121 West Indian manatee, 51 Trachops cirrhosus, 53 Urban sewage, 11, 121, 129 Wet forest, amphibian reproduction Tramp ant, 72 Urbanization, environmental impact in, 184 Transition zone, 102 of, 10 White-collared Seedeater, 149 TRB. See Tempisque River Basin Urbanus dorantes, 105 White-faced capuchin, 49, 50, 51, 138 Tree anteater, 142 Uroderma bilobatum, 53, 55 White-faced Whistling Duck, 155 Tree cutting, near river banks, 120 Uroderma magnirostrum, 55 White Ibis, 148 Trees. See also individual tree names Urotrygon cimar, 128 White-lipped peccary, 50, 51, 55 as bat roosts, 169–70 White-nosed coati, 49 biodiversity of, 9, 18, 19 White-tailed deer, 49, 50, 56, 138 changes in species composition Valverde, Roldan, 204 Wildlife. See also individual animal of, 43–44 Vampire bat, 51, 167–68 names conservation of, 210–24, 268–69 Vampiricide, 168 overview of, 9 forest fragmentation and, 30–36 Vampyressa nymphaea, 53 poaching of, 56, 196 genetic diversity of, 214 Vampyrodes caraccioli, 55 Wildlife Conservation Law (1987), global changes in tree repro- Vampyrum spectrum, 54, 162 204, 285 duction, 38–45 Vanessa cardui, 99 Wildlife Law (1992), 140 phenology and, 17–20, 39–40 Veranillo, 83, 106 Wildlife refuges, 284–85 pollination systems of, 20–27, Vespidae, 88, 107 Barra del Colorado Wildlife 30–34 Vines, 18, 19 Refuge, 286–87 protection zones and, 285–86 Vireo, 138, 151, 157 Dr. Lucas Rodríguez Caballero in watershed, 118 Vireonidae, 157 National Wildlife Refuge, 273 Trichechus manatus, 51 Vismia spp., 167 Ostional National Wildlife Trichilidae, 157 Volcanoes, 8 Refuge, 7 Trigonia rugosa, 22 Refugio Nacional de Fauna Trigoniaceae, 22 Silvestre Gandoca-Manzanillo, Triprios spp., 184 Warblers, 152 196 Trogon elegans, 151 Wasmannia auropunctata, 72 Refugio Nacional de Vida Trogon melanocephalus, 151 Wasmannia spp., 74 Silvestre, 200 Trogons, 151, 153, 154 Wasps Tamarindo Wildlife Refuge, 7, Tropical dry forest. See Dry forests cloud-forest pollination by, 23–25 200 (Costa Rican) dry forest pollination by, 21 Wind Tropical Kingbird, 149 Water extraction, environmental cloud-forest pollination and, 23–25 Tropical wet forest, 102 impact of, 10, 119 dry forest pollination and, 21, 22 Turrialba (Costa Rica), 34 Water hyacinth, 59 Wood-nesting bees, 75 Turtles, 179–80, 181 Water Law No. 276 (1942), 120 Wood Stork, 149 adaptations of to seasonal Water opossum, 50, 51, 138 Woodcreeper, 152, 154 climate, 184 Water protection, environmental Woodpecker, 153 hunting of, 186 law for, 285 Wren, 152, 154 Mayan-Nahual tradition and, Water uses, 7 Wrinkled-faced bat, 52 186 Waterbirds, reproduction of, 153 sea turtles, 68, 194–207 Watershed, 7, 115–23 Two-toed sloth, 50, 51 birds in, 118 Xanthidae, 139 Typha dominguensis, 57, 274 conservation of, 122–23 Xenarthrans, 49 Tyrannus forﬁcatus, 151 groundwater and, 6, 117–18 Xenoglossa spp., 73

340 INDEX frankie index 8/20/03 8:14 PM Page 341

Xiphorhynchus ﬂavigaster, 151 Xylophanes turbata, 86 Zaragoza, 7 Xylocopa spp., 21, 67, 73, 75 Xylopinae, 69 Zebu, 57 Xylocopinae, 73 Xyocopa spp., 69 Zenaida asiatica, 151 Xylophanes chiron, 89 Zenaida macroura, 151 Xylophanes juanita, 86 Zoology Museum (University of Xylophanes lefeburii, 86 Yellow-crowned Night Heron, 148 Costa Rica), 232 Xylophanes pluto, 89 Yellow-naped Parrot, 138, 149 Zygophyllaceae, 270

INDEX 341 frankie colophon 8/20/03 8:14 PM Page 342

PROJECT MANAGEMENT, COPYEDITING, INDEXING, AND COMPOSITION: Princeton Editorial Associates, Inc. TEXT: Scala DISPLAY: Scala Sans and Scala Sans Caps PRINTER AND BINDER: Malloy Lithographing, Inc.