Limnology and Paleolimnology of Hypersaline Lago Enriquillo, Dominican Republic

LIMNOLOGY AND PALEOLIMNOLOGY OF HYPERSALINE LAGO ENRIQUILLO, DOMINICAN REPUBLIC

DAVID GRAY BUCK

A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2004

David Gray Buck

ACKNOWLEDGMENTS

This project would not have been possible without the support of the School of

Natural Resources and Environment and Dr. Stephen R. Humphrey. In addition, the

NSF-IGERT Working Forests in the Tropics program provided funding during the final semester of this project. My thesis work was carried out under the guidance of Dr. Mark

Brenner. His never-ending patience, critical thinking, and encouragement were essential to this project’s completion. In addition, I would like to thank Dr. Michael W. Binford and Dr. Douglas S. Jones for serving on my committee. Dr. David A. Hodell provided guidance and insight throughout the project. Dr. Jason H. Curtis provided instruction on lab techniques for the analysis of stable isotopes and sediment geochemistry and also assisted with data collection in the field. Dr. Carlos A. Alvarez Zarikian assisted with initial ostracod and foraminifer identification and Dr. Fred G. Thompson assisted with gastropod identification. Dr. Mark Pagani and Dr. Jonathan B. Martin assisted with hydrogen isotope and basic water chemistry analysis, respectively. Dr. William M. Last provided mineralogy data. Research funding was provided by a grant from the

University of Florida College of Liberal Arts and Sciences to Dr. Mark Brenner in addition to student research grants from the Geological Society of America and the Gulf

Coast Association of Geological Societies. I am grateful for having been able to use the

facilities of the Florida Institute of Paleoenvironmental Research in UF’s Land Use and

Environmental Change Institute.

iii Work in the Dominican Republic was facilitated by Dr. Andreas Schubert and the staff of the Programa Medioambiental Transfronterizo. Dr. Ramon Sanchez, Franklin

Reynoso, and Matilde Mota of the Subsecretaria de Areas Protegidas y Biodiversidad granted permission for fieldwork. Domingo Morillo and Juan Saldaña of the Instituto

Nacional de Recursos Hidrolicos (INDRHI) kindly shared data on rainfall and lake level fluctuations. I am also grateful for the help of Hermógenes Mendez, Angel Ferrera

Benitez, Lidio Méndez Benitez, Jorge Trinidad Ferreras, and Dolores Méndez Méndez and the guards of Lago Enriquillo National Park.

Finally, I am indebted to my parents for their consistent support and encouragement throughout my life, both academic and extracurricular. And ultimately, the completion of this project would not have been possible without the constant emotional, physical and mental support of my loving wife, Ellie Harrison-Buck.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... iii

LIST OF TABLES...... vii

LIST OF FIGURES ...... viii

ABSTRACT...... ix

CHAPTER

1 INTRODUCTION ...... 1

Site Description ...... 3 Previous Studies in Lago Enriquillo...... 6 Origins of the Enriquillo Basin...... 6 Holocene History of Enriquillo Valley...... 7 Limnological and Environmental Studies of Lago Enriquillo ...... 8 Recent Developments in Lago Enriquillo...... 13

2 PHYSICAL AND CHEMICAL PROPERTIES OF HYPERSALINE LAGO ENRIQUILLO, DOMINICAN REPUBLIC ...... 15

Introduction...... 15 Study Site...... 15 Methods ...... 17 Results and Discussion ...... 18 Temperature and pH...... 18 Salinity and Major Ion Concentrations...... 18 Stable Isotopes...... 20 Microbenthos...... 22 Conclusions...... 22

3 A LATE HOLOCENE RECORD OF CLIMATE VARIABILITY FROM HYPERSALINE LAGO ENRIQUILLO, DOMINICAN REPUBLIC...... 25

Introduction...... 25 Materials and Methods ...... 30 Results...... 34

v Chronology...... 34 Litho- and Biostratigraphy ...... 35 GRA Bulk Density and Magnetic Susceptibility...... 40 Stable isotopes...... 40 Discussion...... 44 Interpretation of Sediment Proxies...... 44 Changing Environments of the Lago Enriquillo Basin ...... 47 Open marine to brackish water environment (~400 BC – 170 AD) ...... 47 High ecosystem variability (170 – 1000 AD) ...... 48 Freshening of Lago Enriquillo (1000 – 1350 AD)...... 49 Lago Enriquillo variability (1350 – present)...... 50 Regional and Global Climate Comparisons ...... 50 Mechanisms for GRA Bulk Density and Magnetic Susceptibility Variability ...51 Comparison to the Yucatan Peninsula...... 56 Regional Climate Linkages – Enriquillo and the Cariaco Basin...... 57 Extra Tropical Teleconnections – GISP2 Ice Core ...... 57 Mechanisms for Late Holocene Climate Variability...... 59 Conclusions...... 62

4 CONCLUSIONS ...... 63

LIST OF REFERENCES...... 66

BIOGRAPHICAL SKETCH ...... 74

LIST OF TABLES

Table page

1.1 Holocene sea-level changes in the Enriquillo Valley...... 8

2.1 Salinity, major ion concentrations and stable isotope ratios of Lago Enriquillo waters, spring waters entering the lake, and Caribbean Sea waters...... 20

3.1 Individual cores retrieved from coring location ENR 13-VI-01 used to develop the composite core...... 31

3.2 Accelerator mass spectrometry radiocarbon dates for samples from Lago Enriquillo, Dominican Republic ...... 31

vii

LIST OF FIGURES

Figure page

1.1 Map showing the location of Hispañola within the Greater Antilles of the Caribbean Basin ...... 2

1.2 Bathymetric map of Lago Enriquillo showing the two distinct basins at 2-m depth contours...... 4

1.3 Monthly rainfall and evaporation data from pluvimetric station Puerto Escondido located within the Lago Enriquillo watershed...... 5

1.4 Elevation of Lago Enriquillo in meters below sea level from1950 to 2003 ...... 11

2.1 Map of Lago Enriquillo on the Caribbean island of Hispañola...... 16

2.2 Water column profiles from the north and south basin of Lago Enriquillo ...... 19

2.3 Changes in salinity and stable isotopes (δ18O, δD) for Lago Enriquillo waters ...... 21

2.4 Elevation of Lago Enriquillo (mBSL) from 1950 to 1963, and timing of hurricanes (H.) and tropical storms (T.S.) passing near the Enriquillo basin ...... 23

3.1 Map of Lago Enriquillo including bathymetry ...... 26

3.2 Age-depth relationship for core ENR 13-VI-01...... 35

3.3 Sediment geochemistry for core ENR 13-VI-01...... 38

3.4 Biostratigraphy in core ENR 13-VI-01 ...... 39

3.5 GEOTEK multisensor core logger data of lithologic variables for core ENR 13-VI-01 ...... 42

3.6 Stable isotope (δ18O) stratigraphy from core ENR 13-VI-01...... 43

3.7 Comparison between sediment geochemistry and lithologic variables ...... 52

3.8 Changes in lithologic variables versus bulk mineralogy...... 55

3.9 Regional comparisons of late Holocene climate variability...... 58

viii

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science

LIMNOLOGY AND PALEOLIMNOLOGY OF HYPERSALINE LAGO ENRIQUILLO, DOMINICAN REPUBLIC

David G. Buck

December 2004

Chair: Mark Brenner Major Department: Interdisciplinary Ecology

I report on the limnology and paleolimnology of hypersaline Lago Enriquillo,

Dominican Republic. The modern lake system was studied during the course of three field visits between June 2001 and March 2003. A combination of salinity, pH, temperature, stable isotopes (δ18O and δD), and major ion concentrations were measured

during these visits. The water column was well mixed in the lake's two basins during

sampling in June 2001, November 2002, and March 2003. Salinity increased from 80 to

106‰ between June 2001 and March 2003, illustrating the variability in lakewater ionic strength on short time scales. Comparison of stable isotope ratios (δD and δ18O) in

freshwater input streams and open lake waters provides evidence of intense evaporation from this closed-basin lake. In addition microfossil assemblages collected from surface

sediment samples reveal that both halophilic and halophobic taxa inhabit the system. An

understanding of the modern limnological variability in the system aids in better

ix interpreting the paleolimnological record of late Holocene climate variability from Lago

Enriquillo.

The paleolimnology of hypersaline Lago Enriquillo, Dominican Republic is inferred from physical, lithologic, biostratigraphic and geochemical data in a 2.70 m sediment core representing the last ~2300 years of deposition. Results suggest highly variable

regional climate dynamics. Shifts in lake water salinity and associated hydrologic

fluctuations reflect changes in evaporation/precipitation ratio (E/P) and are revealed

through biostratigraphy and stable isotope analysis (δ18O) of calcium carbonate shells of

foraminifera, ostracods and gastropods. Variations in the paleolimnological record result

from a dynamic Holocene history that included a period of connection with the Caribbean

Sea prior to 270 BC, followed by a brackish-estuarine period from 270 BC to 450 AD.

Pronounced drying ca. 800 – 1000 AD is inferred from changes in the biostratigraphic

record and is also suggested by positive excursions in δ18O of ostracod and gastropod

shells as well as deviations in sediment bulk density and magnetic susceptibility. The

past ~1000 years (ca. 1000 AD – present) were characterized by high E/P variability as

reflected by rapid changes in microfossil assemblages and δ18O values. High variability

in the Lago Enriquillo paleoclimate record is tied closely to regional patterns of E/P and

tropical climatic regimes. The late Holocene Lago Enriquillo record shows that droughts

on the Yucatan Peninsula (Hodell et al., 2001) and northern South America (Haug et al.,

2003) were pan-Caribbean events related to shifts in the position of the Intertropical

Convergence Zone (ITCZ) and the North Atlantic subtropical high-pressure system that

may also be correlated with solar forcing mechanisms.

x CHAPTER 1 INTRODUCTION

Inland saline water bodies are valuable for studying: the dynamic nature of salt lakes and their aqueous geochemistry (Holdren & Montaño 2002); the presence of unique, salt-tolerant biota (Williams 1998); economically important chemicals such as lithium, borax, sodium carbonates, or zeolites (Eugster & Hardie 1978); sedimentary archives of Holocene climate variability (Dix et al. 1999). According to Eugster and

Hardie (1978), for saline lakes to develop and persist, three basic conditions must be present: 1) the body of water must sit in a hydrologically closed basin; 2) evaporation must exceed the sum of all inflows; and 3) these inflows must be sufficient to sustain the body of water.

Lago Enriquillo, Dominican Republic is a particularly interesting saline water body for study (Figure 1.1). Lago Enriquillo is hypersaline, with historic salinities ranging from 35‰ to over 100‰, depending on rainfall (Buck et al., in press). It is a hydrologically closed basin with a number of groundwater-fed springs and intermittent streams entering the basin, yet experiences extreme evaporative pressure. Its conservation is highly valued for future generations and it contains sedimentary archives useful for understanding late Holocene climate variability in the Caribbean basin.

The purpose of the paper is to expand on the existing knowledge of this unique system. I used a combination of limnological and paleolimnological techniques to: 1) describe the current chemical and physical characteristics of Lago Enriquillo; and 2)

1 2

create a record of late Holocene climate variability for Lago Enriquillo and the greater

Caribbean Basin.

Figure 1.1. Map showing the location of Hispañola within the Greater Antilles of the Caribbean Basin. The Dominican Republic (B) comprises the eastern portion of the island. A closer view (C) shows Lago Enriquillo and its proximity to Laguna Rincon and neighboring Etang Saumatre in Haiti.

The introduction to Lago Enriquillo is based on several studies, originally published in Spanish, which contain valuable information on the lake’s biological and

3 chemical characteristics. These articles are summarized briefly to provide a background of information on the lake that brings together what were originally disparate data sources. Following the introduction, I present water column and sediment data collected in three field excursions between 2001 and 2003. The first section is on the physical and chemical characteristics of Lago Enriquillo with a focus on the chemistry of the water column and 17 springs and 6 streams and canals that discharge into the lake. Stable isotope analyses (δ18O and δD) and major ion concentrations are used to illustrate the extreme evaporative pressure in Enriquillo. The relationship between salinity and δ18O is reported in this section and will be expanded on later as part of the multi-proxy paleolimnological study. Chapter three presents data obtained from a 270 cm sediment core collected from Lago Enriquillo in June 2001. A multi-proxy paleolimnological approach, including sediment geochemistry, biostratigraphy, and stable isotopes, was used to reconstruct late Holocene climate variability in the Enriquillo basin and the rest of the Caribbean basin.

Site Description

Lago Enriquillo is situated in the southwestern corner of the Dominican Republic approximately 4km from the border with Haiti (Figure 1.1). The lake is part of a larger geographic feature, the Neiba/Cul-de-Sac Valley, which stretches from the Bahia de

Neiba in the Dominican Republic to Port-au-Prince, Haiti. The valley, like the majority of Hispañola, is composed of Miocene age limestones that have been metamorphosed and structurally altered during the movement of the Caribbean plate relative to the North

American plate. This tectonic activity has caused much of the uplift and coincident depression in the Neiba Valley. Lago Enriquillo sits in one of these depressions and is

bordered on its north by the Sierra de Neiba and to the south by the Sierra de Bahoruco.

It is separated from the Caribbean by a low 4m rise and separated from the neighboring

basin of Etang Sumatre by a sill that is greater than 100m above sea level in some places

(Mann et al. 1984).

Figure 1.2. Bathymetric map of Lago Enriquillo showing the two distinct basins at 2-m depth contours (modified from Araguás-Araguás et al., 1993).

Lago Enriquillo is the largest lake in the Caribbean, with a surface area of 200km2

and a mean water depth of 6.10m. Currently the lake level is 46m below sea level

(mBSL). The lake is separated into two basins with the northern basin reaching depths of

up to 22.5m and the southern basin reaching a maximum depth of 10.1m (Figure 1.2)

(Schubert 2000a). The watershed of Lago Enriquillo is estimated to be 3500km2. The basin is closed, hydrologically. Water enters the lake via precipitation and several fresh water streams and springs that drain the surrounding Sierras. Water is thought to leave the lake only via evaporation, although some outflow may occur by groundwater seepage.

Average Annual Precipitation versus Evaporation Lago Enriquillo - Pluvimetric Station Puerto Escondido

250

200

150 rainfall

(mm) evaporation 100

0 J FMAMJ J ASOND

Figure 1.3. Pluvimetric station Puerto Escondido is located at N18° 19’ 15”, W71° 34’20”, elevation = 400 mASL. Monthly rainfall values are averaged between 1968 and 1993. Average annual rainfall during this interval totaled 618 mm. Monthly evaporation values were averaged between 1970 and 1990. Average annual evaporation during this interval totaled 1887 mm. Months that did not have data were not included in the calculation. Data provided by INDRHI (Instituto Nacional de Recursos Hidrolicos).

Lago Enriquillo is a hypersaline lake with salinities ranging from 35‰ to 100‰, depending on rainfall. Annual precipitation ranges from 508mm on the southeast shore to 729mm on the northwest shore. Mean daily temperature around the lake ranges between 22.3°C and 33.7°C. Estimated annual evaporation rates can be as great as

2000mm, almost three times that of precipitation (Figure 1.3) (Schubert, 2000a). Most weather affecting Lago Enriquillo enters from the east and follows the long axis of the valley. Although Lago Enriquillo is situated within the rain shadow of the Sierra, there are two distinct rainy seasons. A short rainy season runs from March to April with a longer one from July to September. Rainfall peaks in May and in July-August. During the dry season (December to mid-March) rainfall is less than 20mm/month (Schubert

2000a).

Previous Studies in Lago Enriquillo

Origins of the Enriquillo Basin

Lago Enriquillo and its surrounding watershed have been the focus of a variety of scientific investigations. In a survey of the geography and geology of the Enriquillo valley, Cucurullo (1949) describes “la hoya de Enriquillo” as the lowest elevation on the island of Hispañola. He cites lateral compression in a NNE-SSW direction as responsible for the mountain ranges that occur in series across the island oriented WNW-ESE. The study documents a series of faults oriented along this WNW-ESE axis that border the

Sierra de Neiba and Sierra de Bahoruco. Cucurullo (1949) cites earlier geologic and geographic studies that discuss processes involved in the evolution of the valley, including: a marine transgression during the Miocene and Pliocene; presence of elevated corals and continual uplift of these features; a possible inland seaway connecting the

Neiba/Cul de Sac valley system; and the role that sedimentation and fluvial damming by the Yaque del Sur river played in the isolation of Lago Enriquillo from the Caribbean

Sea.

Recent studies on the geological origins of the Enriquillo valley have focused on the well-preserved sub-aerial fringing coral reef deposits in the Enriquillo valley (Wilson et al. 2000; Curran and Greer 1998; Greer 1997; Stemann and Johnson 1995). Wilson et al. (2000) studied paleoshore erosion associated with the coral reef and the once- submerged limestone cliffs in the valley by examining boring holes of bivalves, sponges and serpulid worms. These borings exhibit characteristics that suggest at least two extended intervals of relatively stable sea-level stands in the valley (Wilson et al. 2000).

Curran and Greer (1998) document large serpulid worm mounds associated with the

Enriquillo reef system and Greer (1997) examined facies changes in the development and

eventual cessation of reef formation in the valley using U/Th dating and three-

dimensional mapping of the coral deposits. Earlier studies on the corals of the Enriquillo

valley provided a valuable chronology for coral development using U/Th dating and

reveal an extended period of marine transgression and consequential coral reef growth

(Mann et al. 1984; Taylor et al. 1985). These dates provide a framework for the

Holocene development of Lago Enriquillo as it was separated from the sea and became a closed-basin, inland, saline lake.

Holocene History of Enriquillo Valley

Subaerial coral reef deposits are found along the perimeter of the Enriquillo valley

and on Isla de Cabritas, an island separating the north and south basins of the lake. The

presence of these corals suggests that a marine incursion occurred in the recent geologic

past lasting long enough and maintaining adequate circulation to allow for the

development of coral beds. The most common species of coral within these deposits are

Montastraea sp. and Siderastrea siderea. Stemann and Johnson (1992) state that the

diversity and ecology of the Enriquillo coral reef deposits are similar to that of modern

Caribbean reefs. A series of radiocarbon and 234U/230Th ages were obtained from these

coral beds and provide a time scale for the marine incursion and development of the coral

beds (Table 1.1).

The 5710±90 yr BP beginning of sedimentation over the reef deposits is also

believed to mark the beginning of the closing of the valley from the sea. Mann et al.

(1984) suggest that the isolation of Lago Enriquillo began with the damming of the valley

caused by increased sediment delivery from the Rio Yaque del Sur that drains the Sierra

de Neiba to the north. The possibility of tectonic uplift along the southern banks of the

lake could have further isolated Lago Enriquillo from the sea (Taylor et al. 1985).

Table 1.1. Holocene sea-level changes in the Enriquillo Valley 2820±40 yr BP to Present Lago Enriquillo is being evaporated and is currently 46 mBSL 5710±90 to 2820±40 yr BP late Holocene sedimentation over the reef with fossil bivalves recording a gradual shift from marine to brackish waters. Thrombolitic stromatolites near present sea level 6200±80 to 4760±40 yr BP mid-Holocene development of a Caribbean fringing reef in a protected environment 8990±60 to 6200±80 yr BP colonization of reef corals above oyster bed (≅34mBSL) 9760±100 to 8990±60 yr BP marine transgression and growth of oysters on an alluvial fan deposit (≅33m BSL) (adapted from Mann et al. 1984; Taylor et al. 1985)

Limnological and Environmental Studies of Lago Enriquillo

Between 1967 and 1977, the Center for Investigation of Marine Biology (CIBIMA) of the University of Santo Domingo organized a program of periodic expeditions to Lago

Enriquillo with the goal of evaluating current limnological conditions. The investigations

included studies of physico-chemical characteristics of the lake water, and flora and

fauna, both in the limnetic and littoral zones (Incháustegui et al. 1978). The earliest records of lake salinity published by CIBIMA (1967) document a lake water salinity of

40.6‰. Between 1967 and 1977 however, salinity increased from 40.6‰ to 79‰.

During the course of a six month investigation between January and June 1977, the

salinity of Lago Enriquillo varied little, between 75 and 79‰. Dissolved oxygen

concentrations in surface waters were very low (0.46-1.68 mg/L) and pH was between

8.35 and 8.45 (Incháustegui et al. 1978).

Wind plays an important role in the ecology and hydrodynamics of Lago

Enriquillo. In four of 11 Secchi disk measurements taken, the Secchi depth (zsd) was

<1.0m with all values <1.80m. Margalef (1986) reported Secchi depths of ~ 1.0m.

Strong trade winds move across the valley and contribute to sediment resuspension and

high turbidity. Examination of littoral deposits suggested that aeolian transport is an

important process for erosion and sedimentation in Enriquillo (Incháustegui et al. 1978).

Biotic aspects of Lago Enriquillo were also documented by Incháustegui and

colleagues (1978). Most biotic activity was concentrated in and around areas of lower

salinity (i.e. mouths of freshwater springs and rivers), although they cite a personal communication regarding the presence of green algae (Cladophora sp.) in the saline lake waters. The dominant green algae in the zones of lower salinity include

Hormonthamnion enteromorphoides and Enteromorpha flexuosa. A monograph published in conjunction with the CIBIMA study focused on contamination and the presence of pathogenic bacteria in Lago Enriquillo (Perez de Incháustegui et al. 1978).

The authors report the presence of pathogenic bacteria, including Salmonella anatum and

Alteromonas putrefaciens, which were sampled from hypersaline lake water as well as from waters adjacent to the Borbollones spring and La Azufrada spring. Under certain conditions these two bacteria produce H2S. This may contribute to the mild sulfur smell

detected in 2002 and 2003 at the Borbollones Sulfur spring and La Azufrada spring, the

latter of which is visited by the largest number of tourists. Other pathogens isolated in

the microbiology analysis included Escherichia coli, Psuedomonas aeruginosa, and

others (Perez de Incháustegui et al. 1978).

In a preliminary analysis of plankton tows Incháustegui et al. (1978) document a

variety of zooplankton including copepods of the family Oncaedae and the rotifer

Brachionus plicatilis. Also noted were the diatoms Pleurosygma strigosum, Amphipora

alata, Nitzchia spp., two species of Navicula, and Amphora cymbifera. Three

cyanophytes were also reported: Microcoleus, Arthrospira and Anacystis (Incháustegui et al. 1978).

The principal fish species living in Lago Enriquillo is Tilapia mossambica, an

African cichlid that was introduced to the lake around 1950 (Incháustegui et al. 1978;

Schubert 2000b). Small fishes of the family Poeciliidae (live-bearing fish) congregate in

the near-shore environment as well as at the mouths of springs and rivers that enter the

lake (Incháustegui et al. 1978). Lago Enriquillo is known for its population of the

American crocodile, Crocodylus acutus. The population is considered to be the largest in the world, although the animals experience considerable reproductive stress because of the very high lake water salinities (Schubert, 2000b).

During the early study of the Enriquillo watershed, Incháustegui et al. (1978) noted that channelization of small streams that once drained into Lago Enriquillo had routed fresh water away from the lake, to agricultural lands. They commented on the potential effect that water diversion could have. Incháustegui and colleagues (1978) cite an 1892 study that reports lake level was 0.61m above sea level. Following the CIBIMA study, lake level was recorded to be 41.95m BSL (Cruz & Duquela-Gonzalez, 2003). Margalef

(1986) expressed some doubt about the reliability of the early measurement.

Nevertheless, it is clear that the lake stage displays pronounced variability. This variability has been documented by the National Institute of Hydrologic Resourses

(INDRHI), in collaboration with other scientists from the Dominican Republic,

International Atomic Energy Agency (IAEA), and US universities. Unpublished data compiled by INDRHI and the Programa Cultura del Agua under the guidance of surveyor

Roberto A. Cruz (Cruz & Duquela Gonzalez, 2003) illustrate the dramatic fluctuations of lake level during the past 50 years (Figure 1.4).

-20 lake level -25

-30

-35

-40 Elevation (mbsl) Elevation -45

-50 1950 1960 1970 1980 1990 2000 Year

Figure 1.4. Elevation of Lago Enriquillo in meters below sea level (mBSL) from1950 to 2003. Lake level data provided by Juan Saldaña of INDRHI and was surveyed by Roberto Cruz and Tesis Duquela-Gonzalez (2003).

The Enriquillo Valley has also received attention from exploration geologists

seeking petroleum deposits. Although no extractive efforts followed these explorations,

important observations were made and recorded in a brief unpublished report presented to

the Institute of Mining, Santo Domingo, by Canadian Superior Oil, Ltd. (Canadian

Superior Oil, 1979). This report states that the surface area of the lake in 1979 had been

reduced by 35% in the previous decade. This loss in surface area correlated with a drop

in lake level of almost 4.6m. The greatest reduction of surface area occurred in the SE

basin where it was estimated the shoreline had receded by as much as 5km when

compared to older topographic maps (Canadian Superior Oil, 1979). This dramatic

response in the SE basin was, in part, a result of its bathymetry, which is characterized by

little slope and a large area of relatively shallow water. Conversely, the NW basin is

deeper, has a much steeper slope, and contains more water, so a drop in lake level does

not expose large areas of the littoral zone (see Figure 1.1).

A visit to Lago Enriquillo in 1983 by Ramon Margalef documented the dynamic

nature of the lake (Margalef, 1986). Margalef (1986) measured salinities between 35.6

and 36.9‰ in the deepest part of the north basin (zmax=23m). The water column was fairly well mixed at that time, with a change in temperature from surface to bottom of only -2.2ºC and a corresponding salinity increase of 1.3‰ (Margalef, 1986). Dissolved oxygen in the surface water at the time of Margalef’s visit was 9.3 mg/L and the water column became anoxic at approximately 16m. Margalef compared relative concentrations of the major ions of Enriquillo waters with seawater, and noted that

Enriquillo waters are related to seawater, but also exhibit characteristics of the local spring waters. For example, elevated concentrations of Mg++ relative to sea water

(2.62:1), when compared with the concentrations of Na+ and Cl- (1.65:1 and 1.78:1,

respectively) led Margalef to conclude that the waters of Lago Enriquillo are not derived

simply from the concentration of seawater. Interaction between spring water (high Ca,

low alkalinity) and the saline lake water (low Ca, high alkalinity) produces a lake water

supersaturated with Ca++, reaching concentrations of 9.0mM (Margalef, 1986). The high

Mg++ concentration is related to dolomite deposits in the watershed and the low

concentration of K+ is perhaps the result of ionic exchange (Margalef, 1986).

In his examination of the flora and fauna of the lake, Margalef (1986) noted many

of the same species that were documented by CIBIMA (Incháustegui et al. 1978). In

plankton samples from lake water, the phytoplankter Chromulina was found in high

concentrations as were several species of dinoflagellates including Cyanomonas sp. and

Oxyrrhis marina. In addition to the diatoms documented by CIBIMA, Margalef (1986) identified specimens from the genera Cyclotella and Mastogloia. He also remarked on

the presence of spherical-shaped isopods as well as abundant miliolid foraminfera, but

did not identify them further.

Lago Enriquillo was also the subject of a recent study by members of the IAEA and

INDRHI that focused on the physio-chemical parameters of the lake as well as extensive

oxygen-18, deuterium and tritium isotopic analyses of lake and spring waters (Araguás

Araguás et al. 1993). Surface water conductivity ranged from 50.0 to 83.75 mS/cm.

Values for pH ranged from 8.01 to 8.14. Major ion concentrations illustrate that

Enriquillo is a Na-Cl dominated system with concentrations of particular ions relative to those of sea water ranging from 1.22:1 for Ca++ to 2.17:1 for Na+ and Cl- and 2.88:1 for

SO4-. Isotopic values of surface waters range from +3.34 to 4.10‰ for oxygen-18 (δ18O)

to 17.3 to 21.9‰ for deuterium (δD). Anomalously low values for both δ18O and δD

from an area in the northwest corner of the northern basin are probably a consequence of

nearby fresh water inputs from Boca de Cachon and Borbollones springs (Araguás

Araguás et al. 1993). Based on these isotopic analyses, Araguás Araguás et al. (1993)

concluded that: 1) there is restricted circulation between the two basins, and 2) there is

larger input of fresh water in the southern basin making it isotopically lighter.

Recent Developments in Lago Enriquillo

Extensive on-going research and conservation efforts in the Enriquillo basin were

officially recognized and honored in 2002 by the UNESCO - Man and Biosphere

program. UNESCO declared 476,700 hectares, including Lago Enriquillo, the

surrounding sierra, and adjacent marine area, the Jaragua-Bahoruco-Enriquillo Biosphere

Reserve (UNESCO 2002). This area is home to a wide variety of marine birds and is also

a resting and feeding location for many species of migratory birds as well as a

reproductive site for the pink flamingo (Phoenicopterus ruber) (Dinerstein et al. 1995).

The lake and surrounding watershed are home to numerous plant and animal species that are endemic to the island of Hispañola and the ecoregion.

The two studies that follow contribute to the growing literature on the ecology and limnology of Lago Enriquillo. The physical and chemical characteristics of Lago

Enriquillo provide an additional data set documenting the water dynamics of the system.

The paleolimnology of Lago Enriquillo illustrates the usefulness of saline waters as archives of Holocene climate data and argues for climatic teleconnections across the

Caribbean and the North Atlantic.

CHAPTER 2 PHYSICAL AND CHEMICAL PROPERTIES OF HYPERSALINE LAGO ENRIQUILLO, DOMINICAN REPUBLIC

Introduction

Lago Enriquillo, Dominican Republic, (18°31.7' N, 71°42.91’ W) is a hypersaline

lake of marine origin. During the middle Holocene, ca. 6000-4800 years before present

(yr BP), Lago Enriquillo was an embayment of the Caribbean Sea and supported a fringing coral reef. The embayment was isolated from the Caribbean Sea between 5000

and 2800 yr BP by tectonic uplift and fluvial damming by the Yaque del Sur River

(Bond, 1935; Mann et al., 1984; Taylor et al., 1985). Today, Lago Enriquillo is a closed

basin lake that is home to a unique flora and fauna, and whose sediment may serve as an

excellent source of paleoenvironmental information. Here I present data on modern

physico-chemical parameters and microfauna distributions that reflect the system’s

sensitivity to variations in rainfall, evaporation, and fresh water inputs. These data will

aid in interpreting past environmental changes as recorded by faunal and geochemical

proxies in sediment cores retrieved from Lago Enriquillo.

Study Site

Lago Enriquillo is the largest lake in the Caribbean. Currently its watershed

encompasses ~3500 km2 and lake level is ~46 m below sea level. Its surface area is ~200

km2, and its estimated mean water depth is 6.10 m (Figure 2.1) (Schubert, 2000a). The

lake has displayed historic salinities ranging from about 35‰ in 1983 (Margalef, 1986) to

>100‰. Annual rainfall near Lago Enriquillo ranges from 508 mm on the SE shore to

15 16

729 mm on the NW shore (Instituto Nacional de Recursos Hidrólicos, unpublished data).

Rainfall maxima occur in late-March and August. Monthly precipitation in the dry season (December to early March) averages <20 mm. Annual rainfall is highly variable, however, and is influenced by tropical depression and hurricane activity in the Caribbean.

Mean daily temperatures around the lake vary between 22.3°C and 33.7°C, and annual

evaporation can be as great as 2000 mm (Schubert, 2000a).

Figure 2.1. Map of Lago Enriquillo on the Caribbean island of Hispañola. Bathymetric map of Lago Enriquillo showing 2-m depth contours (modified from Araguas- Araguas et al., 1993). Letters along the shoreline indicate the approximate location of springs mentioned in the text: (A) La Azufrada, (B) Borbollones, (C) Boca de Cachon, (D) Caoba, (E) La Zurza, (F) El Guayabal.

Methods

I report physical (temperature), chemical (pH, salinity, ion concentrations, stable

oxygen and hydrogen isotopes) and surface sediment data, based on samples collected

from Lago Enriquillo during three visits in June 2001, November 2002, and March 2003.

Water samples for chemical analysis were taken from freshwater springs that discharge into the lake, from the lake surface, and from water-column profiles in the lake’s north and south basins. Water column temperature and pH were measured with a Horiba U-10 meter. Salinity was measured using a SPER Scientific refractometer. Major cations and

anions were measured with a Dionex Model DX 500 ion chromatograph. Chloride concentrations in hypersaline lake waters were determined by titration using 0.1M

AgNO3 and K2CrO4 indicator solution (Mohr method). Total dissolved inorganic carbon

(DIC) was measured with a UIC/Coulometrics Model 5011 Coulometer and 5030

Carbonate Carbon Preparation System.

18 Oxygen isotope ratios (δ O) of 128 water samples were measured by CO2 equilibration using a Micromass Multiprep device interfaced to a VG Prism II mass spectrometer. A subset of 32 samples was measured for deuterium (δD) on a

ThermoFinnigan MAT253 using an H-Device reduction furnace and a CTC PAL autosampler. Data are reported in standard delta notation (δ) relative to Standard Mean

Ocean Water (SMOW).

In addition to collecting limnological data, I also took surface sediment samples at the mouths of fresh water springs as well as from open lake water sites to identify the benthic microfauna of Lago Enriquillo. Surface sediments were processed using standard procedures (Murray, 1991). Samples were washed through a 63 µm sieve, oven-dried

and then dry sieved to separate the sample into >212 µm and >63 µm fractions for identification of microbenthos. Organisms identified included foraminifera, ostracods, and gastropods.

Results and Discussion

Temperature and pH

Surface water temperature in the north basin ranged from 28.6°C to 30.5°C, with the lowest temperature recorded in March 2003 (Figure 2.2a). A subtle temperature difference of < 1°C between the upper and lower waters was observed. The south basin was isothermal at the time of sampling in 2002 and 2003, with temperatures ranging from

27.9°C (March 2003) to 29.7°C (November 2002). Among the three sampling dates,

November was the time of greatest heat storage in the water column. The pH of Lago

Enriquillo waters varied little with time, sampling location, or depth (Figure 2.2b).

Water column temperature and pH results suggest that each basin of Lago

Enriquillo is, separately, well mixed. Strong winds and a large fetch, particularly in the

south basin, probably contribute to polymixis. Although the north basin exhibits subtle

thermal stratification, bottom waters are only ∼1°C cooler than surface waters. Seasonal

differences between the two basins are comparable.

Salinity and Major Ion Concentrations

Lakewater salinity in both basins increased by ~23‰ between June 2001 and

March 2003 (Table 2.1). Salinity in the south basin was consistently ~2‰ higher than in

the north basin. Changes in salinity were accompanied by proportional shifts in

individual ion concentrations. Na+, K+, Mg++, and Ca++ consistently comprised 80%,

1.5%, 16.5%, and 2% of cation milli-equivalents L-1, respectively. Chloride and sulfate

- had relative concentrations of 87.5% and 12.25%, respectively. Bicarbonate (HCO3 ), measured on a subset of water samples, represented only ~0.03% of anion charges.

0 0 A B 2 2 4 4 6 6 8 8 10 10

Depth (m) 12 12 Depth (m) 14 14 June '01 N basin 16 Nov '02 N basin Mar '03 N basin 16 Nov '02 S basin 18 Mar '03 S basin 18 20 20 26 28 30 32 7.8 8 8.2 8.4 Temperature (ºC) pH

Figure 2.2. Water column profiles from the north and south basin of Lago Enriquillo. Temperature (A) and pH (B) profiles were measured in June 2001, November 2002, and March 2003.

Spring waters had salinities < 1‰, but displayed differences in relative ion concentrations. Guayabal spring contained the lowest total ionic charge with Ca++ representing 83% of the total cation milliequivalents (Table 2.1). Boca de Cachon had the highest total ionic charge dominated by Na+ and Cl- ions. Borbollones and Zurza are sulfur springs and emit H2S.

Table 2.1. Salinity, major ion concentrations and stable isotope ratios of Lago Enriquillo waters, spring waters entering the lake, and Caribbean Sea waters. * The concentration factor compares the ionic concentrations of Lago Enriquillo waters to local Caribbean Sea waters (Bahia de Neiba). Because of Enriquillo’s marine origin, this comparison reveals information about the ionic development of the lakewaters since its isolation from the sea. Isotopes Cations Anions 18 salinity δ O δD Na K Mg Ca Cl SO4 HCO3 sample ID ppt (meq/L) (meq/L) (meq/L) (meq/L) (meq/L) (meq/L) (meq/L) Enriquillo lake waters June, 2001 North basin 80.0 4.4 20.9 1077.7 19.3 220.8 26.5 1237.2 175.8 n/d South basin 82.0 4.5 n/d 1086.8 19.7 222.5 26.7 1220.3 166.7 n/d Nov, 2002 North basin 102.0 4.5 22.7 1379.9 26.5 283.0 34.6 1515.4 211.6 5.0 South basin 104.2 4.7 22.6 1306.2 25.1 277.2 35.5 1584.3 222.0 4.4 March, 2003 North basin 104.0 4.8 22.9 1482.6 27.3 302.4 35.4 1632.2 216.0 4.9 South basin 105.9 4.7 23.3 1433.6 26.2 291.4 34.2 1588.7 220.6 4.3 Enriquillo spring waters Borbollones 0.0 -4.1 n/d 1.0 0.1 1.3 3.4 1.0 0.1 4.7 La Zurza-sulfur 1.0 -5.7 n/d 3.5 0.1 2.4 2.6 5.7 0.3 4.4 Guayabal 0.0 -4.1 -17.1 0.2 0.0 0.5 3.5 1.8 0.1 n/d Caoba 0.0 -5.5 -30.5 2.4 0.1 2.1 2.7 3.4 0.2 4.2 Boca de Cachon 1.0 -3.6 -14.7 5.6 0.2 2.5 2.6 9.1 0.3 n/d Caribbean Sea Bahia de Neiba 37.5 1.2 6.9 479.0 9.9 112.3 20.5 550.1 58.9 n/d

Conc. Factor* 2.8 3.9 3.3 3.1 2.8 2.7 1.7 3.0 3.7

Stable Isotopes

2 18 In evaporating water bodies, heavier isotopes of water (H HO and H2 O) are

“enriched” in lake water because of the preferential evaporation of lighter water

1 16 18 2 molecules ( H2 O). Enrichment of O and H (deuterium, or D) provides information on changes in the ratio of precipitation to evaporation in a basin. Therefore, the ratios of

18O to 16O (δ18O) and 2H to H (δD) in waters provide information about the evaporative

history of a water body and can also be used to identify different input sources in water

balance studies.

120 60 A B

100 Salinity = 11.2( δ18O) + 46.865 2 40 r = 0.95 GMWL (Craig, 1961) δD = 8(δ18O) + 10 80

20 60

springs δ D 40 lake waters mixing zone 0 Salinit rain Bahia de Neiba 20 Evaporation Line δD = 4.7(δ 18O) + 0.41 -20

-20 -40 -8 -6 -4 -2 0 2 4 6 -8 -6 -4 -2 0 2 4 6

18 δ O δ18O

Figure 2.3. Changes in salinity and stable isotopes (δ18O, δD) for Lago Enriquillo waters. (A) Oxygen isotopes (δ18O) versus salinity for Lago Enriquillo input waters (rainfall and springs), lake water, and Caribbean sea water. The regression between salinity and δ18O is highly significant (r2 = 0.95). Enrichment of 18O and salinity in lake water reflects intense evaporation from the basin. (B) Solid line is the Global Meteoric Water Line (GMWL), which defines the relation between δ18O and δD in meteoric waters worldwide (δD = 8(δ18O) + 10 (Craig, 1961)). Lago Enriquillo waters (dashed line) deviate from the GMWL because intense evaporation causes non-equilibrium isotopic fractionation.

In Lago Enriquillo waters, salinity and δ18O are strongly correlated (r2 = 0.95) (Fig.

2.3a). Covariance of salinity and δ18O illustrates the enrichment of 18O in Enriquillo

waters due to evaporation. The relationship between δD and δ18O in meteoric waters is

referred to as the Global Meteoric Water Line (GMWL) and is expressed by the equation

δD = 8(δ18O) + 10 (Craig, 1961) (Fig. 2.3b). Lago Enriquillo waters deviate from the

GMWL because of evaporation and specifically, non-equilibrium isotopic fractionation at

the air-water boundary between the atmosphere and Lago Enriquillo waters (Gat, 1980).

Microbenthos

The foraminfera assemblage of Lago Enriquillo is dominated by milliolid species

of the genus Quinqueloculina. The rotallid Ammonia beccarii is also present. The

foraminifera assemblage near freshwater springs contains more taxa, including the

rotallid Criboelphidium poeyanum, the common saline species Ammotium salsum, as well

as a less abundant Rosalina sp. An unknown species, possibly of the genus

Amphistegina, was also encountered. Its presence may be attributed to erosion of

Holocene coral deposits within Enriquillo’s watershed. Ostracods in Lago Enriquillo

include Perissocytheridea rugata, P. bicelliforma, and Cyprideis sp. The gastropod

Heleobops clytus (Thompson & Hershler, 1991) was observed adjacent to springs.

Conclusions

Lago Enriquillo exhibits only minor spatial and temporal differences with respect

to temperature and pH. The measured chemical differences (i.e. pH, major ions, salinity,

stable isotopes) suggest that the two basins are not mixed completely and may have been

hydrologically separated at lower lake stage. The range of salinities in Lago Enriquillo

enables both halophilic and halophobic taxa to inhabit the system. Freshwater hydrobiid

snails (Heleobops clytus) occupy spring discharge areas, whereas salt-tolerant foraminifera (eg., Quinqueloculina spp.) are found in the open lake.

The modern Lago Enriquillo is a dynamic system that is susceptible to short-term

hydrologic variations. Both basins are prone to pronounced shifts in salinity, chemical

composition, and isotopic concentration on short time scales (i.e., seasonal/annual).

Furthermore, long-term changes in Lago Enriquillo’s salinity appear to be related to

climate-driven shifts in freshwater input that are expressed in lake level changes. During

the past 55 years, lake stage has been positively correlated with the passage of hurricanes

and tropical storms in the area (Figure 2.4).

-20 o z

c lake level -25 e a y y 1964 -1964 Cle H.

-30 -1966 Ine H.

-35 -1961 Frances H. 1975 -1975 Elois H. 1958 - T.S. Gerd 1993 - T.S. Cind -40 -1987 Emil H. 1997 - Georges H. Elevation (mbsl) 1979 - David; Frederi H. H. -45

-50 1950 1960 1970 1980 1990 2000 Year

Figure 2.4. Elevation of Lago Enriquillo (mBSL) from 1950 to 1963, and timing of hurricanes (H.) and tropical storms (T.S.) passing near the Enriquillo basin. Lake level data were provided by Roberto Cruz, Tesis Duquela-Gonzalez, and Juan Saldaña of INDRHI. Hurricane and tropical storm data are from the National Oceanic and Atmospheric Administration (NOAA) National Hurricane Center (http://www.nhc.noaa.gov/)

Future research will focus on reconstructing decade to century-scale variability in

lake hydrology during the Holocene through analysis of sedimentary archives retrieved

from Lago Enriqullo. Changes in the abundance of the gastropod H. clytus will serve as a

24 valuable proxy for changes in freshwater input into the system. Past variations in foraminferal assemblages will aid in interpreting past changes in lake water salinity. In addition, oxygen isotope analysis of the carbonate shells of aquatic invertebrates (e.g. foraminifera, ostracoda, gastropoda) collected from sediment cores will aid in the reconstruction of past moisture conditions (Covich and Stuiver, 1974). Understanding the ecology and distribution of microfauna and the modern physico-chemical properties in

Lago Enriquillo will aid in interpreting past changes in the hydrologic budget as observed in sediment cores.

CHAPTER 3 A LATE HOLOCENE RECORD OF CLIMATE VARIABILITY FROM HYPERSALINE LAGO ENRIQUILLO, DOMINICAN REPUBLIC

Introduction

Saline lakes present a unique opportunity to study Holocene paleoclimate because they are sensitive to variations in regional precipitation (Last & Vance, 1997; Eugster &

Hardie, 1978). Lago Enriquillo is a large, closed-basin, hypersaline lake of marine origin located in the Dominican Republic, on the island of Hispañola (Figure 3.1).

Limnological investigations of the lake over the last ~20 years have illustrated this sensitivity with regard to changes in lake water chemistry and lake levels. Published salinity values and associated lake water ion concentrations in Lago Enriquillo range from close to sea water [i.e. ~35‰] (Margalef, 1986) to as high as ~3.0 times that of sea water (Buck et al., in press). In addition, stable isotope studies of meteoric waters, ground waters, and lake waters illustrate the high degree of evaporative loss from the system (Araguás Araguás et al., 1993; Buck et al., in press). A 50-year record of lake level changes documented Lago Enriquillo’s stage variability on decadal time scales

(Cruz & Duquela-Gonzalez, 2003). Lake stage is positively correlated with tropical storm frequency in the basin (Buck et al., in press).

Documented response of the lake to recent climate variability suggested that a sediment record from Lago Enriquillo might preserve an archive of longer-term,

Holocene climate change in the region. Climate in the Caribbean Basin is dominated by the seasonal migration of the Intertropical Convergence Zone (ITCZ) and the North

25 26

Atlantic subtropical high-pressure zone (also referred to as the Azores-Bermuda high)

(Hastenrath, 1984; Malmgren et al., 1998; Giannini et al., 2000; 2001). Variations in the

latitudinal location of these two air masses create the wet/dry seasonal pattern. During

the period when the ITCZ and the North Atlantic high migrate northward (from May to

November), heavy rains occur across the northern hemisphere tropical belt, carried by the

prevailing trade winds.

Figure 3.1. Map of Lago Enriquillo including bathymetry. Lago Enriquillo has a surface area of ~200 km2 and is currently 46m BSL. Mean water depth in the basin is 6.1m with the north basin reaching a maximum depth of 22.5 and the southern basin reaching 10.1m. It is the largest lake in the Caribbean and has a watershed of ~3500 km2. Lake waters are hypersaline, ranging from 35 to >100‰, depending on rainfall variability.

From November to April, the ITCZ shifts to a southerly location and the North Atlantic

high plays a more dominant role in weather patterns in the Caribbean basin bringing drier

conditions to the region (Gray, 1993).

Deglacial and Holocene variations in this regional climate dynamic have been

examined employing paleoclimate archives from marine and lacustrine sediments.

Marine records from the Cariaco Basin, near Venezuela, suggest that the glacial- interglacial transition was characterized by a period of intensified trade wind activity and

subsequent increased upwelling along the southern margin of the basin as a result of meltwater pulses entering the Gulf of Mexico during the retreat of the Laurentide Ice

Sheet (Peterson et al., 1991). The Younger Dryas Chronozone (~12.5 – 11.5 kyr BP) brought a brief return to cooler and drier conditions. Regional climate grew warmer and wetter following this period, probably due to a northward migration of the ITCZ

(Peterson et al., 1991). Warmer, wetter conditions persisted from ~10.5 to 5.4 kyr BP, a

period referred to as the Holocene “thermal maximum” (Curtis and Hodell, 1993; Haug et

al., 2001). This moist period corresponds to a period when the seasonal difference in

solar insolation in the Caribbean region was at a maximum (~8 kyr BP). This probably

caused an enhancement of seasonal weather patterns in the northern hemisphere

Neotropics (Hodell et al., 2001). The Holocene “thermal maximum” was followed by a

period of prolonged drying beginning about 3.5 kyr BP that is documented in lacustrine and marine sequences. Mid- to late Holocene drying is thought to have been a response to a reduction in the seasonal difference in insolation, and a general southward shift in the location of the ITCZ (Haug et al., 2001).

A ~10,500 yr record from Lake Miragoane, Haiti, registers a shift toward aridity

beginning ~3200 yr BP. Drying is inferred from a +1.25‰ shift in the oxygen isotope

composition of ostracod carapaces (Curtis and Hodell, 1993). Palynological data from

Lake Miragoane also suggest greater aridity, with the expansion of xeric forest taxa

Celtis, Bursera, Curatella and Cordia (Higuera-Gundy et al., 1999). The geochemical stratigraphy of a sediment core from Lake Valencia, Venezuela, suggests gradually increasing lake water salinity beginning approximately 3000 yr BP (Bradbury et al.,

1981).

High-resolution paleoclimate records from lakes on the Yucatan peninsula suggest a highly variable late-Holocene climate punctuated by periods of pronounced drought conditions (Fritz et al., 2001). A period of prolonged drought between ~800 and 1000

AD is inferred from oxygen isotopic ratios (δ18O) in both ostracods and gastropods and

sediment geochemistry (Hodell et al., 1995; Curtis et al., 1996; Hodell et al., 2001;

Hodell et al., submitted). Droughts occurred in this region on a frequency of ~208 and 50 year intervals and may be related to variations in solar activity (Hodell et al., 2001). The

Terminal Classic Drought in the Maya area occurred between 770 and 1100 AD and was

relatively strong. Its widespread nature is evident by its detection in a high-resolution record of titanium (Ti) content in marine sediments from the Cariaco Basin, north of

Venezuela (Haug et al., 2003).

Records of late Holocene climate dynamics in the northern Caribbean Basin are less consistent, however, and suggest a combination of locally controlled climate variability and lake level dynamics. In southwest Jamaica, changes in ostracod assemblages record hydrologic variability in Wallywash Great Pond during the last ~10

kyr (Holmes, 1998). An increase in faunal diversity around 1.2 kyr BP is interpreted as a

period of increased hydrologic input into the system that is followed by a return to

assemblages similar to those of the modern system (Holmes, 1998). Although increases

in δ18O of ostracod valves after ~1000 yr BP reflect a transition to slightly drier

conditions, the inferred change in climate is not represented by changes in the ostracod

community assemblage (Holmes, 1998).

Nyberg et al. (2001) used sediment magnetic susceptibility, changes in foraminifer

assemblages, and sediment geochemistry in a 2000-year marine sediment record taken

near Puerto Rico, to explain climate variability in the northeast Caribbean Sea. Time

series analysis of magnetic susceptibility data exhibits similar periodicities to those measured from δ18O and % sulfur (i.e. precipitated gypsum) measured in lake sediment

cores from the Yucatan peninsula (Curtis et al., 1996; Hodell et al., 2001). A positive

excursion in the marine core magnetic susceptibility centered at ~1000 AD, however, is

thought to represent increased transport of terrigenous, ferromagnetic material and a

transition to more humid conditions (Nyberg et al., 2001).

Marine saline ponds on Lee Stocking Island, Bahamas, contain a 1500-year record

that includes both marine and lacustrine sediment (Dix et al., 1999). A shift in

foraminifer assemblages at ~1000 yr BP is interpreted to represent a phase of rapid

marine transgression that may have persisted until ~720 yr BP at which time the

embayment closed. In addition, inputs of marine-derived sediment are preserved in these

ponds and suggest a 300-400 year cycle sea level change related to either hurricane

activity or high-order eustatic variation (Dix et al., 1999).

Paleolimnological data from Lake Miragoane, Haiti, (Curtis and Hodell, 1993)

represent the most complete Holocene paleoclimate record for the northern Caribbean

basin. Its high resolution oxygen isotope profile of ostracod valves illustrates changes in

the ratio of evaporation/precipitation (E/P) in the basin and serves as a proxy for changes

in local/regional climate. Chronological control for this record, however, was compromised by a scarcity of terrestrial organic material for AMS 14C dating.

Radiocarbon dates on ostracod shells were adjusted for possible hard-water lake error

(Deevey and Stuiver, 1964). Additional climate records from this region are needed to

better understand the changing dynamics of circum-Caribbean climate dynamics in the

late Holocene. Lago Enriquillo, Dominican Republic, offered an opportunity to explore

late Holocene Caribbean climate using a multi-proxy paleolimnological approach. Here I

present the sediment geochemistry and lithology, biostratigraphy, and stable isotope

records from a sediment core that illustrates the dynamic nature of Caribbean climate

change during the past ~2300 years.

Materials and Methods

On 13 June 2001, a 270-cm sediment core (ENR 13-VI-01) was retrieved from the

north basin of Lago Enriquillo in approximately 10.5 m of water (Table 3.1). The 71 cm-

long mud-water interface core (MWI-1a) was retrieved using a piston corer (Fisher et al.,

1992) and was sampled at 1.0 cm intervals in the field and stored in whirlpak bags.

Deeper sediments were retrieved using a modified square-rod piston corer (Wright et al.,

1984). Retrieved sediment was transported to the University of Florida and stored at

4°C.

Table 3.1. Individual cores retrieved from coring location ENR 13-VI-01 used to develop the composite core. The composite core stratigraphy was developed using magnetic susceptibilty, bulk density, and %CaCO3 data from each individual core to identify stratigraphic horizons where cores could be spliced together. Nominal Composite Core I.D. Depth Depth ENR MWI-1a 0 - 74 cm 0 - 71 cm ENR MWI-1b 71 - 193 cm 72 - 194 cm ENR Lex Overlap 130 - 228 cm 172 - 270 cm ENR Lex 2 180 - 257 cm 194 - 270 cm

Table 3.2. Accelerator mass spectrometry radiocarbon dates for samples from Lago Enriquillo, Dominican Republic

Composite 14C Calibrated Calendar Sample Name Depth age ± Age Age ± probability and depth (cm) (yrBP) (AD/BC) %

ENR13-VI-01 MWI1A 24-25# 24cm 740 100 690 1260 AD 150 88.5% ENR13-VI-01 MWI1A 41-42# 41cm 760 100 735 1215 AD 185 95.4% ENR13-VI-01 MWI1B 86-87* 86cm 400 60 420 1530 AD 110 95.4% ENR13-VI-01 MWI1B 102-103* 102cm 900 60 810 1140 AD 120 95.4% ENR13-VI-01 MWI1B 146* 146cm 1070 45 990 960 AD 70 95.4% ENR13-VI-01Lex2 201* 214cm 1750 100 1715 235 AD 305 95.4% ENR13-VI-01 Lex2 248* 261cm 2195 35 2220 270 BC 110 95.4%

# Denotes sample of possible aquatic origin * Denotes samples used in age model

Radiocarbon dates for ENR 13-VI-01 were determined by accelerator mass spectrometry (AMS) at Lawrence Livermore National Laboratories. Samples consisted of terrestrial organic material (wood, charcoal), a seed, and one sample of unknown origin (Table 3.2). Radiocarbon ages were calibrated and converted to calendar ages using OxCal Program version 3.9 (Bronk Ramsey, 2001; Bronk Ramsey, 1995) with a

100-year moving average of the tree-ring calibration data set (Stuiver et al., 1998).

Intact sediment core sections were measured for bulk density and magnetic

susceptibility using a GEOTEK Multisensor Core Logger (MSCL). Bulk density was

measured by gamma-ray attenuation (GRA) of the sediment using a sealed 137Cs source.

Magnetic susceptibility was measured using a Bartington loop sensor and is reported in the cgs (centimeter, gram, second) scale. Both measurements were taken simultaneously at 0.5 cm intervals along the core length allowing for stratigraphic correlation between the two data sets.

Elemental analyses of sediments included total carbon (TC), total nitrogen (TN)

and total inorganic carbon (TIC). Samples for elemental analysis were collected at 1 cm

intervals and oven dried for 24-36 hours at 40°C. Samples were then ground with a

mortar and pestle. TC and TN were measured using a Carlo Erba NA 1500 CNS elemental analyzer with autosampler. TIC was measured by coulometric titration

(Engleman et al., 1985) with a UIC/Coulometrics Model 5011 coulometer coupled with a

UIC CM5240-TIC preparation device. Analytical precision of TIC is approximately

±0.5% based on repeated analysis of a calcium carbonate (CaCO3) internal standard.

Organic carbon (OC) was estimated by the difference between TC and TIC.

Oxygen isotope ratios (δ18O) were measured on shells of the foraminiferan

Ammonia beccarii, valves of the ostracod Perissocytheridea rugata and shells of the

gastropod Heleobops clytus. Sediment samples were first disaggregated in a weak (~2%)

H2O2 solution and then washed through a 63µm sieve. After drying, specimens were

picked from the 63µm (foraminifera), 150µm (ostracods) and 212µm (gastropods) fraction under a binocular microscope. Specimens collected for isotopic analysis were then soaked in 15% H2O2 to remove any adhering organic material, cleaned and sonicated

33 in deionized water, and rinsed with methanol before drying. For gastropod samples, shells were broken following the first cleaning and the procedure was repeated a second time to ensure clean shell material. Sample weights for isotopic analysis ranged between

~250 and 600µg. Approximately 35 foraminifera tests were required for each analysis.

Multiple ostracod valves from a sample were ground into a fine powder using a methanol-cleaned stainless steel rod and a fraction of this powder was used for each measurement. Gastropod shells were treated similarly. Measurement of multiple individuals in a single sediment horizon reduces variance inherent in measuring single, short-lived organisms (Heaton et al., 1995).

Isotope samples were reacted in 100% orthophosphoric acid at 70°C using a

Finnigan MAT Kiel III automated preparation device. The evolved CO2 gas was cryogenically purified before being measured online with a Finnigan MAT 252 mass spectrometer. Isotope ratios were compared to an internal gas standard and expressed in conventional delta (δ) notation as a per mil (‰) deviation from Vienna PeeDee

Belemnite. Precision for δ18O samples was ±0.09‰.

Biostratigraphic analysis included foraminfera/ostracod assemblages and gastropod assemblages in core ENR 13-VI-01. Microfossil samples were examined at every 1-cm interval in conjunction with isotope sampling. Changes in foraminfera and ostracod assemblages were noted on a presence/absence basis. Gastropod abundance was determined by counting individuals from the >212µm sediment fraction. Only complete gastropod shells were included in the abundance counts.

Mineralogy from core ENR 13-VI-01 was conducted on a subsample of material from the lower composite cores (Lex OL and Lex 2, see Table 3.1). Samples were

shipped to the University of Manitoba, Department of Geological Sciences

Crystallography Lab. Samples analyzed for mineralogy were air-dried at room temperature, disaggregated in a mortar and pestle, and passed through a 62.5 mm sieve.

Bulk mineralogy and detailed carbonate and evaporite mineralogy were determined using

a Philips PW-1710 powder diffractometer. An automated computer program designed to

identify and match minerals was used to aid in mineral identification. Percentages of the

various minerals were estimated from the bulk mineral diffractograms using the intensity

of the strongest peak for each mineral as outlined by Last (2001).

Results

Chronology

Seven AMS 14C dates were obtained from core ENR 13-VI-01 (Table 3.2). The

two uppermost samples from 24 cm and 41 cm depth yielded the same age of 740±100

and 760±100 14C years BP, respectively. These dates seem old for such shallow samples.

They may represent material that remained in the watershed for some time before being

transported to the lake. It is also possible that these two samples were of aquatic origin

and subject to hard water lake error (HWLE). HWLE is associated with basins that receive 14C-deficient dissolved inorganic carbon from the dissolution of ancient limestone

(Deevey & Stuiver, 1964). Based on dates from paired ostracod and wood samples,

HWLE in Lake Miragoane, Haiti was estimated to be ~1000 years (Curtis & Hodell,

1993). In Lake Punta Laguna, Mexico, HWLE was estimated to be about 1250 years

(Curtis et al., 1996).

The age-depth relation in core ENR 13-VI-01 was estimated assuming a linear

sedimentation rate between five of the seven dated horizons (Figure 3.3, Table 3.2). The basal age of the core is estimated to be 356 calendar years (cal yr) BC based on

extrapolation of the linear sedimentation rate between the two oldest dated horizons.

Dates were assigned to sediments above 86 cm depth (1530 AD) assuming linear

sedimentation between 86 cm and the surface (2001 AD). Mean sampling resolution was

on the order of ~11 years per 1 cm sample.

2000

1500

1000

500 Calendar age (AD/BC)

-500 0 50 100 150 200 250 300

depth

Figure 3.2 Age-depth relationship for core ENR 13-VI-01. Ages for each sample were determined by linear interpolation between dated horizons (Table 3.2) including a value of 2000 calendar year AD at the surface. Ages for sediments below the 261cm sample (270 cal yr BC) were determined by extrapolation between the two basal dates.

Litho- and Biostratigraphy

Inorganic carbon is assumed to be associated primarily with CaCO3 in Enriquillo

sediments and CaCO3 content was estimated as 8.33 x TIC. CaCO3 concentrations in

Lago Enriquillo sediments are highest (>70%) below ~250 BC. Concentrations decrease

upward in two steps, until reaching a mean value of ~50% CaCO3 between 200 – 1500

AD. Concentrations decline again from 1500 - 2000 AD to a minimum value of ~30%

CaCO3 (Figure 3.3a). Percent organic carbon: percent nitrogen (OC/N) ratios have a mean value of 6.28 in the lower part of the core (400 – 145 BC). The mean value

increases to 9.81 between 120 BC and 1100 AD. A brief period of decreased values

(mean = 6.88) occurs between 1100 and 1400 AD. OC/N values then experience two

brief periods of increased values between 1600 and 1750 AD and again around 1800 AD

before decreasing towards to surface (Figure 3.3b). Organic carbon (OC) concentrations

in sediments are lowest in the bottom 25 cm (~400 – 80 cal yr BC). Concentrations then

become highly variable between 80 BC and 1100 AD with a mean concentration of

1.68%. Concentrations increase towards the surface with a maximum concentration of

4.31% (Figure 3.3c).

Stratigraphic variations in the content of CaCO3 and OC are also evident through a

visual analysis of the sediment core (Figure 3.3d). The lower ~15 cm (~360 – 180 BC) of core ENR 13-VI-01 are composed almost entirely of CaCO3-rich sands. A large piece of wood was located within this interval and provided the basal date for core ENR 13-VI-

01. Above this sand layer, CaCO3 concentrations decline coincident with an increase in

OC. A series of spikes in the CaCO3 concentration of the sediments between 238 and 222

cm (~0 – 200 AD) are coincident with a visibly high concentration of gastropod and

bivalve shells. Between 215 and 140 cm (~275 – 985 AD), alternating horizons of

carbonate mud and more organic-rich layers become more regular with at least seven

prominent layers occurring within this ~700 year period. The dominant sedimentary

facies between 140 and 20 cm (985 – 1890 AD), is a CaCO3-rich mud that is interspersed

with horizons of darker, OC-rich layers that represent a ~0.5% increase in OC

37 concentrations. The upper 14 cm (~1920 AD to present) of the core consist of alternating

OC-rich and CaCO3-rich sediments that are represented as darker and lighter bands of sediment, respectively (Figure 3.3d).

The predominant species of ostracod, P. rugata, experiences dramatic fluctuations in concentration throughout the sediment core (Figure 3.4a). It is present in only two distinct sediment intervals in the lower half of the core, occurring between 176 and 161 cm (640 – 800 AD) and 120 and197 cm (170 – 416 AD). Between 136 and 58 cm (~960

– 1684 AD), P. rugata is consistently present. It occurs only sporadically between 15 cm and 58 cm, and is absent from the upper 15 cm of the sediment record.

Foraminifer assemblages are dominated by miliolid species (e.g. Quinqueloculina spp., Triloculina spp., Milliamina fusca) and the rotallid, Ammonia beccarii. A. beccarii has a well-defined salinity tolerance of <67‰ (Bradshaw, 1957), and its presence and/or absence reveals data on salinity variability in the lake water (Figure 3.4b). The bottom- most interval in the sediment core (between 355 – 110 BC) contains two short intervals where A. beccarii is present (345 – 320 BC and 215 – 200 BC). Also present in this bottom interval is the planktonic foraminfer, Globigerina bulloides. A. beccarii is again present in the microfossil record between 234 and 213 cm (~55 – 300 AD). It is then absent from the record for the next 120 cm (~1100 cal years) with only one 2 cm interval containing the foraminifer. Between 91 and 25 cm (~1408 – 1863 AD), A. beccarii, in conjunction with the miliolid assemblage, is consistently present. A.beccarii is absent from the record in the upper 15 cm (1915 to present).

Gastropod biostratigraphy in the sediment core contains two distinct assemblages

(Figure 3.4c). The first assemblage is characterized by a brackish water snail of the

38 genus Battilaria. The second assemblage is defined by the presence of Heleobops clytus, a freshwater, epiphytic gastropod that was first described in samples from the Enriquillo basin (Thompson and Hershler 1991).

abc d 2000 2000

1500 1500

1000 1000 Calendar Age Age Calendar AD(+)/BC(-) 500 500

0 0

Sand -500 -500 Wood 30 40 50 60 70 2 4 6 8 10 12 14 012345 CaCO3 shell %CaCO OC/N %C(org) 3 Fine organic Organic-rich CaCO3 mud

Figure 3.3. Sediment geochemistry for core ENR 13-VI-01. (a) Percent calcium carbonate, (b) organic carbon:nitrogen, and (c) percent organic carbon for ENR 13-VI-01. (d) Coarse lithologic variations in the core also mirror the sediment geochemistry.

H. clytus

050100150

2000 2000

1500 1500

1000 1000

500 500 Calendar Age – AD(+)/BC(-)

0 0

a absence presence b absence presence c -500 -500

15 10 5 0 P. rugata A. beccarii Battilaria sp.

Figure 3.4. Biostratigraphy in core ENR 13-VI-01. (a) The halophilic ostracod P. rugata, (b) the traditionally marine foraminifer A. beccarii, and (c) two gastropods, H. clytus whose habitat is adjacent to fresh water springs, and Battilaria sp., a common brackish water organism.

The transition between these two assemblages creates a well-defined boundary in the microfossil record that is centered between 215 and 208 cm (~275 – 352 AD). Battilaria sp. occurs below the 160 AD horizon with abundances greater than 5 individuals occurring at 106 and 42 AD, as well as 55 BC. H.clytus is present in the microfossil

record in varying abundances between 0 – 208 cm. Peaks in abundance (> 50

individuals) occur at 170, 655, 1115, 1517, 1557, 1609, and 1694 AD.

GRA Bulk Density and Magnetic Susceptibility

Bulk density and magnetic susceptibility were plotted relative to age (calendar

years AD/BC) in core ENR 13-VI-01. GRA bulk density ranges from a high of ~1.9 in

the oldest part of the core to a low of ~1.16 gm/cm3 in the upper-most section of the core

(Figure 3.5a). Sediments exhibit regular variability between 355 BC and 985 AD, with

11 periods of peak bulk density values greater than 1.6 gm/cm3 in this interval. Bulk

density values then exhibit a gradual decline towards the top of the core, displaying only two periods with values greater than 1.6 gm/cm3 in the last ~1000 years. The magnetic

susceptibility record of Enriquillo sediments displays two pronounced periods of high

susceptibility (Figure 3.5b). The first interval occurs between ~355 and 150 BC, and

shows two peaks with values of 8.61 cgs and 12.37 cgs. A second period of high values

occurs between 810 and 985 AD, with a maximum of 14.55 cgs. Six additional peaks,

albeit of lesser magnitude, occur in the interval from 355 BC to 810 AD. After 985 AD, magnetic susceptibility values decrease upwards in the core, with only two values in the

last millennium > 3.0 cgs.

Stable isotopes

P. rugata valves (Figure 3.6a) first appear in the record at ~220 cal yr AD. From

220 to 470 AD, δ18O values average -0.49‰. The next ~ 200 years are devoid of

ostracod valves. Ostracods return to the sediment record ~665 cal yr AD and δ18O values

are highly variable for the next 300 years before reaching maximum values of 1.58‰,

1.45‰, 1.51‰, and 1.24‰ centered at 974, 993, 1018, and 1065 cal yr AD, respectively.

Following this interval, oxygen isotope values decline to a minimum of -0.35‰. Values

then return to an average of 0.58‰ before decreasing abruptly between 1750 -1825 cal yr

AD to a value of -0.79‰. Oxygen isotope concentrations of P. rugata valves then

increase towards the top of the sediment core.

Ammonia beccarii (Figure 3.6b) displays irregular presence over the length of the

sediment record. A. beccarii first appears between ~200 and 385 cal yr BC, during which

time δ18O values averaged -0.67‰. A. beccarii is absent from the microfossil record for

~280 cal yr and then returns to the record around 80 cal yr AD. Oxygen isotope concentrations generally vary about a mean of -1.22‰ for the next 360 cal yr, but the period is also characterized by an abrupt shift to a minimum value of -3.05‰ at 310 cal

yr AD. A. beccarii is absent from the microfossil record between 450 and 1480 cal yr

AD. When it returns, oxygen isotope values vary about a mean δ18O value of -0.04‰.

During the period between 1480 and 1960 cal yr AD, values vary between a maximum value of 0.81‰ at 1600 cal yr AD and minima of -1.93‰ and -0.69‰ centered at about

1690 and 1880 cal yr AD, respectively.

The lowest 20 cm of core ENR 13-VI-01 contains no fossil remains of the gastropod Heleobops clytus. H. clytus (Figure 3.6c) first appears in the microfossil

record at 93 cal yr BC with a δ18O value of 0.92‰. Oxygen isotope values vary about a

mean value of 1.29‰ for the entire record. The record is marked by four distinct minima

at 310, 700, 933 and 1335 cal yr AD. Following the minium at 1335 cal yr AD, δ18O values increase gradually towards the top before the gastropod disappears from the microfossil record in ~1960 cal yr A.D (15.0 cm sediment depth).

2000 a b

1500

1000

500 Calendar Age – AD(+)/BC(-) AD(+)/BC(-) – Age Calendar

-500 1.21.41.61.8 510

GRA bulk density magnetic susceptibility (cgs) (gm/cm 3)

Figure 3.5. GEOTEK multisensor core logger data of lithologic variables for core ENR 13-VI-01. GRA bulk density (a) and magnetic susceptibility (b) for core ENR 13-VI-01. Bulk density and magnetic susceptibility are measured simultaneously on a GEOTEK multisensor core logger.

# of gastropods 050100150 2000 2000

1500 1500

1000 1000 e – AD(+)/BC(-) g

500 500 Calendar A

0 0

a b c -500 -500

-1.5 -1 -0.5 0 0.5 1 1.5 -3 -2 -1 0 -2 -1 0 1 2 3 δ18O (‰, PDB) δ18O (‰, PDB) δ18O (‰, PDB)

Figure 3.6. Stable isotope (δ18O) stratigraphy from core ENR 13-VI-01. Carbonate shell material from the ostracod Perrisocytherridea rugata (a), the foraminifer Ammonia beccarii (b), and the gastropod Heleobops clytus (c). The δ18O of H. clytus is also plotted versus its absolute abundance (see Figure 3.4c). Records are smoothed with a 3-point running mean to illustrate long-term trends in local E/P ratios and regional climate dynamics.

Discussion

Interpretation of Sediment Proxies

Changes in the geochemical composition of lacustrine sediments serve as a proxy

for changes in the type and mode of material transported within a basin. Climate

variability can affect material transport processes through influencing physical and chemical weathering within a basin. In addition, climate has a strong influence on vegetation cover, vegetation type, and soil erosion. The relationship between vegetation cover, vegetation type and soil erosion affects the composition and amount of inorganic and organic material transported into basins (Rosenmeier et al., 2002; Binford et al.,

1987; Engstrom & Wright, 1984).

Changes in the composition of bulk organic matter, expressed as OC/N, provide

information on changes in the proportion of terrestrial versus aquatic organic matter

deposited in the lake basin. Low OC:N values are associated with organisms that are easily decomposed such as phytoplankton while recalcitrant woody debris and other

terrestrial organic material typically have high OC:N ratios. Poor preservation and

diagenetic loss of nitrogen may lead to elevated OC:N ratios (Cohen, 2003).

Carbonate deposited in lakes can be derived from either autochthonous or

allochthonous sources. Autochthonous sources include the burial of shelled organisms

(e.g. ostracods or foraminifera) or bio-induced precipitation during CO2 removal in

photosynthesis. Alternatively, allochthonous sources of carbonate can be delivered via

erosion and deposition of material from the lake’s watershed. Changes in carbonate

concentrations of sediments can elucidate changes in productivity, water chemistry and

regional climate factors.

Microfossil assemblages can serve as biotic proxies of environmental change.

Many organisms have well constrained habitat preferences and environmental

requirements that aid in the determination of past conditions. Salinity is considered a

primary determining factor in organismal distributions, but other biotic and abiotic

factors can also exert strong influence on their abundance (e.g food availability,

competition, predation, temperature, dissolved oxygen, pH) (Murray, 1991; Herbst,

2001). Paleolimnological reconstructions are best served when multiple invertebrate

groups are used (Crisman, 1978).

Physical properties of sediments (e.g. GRA bulk density and magnetic

susceptibility) serve as an indirect measure of litho-facies characteristics. Bulk density provides information on changes in lithology and porosity. Magnetic susceptibility

represents the degree to which a material can be “magnetized” by an external magnetic

field and is a function of the composition and concentration of magnetic minerals in the

sediment. Changes in these two physical properties can be related to both internal

processes such as chemical redox reactions that result in magnetic mineral formation

(Cohen, 2003), or climatic processes such as increased terrigenous input resulting from either wetter conditions (Nyberg, 2001) or increased Aeolian transport of terrigenous material during arid periods (Street-Perrott et al., 2000). Changes in sediment bulk density and magnetic susceptibility can reveal important information about internal and external processes within the catchment (Oldfield et al., 1983).

Stable isotope analysis of calcite-secreting organisms has been used extensively as an effective proxy for inferring past environmental change during the Holocene (Holmes,

1998; Curtis et al 1996; Curtis and Hodell, 1993; Covich and Stuiver, 1974). The oxygen

isotopic composition of authigenic calcite and aragonite (e.g. foraminifera, ostracod and

gastropod shells) is dependent on the temperature and oxygen isotopic ratio (18O/16O) of the host (i.e. lake) water (Craig, 1965). It is believed that temperature during the

Holocene has been relatively stable (Brenner el al. 2002) and long-term temperature variations in the tropics have been relatively small (Curtis and Hodell, 1993). Therefore, assuming that carbonate-secreting organisms produce their shells in equilibrium with lake water, the changing isotopic signature of the calicite or aragonite reflects the changes in the isotopic signature of the host water (Stuiver, 1970).

16 Isotope fractionation in the host water occurs because the lighter isotope, H2 O,

18 has a higher vapor pressure than H2 O and is preferentially incorporated into the gaseous

phase during evaporative events, leaving the water enriched in 18O (Fontes and

Gonfiantini, 1967; Curtis and Hodell, 1993). The isotopic signature of the host water is

thus controlled by climate conditions, namely the ratio between evaporation and

precipitation (E/P) (Curtis and Hodell, 1993). In addition, it has been shown that changes

in salinity can affect δ18O of carbonate shells (Chivas et al., 1985; 1986; Curtis and

Hodell, 1993). Discerning the effects of salinity versus changing E/P ratios on the δ18O of carbonate shells requires further information on chemical concentrations within the carbonate shell material (i.e. Sr/Ca and Mg/Ca ratios) (Chivas et al., 1986). Studies of

Lago Enriquillo waters have shown that δ18O and salinity covary up to salinities of 108‰

(r2 = 0.95) (Buck et al., in press). Changes in the δ18O of shell material in Enriquillo

probably represent changes in both the δ18O and salinity of lake water.

Changing Environments of the Lago Enriquillo Basin

Open marine to brackish water environment (~400 BC – 170 AD)

The base of core ENR 13-VI-01 (~385 – 270 BC) contains a high concentration of

CaCO3 (~70%) (Figure 3.4b). This section of the core also contains the foraminifer, G. bulloides. Isotope values obtained from A. beccarii are relatively light (average δ18O = -

0.67‰). G. bulloides is a planktonic foraminifer that is most abundant in cool upwelling environments with salinities between 33 and 36‰ (Hilbrecht, 1996). The presence of G. bulloides, together with the high CaCO3 content of the sediments and the isotopically

light signature of A. beccarii, indicate an open marine environment with salinities near

modern Caribbean values (i.e. ~35‰).

Studies of Holocene corals deposited in the Enriquillo basin during a marine

transgression yielded upper U-Th dates of ~3000 yr BP (~1050 cal yr BC) (Mann et al.,

1984; Taylor et al., 1985). If basal sediments of core ENR 13-VI-01 contain a hiatus,

part of the record spanning the transition from an open marine to a brackish environment

may be missing.

It has been suggested that the transition from an open marine to a brackish water

environment was the result of fluvial damming caused by the Yaque del Sur River that

flows southeast into the Neiba/Cul de Sac Valley (Mann et al., 1984). Although a precise

mechanism for basin isolation cannot be inferred from the sediment record, the transition

is recorded in the sediment geochemistry and associated microfossil assemblages near the

base of the core, about 270 ±110 cal yr BC. Organic carbon content (Figure 3.4c)

increases rapidly above this basal date, from < 1% to ~ 3%. Declines in bulk density and

magnetic susceptibility mark the horizon where wood was sampled for the basal AMS

14C date.

Presence of the gastropod Battilaria sp. between 310 BC and 160 AD (Figure 5c)

indicates a period of reduced marine influence and a transition to estuarine/brackish water

conditions. The halophobic gastropod H. clytus is present in this interval, but in very

small numbers. The δ18O values from this interval suggest a period of gradual freshening

of the waters and a reduction in E/P (Figure 3.7b, 3.7c).

High ecosystem variability (170 – 1000 AD)

Shifts in the microfossil assemblage around ~170 AD suggest a transition from

brackish waters to an environment more consistent with the hypersaline modern lake

ecosystem. The modern lake environment is characterized by hypersaline lake waters

that receive fresh water inputs from a series of springs and streams. This combination of

saline and fresh waters enables both halophobic and halophilic taxa to reside in the lake.

In addition to the previously noted foraminifera and ostracods, the freshwater gastropod

H. clytus also characterizes the carbonate microfossil assemblage (Buck et al., in press).

A pronounced shift in the gastropod assemblage around ~170 AD marks a hydrologic transition from brackish waters to an environment more consistent with the hypersaline

modern lake ecosystem. At this time, H. clytus comes to dominate the >212µm

microfossil assemblage, and the gastropod Battilaria sp. all but disappears (Figure 3.5c).

This transition is also marked by δ18O minima for all three species measured (Figure 3.7).

Following this transition from an estuarine to a hypersaline lacustrine environment,

paleo-proxies indicate periods of changing lake water salinities that affect

microinvertebrate assemblages and sediment lithology. Following the isotopic minimum

at ~170 AD, δ18O values of the ostracod P. rugata increase quickly by >1.0‰. This

increase could not be verified isotopically with foraminifera because of their scarcity in

the record (Figure 3.5b). The period between ~170 and 400 AD has only four samples

containing A. beccarii. A. beccarii has a limited salinity range in which it can reproduce

(16-67%) (Bradshaw, 1957). Its near absence during this interval suggests that salinities

increased to values >67%. The δ18O values of H. clytus also indicate increased E/P and

salinity (Figure 3.7c). E/P and salinity continued to increase until ~670 AD when

pronounced freshening occurred. This freshening event is represented by a peak in H.

clytus abundance (>50 individuals) that is coincident with a decrease in δ18O of >2.0‰

(Figure 3.5c). δ18O values for P. rugata also decline between 650 and 720 AD.

Carbonate shell material in the sediment core appears to be altered in the interval between ~700 and 1000 AD. Carbonate shells of both ostracods and gastropods showed pockets of erosion on the outer shell surface. Ostracods also appeared bleached. Few levels within this sediment interval contained carbonate shell material that did not appear diagenically altered. However, in the few intervals that did contain unaltered samples, the δ18O records of P. rugata and H. clytus suggest three excursions to drier, more saline

conditions in Lago Enriquillo between 700 and 1000 AD. The three excursions are

centered between 725 and 775 AD, around 830 AD, and 966 – 999 AD.

Freshening of Lago Enriquillo (1000 – 1350 AD)

Between 1000 and 1300 AD, Lago Enriquillo experienced a prolonged period of

wet, less saline conditions. This is inferred primarily from the decreasing δ18O of P.

rugata and H. clytus. Oxygen isotope ratios of P. rugata declined by nearly 1.5‰ and H. clytus declined by almost 3.0‰ (Figure 3.7a, 3.7c). From ~1000 to 1100 AD there was also an increase in the sedimentation rate in Lago Enriquillo that coincided with a subtle decline in the TOC/TN ratio. This decline suggests a greater contribution of

autochthonous carbon (i.e. phytoplankton) to the sediment organic matter, which may

have been a consequence of greater algal production under less saline conditions.

The decline in δ18O values continued from 1100 to 1350 AD. This period was

characterized by core lithology >50% CaCO3 and declining organic carbon content

(Figure 3.4). The isotope minimum measured in the shell material of H. clytus at ~ 1350

AD coincides with a slight increase in the abundance of this gastropod (Figure 3.5c).

Lago Enriquillo variability (1350 – present)

Following the isotopic minimum at 1350 AD, Lago Enriquillo transitioned into a

mode of consistent, yet damped variability with regard to sediment geochemistry, lake

water salinity, and inferred E/P. Bulk density and magnetic susceptibility variations are

reduced during this interval. A return to lake water salinities at or slightly higher than

seawater is marked by the return of the foraminifer A. beccarii in the biostratigraphic

record. Organic carbon concentrations and OC/N ratios remained high between 1500 and

1850 AD, suggesting increased contribution of terrestrially-derived organic matter. A

decline in OC/N ratios after 1850 AD suggests an increased contribution of

phytoplankton to sediment organic material. In addition, the gastropod H. clytus

becomes more abundant in this interval. Between 1350 and 2000 AD, a total of 1377

individuals were counted compared to only 984 individuals counted from the entire sediment core below this interval (130 BC – 1350 AD). Four of seven peaks in

abundance (>50 individuals/sample) occur in this time interval. This suggests a greater

and more consistent contribution of fresh water into Lago Enriquillo.

Regional and Global Climate Comparisons

The geochemical, biostratigraphic, and isotope records from Lago Enriquillo

illustrate this system’s sensitivity to local climate variations within its watershed. The

GRA bulk density and magnetic susceptibility records from this unique inland saline lake

may also reflect Holocene climate records. Comparisons between these two sediment

variables from Lago Enriquillo and other lacustrine and marine proxy climate records

within the Caribbean basin suggest consistent responses to changes in larger regional

climate regimes.

Mechanisms for GRA Bulk Density and Magnetic Susceptibility Variability

Bulk density measurements reflect the lithology and porosity of the sediments

Magnetic susceptibility is an indicator of lithologic change, revealing changes in the

magnetic character of sediments and also responds to changes in sediment particle size.

Comparison of these two variables with other sediment geochemical data helped

elucidate potential large-scale drivers of environmental change in the Lago Enriquillo

watershed and the larger Caribbean.

A comparison between bulk density, magnetic susceptibility, organic carbon

concentrations and CaCO3 concentrations suggests a series of complex relationships

between these lithologic and geochemical variables in ENR 13-VI-01. Figure 3.7

compares these four proxies. The shaded gray bands (numbered 1 through 5) correspond

with approximately 200 year time intervals surrounding periods of higher bulk density and magnetic susceptibility. Placement of these bands captures the two greatest excursions in bulk density and magnetic susceptibility centered around ~ 300 BC and

~900 AD. Each band is separated from other bands by ~200 years (Figure 3.7).

2000 2000

a b c d

1500 1500

1000 1000

)

- 4 ( /BC ) + (

Calendar Age Age Calendar AD 3 500 500

0 0

-500 -500 1.2 1.4 1.6 1.8 1234 30 40 50 60 70 510 %C(org) %CaCO GRA bulk density 3 magnetic susceptibility (cgs) (gm/cm3)

Figure 3.7. Comparison between sediment geochemistry and lithologic variables. This figure includes a comparison between (a) GRA bulk density, (b) organic carbon concentration, (c) CaCO3 concentration, and (d) magnetic susceptibility. In addition, 5 separate periods of high bulk density and magnetic susceptibility are highlighted. These bands represent ~200 years of the sediment record and are separated by approximately 200 years as well. Within these periods of high bulk density and, to a lesser extent, magnetic susceptibility the relationships between OC and CaCO3 can be illustrated.

During periods of high bulk density and high magnetic susceptibilty, organic

carbon (OC) decreases implying a reduction in the input of organic matter during these

periods (Figure 3.7a,b,d). Declines in OC concentrations are coincident with declines in

OC:N ratio (see Figure 3.3), suggesting a decrease in terrestrial inputs to the organic

carbon pool. In addition, CaCO3 concentrations closely parallel the variability in bulk density suggesting increased deposition of fine grained carbonate clay material during

these intervals. This increase in CaCO3 is most likely not related to increased input from

erosion in the watershed because of the out of phase relationship between CaCO3 and organic carbon (Figure 3.7b,c).

The alternating dark, organic-rich and grey, CaCO3-rich sediments (Figure 3.7b, c) may indicate reduced conditions at the sediment-water interface. It is possible for sulfate reduction to occur in saline environments (e.g. estuarine and marine environments) that

-2 have abundant sulfate (SO4 ) in the water column (Berner, 1980; Passier et el., 2001;

Otero & Macias, 2003).

Sulfate reduction occurs under anoxic conditions and is mediated by obligate

anaerobic sulfate reducing bacteria. During sulfate reduction, two byproducts are

produced:

-2 - 2CH2O + SO4 ---> H2S + 2HCO3

The two end products are then available for further reactions at the sediment-water

-2 interface. The bicarbonate ion can dissociate further to a carbonate ion (CO3 ) and

become available for bonding with divalent cations in the water column (e.g. Ca+2 and

+2 Mg ). Sulfide ions, produced via further reduction of H2S, can also bond with cations in

the water column. In saline environments under anoxic conditions, it is likely that these

sulfide ions will bond with reduced iron (Fe+2) in the pore waters and anoxic water

column. Pyrite (FeS2) is ultimately formed although intermediate compounds can also be

present (e.g. FeS). In marine sediments with alternating organic-rich and organic-poor

horizons, the transitions between horizons can contain minerals of high chemical remnant

magnetization (CRM) (Garming et al., 2004). A connection between pyrite formation

and magnetization may also exist in the Enriquillo sediments (Figure 3.8c).

Microscopic inspection of sediments from intervals with high bulk density did

confirm the presence of pyrite minerals. Although mineralogical data for the entire ENR

13-VI-01 core are not available, the association between pyrite and bulk density parameters in the interval between 356 BC and 630 AD (Figure 3.8a) suggests the two variables are correlated positively. Mineralogical data from the same sediment interval also suggest a relationship between quartz (SiO2) content and both bulk density and

magnetic susceptibility (Figure 3.8b, d). No quartz-bearing deposits, however, are recorded in the watershed. Consequently, I attribute the presence of quartz in these

deposits to two possible mechanisms, both of which suggest arid conditions. Intervals

with high bulk density and low carbonate microfossil preservation contained high concentrations of siliceous sponge spicules. Lack of carbonate in these intervals, in conjunction with the presence of recalcitrant sponge spicules, suggests reducing, low-pH

conditions which would dissolve biogenic carbonate. High concentrations of detrital

quartz may be attributed to periods of increased aeolian transport. Significant amounts of

dust from North Africa have been documented in the Caribbean over the past 40 years

(Prospero & Lamb, 1997). These dust plumes are inversely correlated with regional

rainfall (Prospero & Lamb, 1997) and may also be correlated with the North Atlantic

Oscillation (Moulin et al., 1997). Long-term records of aeolian transport have also been

documented for the Holocene and are correlated with periods of high E/P, i.e. drier

conditions in West Africa (Street-Perrott et al., 2000).

GRA bulk density (gm/cm3) magnetic susceptibility (cgs)

1.2 1.4 1.6 1.8 1.2 1.4 1.6 1.8 24681012 4 6 8 10 12

600 600

400 400

200 200

0 0 Calendar Age – AD(+)/BC(-) Calendar Age –

-200 -200

a b c d -400 -400

04 8 0246810 048 246 810 Pyrite (%) Quartz (%) Pyrite (%) Quartz (%)

Figure 3.8. Changes in lithologic variables versus bulk mineralogy. Variations in the bulk density record are related to pyrite (a) and quartz content (b). Magnetic susceptibility is also related to variations in sediment pyrite formation (c) and the quartz fraction (d). Pyrite formation is a proxy for anoxic bottom waters and increased salinity in the basin. Quartz is a proxy for Aeolian dust transport. (Pyrite and quartz are plotted as dashed lines.)

Variations in bulk density and magnetic susceptibility are related to both pyrite and quartz concentrations in the mineral fraction of sediments from Lago Enriquillo. The combination of these proxies suggests that high bulk density reflects periods of anoxia and sulfur reduction in the water column caused by high salinity and possibly periods of increased aeolian dust transport resulting from arid conditions across the equatorial

Atlantic. Assuming this to be the case, the record of bulk density can be used to relate the paleoclimate record from Lago Enriquillo to other Caribbean records to investigate the regional dynamics of climate variability.

Comparison to the Yucatan Peninsula

Paleolimnological data from the Yucatan Peninsula, Mexico, revealed several phases of drought during the last 2600 years (Curtis et al., 1996; Hodell et al., 2001; in press). The periodicity of gypsum precipitation in a core from Lake Chichancanab suggests a strong linkage between solar forcing mechanisms and climate variability in the region (Hodell et al., 2001). In addition, the oxygen isotope record from Punta Laguna in the northeast Yucatan is in close agreement with the gypsum record from Chichancanab

(Curtis et al., 1996). The bulk density record from Lago Enriquillo closely mirrors both the percent sulfur record from Chichancanab (Figure 3.9a) and the δ18O record from

Punta Laguna (Figure 3.9b) suggesting similar forcing mechanisms. In particular, two periods of high gypsum content (Chichancanab), positive excursions in δ18O (Punta

Laguna), and increases in bulk density (Enriquillo) occurred between 475 and 250 BC and 750 and 1050 AD. The cyclicity of droughts observed in the Yucatan (~208 and 50 year) is apparent in the Enriquillo bulk density record as well.

Regional Climate Linkages – Enriquillo and the Cariaco Basin

A marine record from the Cariaco Basin off Venezuela reveals a high-resolution

record of climate variability that is closely correlated with the records from the Yucatan

and Enriquillo (Figure 3.9c) (Haug et al., 2003). The Cariaco record infers periods of dry

and moist climate via titanium (Ti) concentrations in the marine sediments. Titanium is

used as a proxy for fluvial transport of terrestrial material during moist periods.

Consequently, Ti concentrations and rainfall are thought to be positively correlated in the

Cariaco record. However, the Cariaco record is dated relative to an assumed correlation

with the Medieval Warm Period (between 1000 and 1300 AD) and consequently “floats

in time”. Although the exact timing of drought in the Cariaco record differs slightly from

the Yucatan records and the Enriquillo record, the frequency and scale of drought events are comparable.

Extra Tropical Teleconnections – GISP2 Ice Core

The Lago Enriquillo record also compares with extra-tropical records from the

North Atlantic. The δ18O record from the Greenland GISP2 ice core is a >3000 m proxy

record of climate variability. The upper ~500 m of the GISP2 record correspond to the past ~2500 years and reveal close correlation with the Lago Enriquillo bulk density record between ~700 and 2000 AD (Figure 3.9d). Variations in the oxygen isotopic ratio of the Greenland ice record are positively correlated with variations in atmospheric

temperature (i.e. ⇑ Temp = ⇑ δ18O) (Stuiver et al., 1995).

δ18O ( ‰.PDB) % S Cytheridella ilosvayi Tit anium ( %) GISP2 δ18O ( ‰, PDB) 051015 -2 -1 0 1 0.2 0.15 0.1 -36 -35 -34

2000 2000

1500 1500

1000 1000

500 500

0 0 Calendar Age – AD(+)/BC(-) Calendar Age –

a b c d -500 -500

1.2 1.4 1.6 1.8 1.2 1.5 1.8 1.21.41.61.8 11.41.8 GRA bulk density GRA bulk density GRA bulk density GRA bulk density (cm/gm3) (cm/gm3) (cm/gm3) (cm/gm3)

Figure 3.9. Regional comparisons of late Holocene climate variability. A comparison between Lago Enriquillo bulk density and: (a) % sulfur from Lake Chichancanab, Mexico (Hodell, et al., 2001); (b) δ18O record from Punta Laguna, Mexico (Curtis et al., 1996); (c) the % titanium record from the Cariaco Basin in the southern Caribbean Sea (Haug et al., 2003); and (d) the δ18O record from the GISP2 ice core (Stuiver et al., 1997). Panels a through c illustrate the strong regional climatic similarities between the Yucatan, Lago Enriquillo, and the Cariaco Basin. Panel d illustrates periods of climatic teleconnectivity between the Caribbean Basin and high northern latitudes.

Examination of the post-818 AD interval revealed a relation between solar activity

(determined through variations in the production of cosmogenic 14C) and the GISP2 δ18O record (Stuiver et al., 1997). Periods of low solar activity (i.e. high ∆14C) are correlated with low δ18O values (i.e. cooler temperatures) in the Greenland ice core that include a

20-40 year lag due to ocean-atmosphere interactions (Stuiver et al., 1997, see figure 4).

The comparison between the GISP2 and Enriquillo records suggests that circum-

Caribbean and North Atlantic climate was affected by solar variability during the last millennium.

Mechanisms for Late Holocene Climate Variability

The mechanisms by which solar activity might affect both circum-Caribbean and

polar climate variability are poorly understood. Modeling experiments using a 0.25%

reduction in solar activity suggest a low latitude cooling response (Rind & Overpeck,

1995). Low latitude cooling can lead to a reduction in the intensity of Hadley cell

circulation across the tropics, thus reducing atmospheric moisture in the tropics and heat

transport to higher latitudes. In addition, it has been hypothesized that increases in ∆14C may result in minor increases in precipitation in the North Atlantic (Magny, 1993 – as cited in Stuiver & Braziunus, 1993). The combination of reduced heat transport and increased precipitation in the North Atlantic causes perturbations to the North Atlantic salt oscillator (Broecker et al., 1990) and ultimately to North Atlantic Deep Water

(NADW) production and regional climate dynamics including a cooler polar region

(Stuiver & Braziunus, 1993).

Additional mechanisms linking ∆14C to climate variability that have been suggested

include changes in the amount of ultraviolet light reaching the Earth’s upper atmosphere

and cascading effects on ozone production, upper-atmospheric temperatures, and possibly

cloud formation (Haigh, 1994; Van Geel et al., 1999). The relationship between ∆14C

and Holocene climate becomes more tenuous on larger timescales. Between ~500 and

2000 AD, the δ18O and gypsum concentration records from Yucatan lakes and the isotope

record from Greenland are in phase with ∆14C. However, prior to 500 AD, the variability

in both solar activity (∆14C) and these paleo-proxies are completely out of phase with one

another. (Stuiver et al., 1997; Hodell et al., 2001). Although correlations between long-

term paleoclimate records from the circum-Caribbean and North Atlantic have been

established using both terrestrial (Hillesheim et al., submitted) and marine archives

(Peterson et al., 2000), establishing that they respond to solar variability has proven more

difficult. Comparisons between long-term proxy climate records and ∆14C will be better

served by taking into account both solar-atmospheric and ocean-atmospheric production and diffusion of 14C. (Stuiver & Braziunus, 1993).

Solar forcing mechanisms and their potential control on moisture and heat transport

from the tropics also affects the dominant tropical weather pattern in the Intertropical

Convergence Zone (ITCZ). Variations in ocean circulation patterns from the tropical

Atlantic into the Gulf of Mexico, that are linked with the seasonal migration of the ITCZ, exhibit cycles similar to century-scale solar variability (Poore et al., 2004). Changes in salinity and associated foraminifer communities in the Cariaco basin are also related to the migration of the ITCZ that is in turn correlated to solar variability (Black et al., 2004).

During the later Holocene (ca. 3500 BP), it is thought that a general shift southward of the ITCZ caused drier conditions in the Caribbean basin in response to changing seasonal differences in insolation (Haug et al., 2001; Curtis & Hodell, 1993). This

61 orbitally-driven shift in insolation caused increased seasonality in the Southern

Hemisphere and a corresponding decrease in seasonality in the Northern Hemisphere.

The lower seasonality in the Northern Hemisphere decreased the likelihood of “pulling” the ITCZ northward off of the equator (Haug et al., 2001).

The migration of the ITCZ is also affected by SST in the equatorial Pacific. As

SST increases in the eastern Pacific (during an ENSO year, for example) sea level pressure (SLP) also decreases. This changes the pattern of surface flow in the Caribbean and results in increased SLP over the Caribbean. Increased trade wind activity follows the pressure gradient between the tropical Atlantic and the equatorial Pacific and results in a decrease in SST (Giannini et al., 2000). ENSO events effectively normalize temperatures across the equator that reinforces the southward shift of the ITCZ, bringing dry conditions to the circum-Caribbean (Maslin & Burns, 2000; Haug et al., 2001).

An additional driver of Holocene climate variability in the Caribbean basin is the

North Atlantic Oscillation SST tripole. The NAO SST tripole includes the North Atlantic dipole and the tropical Atlantic pressure system (aka Bermuda-Azores high) located between 0° and 15° N (Visbeck et al., 2003). These two SST/SLP phenomena can have competing and sometimes complementing effects on rainfall in the Caribbean (Giannini et al., 2000). Rainfall in the north Caribbean and the NAO dipole are inversely correlated

(Malmgren et al., 1998). Changes in the tropical Atlantic high pressure system are closely correlated with ENSO (Latif & Grötzner, 2000; Giannini et al., 2000). An increase in SLP in the tropical Atlantic results in increased trade wind activity, reduced convective activity, and drying in the Caribbean. The role of ENSO in determining SLP in the tropical Atlantic is complicated by changing SST across the NAO dipole.

Variability in Caribbean climate regimes is the result of complex interactions

between large scale ocean-atmospheric connections that may also be sensitive to

variations in solar activity. The agreement between lacustrine, marine, and glacial ice

records suggests that climate variability during the past ~2300 years occurred in response

to similar forcing mechanisms that included periods of large-scale excursions in both

atmospheric, oceanic, and potentially solar fields.

Conclusions

Lago Enriquillo’s hypersaline lake water chemistry is highly sensitive to changes in

local and regional climate. The ~2300 year sediment record from Lago Enriquillo

provides a high resolution, multi-proxy record of climate variability in the Caribbean

basin. The Enriquillo record is consistent with Holocene climate records from the

Yucatan peninsula (Curtis et al., 1996; Hodell et al., 2001; in press) and the Cariaco basin

(Haug et al., 2003). The agreement between lacustrine and marine records from the

Caribbean suggests that these climatic excursions were regional in nature. In addition, comparisons between the Enriquillo record and Greenland GISP2 ice core record argue for teleconnections between the Caribbean and the North Atlantic. The shift in these records at ~500 AD from being out of phase to in phase with one another suggests dynamic and changing interactions between the polar and tropical regions of the Atlantic.

Forcing mechanisms for this climate variability include solar variability, ENSO, and the

NAO SST tripole that result in shifts in the location of the ITCZ.

CHAPTER 4 CONCLUSIONS

Lago Enriquillo is a unique and dynamic inland saline lake. With this study, I have documented the lake’s variability over both short and long time periods. Short-term variability is observed in lake water salinity, major ion concentrations and stable isotopes

and is controlled largely by changes in E/P. Lake water salinity increased from 80 to

106ppt between June 2001 and March 2003 while ion concentrations increased by almost

20 percent. In addition, lake level decreased by ~1.8m during the same interval of time.

Lake level has varied significantly during the past 50 years (Cruz and Duquela-Gonzalez,

2003) and is closely correlated with tropical storm and hurricane frequency across the

basin. Isotopic analysis (δ18O and δD) of Lago Enriquillo lake waters illustrated the

effects of intense evaporation in the system. Evaporation of lake waters also resulted in a

positive linear response between salinity and δ18O. The microbenthic community of

Lago Enriquillo contains both halophilic and halophobic organisms, including foraminifera, ostracods, and gastropods. Habitat preference of these organisms, in combination with the other modern physical and chemical characteristics ultimately allowed for better interpretation of the paleolimnological record retrieved from Lago

Enriquillo.

The paleolimnological record from core ENR 13-VI-01 spans the past ~2300 years and reveals the lake’s sensitivity to climate variability over long time periods. A combination of sediment physical, mineralogical and geochemical properties, biostratigraphy, and stable isotope analysis was employed in this multi-proxy

63 64

paleolimnological study. During the middle Holocene, Lago Enriquillo was an

embayment of the Caribbean Sea and supported a fringing coral reef. The embayment

was cut off from the sea and transitioned to an environment synonymous with estuarine

conditions. The lake experienced a prolonged period of high variability between 170 and

1000 AD. This period of variability was punctuated by periods of high salinity and E/P

as inferred from the biostratigraphic and stable isotope records. The lake experienced a

brief freshening period between 1000 and 1350 AD that coincides with the Medieval

Warm Period that is recorded in other records from the circum-Caribbean (Curtis et al.,

1993; Hodell et al., 2001; Haug et al., 2003). Following this brief period of increased moisture availability, the lake was again influenced by the arid climate conditions that persist today.

The Lago Enriquillo record is also well correlated with other regional climate records. Together these records present a late Holocene climate history that is characterized by pan-Caribbean episodes of high E/P that occur with a ~208 and 50 year cyclicity. Mechanisms for this pan-Caribbean climate pattern focus on the migration of the ITCZ and how its position across the equatorial Atlantic belt can be affected by larger regional climate signals including ENSO, the NAO tripole, and solar variability.

Future research in the region will seek to construct additional paleolimnological records from the neighboring basins of Laguna Rincon and Etang Saumatre. In addition, a remote sensing project designed to monitor lake level fluctuations over time will benefit the existing survey of lake levels and better strengthen the proposed connection between tropical storm frequency and Lago Enriquillo lake level. Finally, greater attention must be focused on the role that current watershed management practices play in the ecology

and conservation of Lago Enriquillo. Currently, extensive irrigation canals and water storage facilities are under construction in the watershed that may ultimately upset the lake’s water balance. A complete survey of water control structures and volume

estimates for water diverted from the lake will benefit future natural resource managers and residents in the basin.

LIST OF REFERENCES

Araguás Araguás, L., Michelen, C., & Febrillet, J.,1993. Estudio de la dinamica del lago Enriquillo: informe de avance. Project DOM/8/006, International Atomic Energy Agency, Vienna, Austria.

Berner, R.A., 1980. Early Diagenesis – a theoretical approach. Princeton: Princeton University Press, 241p.

Binford, M.W., Brenner, M., Whitmore, T.J., Higuera-Gundy, A., Deevey, E.S., & Leyden, B.W., 1987. Ecosystems, paleoecology, and human disturbance in subtropical and tropical America. Quaternary Science Review 6: 115-128.

Black, D.E., Thunell, R.C., Kaplan, A., Peterson, L.C., & Tappa, E.J., 2004. A 2000-year record of Caribbean and tropical North Atlantic hydrographic variability. Paleoceanography 19: PA2022.

Bond, R.M., 1935. Investigations of some Hispaniolan lakes. II. Hydrology and hydrography. Archiv für Hydrobiologie 28: 137-161.

Bradbury, J.P., Leyden, B., Salgado-Labouriau, M., Lewis, W.M. Jr., Schubert, C., Binford, M.W., Frey, D.G., Whitehead, D.R., & Weibezahn, F.H., 1981. Late Quaternary environmental history of Lake Valencia, Venezuela. Science 214: 1299-1305.

Bradshaw, J.S., 1957. Laboratory studies on the rate of growth of the foraminifer ‘Streblus beccarii (Linne) var. tepida (Cushman)’. Journal of Paleontology 31: 1138-1147.

Brenner, M., Rosenmeier, M.F., D.A. Hodell, and J. H. Curtis. 2002. Paleolimnology of the Maya Lowlands: long-term perspectives on interactions among climate, environment, and humans. Ancient Mesoamerica 13:141-157.

Broecker, W.S., Bond, G., Klaus, M., Bonani, G., & Wolfli, W., 1990. A salt oscillator in the glacial Atlantic? 1. The concept. Paleoceanography 5: 469-477.

Bronk Ramsey, C., 1995. Radiocarbon Calibration and Analysis of Stratigraphy: The OxCal Program. Radiocarbon 37 v2: 425-430.

Bronk Ramsey, C., 2001. Development of the Radiocarbon Program OxCal. Radiocarbon 43 v2a: 355-363.

66 67

Buck, D.G., Brenner, M., Hodell, D.A., Curtis, J.H., Martin, J.B., & Pagani, M., in press. Chemical and physical parameters of hypersaline Lago Enriquillo, Dominican Republic. Verh. Internat. Verein. Limnol.

Canadian Superior Oil, Ltd., 1979. Bathymetry of Lake Enriquillo. Unpublished report presented to the Institute of Mining, Santo Domingo, D.R. 5p.

Chivas, A.R., Dedeckker, P., & Shelley, J.M.G., 1985. Strontium content of ostracods indicates lacustrine palaeosalinity. Nature 316: 251-253.

Chivas, A.R., Dedeckker, P., & Shelley, J.M.G., 1986. Magnesium and strontium in non- marine ostracod shells as indicators of palaeosalinity and palaeotemperature. Hydobiologia 143: 135-142.

Cohen, A.S., 2003. Paleolimnology: the history and evolution of lake systems. New York: Oxford University Press, 500p.

Covich, A. & Stuiver, M., 1974. Changes in Oxygen 18 as a measure of long-term fluctuations in Tropical lake levels and molluscan populations. Limnology and Oceanography 19.4: p. 682-691.

Craig, H., 1961. Isotopic variations in meteoric waters. Science 133: 1702-1703.

Craig, H., 1965. The measurement of oxygen isotope paleotemperatures, in: E. Tongiorgi (ed), Stable Isotopes in Oceanographic Studies and Paleotemperatures. Italy: Consiglio Nazionale delle Richerche, p. 161-182.

Crisman, T.L., 1978. Interpertations of past environments from lacustrine animal remains, in: D. Walter (ed), Biology and Quaternary Environments. Canberra: Australian Academy of Sciences, 264p.

Cruz, R. & Duquela-Gonzalez, T., 2003: Niveles del Lago Enriquillo. – Instituto Nacional de Recursos Hidrolicos, Santo Domingo, unpublished data.

Cucurullo, O. Jr., 1949. La hoya de Enriquillo. Santo Domingo: Universidad de Santo Domingo, Instituto Geografico y Geologico 64:3, 40p.

Curran, H.A. & Greer, L., 1998. Giant serpulid worm mounds cap a mid-Holocene fringing reef sequence, Enriquillo Valley, Dominican Republic. Abstracts with Programs--Geological Society of America 30:7, 334-335.

Curtis, J.H., & Hodell, D.A., 1993. An isotopic and trace element study of ostracods from Lake Miragoane, Haiti: A 10,500 year record of paleosalinity and paleotemperature changes in the Caribbean. American Geophysical Union Geophysical Monograph 78: 135-152.

Curtis, J.H., Hodell, D.A., & Brenner, M. 1996. Climate variability of the Yucatan peninsula (Mexico) during the past 3500 years, and implications for Maya cultural evolution. Quaternary Research 46: 37-47.

Deevey, E.S. & Stuiver, M., 1964. Distribution of natural isotopes of carbon in Linsley Pond and other New England lakes. Limnology and Oceanography 9: 1-11.

Dinerstein, E., Olson, D.M., Graham, D.L., Webster, A.L., Primm, S.A., Bookbinder M.P., & Ledec, G., 1995. A conservation assessment of the terrestrial ecoregions of Latin America and the Caribbean. Washington, D.C.:The World Bank, 150p.

Dix, G.R., Patterson, R.T., & Park, L.E., 1999. Marine saline ponds as sedimentary archives of late Holocene climate and sea-level variation along a carbonate platform margin: Lee Stocking Island, Bahamas. Palaeogeography, Palaeoclimatology, Palaoecology 150: 223-246.

Engleman, E.E., Jackson, L.L., & Norton, D.R., 1985. Determination of carbonate carbon in geological materials by coulometric titration. Journal of Great Lakes Research 2: 307–323.

Engstrom, D. R., & Wright, H.E., 1984. Chemical stratigraphy of lake sediments as a record of environmental change. In: E. Y. Haworth and J. W. G. Lund, (eds.), Lake Sediments and Environmental History, London: Leicester University Press, p. 11- 67

Eugster, H.P. & Hardie, L.A., 1978. Saline Lakes, in Lerman, A. (ed), Lakes: Chemistry, Geology, Physics. New York: Springer-Verlag, p. 237-293.

Fisher, M., Brenner, M., & Reddy, K.R., 1992. A simple, inexpensive piston corer for collecting undisturbed sediment/water interface profiles. Journal of Paleolimnology 7: 157-161.

Fontes, J.C. & Gonfiantini, R., 1967. Comportement isotopique au cours de l’evaporation de deux basins sahariens. Earth and Planetary Science Letters 3: 258-266.

Fritz, S.C., Metcalfe, S.E., & Dean, W., 2001. Holocene climate patterns in the Americas inferred from paleolimnological records, in: Markgraf, V. (ed), Interhemispheric Climate Linkages. San Diego: Academic Press, p. 241-263.

Garming, J.F.L., de Lange, G.J., Dekkers, M.J., & Passier, H.F., 2003. Changes in magnetic parameters after sequential iron phase extraction of eastern Mediterranean sapropel S1 sediments. Studia Geophysica et Geodaetica 48: 345-362.

Gat, J.R., 1980: Isotope Hydrology in very saline lakes, in: Nissenbaum, A. (ed.), Hypersaline Brines and Evaporitic Environments. New York: Elsevier, p. 1-7.

Giannini, A., Kushnir, Y., & Cane, M.A., 2000. Interannual variability of Caribbean rainfall, ENSO, and the Atlantic Oscillation. Journal of Climate 13: 297-311.

Giannini, A., Cane, M.A., & Kushnir, Y., 2001. Interdecadal changes in the ENSO teleconnection to the Caribbean region and the North Atlantic Oscillation. Journal of Climate 14: 2867-2879.

Gray, C. R., 1993. Regional meteorology and hurricanes, in Maul, G.A. (ed.): Climatic Change in the Intra-Americas Sea. London: Edward Arnold, p. 87–99.

Greer, L., 1997. Facies variability and three dimensional reef structure of the Enriquillo fringing reef, Dominican Republic. AAPG Bulletin 81: 1774-1775.

Haigh, J.D., 1994. The role of stratospheric ozone in modulating the solar radiative forcing of climate. Nature 370: 544-546.

Hastenrath, S., 1984. Interannual variability and the annual cycle: mechanisms of circulation and climate in the tropical Atlantic sector. Monthly Weather Review 112: 1097–1107.

Haug, G.H., Hughen, K.A., Sigman, D.M., Peterson, L.C., & Röhl, U., 2001. Southward migration of the Intertropical Convergence Zone through the Holocene. Science 293: 1304-1308.

Haug, G.H., Gunther, D., Peterson, L. C., Sigman, D. M., Hughen, K. A., Aeschlimann, B., 2003. Climate and the Collapse of Maya Civilization. Science 299: 1731-1735.

Heaton, T.H.E., Holmes, J.A., & Bridgewater, N.D., 1995. Carbon and oxygen isotope variations among lacustrine ostracods: Implications for paleoclimate studies. The Holocene 5: 428-434.

Herbst, D.B., 2001. Gradients of salinity stress, environmental stability and water chemistry as a template for defining habitat types and physiological strategies in inland salt waters. Hydrobiologia 466: 209-219.

Higuera-Gundy A., Brenner, M., Hodell, D.A., Curtis, J.H., Leyden, B.W., & Binford, M.W., 1999. A 10,300 14C record of climate and vegetation change in Haiti. Quaternary Research 52: 159-170.

Hilbrecht, H., 1996. Extant planktonic foraminifera and the physical environment in the Atlantic and Indian Oceans. Zürich: Mitteilungen aus dem Geologischen Institut der Eidgen Technischen Hochschule und der Universität Zürich, Neue Folge. No. 300, 93p.

Hillesheim, M.B., Hodell, D.A., Leyden, D.W., Brenner, M., Curtis, J.H., Anselmetti, F.S., Ariztegui, D., Buck, D.G., Guilderson, T.P., Newell, S.D., Rosenmeier, M.F., & Schnurrenberger, D.W. submitted. Lowland Neotropical climate change during the Last Deglacial and Early Holocene. Journal of Quaternary Science.

Hodell, D.A., Curtis, J.H., & Brenner, M., 1995. Possible role of climate in the collapse of Classic Maya civilization. Nature 375: 391-394.

Hodell, D. A., Brenner, M., Curtis, J.H., Guilderson, T., 2001. Solar forcing of drought frequency in the Maya Lowlands. Science 292: 1367–1369.

Hodell, D.A., Brenner, M. & Curtis, J.H., in press. Terminal Classic Drought in the Northern Maya Lowlands Inferred from Multiple Sediment Cores in Lake Chichancanab (Mexico). Quaternary Science Reviews.

Holdren, G.C. & Montaño, A., 2002. Chemical and physical characteristics of the Salton Sea, California. Hydrobiologia 473: 1-21.

Holmes, J.A., 1998. A late Quaternary ostracod record from Wallywash Great Pond, a Jamaica marl lake. Journal of Paleolimnology 19: 115-128.

Incháustegui, S., Gutiérrez, W., Rivas, V., Álvarez, V., Nunez de Ricart, N., & Bonnelly de Calventi, I., 1978. Notas sobre la ecología del Lago Enriquillo, República Dominicana en 1977, 307-342, in, Idelisa Bonnelly de Calventi (ed), Conservacion y Ecodesarollo, Centro de Investigaciones de Biología Marina. Santo Domingo: Universidad Autónoma de Santo Domingo, 531p.

Last, W.M., Vance, R.E., 1997. Bedding characteristics of Holocene sediments from salt lakes of the northern Great Plains, Western Canada. Journal of Paleolimnology 17: 297-318.

Last, W.M., 2001. Mineralogical analysis of lake sediments. In: Last, W.M., Smol, J.P. (Eds.), Tracking Environmental Change Using Lake Sediments. Volume 2: Physical and Geochemical Techniques. Dordrecht: Kluwer Academic Publishers, p. 143– 187.

Latif, M., & Grotzner, A., 2000. The equatorial Atlantic oscillation and its response to ENSO. Climate Dynamics 16: 213–218.

Magny, M., 1993. Solar influences on Holocene climatic changes illustrated by correlations between past lake-level fluctuations and the atmospheric 14C record. Quaternary Research 40: 1-9.

Malmgren, B.A., Winter, A., & Chen, D. 1998. El Niño-Southern Oscillation and the North Atlantic Oscillation control of climate in Puerto Rico. Journal of Climate 11: 2713-2718.

Mann, P., Taylor, F.W., Burke, K., & Kulstad, R., 1984. Subaerially exposed Holocene coral reef, Enriquillo Valley, Dominican Republic. Geologic Society of America Bulletin 95: 1084 -1092.

Margalef, R., 1986. Limnología del lago Enriquillo (República Dominicana). Oecologia aquatica, 8: 1-10.

Maslin, M.A. & Burns, S.J., 2000. Reconstruction of the Amazon Basin effective moisture availability over the past 14,000 years. Science 290: 2285-2287.

Masson, M.A., 2002. Community economy and the mercantile transformation. In: Masson, M.A. & Freidel, D.A., eds. Ancient Maya Political Economies. Lanham: AltaMira Press, p. 335-364.

Moulin, C., Lambert, C.E., Dulac, F., & Dayan, U., 1997. Control of atmospheric export of dust from North Africa by the North Atlantic Oscillation. Nature 387: 691-694.

Murray, J.W., 1991. Ecology and paleoecology of benthic foraminifera. New York: Longman Scientific & Technical, 397p.

NOAA, 2004: NHC/TPC Archive of Past Hurricane Seasons. – National Oceanic and Atmospheric Administration (http://www.nhc.noaa.gov/pastall.shtml - last accessed 06/15/2004)

Nyberg, J., Kuijpers, A., Malmgren B.A., & Kunzendorf, H., 2001. Late Holocene changes in precipitation and hydrography recorded in marine sediments from the northeastern Caribbean Sea. Quaternary Research 56, 87-102.

Oldfield, F., Barnosky, C., Leopold, E.B., & Smith, J.P., 1983. Mineral magnetic studies of lake sediments. Hydrobiologia 103: 37-44.

Passier, H.F., de Lange, G.J., & Dekkers, M.J., 2001. Magnetic properties and geochemistry of the active oxidation front and the youngest sapropel in the eastern Mediterranean Sea. Geophysical Journal International 145: 604-614.

Perez de Incháustegui, J., Castro de Checo, N., Bonnelly de Calventi, I., 1978. Bacterias patogenas en el Lago Enriquillo, in, Idelisa Bonnelly de Calventi (ed), Conservacion y Ecodesarollo, Centro de Investigaciones de Biología Marina. Santo Domingo: Universidad Autónoma de Santo Domingo, p. 345-348.

Peterson, L.C., Overpeck, J.T., Kipp, N.G. & Imbrie, J., 1991. A high-resolution late Quaternary upwelling record from the anoxic Cariaco Basin, Venezuela. Paleoceanography 6: 99-119.

Peterson, L.C., Haug, G.H., Hughen, K.A. & Röhl, U., 2000. Rapid changes in the hydrologic cycle of the tropical Atlantic during the last glacial. Science 290: 1947- 1951.

Poore, R.Z., Quinn, T.M., & Verardo, S., 2004. Century-scale movement of the Atlantic Intertropical Convergence Zone linked to solar variability. Geophysical Research Letters 31: L12214.

Prospero, J.M., Lamb, P.J., 2003. African droughts and dust transport to the Caribbean: climate change implications. Science 302: 1024-1027.

Rind, D. & Overpeck, J., 1995. Modeling the possible causes of decadal-to-millennial scale variability. In, The natural variability of the climate system on decade-to- century time-scales. Washington, D.C.: National Academy of Sciences Press, pp. 187-197

Rosenmeier, M.F., Hodell, D.A., Brenner, M., Curtis, J.H., Martin, J.B., Anselmetti, F.S., Ariztegui, D., & Guilderson, T.P., 2002. Influence of vegetation change on watershed hydrology: implications for paleoclimate interpretation of lacustrine δ18O records. Journal of Paleolimnology 27: 117-131.

Schubert, A., 2000a. El Lago Enriquillo: Patrimonio Natural y Cultural del Caribe. Santo Domingo: Banco Central de la República Dominicana, 50p.

Schubert, A., 2000b. Monstuos Simpaticos: los cocodrilos del Lago Enriquillo. Santo Domingo: Banco Central de la República Dominicana, 40p.

Stemann, T.A. & Johnson, K.G., 1992. Coral Assemblages, biofacies, and ecological zones in the mid-Holocene reef deposits of the Enriquillo Valley, Dominican Republic. Lethaia 25: 231-241.

Stemann, T.A. & Johnson, K.G., 1995. Ecologic stability and spatial continuity in a Holocene reef, Lago Enriquillo, Dominican Republic. Abstracts with Programs - Geological Society of America 27: 166.

Street-Perrott, F.A., Holmes, J.A., Waller, M.P., Allen, M.J., Barber, N.G.H., Fothergill, P.A., Harkness, D.D., Ivanovich, M., Kroon, D., & Perrott, R.A., 2000. Drought and dust deposition in the West Africa Sahel: a 5500-year record from Kajemarum Oasis, northeastern Nigeria. The Holocene 10.3: 293-302.

Stuiver, M., 1970. Oxygen and Carbon isotope ratios of fresh-water carbonates as climatic indicators. Journal of Geophysical Research 75: 5247-5257.

Stuiver, M. & Braziunus, T.F., 1993. Sun, ocean, climate and atmospheric 14CO2: an evaluation of causal and spectral relationships. The Holocene 3: 289-305.

Stuiver, M., Grootes, P.M., & Braziunas, T.F., 1995. The GISP2 18O climate record of the past 16,500 years and the role of the sun, ocean and volcanoes. Quaternary Research 44: 341-354.

Stuiver, M., T.F. Braziunas, P.M. Grootes, and G.A. Zielinski. 1997. Is there evidence for solar forcing of climate in the GISP2 oxygen isotope record? Quaternary Research 48:259-266.

Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Hughen, K.A., Kromer, B., McCormac, F.G., van der Plicht, J., Spurk, M., 1998. INTCAL98 Radiocarbon Age Calibration 24,000 – 0 cal BP. Radiocarbon 40: 1041–1083.

Taylor, F.W., Mann, P., Valastro, S. Jr., & Burke, K., 1985. Stratigraphy and Radiocarbon Chronology of Subaerially Exposed Holocene Coral Reef, Dominican Republic. Journal of Geology 93: 311-332.

Thompson, F.G. & Hershler, R., 1991. New hydrobiid snails (Mollusca: Gastropoda: Prosobranchia: Truncatelloidea) from North America. Proceedings of the Biological Society of Washington 104: 679-683.

UNESCO, 2002. “Biosphere Reserve Information: Jaragua-Bahoruco-Enriquillo”, http://www2.unesco.org/mab/br/brdir/directory/biores.asp?code=DOM+01&mode= all, UNECSO - MAB Biosphere Reserve Directory. (accessed 1/29/04).

Van Geel, B., Raspopov, O.M., Renssen, H., van der Plicht, J., Dergachev, V.A. & Meijer, H.A.J., 1999. The role of solar forcing upon climate change. Quaternary Science Reviews 18: 331-338.

Visbeck, M., Chassignet, E.P., Curry, R., Delworth, T., Dickson, B., & Krahman, G., 2003. The ocean’s response to North Atlantic Oscillation variability. American Geophysical Union Geophysical Mongraph 34: 113-145.

Williams, W.D., 1998. Salinity as a determinant of the structure of biological communities in salt lakes. Hydrobiologia 381: 191-201.

Wilson, M.A., Curran, H.A., & Greer, L., 2000. Rocky shore bioerosion and encrustation as a key to assessing sea-level change; an example from an emergent mid-Holocene limestone coast, Lago Enriquillo, the Dominican Republic. Abstracts with Programs - Geological Society of America 32: 96.

Wilson, S.M., 2001. The prehistory and early history of the Caribbean. In: Woods, C.A. & Sergile, F.E., eds. Biogeography of the West Indies: patterns and perspectives, 2nd edition. Boca Raton: CRC Press, p. 519-527.

Wright, H.E., Jr., Mann, D.H., & Glaser, P.H., 1984. Piston corers for peat and lake sediments. Ecology 65: 657-659.

BIOGRAPHICAL SKETCH

David Buck was born January 8, 1974, in Hartford, Connecticut. His family moved to Charlotte, North Carolina, where much of his extended family still lives. David graduated from West Charlotte High School in June of 1992. He attended the University of North Carolina at Chapel Hill from 1992 to 1996 and graduated with a bachelor’s degree in Latin American Studies and a minor in geology. Following college, David lived and worked in Prescott, Arizona, for two years before getting a job that would take

him to Belize, Central America. He worked as the field station manager for Monkey Bay

Wildlife Sanctuary for two years during which time he was responsible for facilitating

international study abroad programs that focused on the natural and cultural history of the

Maya region. It was in Belize that David met his wife, Ellie, an archaeologist from

Boston University.

David began the master’s program in interdisciplinary ecology at the University of

Florida in the fall of 2001. His research at UF has focused on acquatic ecology and late

Holocene climate variability. While at UF, David has had the opportunity to work on

projects in the Dominican Republic, Guatemala, and Florida. In addition to his academic

work, David still maintains an active interest in ecology and conservation issues in the

Neotropics.