A GAP ANALYSIS OF U.S. VIRGIN ISLANDS May 15, 2013 Final Report
Photo credit Alberto López
A GEOGRAPHIC APPROACH TO PLANNING FOR BIOLOGICAL DIVERSITY U.S. Geological Survey USDA Forest Service International Institute of Tropical Forestry Caribbean Landscape Conservation Cooperative
1
THE U.S. VIRGIN ISLANDS GAP ANALYSIS PROJECT FINAL REPORT May 9, 2013 William A. Gould (PI) Research Ecologist, USDA Forest Service, International Institute of Tropical Forestry, Río Piedras, Puerto Rico 00926-1119
Mariano C. Solórzano-Thillet, Biologist and GIS Specialist Gary S. Potts, Remote Sensing Specialist Maya Quiñones, Cartographic Specialist USDA Forest Service, International Institute of Tropical Forestry
Jessica Castro-Prieto, Ecologist IGERT Fellow-Department of Environmental Sciences University of Puerto Rico-Rio Piedras Campus
Lisa D. Yntema U.S. Virgin Islands
Contract Administration Through: USDA Forest Service, International Institute of Tropical Forestry, Río Piedras, Puerto Rico 00926-1119 Submitted by: William A. Gould Research Performed Under: Agreement No. 06HQPG0014 USDA Forest Service, International Institute of Tropical Forestry, 1201 Calle Ceiba, Río Piedras, Puerto Rico 00926-1119 U.S. Geological Survey, Biological Resources Division, Gap Analysis Program
LIST OF PROJECT AFFILIATES U.S. Forest Service International Institute of Tropical Forestry United States Geological Survey Biological Resources Division
Suggested citation: Gould WA, Solórzano MC, Potts GS, Quiñones M, Castro-Prieto J, Yntema LD. 2013. U.S. Virgin Islands Gap Analysis Project – Final Report. USGS, Moscow ID and the USDA FS International Institute of Tropical Forestry, Río Piedras, PR. 163 pages and 5 appendices.
2
TABLE OF CONTENTS
DEDICATION 4 ACKNOWLEDGEMENTS 5 EXECUTIVE SUMMARY 6 INTRODUCTION 11 LAND COVER CLASSIFICATION AND MAPPING 26 PREDICTED ANIMAL SPECIES DISTRIBUTIONS AND SPECIES RICHNESS 71 LAND STEWARDSHIP 95 ANALYSIS BASED ON STEWARDSHIP AND MANAGEMENT STATUS 116 CONCLUSIONS AND MANAGEMENT IMPLICATIONS 157 PRODUCT USE AND AVAILABILITY 158 GLOSSARY 162 GLOSSARY OF ACRONYMS 163
LIST OF APPENDICES
Appendix 1. List of example GAP applications. Appendix 2. Terrestrial vertebrate species of the U.S. Virgin Islands. Appendix 3. Terrestrial vertebrate species accounts and bibliography. Appendix 4. Species occurrence maps. Appendix 5. Species predicted distribution maps.
3
DEDICATION
The U.S. Virgin Islands are the eastern most extension of U.S. lands and seas. They are small in area and number of inhabitants, but blessed with the beauty of the Caribbean land, seas and sky and the warmth of the people. The three larger islands and a number of small cays have a wealth of natural resources, including rare and endangered plants and animals; endemic species; forests, mangroves, beaches, coves, cays, seagrass beds and reefs — all habitats vital to resident and migratory birds and marine species. Land and seascape stressors of climate change, sea level rise and land use issues create a complex arena for managing natural resources. Assessing biodiversity and the state of the environment is critical to making sound decisions to preserve, conserve, and augment the wealth of natural resources for future generations. In that spirit, this report is dedicated to all those who devote time, energy, and ideas to the understanding, preservation, conservation, restoration, and enjoyment of the beauty and natural resources of the U.S. Virgin Islands.
William Gould Principle Investigator
4
ACKNOWLEDGEMENTS
This work has been accomplished with the assistance of a great number of people. Their support and enthusiasm for conservation has been inspirational. I would like to first thank Jaime Collazo for bringing the GAP approach to the region and collaborating with the International Institute of Tropical Forestry. The support provided the U.S. Forest Service International Institute of Tropical Forestry was very important. Thanks to Institute Director Ariel Lugo. Kevin Gergley, Program Manager of the USGS Biological Resources Division, National Gap Program in Moscow Idaho has strongly supported this work. The staff at the National Gap office, including Jocelyn Aycrigg, Jill Maxwell, Anne Davidson, and Nicole Coffey, have been a great and patient help.
We have collaborated and received expertise and support from several people working in the Virgin Islands: including Brian Daley and Marcia G. Taylor at the University of the Virgin Islands; Laurel Brannick, Christy McManus and Zandy Hillis-Starr with the National Park Service; Carol Cramer-Burke with the St. Croix Environmental Association; people at the Island Resources Foundation; Renata Platenberg, Jennifer Valiulis, and Judy Pierce at the Department of Planning and Natural Resources Division of Fish and Wildlife; Claudia Lombard at the U.S. FWS Sandy Point National Wildlife Refuge; James Byrne and Shawn Margles at the Nature Conservancy; the U.S. Geological Survey National Wetlands Research Center; Magen’s Bay Authority; Brittany Barker, Ph.D. Candidate from the University of New Mexico; independent researchers and stakeholders, including Lisa Yntema, expert on birds from St. Croix; Russell Slatton from Geographic Consulting LLC; and Mario Francis from the Junior Gardening & Ecology Academy. Collaborators have help us coordinate meetings and provided information and suggestions to improve our work. They have also reviewed our products, supported our efforts, and shared their expertise as well as species occurrence and GIS data with us. We thank them all and hope these products will be useful in their work.
Scientists at IITF have been a source of feedback, expertise and encouragement including Director Ariel Lugo, Project Leader Grizelle González, and other scientists Wayne Arendt, Eileen Helmer, Jean Lodge, Frank Wadsworth, Pete Weaver, and Joe Wunderle. I would like to thank all the people in the IITF GIS and Remote Sensing Laboratory for their support and enthusiasm, including Olga Ramos-González and Maya Quiñones.
Finally, I would like to thank the primary Institute GAP staff who shared their time, enthusiasm, and expertise to make this project successful, including Gary Potts, Mariano Solórzano, Maya Quiñones, and Jessica Castro Prieto.
5
EXECUTIVE SUMMARY
The U.S. Virgin Islands Gap Analysis Project has developed and compiled information of the U.S. Virgin Islands’ landcover, vertebrate occurrences, and natural history information, and land stewardship. It is based on methods developed by the National Gap Analysis Program to determine the degree to which animal species and natural communities are represented in the current mix of conservation lands. Those species or communities not well represented are considered conservation “gaps.” The purpose of the U.S. Virgin Islands Gap Analysis Project is to provide geographic and ecological information on the status of not only threatened or rare species, but the common species of the U.S. Virgin Islands. This provides land managers, government planning and policy makers, scientists, students, and the general public with information to make better decisions regarding land management and conservation. The U.S. Virgin Islands Gap Analysis project has four major components: Landcover mapping, documentation of vertebrate species distributions, documentation of land stewardship practices with respect to conservation, and an integrated analysis of these three elements. The databases of species occurrence, land cover, and stewardship for the USVI GAP are being integrated with the Puerto Rico Gap Analysis Project (PRGAP) (Gould et al. 2008) and the Puerto Rico-USVI Integrated Terrestrial-Aquatic Gap (Integrated Gap) (Gould et al. 2010) to allow regional analyses of terrestrial, freshwater, and marine habitats and biodiversity.
The U.S. Virgin Islands (USVI) are located in the Caribbean just east of Puerto Rico in the northwestern most section of the Lesser Antilles. The USVI cover 350 km2 and include St. Thomas, St. John, St. Croix, and a number of cays. They harbor relatively high numbers of species and substantial levels of endemism, particularly among the reptiles.
USVI GAP includes 153 species of terrestrial vertebrates: 118 birds, 21 reptiles, 7 amphibians and 8 mammals. These include endemic, breeding resident, breeding migrant, established exotic and nonbreeding migrant species. The majority are breeding residents. Breeding migrants include birds and marine turtles - which use terrestrial habitat for nesting. Ten to 20 percent of the amphibians and reptiles are endemic. We are following the traditional GAP approach, developing geospatial information and databases on land stewardship, species occurrence, and land cover.
Landcover Traditionally, GAP projects have relied on satellite imagery from Landsat 5 TM and Landsat 7 ETM+ to provide the spatial and spectral information to derive land cover habitat maps at 30m spatial resolution. Puerto Rico GAP (Gould et al. 2008) incorporated the Landsat 7 ETM+ 15m panchromatic band to enhance the spatial resolution and infrared bands in order to improve the delineation of habitats at the sub-pixel level in complex tropical landscapes. Current Landsat 7 ETM+ imagery and scene acquisition is limited by the scan line correction (SLC) error, horizontal lines with no data that appear across the entire image since July 2003, and the Long Term Acquisition Plan (LTAP), the
6
use of a set of criteria that includes cloud-cover forecasts (Landsat Project Science Office, 1998) to guide Landsat image collections. These limitations make the collection of new images and the use of existing images for tropical humid regions with a high potential for cloud cover difficult.
For the USVI GAP, we used imagery from the Advanced Land Imager (ALI) onboard the Earth Observation 1 (EO-1) satellite. EO-1 was launched by the U.S. Geological Survey (USGS) in 2000 as a one year technical mission for data continuity assessment for the Landsat pro-grams
We have classified 67.2% of the USVI as predominantly woody vegetation and 6.9% as grassland or herbaceous agriculture. We have classified 21.5% as developed land and about 1% water and 1.9% natural barrens. Twenty-six land cover units are described dominated by woody vegetation. Of these, upper elevation and gallery moist forests, shrublands and woodlands cover 4.6%, dry forests, shrubland and woodland cover 60.7%, and flooded mangrove forests cover 1% of the islands. Closed forests cover 20.6%, woodlands 12.5%, closed canopy shrublands 28.5% and open canopy shrublands 5.6% of the area.
Vertebrates The U.S. Virgin Islands terrestrial vertebrate biodiversity is composed of over 294 species. These include breeding and non-breeding residents, non-breeding and breeding migratory, vagrant or accidental species, and established exotics. We selected a group of 153 species of birds, mammals, reptiles and amphibians for the USVI GAP Analysis. We left out those species that are very rare non-breeding migrants and vagrants as well as extirpated species such as the tree frog Eleutherodactylus schwartzi. In addition to the breeding species we included a group of non-breeding residents, breeding and non- breeding migratory species of importance for which there was occurrence data available. We also included some non native species (established exotic species), which may have a potential impact on the native biota, such as the mongoose.
Species geographic distributions were mapped using a network of 2 km2 hexagons that cover the U.S. Virgin Islands. Each hexagon was attributed with the species probability of occurrence according to one of eight categories on the likelihood that the species occurs in the hexagon. Species probability of occurrence information is derived from published literature, unpublished datasets, museum records, and expert opinion. A species
7
record of occurrence was confirmed when associated with a credible observation, including a record of the location, the observation date, and the observer’s name.
Species habitats were predicted using models developed from the literature and expert review. Species distributions were mapped by identifying predicted habitat within the species geographic range. The resulting 153 maps of species predicted distribution are a result of the integration of information from both the vertebrate database and landcover mapping. Predicted distributions are in raster format with a resolution of 10 m2. We modeled the habitat distribution of 118 bird species, 21 reptiles, 6 amphibians, and 8 mammal species that occur in the USVI.
Combining information from all species allows a prediction of how many and which species occur for any given area of Puerto Rico down to a 10 m2 area. We developed species richness maps using this information in order to assess patterns of biodiversity. The resulting patterns indicate the forested parts of the landscape have the highest species richness, represented by as many as 73 species. Urban and barren areas have the lowest species richness, represented by as few as 7 species. Individual taxonomic groups each show distinct patterns.
Land stewardship Land use and development pressures are occurring in nearly all parts of the world. There is increasing need for land managers to prioritize and justify conservation efforts at both local and regional scales. An important component of these efforts is the regional analysis of the state of conservation management. This includes identifying whether current management areas encompass the extent and diversity of land areas available for conservation, whether currently managed areas adequately reach conservation goals, and what areas are most in need of conservation.
The GAP program currently uses a scale of 1 to 4 to denote relative degree of maintenance of biodiversity for stewardship areas. A status of "1" denotes the highest, most permanent level of maintenance, and "4" represents the lowest level of biodiversity management, or unknown status. The four status categories can generally be defined as follows (after Scott et al. 1993, Edwards et al. 1995, Crist et al. 1995):
Status 1: An area having permanent protection from conversion of natural land cover and a mandated management plan in operation to maintain a natural state. Status 2: As above but which may receive use or management practices that degrade the quality of existing natural communities. Status 3: As above for the majority of the area but subject to extractive uses of either a broad, low-intensity type or localized intense type. Status 4: Lack of irrevocable easement or mandate to prevent conversion of natural habitat types to anthropogenic habitat types. Also includes those tracts for which the existence of such restrictions or sufficient information to establish a higher status is unknown.
8
We identified a set of 89 stewardship areas with potential biodiversity protection. Sixty- nine were classified with Status 1, 2 or 3. These represent 13% of the total land area of the U.S. Virgin Islands. Land protection varies greatly from island to island. While St. John provides protection to 56.6% of its territory, the other two main islands protect less than 6%. From the 69 protected areas, 19 (27.5%) of them were classified in Status 1; 23 areas (33.3%) in Status 2 and 27 areas (39.1%) in Status 3. However, protected areas in Status 1, the highest level of protection, represented 75.3% of the total because of the relatively large size of some of the areas including the Virgin Islands National Park, Sandy Point Wildlife Refuge and Buck Island Reef National Monument. Only five protected areas had management plans and two were in the process of development.
Protected land ownership is shared among 10 public and private institutions and individuals, with the federal government as the primary owner (71.2 %), followed by the USVI government (14.9 %), non-governmental organizations (7.1 %) and private institutions or individuals (3.2 %). Similarly, 71.2 % of the protected area is managed by federal agencies, 14.8 % by USVI governmental agencies, 10.2% by non-governmental organizations and the remaining 3.7% is co-managed between the USVI government and federal agencies. Protected areas management was primarily shared among 11 public and private institutions, where the U.S. National Park Service, the USVI DSPR, the USVI Department of Planning and Natural Resources (Division of Fish and Wildlife) and The Nature Conservancy represent the main protected area managers in the U.S. Virgin Islands.
Gap Analyses - landcover Thirty percent of the Virgin Islands’ forests are protected in areas with Gap Status 1 or 2. This includes 22% of the dry forests, 36% of the moist forests, and 30% of the flooded forests, typically mangroves.
Gap Analyses - vertebrates Four species have less than 1% of their habitat protected. In terms of the amount of protected habitat, these species appear to be some of the most vulnerable, although this is only one of many factors that affect the long-term survival of a species. Three birds and one reptile fall within this range of habitat protection. Two of the birds are associated with wetlands or water-bodies in general: The least grebe and the purple gallinule. Habitats such as wetlands and salt ponds in U.S. Virgin Islands are under great threat of degradation or destruction as has occurred in the past (McNair et al. 2006). The other bird, the Antillean nighthawk, is a breeding migrant whose habitat is more represented in the landscape than the other three species (13.7%) but is a species of greatest concern for the DPNR Division of Fish and Wildlife (Plantenberg et al. 2005). The last one, a reptile, the fat-tailed gecko is an introduced species and represents little concern to local environmental agencies.
Thirty-seven species have 1% to less than 10% of their habitat protected. This group of species includes birds, reptiles and amphibians; and some of these species habitats cover a significant amount of land area. Such is the case of the zenaida dove (82.2% of the islands), the smooth-billed ani (53.2%) and the St. Croix anole (47.2%). It could be said
9
that these species are relatively widespread and as a consequence they have very little protection of their habitat (5.1% and 3.3% respectively). There are some species in this range of protection, however, whose habitat covers very little of the territory, even less than 1%, including some sandpipers, the leatherback sea turtle, the clapper rail, the Caribbean coot and others. These latter species share in common certain habitats, such as sandy beaches or wetlands. In addition, the endangered Virgin Islands boa, Epicrates monensis granti, is also in this range, and its habitat (8.4% of the islands or 2962 ha) currently enjoys very little protection (1.7%).
Forty-one species have 10% to less than 20% of their habitat protected. Most of these species associated habitats cover a significant amount of land area, 24% in average, but up to 90.4% of the territory in the case of the Velvety free-tailed bat. In contrast, the least sandpiper habitat only covers 1.5% of the territory and some other four species habitats represent less than 1%.
Fifty-six species have 20% to less than 50% of their habitat protected. These species habitats cover between 0.2% (green sea turtle) and 36.6% (black-whiskered vireo) of the islands territory, 6.7% in average and a standard deviation of 6.4%. In other words, the amount of area covered by the habitats associated with these species shows much less variation when compared to the previous two range categories of habitat protection.
Thirteen species have more than 50% of their habitat protected. Most of these species habitats cover less than 1% of the islands territory. This is the case of some marine birds - three terns, two boobies and the brown noddy – whose habitat is mostly located in off- shore cays which have been designated marine sanctuaries by the DPNR Division of Fish and Wildlife. Although this category encompasses mostly birds there is also one reptile: St. Croix Ground Lizard. This latter has been extirpated from St. Croix’s main island and is now relegated to the four cays surrounding St. Croix. It is listed as endangered under the US Endangered Species Act. Two species in this protection range, the Puerto Rican Flycatcher and the Black-throated Blue Warbler, cover more territory than the rest, 11.5% and 12.8% respectively.
Fifty-eight species are listed as either federally threatened or endangered or given partial status, are locally listed by the USVI Indigenous Species Act as locally threatened, locally endangered, data deficient or special concern, or are classified as species of concern or species of greatest concern by the DPNR Division of Fish and Wildlife. Most cover less than 10% of the islands territory and 18 cover less than 1%. Only two cover more than 50% of the territory: the Antillean fruit-eating bat and the white-crowned pigeon. However, 48 of these species have more than 10% of their habitat under protection. There are some results that stand out. For example, the Virgin Island tree boa only has 1.6% of its predicted habitat under protection, and the endemic yellow mottled coqui, only 2.4% protected.
10
INTRODUCTION
How this report is organized This report is a summation of a scientific project. While we endeavor to make it understandable for as general an audience as practicable, it reflects the complexity of the project it describes. A glossary of terms is provided to aid the reader in its understanding, and for those seeking a detailed understanding of the subjects, the cited literature should be helpful. The organization of this report follows the general chronology of project development, beginning with the production of the individual data layers and concluding with analysis of the data. It diverges from standard scientific reporting by embedding results and discussion sections within individual chapters. This was done to allow the individual data products to stand on their own as testable hypotheses and provide data users with a concise and complete report for each data and analysis product.
We begin with an overview of the Gap Analysis Program mission, concept, and limitations. We then present a synopsis of how the current biodiversity condition of the project area came to be, followed by land cover mapping, animal species distribution prediction, species richness, and land stewardship mapping and categorization. Data development leads to the Analysis section, which reports on the status of the elements of biodiversity (plant communities and animal species), for the U.S. Virgin Islands. Finally, we describe the management implications of the analysis results and provide information on how to acquire and use the data.
A number of the figures and maps in this report are also available as more detailed and high resolution maps and as research publications with more in-depth analyses.
The Gap Analysis Program Mission The mission of the Gap Analysis Program is to prevent conservation crises by providing conservation assessments plant communities and native animal species and to facilitate the application of this information to land management activities. This is accomplished by meeting the following five objectives:
1) Map current land cover, 2) Map the predicted distribution of selected terrestrial vertebrates, 3) Document the representation of land cover types and animal species in areas managed for the long-term maintenance of biodiversity, 4) Make PRGAP project information available to the public and those charged with land use research, policy, planning, and management and 5) Build institutional cooperation in the application of this information to state and regional management activities.
To meet these objectives, it is necessary that GAP be operated at the state, commonwealth, or regional level while maintaining consistency with national standards.
11
Participation by a wide variety of cooperators is necessary and can lead to understanding and acceptance of the data and the development of relationships that will lead to cooperative conservation planning.
State objectives for GAP The U.S. Virgin Islands Gap Analysis Project had a number of additional objectives. We adapted methodology of the National Gap Program to meet the needs of a small archipelago of tropical islands with a unique social, economic, and ecological environment. The USVI environment includes diverse bioclimatic zones, considerable ecological variation over small distances, a high degree of development pressure, a long history of ecological research but few published descriptions of natural vegetation at the plant community level, and the majority of the land management for conservation being done by the National Park Service.
Our State-level objectives were to: 2 1) Develop a coverage of 2 km hexagons (Map 1) for use in occurrence and range mapping that accommodates the scale of landscape variation in the USVI (Gould et al. 2007a). 2) Develop contacts and collaborations with conservation agencies and groups in Puerto Rico in order to receive input and communicate our findings in ways useful to immediate conservation concerns. 3) Develop databases of species occurrences, scientific literature related to species habitats and distributions, and species natural history information that are dynamic and will have a useful life beyond the completion of USVIGAP. 4) Compile and develop new information on natural vegetation at the plant community level, and crosswalk these descriptions with other descriptive hierarchies. 6) Develop a project that can serve as a foundation for the development of an integrated Puerto Rico-United States Virgin Islands Gap Project and Caribbean Gap Analysis Project.
The Gap Analysis Concept The Gap Analysis Program (GAP) brings together the problem-solving capabilities of federal, state, and private scientists to tackle the difficult issues of land cover mapping, animal habitat characterization, and biodiversity conservation assessment at the state, regional, national, and international levels. The program seeks to facilitate cooperative development and use of information. Throughout this report we use the terms "GAP" to describe the national program, "GAP Project" to refer to an individual state or regional project, and "gap analysis" to refer to the gap analysis process or methodology.
Much of the following discussion was taken verbatim from Edwards et al. 1995, Scott et al. 1993, and Davis et al. 1995. The gap analysis process provides an overview of the distribution and conservation status of several components of biodiversity. It uses the distribution of actual vegetation and predicted distribution of terrestrial vertebrates and, when available, invertebrate taxa. Digital map overlays in a GIS are used to identify individual species, species-rich areas, and vegetation types that are unrepresented or under-represented in existing management areas. It functions as a preliminary step to the
12
more detailed studies needed to establish actual boundaries for planning and management of biological resources on the ground. These data and results are then made available to the public so that institutions as well as individual landowners and managers may become more effective stewards through more complete knowledge of the management status of these elements of biodiversity. GAP, by focusing on higher levels of biological organization, is likely to be both cheaper and more likely to succeed than conservation programs focused on single species or populations (Scott et al. 1993).
Biodiversity inventories can be visualized as "filters" designed to capture elements of biodiversity at various levels of organization. It is postulated that 85-90% of species can be protected by the coarse filter without having to inventory or plan reserves for those species individually. A fine filter is then applied to the remaining 15-10% of species to ensure their protection. Gap analysis is a coarse-filter method because it can be used to quickly and cheaply to assess the other 85-90% of species. GAP is not designed to identify and aid protection of elements that are rare or of very restricted distribution; rather it is designed to help "keep common species common" by identifying risk far in advance of actual population decline. These concepts are further developed below.
The intuitively appealing idea of conserving most biodiversity by maintaining examples of all natural community types now widely recognized and numerous approaches to the spatial identification of biodiversity in association with natural community type have been described (Kirkpatrick 1983, Margules et al. 1988, Pressey and Nicholls 1989, Nicholls and Margules 1993). Furthermore, the spatial scales at which organisms use the environment differ tremendously among species and relates to differences in body size, food habits, mobility, and other factors. Hence, no coarse filter will be a complete assessment of biodiversity protection status and needs. However, species that fall through the pores of the coarse filter, such as narrow endemics and wide-ranging mammals, can be captured by the safety net of the fine filter. Community-level (coarse-filter) protection is a complement to, not a substitute for, protection of individual rare species. Gap analysis is essentially an expanded coarse-filter approach (Noss 1987) to biodiversity protection. The land cover types mapped in GAP serve directly as a coarse filter, the goal being to assure adequate representation of all native vegetation community types in biodiversity management areas. Landscapes with great vegetation diversity often are those with high edaphic (soil or substrate) variety or topographic relief. When elevational diversity is very great, a nearly complete spectrum of vegetation types known from a biological region may occur within a relatively small area. Such areas provide habitat for many species, including those that depend on multiple habitat types to meet life history needs (Diamond 1986, Noss 1987). By using landscape-sized samples (Forman and Godron 1986) as an expanded coarse filter, gap analysis searches for and identifies biological regions where unprotected or under-represented vegetation types and animal species occur.
More detailed analyses were not part of this project, but are areas of research that GAP as a national program is pursuing. For example, a second filter could combine species distribution information to identify a set of areas in which all, or nearly all, mapped species are represented. There is a major difference between identifying the richest areas
13
in a region (many of which are likely to be neighbors and share essentially the same list of species) and identifying areas in which all species are represented. The latter task is most efficiently accomplished by selecting areas whose species lists are most different or complementary. Areas with different environments tend to also have the most different species lists for a variety of taxa. As a result, a set of areas with complementary sets of species for one higher taxon (e.g., mammals) often will also do a good job representing most species of other higher taxa (e.g., trees, butterflies). Species with large home ranges, such as large carnivores, or species with very local distributions may require individual attention. Additional data layers can be used for a more holistic conservation evaluation.
These include indicators of stress or risk (e.g., human population growth, road density, rate of habitat fragmentation, and distribution of pollutants) and the locations of habitat corridors between wild lands that allow for natural movement of wide-ranging animals and the migration of species in response to climate change.
General Limitations Limitations must be recognized so that additional studies can be implemented to supplement GAP. The following are general project limitations; specific limitations for the data are described in the respective sections:
1. GAP data are derived from remote sensing and modeling to make general assessments about conservation status. Any decisions based on the data must be supported by ground-truthing and more detailed analyses.
2. GAP is not a substitute for threatened and endangered species listing and recovery efforts. A primary argument in favor of gap analysis is that it is proactive: It seeks to recognize and manage sites of high biodiversity value for the long-term maintenance of populations of native species and communities before they become critically rare. Thus, it should help to reduce the rate at which species require listing as threatened or endangered. Those species that are already greatly imperiled, however, still require individual efforts to assure their recovery.
3. GAP data products and assessments represent a snapshot in time generally representing the date of the satellite imagery. Updates are planned on a 5-10 year cycle, but users of the data must be aware of the static nature of the products.
4. GAP is not a substitute for a thorough national biological inventory. As a response to rapid habitat loss, gap analysis provides a quick assessment of the distribution of vegetation and associated species before they are lost, and provides focus and direction for local, regional, and national efforts to maintain biodiversity. The process of improving knowledge in systematics, taxonomy, and species distributions is lengthy and expensive. That process must be continued and expedited, however, in order to provide the detailed information needed for a comprehensive assessment of our nation's biodiversity. Vegetation and species distribution maps developed for GAP can be used to make such surveys more cost-effective by stratifying sampling areas according to expected variation in biological attributes.
14
The Study Area
A Brief Description of the U.S. Virgin Islands
The U.S. Virgin Islands (USVI) are located at the northwest of the Lesser Antilles, a set of small islands in the eastern Caribbean basin. It is composed of three main islands, St. Thomas, St. John and St. Croix and more than 60 recognized cays and off-shore rocks (Fig. 1 and 2; Table 1). In all, the islands have a land area of 346 km2. The islands have a lot in common in terms of geology, climate, and ecology (Acevedo-Rodriguez 1996).
St. Thomas and St. John, on the north, are geologically part of the Puerto Rican bank and were connected in the past, but not St. Croix, located some 60 km to the south. It has been estimated that the northern islands separation occurred some 8,000 to 10,000 years ago, at the end of the last Ice Age. St. Croix on the other hand, either has been isolated for much longer or was never connected to the other islands (Heatwole et al. 1981; Wiley and Vilella 1998). A sea channel more than 3,600 meters deep separates St. Croix from the other islands and a distance of 64 Km. The USVI are of volcanic origin but also contain some limestone derived soils particularly in St. Croix (Rankin 2002; NRCS 2000).
The climate of the islands is classified as subtropical (Figure 6; Ewel and Whitmore 1973) because, even though they are located south of the Tropic of Cancer, the annual mean temperature of at sea level is lower than 24ºC (Wiley and Vilella 1998). The surrounding seas have a cooling effect on the islands. The average rainfall ranges from 750 mm in the coastal areas and up to1,400 mm in the higher elevations (Corven 2008; Wiley and Vilella 1998). St. Thomas and St. John receive between 30 to 60 inches of rain annually, while St. Croix receives between 20 and 50 depending on the location. Western areas of St. Croix tend to receive more rainfall than the eastern parts of the island (NRCS 2000). Two lifezones (sensu Holdridge 1967) have been described for USVI (Fig. 2) including subtropical dry, and subtropical moist forest zones (Ewel and Whitmore 1973). The region is dominated by the easterly trade-winds (NRCS 2000) which combined with the regions hilly topography creates an orographic effect. The later consists of moist air being lifted over mountainous terrains promoting the formation of periodical rainfall, probably the major cause of precipitation in this region. Rainfall is heaviest from August to December and generally along the northern and central portions of the islands. The eastern parts of the islands receive less rainfall (VIWRRI 2006). In terms of temperatures, there is less than 10 degrees F difference between the mean temperatures of the coolest and warmest months. The highest temperatures are in August or September, and the lowest are in January or February. In addition, the climate is characterized by high evaporation rates due to continuous wind currents and the warm temperatures (NRCS 2000).
The landcover of the USVI had a similar history as the Puerto Rican landscape. During the 17 and 18 centuries the colonization process by European settlers, principally the
15
Danes, brought about an era of agricultural plantations. Actually, one of the main drivers of landcover change was the plantation of sugar cane (Dookhan 1974). But other crops were also important, such as cotton, tobacco and coffee (Acevedo-Rodriguez 1996). In addition, trees were harvested for building materials. Acevedo- Rodriguez (1996) tells us that during the second half of the 18 century sugar became the most important crop in St. John and at certain point in time 60 percent of the island was under sugar cultivation.
Later on, during the early decades of the 19 century, the sugar economy of the region collapsed and other economic activities began or grew in importance such as wood cutting for charcoal production (Acevedo-Rodriguez 1996). During the late 20th century around 16 percent of the total land area of St. John was used for agricultural purposes, most of which represented pastures and grazing livestock (NRCS 2000). Eventually, the agricultural era gave way to the industrialized and urbanization era.
St. John, the smallest island with 50 km2, is characterized by rugged topography, thin soils, and limited water resources. Over 80 per cent of the island has slopes greater than 30 percent (Anderson, 1994). Of the three islands, it is the one with the least amount of flat terrain or plains with only 5.6%, and the most abundant steep slopes (Figure 4). In addition, it is the least disturbed of the USVI, with a total of 4.4% built-up area (Figure 7), and is recognized for the largest protected area, the Virgin Islands National Park established in 1956. The island, however, is increasingly at risk from urban development issues, such as the erosion caused by road construction (MacDonald et al. 1997). It’s highest pick, Bordeaux Mountain, has an elevation of 387 m. St. Thomas is also a very hilly island with limited plains or flat areas, and has a maximum elevation of 477 m (Corven 2008). Steep slopes in St. Thomas account for 28 percent of the land (Figure 3). St. Croix is the largest island, 22 kilometers across, with a total area of 53,480 acres (NRCS 2000).
Figure 1. St. Thomas and St. John and surrounding cays. Refer to Table 1 for island names.
16
Figure 2. St. Croix and surrounding cays.
Table 1. Islands, cays and rocks names of the U.S. Virgin Islands.
Island # Island # Island # Island # Bovoni Cay 1 Inner Brass Island 18 Sula Cay 35 Mingo Cay 52 Buck Island (STT) 2 Kalkun Cay 19 Thatch Cay 36 Perkins Cay 53 Capella Island 3 Little Flat 20 Turtledove Cay 37 Ramgoat Cay 54 Cas Cay 4 Little Hans Lollik Island 21 Water Island 38 Rata Cay 55 Cockroach Cay 5 Little St James Island 22 Welk Rocks 39 St. John 56 Cricket Rock 6 Little St Thomas Island 23 West Cay 40 Steven Cay 57 Current Rock 7 Lizard Rocks 24 Blunder Rock 41 Trunk Cay 58 Dog Island 8 Outer Brass Island 25 Booby Rock 42 Two Brothers 59 Dog Rocks 9 Patricia Cay 26 Carval Rock 43 Waterlemon 60 Dutchcap Cay 10 Pelican Cay 27 Cinnamon Cay 44 Whistling Cay 61 Fish Cay 11 Rotto Cay 28 Cocoloba Cay 45 Buck Island (STC) 62 Flat Cay 12 Saba Island 29 Congo Cay 46 Green Cay 63 Frenchcap Cay 13 Sail Rock 30 Flanagan Island 47 Protestant Cay 64 Great St. James 14 Salt Cay 31 Grass Cay 48 Ruth Cay 65 Green Cay (STT) 15 Savana Island 32 Henley Cay 49 St Croix 66 Hans Lollik Island 16 Shark Island 33 Leduck Island 50 Hassel Island 17 St. Thomas 34 Lovango Cay 51
U.S. Virgin Islands landscape The landscape of the U.S. Virgin Islands is made up of a variety of ecosystems including forests, woodlands, shrublands, grasslands, wetlands, rocky shores, sandy beaches and urban environments (figure X landcover). This habitats harbor many species of amphibians, reptiles, birds and mammals (Rice et al. 2002; Schwartz, A. and W. R. Henderson 1991; ). A number of studies have come up with slightly different results of
17
the forest cover in the U.S. Virgin Islands. Nowadays it represents more than 50 percent of the islands (Daley 2010; Brandeis and Oswalt 2007). Our calculations from the U.S. Virgin Islands Gap Landcover show a 67.2 percent cover including forests, woodlands, and shrublands (Table 4) This appears to be a positive outcome for forest habitats given the fact that by the beginning of the 20th century about 90% of the forested lands had been cleared for agriculture, wood production or other uses. In1976, Somberg (1976) estimated the forest cover of the three major islands to be 45.1 percent or 35,000 acres.
Figure 3. Physiography of the St. Thomas (Gould et al. 2007b).
Figure 4. Physiography of the St. John (Gould et al. 2007b).
18
Figure 5. Physiography of the St. Croix (Gould et al. 2007b).
Nevertheless, the majority of forested lands in the USVI are second-growth as a result of its land use history (Wiley and Vilella 1998; Daley 2010; Brandeis and Oswalt 2004). Today’s forests in the USVI are a biologically diverse mix of native and exotic species. However, as Daley (2010) points out for St. Croix, younger forest stands, not older than 50 years, are primarily dominated by the exotic leguminous tree Leucaena leucocephala, the most common tree in the USVI. This prospect is not likely to change in many decades to come as it appears that secondary forest succession in the USVI is quite a slow process that may even become stalled in highly impacted areas (Daley 2010). From his study of the forest cover in St. Croix, Daley (2010) found a 56% cover in both 1992 and 2002, while Brandeis and Oswalt (2007) found an overall 60% cover for the USVI in 2004. Daley tells us that during his 10 year period of study, forest cover remained stable through a 1,500 ha gain and loss in some areas: some abandoned grasslands or cultivated pasture land, gave way to early stage forests while some secondary forest was lost to urban development and other uses. There is also a trend towards forest fragmentation and urban development. In addition, grasslands decreased substantially, partly because some patches were converted to urban development.
St. John is not only the island with the highest forest cover (followed by St. Thomas and St. Croix respectively) but in addition, it’s the one that has the most mature forest cover, some 20% (Brandeis and Oswalt 2004; Daley 2010). The forest recovery in St. John has been attributed to its mountainous rugged terrain, an earlier abandonment of plantation agriculture and the establishment of the Virgin Islands National Park in 1956 (Platenberg et al. 2005). The other two islands lost more than 10 percent of their forest cover over a 10 year period starting in 1994 (Brandeis and Oswalt 2004). Differences in land use
19 through time have created different forest landscapes in each of the major islands (Chakroff 2010). While St. Thomas natural habitats are having more pressure from urban development and the tourism industry in recent years, St. Croix has also had more pressure from agricultural activities such as cattle grazing.
Figure 6. The lifezones of the U.S. Virgin Islands (modified from Ewel and Whitmore 1973).
20
Other habitats in the U.S. Virgin Islands have also been impacted through time. According to studies by the Virgin Islands Conservation Data Center in St. Croix, coastal development has increased at a fast pace over the last 40 to 50 years and this has led to the degradation of half the mangrove cover and associated saline wetlands (McNair et al. 2006). The construction of an industrial complex and oil refinery brought about the destruction of one of the most important wetlands in St. Croix, Krause Lagoon, in the early 60’s (McNair et al. 2006; Platenberg et al. 2005). About 11.6 percent of the land in the USVI is developed or built-up (Figure 7).
Wildlife
The Caribbean is considered a global biodiversity hotspot (Myers et al. 2000) and the U.S. Virgin Islands fosters a great number of native vertebrate species. Besides the abundant number of resident species, the islands are also important stop-overs for a great number of migratory species including birds and reptiles. Many warblers, for example, visit the islands for part of the year, and sea turtles, such as the leatherback, nest on their sandy beaches (Corven 2008; Platenberg et al. 2005)
More than 210 birds have been recorded from USVI, must of which are Nearctic or Intratropical migrants and some 30 species are vagrant (Corven 2008). According to Corven (2008) some 60 species of birds are breeding residents, although other sources indicate they are as little as 17 (Platenberg et al. 2005). Birdlife International identified nine Important Bird Areas in the USVI (Corven 2008). The only native mammals in the USVI are bats, of which there are six species, and none of which is endemic. In contrast, there are 10 species of introduced mammals, some of which are domestic animals that have become feral. These include the domestic dog and cat, the small Indian mongoose, burros, pigs, the White-tail Deer, goat, roof rats, Norway rat and the house mouse. Cays are of particular importance to the native reptile species, some of which have been extirpated from the main islands (Heatwole et al. 1981; McNair 2003; Wiley and Vilella 1998). Such is the case of the St. Croix’s ground lizard which now inhabits three cays, off of St. Croix.
Conservation issues and threats to biodiversity
A number of vertebrate species and plants of the USVI are listed as threatened or endangered under the U.S. Endangered Species Act list. There are six reptiles, including four marine sea turtles, the St. Croix Ground Lizard, and the Virgin Islands Tree Boa; one amphibian, the Puerto Rican Crested Toad that has apparently been extirpated; and four birds, including the Brown Pelican, the Piping Plover, the Roseate Tern, and finally the White-necked Crow that appears to have been extirpated from the USVI (Platenberg et al. 2005).
Some of the conservation issues or threats of greatest concern in the USVI include urban development, introduced or exotic species, and climate change (Platenberg et al. 2005).
21
Urban development and land use change from natural habitats to housing, hotels, paved or unpaved roads, golf courses, or other uses, represent a current threat to the USVI habitats and wildlife (Chakroff 2010; Platenberg et al. 2005). Constructions not only eliminate natural cover and create impervious surfaces but also pose the problem of erosion and sediment runoff that causes a detrimental effect on soils and coastal and marine habitats.
Figure 7. Developed or built up areas of the U.S. Virgin Islands. The red areas represent low, medium and high density urban development as calculated from the U.S. Virgin Islands Gap Landcover shown in Figure X
22
In St. John, studies from the 1990’s showed an increase in sediment delivery to streams while at the same time an expansion of the unpaved road network was noticed. Rats, the mongoose, white-tailed deer, feral donkeys, hogs, cats and dogs have been associated with negative impacts towards native vertebrates and also the native vegetation (Platenberg et al. 2005; Wiley and Vilella 1998). Amphibians such as the Cuban Tree Frog and the Cane Toad have been known to have a deleterious effect on the fauna of other islands and are believed to be having a detrimental impact on the USVI native wildlife (Platenberg et al. 2005). Feral cats and dogs as well as the mongoose are having a considerable negative impact on the native fauna and have been implicated in the extirpation and/or lower numbers of native species populations.
References
Acevedo-Rodriguez P. 1996. Flora of St. John: U.S. Virgin Islands. Memoirs of the New York Botanical Garden. (78) 581 pp.
Anderson DM. 1994. Analysis and modeling of erosion hazards and sediment delivery on St. John, U.S. Virgin Islands. U.S. National Park Service Water Resources Division, Technical Report NPS/NRWRD/NRTR-94/34. Fort Collins, Colorado.
Brandeis, T.J. and Oswalt, S.N. 2004. The status of Virgin Islands forests. Forest Service U.S. Department of Agriculture Southern Research Station Resource Bulletin SRS-122.
Corven J. 2008. U.S. Virgin Islands. In: Important Bird Areas in the Caribbean. Wege D.C. and Anadon-Irrizarry V. Editors. Birdlife International. BirdLife Conservation Series (15) 348 pp.
Dookhan, I. 1974. A History of the Virgin Islands of the United States. Second Edition. Canoe Press. 321 pp.
Ewel, J. J., and J. L. Whitmore. 1973. The ecological life zones of Puerto Rico and the U.S. Virgin Islands. USDA Forest Service, Res. Pap. ITF-18. Río Piedras, Puerto Rico.
Gannon, M.R., A. Kurta, A. Rodríguez-Durán, and M.R. Willig. 2005. Bats of Puerto Rico – An Island Focus and a Caribbean Perspective. Texas Tech University Press. 239 pp.
Heatwole H., Levons R., and Byer M.D. 1981. Biogeography of the Puerto Rican Bank. Atoll Research Bulletin. No. 251.
Holdridge, L.R . 1967. Life zone ecology . Tropical Science Center. San Jose, Costa Rica.
23
Kirkpatrick, J.B. 1983. An iterative method for establishing priorities for the selection of nature reserves: An example from Tasmania. Biological Conservation 25:127- 134.
L.H. MacDonald, D.M. Anderson, and W.E. Dietrich. 1997. Paradise threatened: land use and erosion on St. John, U.S. Virgin Islands Environmental Management, 21 (6) (1997), pp. 851–863
McNair, D. 2003. Population estimate, habitat associations and conservation of the St. Croix Ground Lizard Ameiva polops at Protestant Cay, United States Virgin Islands. Caribbean Journal of Science 39(1): 94-99
McNair, D. B., L. D. Yntema, and C. Cramer-Burke. 2006. Use of waterbird abundance for saline wetland site prioritization on St. Croix, United States Virgin Islands. Caribbean Journal of Science 42:220-230.
Mac, M.J., Opler, P.A., Puckett Haecker, C.E., and Doran, P.D. 1998. Status and trends of the nation's biological resources. 2 vols. U.S. Department of the Interior, U.S. Geological Survey, Reston, Va.
Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B. & Kent, J. 2000. Biodiversity hotspots for conservation priorities. Nature. 203: 853–858.
Platenberg, R.J., F. E. Hayes, D. B. McNair and J. J. Pierce. 2005. A comprehensive wildlife conservation strategy for U.S Virgin Islands. Division of Fish and Wildlife, St. Thomas. 251pp.
Rankin, D.W. 2002. Geology of St. John, U.S. Virgin Islands. U.S. Government Printing Office, Washington. U.S. Geological Survey Professional Paper 1631
Rice, K.G., Waddle, J.H., Corckett, M.E., Carthy, R.R., and Percival, H.F. 2005. Herpetofaunal Inventories of the National Parks of South Florida and the Caribbean: Volumen II. Virgin Islands National Park, U.S. Geological Service Report Series 2005–1301, 45 pp.
Rivera, L.H., Frederick, W.D., Farris, C. Jensen, E.H., Davis, L., Palmer, C.D., Jackson, L.F. and McKinzie, W.E. 1970. Soil survey of the Virgin Islands of the United States. Washington, DC: U.S. Department of Agriculture, Soil Conservation Service, 123 pp.
Schwartz, A. and W. R. Henderson 1991). Amphibians and Reptiles of the West Indies. Descriptions, Distributions, and Natural History, Univ. Press Florida, Gainesville, FL. pp. 720
U.S. Department of Agriculture, Natural Resources Conservation Service [NRCS]. 2000. Soil Survey of the United States Virgin Islands. Edited by John R. Davis in
24
cooperation with the Virgin Islands Department of Planning and Natural Resources; the Virgin Islands Cooperative Extension Service; and the United States Department of Interior, National Park Service
University of the Virgin Islands Water Resources Research Institute [VIWRRI]. 2006. Evaluating the sediment retention function of salt pond systems in the U.S. Virgin Islands. Prepared by Denise S. Rennis, Colin M. Finney and Barry E. Devine. Partially funded by the United States Department of the Interior U. S. Geological Survey.
Wiley, J.W., and Vilella, F.J. 1998. Caribbean Islands. in Mac, M J, Opler, P A, Haecker, C E P & Doran., P D eds Status and Trends of the Nation’s Biological Resources. USGS, Reston.
25
U.S. Virgin Islands GAP Analysis Project Land Cover Mapping
Principle Investigator - William A Gould
Remote Sensing Analysts and Cartographic Technicians - Gary S Potts, Maya Quiñones, Mariano Solórzano
International Institute of Tropical Forestry Jardín Botánico Sur1201 Calle Ceiba Río Piedras PR 00926-1119
26
1. Introduction
A part of the U.S. Virgin Islands GAP analysis project is to create a new recent land cover map within the mapping criteria set by the GAP project.
The U.S. Virgin Islands (18˚ 20’N, 64 ˚50’W) are situated in the eastern extreme of the Greater Antilles and include St. Croix, St. Thomas, St. John and Water Island and are surrounded by numerous cays and rocks. The climate is considered sub-tropical with almost constant easterly trade winds, with the rainy season extending from May to November. The islands exhibit a rugged topography with a maximum elevation of 474m (1556 feet) at Crown Mountain on St. Thomas.
The purpose of creating a land cover map of the U.S. Virgin Islands through remote sensing imagery is to provide habitat information that can be used to model vertebrate species distributions as part of the combined Puerto Rico and the U.S. Virgin Islands GAP analysis projects.
2. Initial Project Assessment - Remote Sensing A. Landsat 7 ETM+ failure and limitations for land cover mapping
Traditionally, GAP analysis projects across the U.S. have relied extensively on satellite imagery from Landsat 5 Thematic Mapper (TM) and Landsat 7 Enhanced Thematic Mapper Plus (ETM+) to provide the spatial and spectral information to derive land cover habitat maps at 30m spatial resolution.
Current Landsat 7 ETM+ imagery and scene acquisition is limited by both the scan line correction (SLC) error (fig. 1) since May 31st 2003 - horizontal lines with no data across the majority of a scene except for a small central area - and the Long Term Acquisition Plan (LTAP), the use of a set of criteria that includes cloud-cover forecasts to guide Landsat image collections (Landsat Project Science Office 1998). Although Landsat 5 TM is still operational and providing global imagery it is limited by the sensor’s age (Landsat 5 launched in 1984) and inadequate scene collections around Puerto Rico and the U.S. Virgin Islands. An imagery search returned no scene acquisitions for path 4 row
27
47 (covering St. Thomas and St. John) and path 4 row 48 (covering St. Croix) since 1992 (USGS Global Visualization Viewer 2011; USGS EarthExplorer 2011).
These limitations make the collection of new images and the use of existing images for tropical humid regions with a high potential for cloud cover difficult. Therefore, land cover mapping of the U.S. Virgin Islands with Landsat 7 ETM+ or Landsat 5 TM images was not a practical option.
Figure 1. 30m true color (bands 3,2,1) Landsat 7 ETM + SLC error image of St. John 09/23/2007
An Initial assessment of available satellite imagery from a number of sensors was undertaken with consideration of their temporal, spectral, and spatial resolution. The sensors evaluated included: IKONOS, QuickBird, Système Probatoire d'Observation de
28
la Terre 5 (SPOT 5), the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), and the Moderate Resolution Imaging Spectrometer (MODIS) onboard Terra and Aqua, in addition to hyperspectral imagery such as EO-1 Hyperion. Primary consideration of the GAP analysis habitat mapping requirements, the technical and mapping capabilities of each satellite, in addition to cost, present mapping schedules, sensor degradation and performance issues e.g. ASTER’s five short wave infrared 30m bands have been unusable since 2008 due to saturation and stripping anomalies (NASA Jet Propulsion Laboratory 2008), pre-processing requirements, and the availability of recent archived imagery all influenced the final decision for the mapping structure of the U.S. Virgin Islands GAP.
Additional challenges for the U.S. Virgin Islands Gap land cover mapping is the persistent cloud and associated cloud shadows within satellite imagery that is a significant factor for a project based in a humid sub-tropical region. Cloud and cloud shadow significantly reduce the amount of available imagery for land cover and land use mapping in tropical regions and is a consistent issue (Martinuzzi et al. 2007; Helmer and Ruefenacht 2007).
A. Remote Sensing Platform – EO-1 ALI
For the U.S. Virgin Islands GAP, alternative imagery was used from the Advanced Land Imager (ALI) sensor onboard the Earth Observing-1 (EO-1) satellite. EO-1 was originally launched by the NASA in 2000 as a one year technical mission to validate technologies for the acquisition of multispectral images of the earth consistent with the requirements of the Landsat program (Hearn 2002) with the ALI and Hyperion sensors onboard. However, the mission was originally extended for one year due to remote sensing research interest and use of data in the scientific community. As of September 2011, EO- 1 ALI was still operational and collecting data as an extended mission.
EO-1 ALI holds a number of advantages over Landsat 7 ETM+ when considering its mapping capabilities, but it is subject to a few limitations and disadvantages due to the nature of the EO-1 program as a technical validation mission.
29
ALI offers improved spectral resolution over the Landsat 7 ETM+ (table 1) with an additional 3 bands, increasing the number of bands to 9 covering blue to short wave infrared (SWIR) wavelengths compared to the 6 bands of the ETM+. Thenkabail et al. (2004) found the ALI sensor and the use of the 9 multispectral bands improved land cover accuracies and sensitivity for nine complex African forest land cover classes in comparison with IKONOS and ETM+.
ALI ETM+ Band Wavelength (µm) Spatial Band Wavelength Spatial (alternative Resolution (m) (µm) Resolution name) (m) 1 (Pan) 0.48-0.69 10 Pan 0.52-0.9 15 2 (MS-1 1p) 0.433-0.453 30 3 (MS-1) 0.45-0.515 30 1 0.45-0.515 30 4 (MS-2) 0.525-0.605 30 2 0.525-0.605 30 5 (MS-3) 0.63-0.69 30 3 0.63-0.69 30 6 (MS-4) 0.775-0.805 30 4 0.75-0.9 30 7 (MS-5 – 4p) 0.845-0.89 30 8 (MS-6 – 5p) 1.2-1.3 30 9 (MS – 5) 1.55-1.75 30 5 1.55-1.75 30 No Thermal 6 60 10 (MS – 7) 2.08-2 35 30 7 2.09-2.35 30 Table 1. ALI v ETM+ Band characteristics and comparison.
The additional SWIR band 5p, within the range 1.2 to 1.3 µm, provides additional information useful for agricultural, forestry and vegetation applications (Bryant et al. 2003; Chander et al. 2004). The ETM+ band 4 – near infrared (NIR) - was split into two bands on the ALI to avoid a strong water absorption feature between 0.805 – 0.845µm that occurs in the middle of Landsat 7 ETM+ band 4 (Elmore and Mustard 2003). Therefore, ALI bands 4 and 4p are less sensitive to variations in atmospheric water and can be considered a significant advantage when mapping a humid tropical region where corrections are necessary for the effects of absorption and emission from water vapor which interferes and influences emitted or reflected surface radiances.
30
Another important difference between the sensors is the radiometric resolution. ETM+ is an 8-bit sensor while the ALI sensor exhibits a finer radiometric resolution at 16-bit, thus it is more sensitive to detecting small differences in reflected and emitted energy. Donegan and Flynn (2004) found that the ALI instrument is less susceptible than ETM+ to saturation within the SWIR. This is valuable when mapping a complex heterogeneous landscape such as the U.S. Virgin Islands where a number of surface features will exhibit high reflectance values including urban, barren, exposed soil, coastal rocks, pastures, agricultural and abandoned agriculture lands and coastal scrub with sparse vegetation, all features that can exceed the dynamic range and might saturate ETM+. In addition, the signal-to-noise ratio (SNR) is between four and ten times larger than the ETM+ in the bands common to both sensors (Chander et al. 2004).
Although the multispectral bands of the ALI and ETM+ both provide imagery at 30m spatial resolution, the panchromatic bands on each sensor are different. The spatial resolution of The ETM+ panchromatic band is 15m while the ALI panchromatic band provides an improved spatial resolution of 10m. The Puerto Rico GAP analysis project (Gould et al. 2008) incorporated the Landsat 7 ETM+ 15m panchromatic band to enhance the spatial resolution of the 30m visible and infrared bands in order to improve the delineation of habitats at the sub-pixel level in complex tropical landscapes. For the U.S. Virgin Islands GAP the ALI panchromatic band offered the same opportunity but at an improved 10m spatial resolution using panchromatic sharpening techniques.
The ALI panchromatic band attenuates beyond red in order to allow more accurate identification of vegetation than is possible with the ETM+ panchromatic band, which extends into the NIR (Chander et al. 2004). This modification of the panchromatic also enhances the contrast between vegetation and non-vegetation regions (Lencioni et al. 2005). An advantage of the reduced band width of the 10m ALI panchromatic band when pan-sharpening imagery to improve the spatial resolution of the 30m multispectral band is related to surface adjacency effects, complicated multiple scattering in the atmosphere- land surface system (Liang et al. 2001). These adjacency effects, particularly in vegetated regions increase with wavelength, with a major influence in the near and mid-infrared
31
region (Cattrall and Thome 2002), however this is reduced with ALI as these effects are minimal in the visible but will be incorporated when pan-sharpening ETM+ images.
The main limitations of ALI are related to image collections and the sensor’s swath. The ETM+ has a ground swath of 185km while ALI has a ground swath of just 37km. While the ETM+ can image St. Thomas and St. John within the same scene on path 4 row 47, ALI needs two separate image collections to cover the same area on different dates. However, in order to increase the temporal resolution from the 16 day revisit time, ALI has off-NADIR pointing capabilities for image acquisitions. Another disadvantage of the EO-1 ALI sensor is that no significant archive of imagery covering the U.S. Virgin Islands was readily available – only a few images compared to hundreds that the Landsat series can offer. Additionally, all data collections for EO-1 ALI have to be scheduled in advance through a Data Acquisition Request (DAR), with a potential of three images collected per DAR based upon a 20% cloud cover threshold criteria. These collections have the potential to conflict with other ALI or Hyperion DAR requests within the region and may be subjected to a cloud cover assessment.
The lack of a thermal band on the ALI sensor is a notable exclusion when compared to the Thematic Mapper (TM) and ETM+ sensors when mapping in the humid tropical regions. A number of cloud and cloud shadow masking techniques and algorithms rely extensively on the thermal band of the ETM+ or TM to create accurate and efficient cloud and cloud shadow masks. Martinuzzi et al. (2007) highlights the importance of the thermal band in Landsat TM and ETM+ sensors for the creation of cloud and cloud shadow free images in tropical and other landscapes.
B. U.S. Virgin Islands Data Acquisition
At the end of 2006, three Data Acquisition Requests (DARs) were requested with the USGS, one for St. Thomas, one for St. John and one for St. Croix. Attempts were made to improve the amount of available usable imagery by trying to overlap and include other islands in the primary collection scene with the strategic placing of each scenes central point. Initially, the aim was to collect imagery during the dry season – January to March/April – in order to improve the chances of cloud and cloud shadow free imagery. However, due to scheduling conflicts the initial image acquisitions started in March 2007 and the 3 DARs were fulfilled in September 2007.
32
Image ID Acquisition Season Island Look Cloud Cover Date Angle (%) EO1A0040472007081110KZ 03/22/2007 Dry St. -14.005 70 Thomas EO1A0040472007089110KY 03/30/2007 Dry St. Croix 20.991 90 EO1A0040472007109110PA 04/19/2007 Wet 4.3332 50 EO1A0040472007114110PZ 04/24/2007 Wet St. John -1.1307 70 EO1A0040472007119110KZ 04/29/2007 Wet St. -7.5514 40 Thomas EO1A0040472007155110KW 06/04/2007 Dry/Wet 20.488 40 EO1A0040472007165110PA 06/14/2007 Dry/Wet 6.0103 40 EO1A0040472007175110KV 06/24/2007 Dry/Wet -11.166 40 EO1A0040472007206110KW 07/25/2007 Wet 20.721 50 EO1A0040472007226110KZ 08/14/2007 Wet -12.629 40 EO1A0040472007239110PB 08/27/2007 Wet 7.7806 80 EO1A0040472007257110KY 09/14/2007 Wet 21.886 30 Table 2. List of EO-1 ALI scenes for the U.S. Virgin Islands collected in 2007 from the DARs
3. U.S. Virgin Islands GAP Land Cover Mapping Process
A. U.S. Virgin Islands site visits and field reconnaissance.
In 2007 a number of visits were made to the U.S Virgin Islands, travelling to St. Thomas and St. John in August 2007 and St. Croix in November 2007. The purpose was to meet with local scientists and experts for information and collaboration for the U.S. Virgin
33
Islands GAP project. The primary objective for the land cover component was to visit a range of vegetation habitats, understand the vegetation differences between the three main islands, and observe the forest types and their structure. Detailed photography, land cover notes and ground control points (GCPs) taken with a survey grade GPS receiver on a wide range of vegetation types and built up areas were collected for reference purposes.
Figure 2. Flooded salt flat near Coral Bay with fringing Mangrove - St. John
34
Figure 3. Inner and Outer Brass Islands – St. Thomas
B. Ancillary Imagery
At the start of the U.S. Virgin Islands GAP analysis project, digital aerial orthophotos for the U.S. Virgin Islands were available from 1999 and 2004. The 1999 aerial photos were scanned at a resolution of 500 dots per inch (DPI) yielding 2.4 by 2.4 meter pixels for the 1:48000 scale photography (NOAA 1999). The 2004 aerial photos are provided with a resolution of 1 by 1 meter in true color and with 8-bit radiometric resolution (United States Army Corps of Engineers 2004). Later, during the project digital aerial orthophotos with 0.3 by 0.3 meter pixel from 2006/2007 (United States Army Corps of Engineers 2007) became available.
Although the aerial photos provide high resolution imagery for image interpretation and classification techniques, none of the available data sets provided complete coverage of the U.S. Virgin Islands, with aerial photo coverage from cays or rocky outcrops missing
35
(e.g. Frenchcap cay, Sail rock, Cockroach cay, and cricket rock). Also, some areas were obscured due to cloud and cloud shadows.
Additionally, due to the high percentage of cloud and cloud shadows identified within the initial available ALI imagery during the completion of the DARs, a request was issued to the USGS Landsat project team for an increase in Landsat 7 ETM+ scenes for path 4 row 47 and path 4 row 48 from 2007.
Other data that became available to the project was discrete LiDAR data collected in January and February 2004 by 3001 Inc. and the U.S. Army Corps of Engineers, using a Leica Geosystems ALS 40 sensor (3001 Inc. 2004). The LiDAR data became available in 2008 with raw point cloud data provided by the contractor in .xyz format with a point spacing of 2.76 meters and included multiple return data, primarily first and last returns with some intermediate returns. The LiDAR data extent covered the main islands of St. Thomas, Water Island, and St. John and some of the surrounding cays. For St. Croix the LiDAR data was limited to just the east and west ends of the Island, the east end included data for Buck Island.
4. Data Preparation and Processing
All data preparation and processing of satellite imagery and LiDAR was completed at the I.I.T.F. GIS and Remote Sensing Laboratory using four main image processing software: ERDAS imagine 9.3 and IDRISI Taiga for the satellite imagery and ArcGIS 9.3 and FU.S.ION LDV 2.70 processing software (McGaughey 2009) for processing the LiDAR data. Image imports, reprojection, image geo-registration, radiometric normalizations, atmospheric corrections, data analysis and organization, image classification and post processing were all completed with the listed software.
A. ALI Date Structure
36
EO-1 ALI imagery is provided as a level 1Gst product with scaled 16-bit radiance values in a hierarchical data format (HDF) with the bands written as band sequential (BSQ). The level 1Gst is terrain corrected through the use of a DEM product from the Shuttle Radar Topography Mission (SRTM). This corrects parallax error due to topographic relief and improves the overall band-to-band registration (USGS 2010). The ALI product are distributed with a Universal Transverse Mercator (UTM) map projection and WGS84 horizontal datum, with units in meters.
B. Image Import and Radiometric Calibration
Each ALI scene was imported using IDRISI Taiga, and the distributed projection information was added to each band. The digital values in each band in the Level 1G ALI data products are provided in scaled at-sensor radiance values that represent 16-bit absolute radiances, with the data stored as 16-bit signed integers. The units represent mW/cm2 SR μm (USGS 2010). ALI images were converted to obtain the original at satellite absolute radiance values using Eq. (1)
L = (Digital Number * scaling factor) + offset (1)
L= radiance (mW/cm2 SR μm)
ALI Band Spectral Range Scaling Factor Offset (µm) 1 0.048-0.69 0.024 -2.2 2 0.433-0.453 0.045 -3.4 3 0.45-0.515 0.043 -4.4 4 0.525-0.605 0.028 -1.9 5 0.63-0.69 0.018 -1.3 6 0.775-0.805 0.011 -0.85
37
7 0.845-0.89 0.0091 -0.65 8 1.2-1.3 0.0083 -1.3 9 1.55-1.75 0.0028 -0.6 10 2.08-2.35 0.00091 -0.21 Table 3. ALI scaling and offset values for radiance rescaling
C. Atmospheric Correction The atmospheric correction of the ALI imagery, with each band individually corrected from a top of atmosphere (ToA) radiance to a surface reflectance value, was completed within the IDRISI Taiga software using the inbuilt ATMOSC module. This correction accounts for scattering and absorption as well as reducing the influence of haze and thin clouds within the scene. Atmospheric effects include molecular and aerosol scattering and absorption by gases, such as water vapor, ozone, oxygen and aerosols (Liang et al. 2001). The ATMOSC full radiative transfer model was used.
Inputs into the ATMOSC correction included the date and time of the image collection, the wavelength of the band center, the ALI offset and scaling factors, the satellite viewing angle, the sun elevation angle, spectral solar irradiance (calculated from the time of the image collection and the center wavelength of the input band), an estimation of the optical thickness of the atmosphere, a spectral diffuse sky irradiance input can also be selected for each band but due to the complexity in extracting the value a default value of 0 was selected.
A Dark object subtraction value (DOS), also known for as the DN haze value was calculated for each band. The DOS method assumes that within a satellite image there exist features that have near-zero percent reflectance (i.e. water, dense forest, shadow) (Schroeder et al. 2006). The signal recorded by the sensor from those features is solely a result of atmospheric scattering (path radiance), which must be removed (Chavez 1996). With high amounts of cloud and cloud shadow within the ALI imagery, there were plenty of cloud shadow regions over the deep ocean or forested areas in topographic shadow for the calculation of each value. D. Reprojection.
38
The U.S. Virgin Islands GAP analysis project is an extension of the PRGAP analysis project (Gould et al. 2008). All geospatial data managed within the project is projected to State Plane (Lambert Comformal Conic), North American Datum 1983 (NAD 83), meters. After initial import of the normalized ALI scenes and layer stacking of the multispectral bands into ERDAS Imagine 9.3, the multispectral bands (bands 02 – 10) and the panchromatic band were reprojected to State Plane, NAD 83 using a rigorous transformation with nearest neighbor resampling (NN).
For all data processing that requires resampling, i.e. reprojection, panchromatic sharpening and geo-registration, the resampling was completed using the NN method. The NN preserves the spectral integrity of the data but may introduce some spatial discontinuities, compared to the cubic convolution (CC) resampling technique that maintains the spatial detail but distorts the spectral values (Vijayaraj 2004).
E. Sea Mask
Using the 2004 digital aerial orthophotos and photogrammetry interpretation techniques a coastline was manually digitized and stored as a vector file. For areas with cloud cover along the coast, the 1999 aerial photos were used. This resulted in a detailed high resolution visually interpreted wet and dry coastline. To mask the sea within each ALI scene the coastline was buffered by 20m in order not to mask out any coastal features (e.g. rocky outcrops, sand) that might not perfectly match the coastline file as well as to account for the spatial resolution of the ALI images.
F. Panchromatic Sharpening
Each ALI scene was then pan-sharpened, where the 10 meter panchromatic band is fused with the 30 meter multispectral bands in order to improve their spatial resolution by adding the spatial detail within the panchromatic band while attempting to maintain the
39 spectral integrity and information of the multispectral bands. Principle Component Analysis (PCA), Image Hue Saturation (IHS), Multiplicative and Brovey Transform panchromatic sharpening techniques were evaluated in ERDAS Imagine. Multiplicative and Bovey Transform techniques maintain the spatial detail of the panchromatic band and are useful techniques for urban studies and applications, where features have sharp boundaries, while IHS techniques are suited for visual interpretation (ERDAS 2010). Van Tu (2005) identifies the use of the PCA method for areas with fuzzy boundaries, such as in vegetation classifications, as it preserves the original scene’s radiometry of the data in the output bands.
G. Geo-registration
The 2004 digital aerial orthophotos of the U.S. Virgin Island in State Plane, NAD 83 served as the main base map for registration of images and products. Each ALI scene was geo-registered using the State Plane projection, NAD83 datum and meters (linear unit), using a first order polynomial wrapping equation which included a selection of known ground control points (GCPs) within the ALI imagery and the 2004 aerial photo reference imagery and include a 30m digital elevation model (DEM) from the USGS. The image was resampled using a nearest neighbor transformation and any GCPs with a high root mean square error (RMSE) were removed so that only GCPs within 1m RMSE were applied in the resampling.
F. Cloud and Cloud Shadow Masks
The next processing challenge was to create a cloud and cloud shadow free image from each scene prior to image classification. The difficulty in mapping cloud and cloud shadows in a subtropical region is related to the heterogeneity and intricacy of the landscape. Forest types (e.g. moist forests, mangroves, and forested regions) in topographic shadow can exhibit spectral similarities to cloud shadows, while surface targets of high reflectance (e.g. urban, barren and sand) have spectral signatures similar to cloud.
Before image classification, the removal of clouds and cloud shadow is important to decrease misclassification and reduce their influence on the spectral signatures of ground
40
targets. Fundamentally, the cloud and cloud shadow regions are areas of no data and need to be removed and subsequently replaced with cloud and cloud shadow free data.
Many cloud and cloud shadow detection methods are typically based on image contrast, in the visible bands and the thermal bands, between the high reflectance and cold cloud cover and the lower reflectance and warmer surface features. Due to the lack of a thermal band on the ALI sensor that would provide useful temperature information to aid masking cloud cover, alternative cloud masking techniques were needed.
Three masks were created using Principle Component Analysis (PCA) and taking the difference from the first and second principle components on the visible (blue, green and red bands) and the near infrared (NIR) to short wave infrared (SWIR) bands separately. A cloud mask was created using PCA on the visible bands and a cloud shadow mask was created using PCA on the NIR to SWIR bands. The cloud mask captured most of the urban pixels due to spectral similarity with cloud pixels. Using inverse PCA on the visible bands, an urban mask was created that could be subtracted from the cloud mask.
5. Land Cover Image Classification A. Mapping Considerations
Due to some of the mapping limitations previously outlined i.e. scene look angle, date of collection and seasonality, if achievable, the base ALI image for each island was the ALI scene with the least amount of cloud cover and with a limited sensor look angle. Additionally, for areas within the scene where no data is available due to cloud and cloud shadow, the ALI images used to fill in these areas with data were as close to the date of the base image as possible.
B. Image segmentation
For classification of the ALI imagery, a combination of geological information and Holdridge life zones – moist and dry - were used for simple image segmentation to create areas of similar geoclimatic zones in order to reduce spectral variance that can confuse land cover types during the classification process. Image segmentation is a common
41
process in data with high and moderate spatial resolution in heterogeneous landscapes and where spectral confusion in the classification is frequent among many of the land cover classes.
C. Self-Organizing Map (SOM) Artificial Neural Network Classification
Once segmented, each section was imported to IDRISI Taiga and classified using an unsupervised Kohonen’s Self-Organizing Map (SOM) neural network (Kohonen 1990). ANN approaches have a distinct advantage over statistical classification methods in that they are non-parametric and require little or no a priori knowledge of the distribution model of input data (Benediktsson and Sveinsson 1997).
The SOM neural network was initially tested with a number of input parameters set for the 10 nodes of the input layer. Initial learning rates were tested based on Li and Eastman (2006) whom extracted 12 land cover/use classes from a 3 band (green, red, infrared) SPOT HRV (High Resolution Visible) image with a minimum learning rate of 0.5 and a maximum learning rate of 1.0 with 15 x 15 layers in the neuron layer. After initial testing and visual assessment of the unsupervised classification clusters the learning rates and neuron layer value were adjusted. Hugo et al. (2006) outlined classifying high spectral dimensional satellite imagery from MERIS, using a SOM and 13 MERIS bands to map 19 land cover classes, with an initial maximum learning rate set at 0.7 and a 25 x 25 neuron layer. For the ALI imagery, the layers in the neuron layer were expanded from 15 x 15, to 30 x 30 then 60 x 60, with the maximum learning rate at 0.7 and the minimum learning rate set at 0.1. The K-means clustering rule was selected with 200 set as the maximum output of clusters.
The input for the neural network included the nine spectral bands of each ALI image as well as a Soil-Adjusted Total Vegetation index (SATVI) product. The SATVI is calculated using the red (ALI band 5) and two SWIR bands (ALI bands 8 and 9) from the following equation Eq. (2):
9 5 10 SATVI = (1 + ) 9 + 5 + 2 퐵푎푛푑 − 퐵푎푛푑 퐵푎푛푑 퐿 퐵푎푛푑 퐵푎푛푑 퐿
42
L is constant (related to the slop of the soil-line in the feature space plot) that is usually set to 0.5
The SATVI is a combination of the Soil-Adjusted Vegetation Index (SAVI) and the Normalized Difference Senescent Vegetation Index (NDSVI) and is sensitive to green and senescent vegetation (Marsett et al. 2006). This provides information regarding woody material, dry or dormant vegetation that may be under represented due to the lack of seasonal imagery to accurately map vegetation dynamics. The drier exposed regions of the U.S. Virgin Islands, with low tree canopy cover, dry shrub, scrub and grassland may be conservatively mapped, even at 10m spatial resolution. Additionally, the SATVI accounts for reflectance from the soil background which is typical in drier sub-tropical disturbed and fragmented landscapes which can influence the apparent vegetation reflectance signature. Adding the product as an additional band also helps reduce some of the noise created by variation of topographic illumination within a scene as well as additional shadowing caused by the viewing angle of the satellite at the time of the image collection.
D. Classification of the ALI data
For the St. Croix land cover classification map, three EO-1 ALI images were used. The best cloud free image was used as a base image, this being an ALI image from the 14th of June 2007. ALI images from the 24th of June 2007 and the 14th of September 2007 were used to fill in any missing data due to cloud cover and cloud shadow. This provided an 88.57% cloud and cloud shadow free image for St. Croix.
A similar procedure was followed for St. Thomas and St. John. For St. Thomas three EO-1 ALI images from the 29th of April 2007, the 24th of June 2007 and the 14th of August 2007 provided 90.01% cloud and cloud shadow free imagery for classification. For St. John, four EO-1 ALI images from the 14th and 24th of June 2007, 29th of April 2007 and the 14th of August 2007 provided 96.37% cloud and cloud shadow free imagery for classification.
E. Use of Landsat 7 ETM+ and ASTER imagery for areas of no data
43
Unfortunately, the ALI DARs did not provide complete cloud free imagery for any of the islands, even with the use of adjacent scenes without the restriction of seasonality. A search for additional imagery was conducted, and images from Landsat 7 ETM+ and ASTER were acquired to fill in these areas of no data. The Landsat 7 ETM+ and ASTER data followed the same processing and classification flow as the ALI where possible, with their respective calibrations applied. The six multispectral bands (1-5 and 7) of the ETM+ imagery was pan-sharpened to 15m spatial resolution using PCA with NN resampling in ERDAS Imagine. For ASTER, which has no panchromatic band, just the three visible 15m bands were used due to the unusable 30m SWIR bands. A Normalized Difference Vegetation Index (NDVI) product was created for each sensor and used as an additional band in the SOM neural network classification
For St. Croix the remaining 11.43% of missing data was taken from two Landsat 7 ETM+ images from the 31st of January 2009 and the 16th February 2009 that had little cloud cover. The two images were mosaiced, with the January image acting as the base file and the February image providing the data missing due to the SLC error. For St. Thomas the remaining 10% of missing data was provided by an ASTER image from the 30th of December 2008 and two Landsat 7 ETM+ images from the 9th of September 2007 and the 8th of August 2008. For St John, the remaining 3.63% of missing data was taken from an ASTER image from the 28th of December 2007. The resulting classification from each sensor was at 15m spatial resolution, so the data was resampled to 10m and manually edited to match the spatial resolution of the ALI classification.
F. Primary Land Cover Classification
The unsupervised SOM neural network classification produced a number of classes, with the maximum possible amount for one segment being 200 classes. The classes were then refined into useful land cover types.
This was completed through visual interpretation using field information and photography of vegetation types from site visits to the U.S. Virgin Islands in 2007, the aerial photography from 1999, 2004, and 2007, and comparing classification results to previous land cover maps (Conservation Data Center 2000; Kennaway et al. 2008). With
44
the additional use of reports focusing on specific areas within the U.S. Virgin Islands (Daley 2009; Dammann and Nellis 1992; Weaver 2006a; Weaver 2006b) all providing useful information for class interpretation.
The primary unsupervised land cover classes were recoded to closed forest, open forest, closed canopy forest shrubland, open canopy forest shrubland and scrub, grasslands, maintained grasslands, urban, rock, water, and mangrove.
G. LiDAR data processing and products
Using FU.S.ION/LDV 2.70 LiDAR processing software the raw LiDAR .xyz format files were imported and converted to an .lda format for data processing compatibility in FU.S.ION/LDV. First, using the Groundfilter and GridSurfacecreate algorithms, a digital terrain model (DTM) was created with 10 meter pixel size. Both algorithms have parameters that can be set to improve the creation of the DTM, the primary aim is to remove vegetation and building points above the ground for a reliable bare earth surface product.
Canopy height and canopy cover products were also extracted from the LiDAR data. The CanopyModel algorithm uses the LiDAR DTM product to extract tree canopy heights using the return for the highest elevation for a given cell size. The Cover algorithm extracts a percentage canopy cover for a given cell size. The CanopyModel and Cover algorithms have a number of input parameters than can be adjusted for the creation of each product. The tree canopy height model was created with a 10 meter grid cell size, while the percentage canopy cover product had a 15 meter grid cell size.
H. Incorporation of ancillary data sources
A number of ancillary layers were used to refine and stratify the classification to provide a detailed land cover classification of habitat. The layers helped to remove spectral confusion that was present in some of the classes, and combinations of layers helped to delineate specific habitats based on a set of modeling criteria. These layers included products such as wetland delineations from the National Wetland Inventory (NWI), geological information from the USGS and the National Park Service (NPS) Geologic
45
Resources Inventory (GRI) program, 30 meter Digital Elevation Models from the USGS, National Hydrography Datasets (NHD) from the USGS and U.S. Environmental Protection Agency (U.S.EPA) NOAA Environmental Sensitivity Index (ESI), Holdridge Ecological Life Zones (Ewel and Whitmore 1973), the Natural Resources Conservation Service Soil Survey of the U.S. Virgin Islands, the 1999, 2004 and 2007 aerial photographs as well as spectral interpretation from the ALI imagery and a number of derived products created at the I.I.T.F. GIS and Remote Sensing laboratory such as bare earth surfaces, canopy cover, canopy height, landforms, coastline, slope, aspect and watersheds.
I. Refining classes, mapping Issues and confusion
After the initial unsupervised classification, it was evident that there was a degree of spectral separation of a maintained grassland class in and around developed areas in comparison to dry scrub, abandoned or regenerating scrub and grassland classes in coastal areas and agricultural areas, although some manual editing of this class through the use of aerial photography was required.
Some manual editing of confused classes was necessary to refine and improve the land cover classification. Due to the spatial resolution of the ALI imagery, shadows from buildings, topographic variations and in rocky coastal regions resulted in class confusion. Typically some of the pixels falling in these areas were classified as water or closed forest, some closed forest at elevation had mangrove pixels and some open forest in shaded areas was identified as a closed forest habitat. Coastal sand and rock had some confusion with urban classes, likewise with some recently clear cut grassland, bare soils and barren areas. Agricultural classes are similar spectrally to a number of classes so these were mapped manually through interpretation of aerial photography.
To map moist and dry land cover classes a combination of ancillary layers were used. The Holdridge life zones were used as a coarse guide, then additional information including elevation ranges from the USGS 30m DEM and LiDAR derived 10m DEM, slope and aspect, interpretation of aerial photography from 1999, 2004, 2007, and LiDAR canopy height and canopy cover products all provided information to adjust the moist and
46
dry boundary. The wetland classes and seasonally flooded forest and shrubland classes were mapped using the NWI as a guide, then the soil survey data, the 10m LiDAR DEM and interpretation of aerial photography guided class boundaries. The ESI and aerial photo interpretation guided the mapping of the coastal classes, while the landforms, LiDAR products and aerial photography helped map the gallery forest classes. The urban class was filtered to represented low, medium and high urban development.
The LiDAR data was not integrated further into the classification process due to evidence that small footprint discrete LiDAR at this resolution does not penetrate closed forest canopy effectively limiting the usability of derived products in land cover classification (Drake et al. 2002; Kenneway et al. 2008; Martinuzzi et al. In Press). Additionally, through assessment of the 2004 and 2007 aerial photography, it was evident in some areas that the LiDAR information did not match the 2007 to 2009 satellite imagery due to significant land cover change. On all of the islands some vegetated areas had been cleared and were in the process of being developed (e.g. roads or buildings). While on St. Croix the pattern of clearing abandoned pastures dominated by scrub and shrub species for agriculture was seen in many areas in the 2007 aerial photos. While some pastures identified in the 1999 and 2004 aerial photos had been abandoned. Newly forested land on St. Croix is, in fact, abandoned pastures covered in a single age-class of tan-tan (Daley 2009). However, the LiDAR was useful to improve the mapping of wetland habitats at finer spatial scales, as in relatively flat coastal areas with sparser vegetation the extraction of an accurate bare earth surface from the LiDAR data is easier.
The final land cover classification for St. Croix consisted of forty nine land cover classes at 10m spatial resolution (Figure 6 below) with forty seven of those classes appearing on either St. Thomas (Figure 4) or St. John (Figure 5). The classification identifies the major land cover habitats with a few land use classes incorporated (i.e. Urban and Agricultural classes).
6. Final Classification
U.S. Virgin Islands GAP Land Cover Classes
47
Figure St. 4. and Thomas Water Island spatial land10m resolution cover (legend below)
48
Figure St. John 5. 10 meter resolutionspatial land cover (legend map below)
49
Figure St. Croix 6. 10 meter resolution spatial lan
d cover map (legend below) (legend map cover d
50
51
Land Cover Classes Area Hectares Percent U.S. Virgin Island terrestrial extent 35163.0 100.0 Forest, Woodland and Shrubland 23621.7 67.2 Forest 7228.2 20.6 Woodland 4408.2 12.5 Closed Canopy Shrubland 10020.4 28.5 Open Canopy Shrubland 1965.0 5.6
Dry 21340.1 60.7 Dry Forest 5343.7 15.2 Dry Woodland 4288.5 12.2 Dry Closed Canopy Shrubland 9744.9 27.7 Dry Open Canopy Shrubland 1963.0 5.6
Alluvial Substrates Dry Alluvial Semideciduous Forest 147.2 0.4 Dry Alluvial Shrubland 1801.2 5.1 Dry Alluvial Evergreen Gallery Forest 74.7 0.2 Dry Alluvial Woodland 506.7 1.4 Dry Alluvial Open Shrubland 828.1 2.4 Calcareous Substrates Dry Limestone Semideciduous Forest 111.3 0.3 Dry Limestone Shrubland 1624.6 4.6 Dry Limestone Evergreen Gallery Forest 117.3 0.3 Dry Limestone Woodland 366.4 1.0 Dry Limestone Open Shrubland 200.5 0.6 Noncalcareous Substrates Dry Noncalcareous Semideciduous Forest 4042.1 11.5 Dry Noncalcareous Shrubland 6319.1 18.0 Dry Noncalcareous Evergreen Gallery Forest 851.1 2.4 Dry Noncalcareous Woodland 3415.5 9.7 Dry Noncalcareous Open Shrubland 934.4 2.7
Moist 1612.9 4.6 Moist Forest 1404.7 4.0 Moist Woodland 119.6 0.3 Moist Closed Canopy Shrubland 86.5 0.2 Moist Open Canopy Shrubland 2.0 <0.1
Alluvial Substrates Lowland Moist Alluvial Evergreen Gallery Forest 84.5 0.2 Noncalcareous Substrates Lowland Moist Noncalcareous Evergreen Forest 414.7 1.2 Lowland Moist Noncalcareous Shrubland 86.5 0.2 Lowland Moist Noncalcareous Evergreen Gallery Forest 905.5 2.6
52
Lowland Moist Noncalcareous Woodland 119.6 0.3 Lowland Moist Noncalcareous Open Shrubland 2.0 <0.1 Flooded Forest 668.7 1.9 Seasonally Flooded Non Saline Forest 125.6 0.4 Seasonally Flooded Non Saline Shrubland 49.8 0.1 Seasonally Flooded Saline Forest 13.9 <0.1 Seasonally Flooded Saline Shrubland 139.1 0.4 Mangrove Forest and Shrubland 340.3 1.0 Grasslands 2518.2 7.2 Dry Grasslands 2441.7 6.9 Dry Grasslands and Pastures 2441.7 6.9 Moist Grasslands 1.2 <0.1 Moist Grasslands and Pastures 1.2 <0.1 Wet Grasslands 64.0 0.2 Seasonally Flooded Herbaceous Non-Saline Wetlands 11.8 <0.1 Seasonally Flooded Herbaceous Saline Wetlands 52.2 0.1 Flooded Grasslands 11.3 <0.1 Emergent Herbaceous Saline Wetlands 0.3 <0.1 Emergent Herbaceous Non-Saline Wetlands 11.0 <0.1 Agriculture 223.8 0.6 Hay and Row Crops 215.0 0.6 Woody Agriculture 8.8 <0.1 Natural Barrens 665.1 1.9 Riparian and other Natural Barrens 4.5 <0.1 Riprap 17.8 0.1 Rocky Cliffs and Shelves 368.4 1.0 Salt and Mud Flats 104.9 0.3 Gravel beaches 36.8 0.1 Fine to medium grained sandy beaches 94.5 0.3 Mixed sand and gravel beaches 38.2 0.1 Artificial Barrens 202.2 0.6 Artificial Barrens 202.2 0.6 Developed Areas 7568.7 21.5 Maintained Grassland 3475.3 9.9 Low-Density Urban Development 841.9 2.4 Medium-Density Urban Development 2116.6 6.0 High-Density Urban Development 1134.9 3.2 Water 363.5 1.0 Aquaculture 0.8 <0.1 Fresh Water 40.4 0.1 Salt Water 322.3 0.9 Table 4. The U.S. Virgin Islands GAP Analysis land cover classes, their area and percentage of the U.S. Virgin Islands total area.
53
A. U.S. National Vegetation Classification Land Cover Product
A secondary land cover map product was created using the original U.S. Virgin Island GAP land cover map (above). The U.S. National Vegetation Classification Standard (NVCS) is a scheme for classifying vegetation across the United States.
The NVC is a hierarchical system designed to classify existing vegetation (i.e. plant cover, floristic composition, and vegetation structure documented to occur in a specific area at a specific time) on the basis of both physiognomic and floristic criteria (FGDC, 2008). Recent revisions to the NVC have attempted to integrate various land cover types dominant in the Caribbean that previously have not been well represented, these classes can be identified in the U.S. Virgin Islands land cover legend and table.5 below.
The U.S. Virgin Islands GAP land cover map classes were recoded to match the U.S. NVC. The original 49 classes were recoded into 23 classes. The NVC cover types were attributed to the lowest level in the classification and varied by land cover type, some to group, some to macrogroup and some to division. If available the recoded U.S.NVC Map Unit has information on Class, Subclass, Formation, Division, Marcogroup and Group.
54
Figure St. 4. and Thomas Water Island spatial land10m resolution cover (legend below)
55
Figure St. John 4. 10m spatial land resolution cover (legend below)
56
Figure St. Croix 4. 10m spatial land resolution cover (legend below)
57
58
Table 5. The U.S. Virgin Islands NVC land cover classes, their area and percentage of the U.S. Virgin Islands total area.
Land Cover Classes Area Hectares Percent U.S. Virgin Island terrestrial extent 35163.0 100.0 Forest, Woodland, Shrubland and Grassland 26139.8 74.3 Forest and Woodland 11825.5 33.6 Shrubland and Grassland 14314.6 40.7
Forest and Woodland G585 Caribbean Dry Broadleaf Forest 8589.1 24.4 G455 Caribbean Seasonal Evergreen Moist Lowland Forest 1162.7 3.3 G454 Caribbean Moist Lowland-Submontane Forest Group 1404.7 4.0 G002 Caribbean Hardwood Swamp Group 328.4 0.9 G004 Caribbean Mangrove Tidal Swamp Group 340.3 1.0 Shrubland and Grassland D094 Caribbean & Central American Lowland Shrubland, Grassland & 14303.3 40.7 Savanna M041 Caribbean & Central American Freshwater Marsh 11.3 <0.1 Agriculture 223.8 0.6 Agricultural Land 223.8 0.6 Natural Barrens 665.1 1.9 Riparian and other Natural Barrens 4.5 <0.1 Riprap 17.8 0.1 Rocky Cliffs and Shelves 368.4 1.0 Salt and Mud Flats 104.9 0.3 Gravel beaches 36.8 0.1 Fine to medium grained sandy beaches 94.5 0.3 Mixed sand and gravel beaches 38.2 0.1 Artificial Barrens 202.2 0.6 Artificial Barrens 202.2 0.6 Developed Areas 7568.7 21.5 Maintained Grassland 3475.3 9.9 Low-Density Urban Development 841.9 2.4 Medium-Density Urban Development 2116.6 6.0 High-Density Urban Development 1134.9 3.2 Water 363.5 1.0 Aquaculture 0.8 <0.1 Fresh Water 40.4 0.1 Salt Water 322.3 0.9
59
7. Accuracy Assessment - In process 8. Discussion and Project Limitations
Mapping the land cover of the U.S Virgin islands provided a number of challenges. These can be attributed to the complexity and variation of the landscape between the three main islands, the use of satellite imagery for the mapping process based on the location of the islands, and the use of a range of ancillary data for refining the land cover types.
The U.S. Virgin Islands offer difficulty in land cover mapping due to the intricacy and heterogeneity of the landscape, land use history, current land management practices and anthropogenic influences. The islands offer a range of vegetation types, from various closed forest habitats such as moist closed canopy forests dominated by evergreen species, to mixed drier or drought deciduous closed and open forest types. While closed and open canopy shrub and scrub land dominate some of the exposed and drier regions. Red, black, white and buttonwood mangrove species (Thomas and Devine 2005) are found in many of the coastal wetlands. Some of the most diverse habitats in the U.S. Virgin Islands exist in the watercourses (Gardner 2008), which contain distinct forest types such as gallery moist forest and gallery shrub land and freshwater habitats. Additionally, agricultural practices in St. Croix offer their own distinct land cover associations.
One major limitation was the lack of ability to account for seasonal variation in many of the land cover types that can be evaluated through analysis of the temporal variation in the ALI spectral signatures for many of the land covers. Mapping of many of the woodland, shrub and scrubland types that are dominated by deciduous species would benefit from a range of cloud free data with leaf-on and leaf-off information. Reese et al. (2002) outlines higher classification accuracy for forested and agricultural cover types when using imagery acquired in the spring, summer and fall from the same year.
60
Additionally, the wetland environments are dynamic coastal environments that exhibit sensitivity to the wet and dry seasons.
An additional challenge was mapping moist and dry forest boundaries between the three islands. The Holdridge life zones mapped broad general areas of each of the islands with significant moist zones and dry zones. Anthropogenic impacts such as disturbance and development and natural impacts such as hurricanes (Hugo in 1989 and Marilyn in 1995) and tropical storms have severely degraded the moist forest habitat across the U.S. Virgin Islands with Thomas and Devine (2005) arguing that the moist forest habitat may no longer exist on St. Croix and may have transitioned into drier forests types. On St. Thomas, around Crown Mountain, development and clearing in the moist Holdridge life zone, evident in the satellite imagery, aerial photography and LiDAR products, suggests that natural disturbance has affected the species composition and degraded the moist forest habitat into drier or transitional forest habitats in a number of locations. As of 2010 the population figure for the U.S. Virgin Islands was 50,601 for St. Croix, 51,634 for St, Thomas and 4,170 for St. John (U.S Census Bureau 2010). Croven (2008) highlights the densely populated St. Thomas as exhibiting the most severe and extensive loss and degradation of natural habitats, but with the other islands also impacted. St. John is the least developed island, with the Virgin Islands National Park established in 1956, offering protection to many vegetation habitats. Approximately 56% of St. John is contained within the Virgin Islands National Park (2,816 ha of land and 2,287 ha of water), comprising forests and watersheds extending into the marine environment (Platenberg and Boulon 2006).
Mapping moist forest types is usually aided by ancillary information such as precipitation and evapotranspiration data with understanding of the climate variability for a region. Unfortunately, the U.S. Virgin Islands has a limited number of weather stations with long term archived data. The average rainfall is about 55 inches and varies widely from island to island and across each island, this variability affects plant community distribution (Thomas and Devine 2005). Analysis of the climatic variability between the watersheds across each island that is useful for vegetation mapping is currently limited. Daly (2006)
61
outlines the lack of high-quality, high-resolution precipitation data on St. Croix to produce accurate rainfall patterns with high accuracy.
The transitional forest boundary types, that can provide unique habitats for some animal species are spectrally complicated to discriminate without extensive ground data and multi-temporal image analysis. Brandeis and Oswalt (2005) identify that the boundary between subtropical dry forest and subtropical moist forest is usually indistinct, with gradual changes in forest structure. Mapping such habitats was considered beyond the scope of this project.
Although the ALI sensor has a limited swath at 37km, it is useful for providing high resolution imagery for small scale island mapping, such as the individual islands of the U.S. Virgin Islands. Spectrally, spatially and radiometrically the ALI sensor showed potential to produce high quality land cover classification maps with a high degree of accuracy. White (2003) found ALI superior for discriminating among several classes, including thicket, cropland, cleared land, and mangrove, with higher classification accuracies compared to ETM+ and TM.
Spectral confusion and mixed pixels are a regular issue in heterogeneous, fragmented landscapes. The panchromatic sharpening technique improves the spatial resolution of the multispectral bands from 30 meters to 10 meters and can significantly reduce the level of mixing, this reduces the amount of different land cover types that may influences a pixel’s signature, but it can create a larger amount of smaller mixed pixels. The amount of mixed pixels can potentially increase and influence the classification process. Statistical analysis and evaluations of subsetting a single pixel into nine pixels for ALI imagery is not documented. Pan-sharpening of Landsat 7 ETM+, where a single pixel will be enhanced into four pixels has been well documented and evaluated through a range of pan-sharpening techniques (Eshtehardi and Ebadhi 2008; Lui, 2000; Memarsadeghi et al. 2006).
Although confused classes from spectral mixing and mixed pixels in a complex sub- tropical landscape can add to the misclassification when mapping with satellite imagery, some misclassified pixels may be attributed to other factors rather than classification
62 error. As previously outlined, ALI images acquired within nadir were preferable but the collections had a significant range of sensor look angles (Table 2). While cross track pointing can provide flexibility for image acquisitions, an increase look angle may have adverse impacts on the geometric characteristics of a particular scene (i.e. pixel distortion and band-to-band alignment) (USGS 2010). These impacts also increase slightly with collections close to the equator. This distortion was noted within the ALI imagery in some areas with significant topographic variation such as with St. Thomas, St. John and the north west of St. Croix. However, this was only visible in a few regions where the look angle was high.
Another issue related to topography. Imagery collected in mountainous regions or in areas with varied topography is subjected to a radiometric distortion known as the Topographic effect. This results from illumination differences due to the angle of the sun and terrain in relation to the sensor. This effect causes a high variation in the reflectance response for similar vegetation types (Riaño et al. 2003). The U.S. Virgin Islands are considered a mountainous terrain with steep slopes common, 50% of the land exceed slopes of 25 to 35% (Thomas and Devine 2005). Topographic normalization attempts to compensate the different solar illuminations due to the irregular shape of the terrain. However, Bruce and Hibbert (2004) outline the complexity of this correction and the need of a DEM with at least equal or better spatial resolution than the imagery to be corrected, while prefect registration between the DEM and imagery is essential for an accurate correction. The only DEM with a reported accuracy was the 30 meter USGS DEM that was considered too coarse, while the LiDAR derived bare earth product did not provide complete coverage of the mapping area. Additionally, although the LiDAR data provided a higher resolution DEM, the coarse resolution meant that in densely vegetated areas, usually areas where the correction is important if it falls in shadow, the vegetation density limits the ability to accurately extract ground elevations as the LiDAR pulse may not penetrate to the forest floor.
Other considerations relate to the use of a variety of ancillary layers for mapping. Their accuracy, error and uncertainty can potentially influence the accuracy of the refined land
63 cover classes. Mapping with ancillary data with different spatial scales e.g. 30m USGS DEM for landforms limits its use for mapping 10m spatial habitats with the ALI imagery.
References
Beck, R. 2003. EO-1 User Guide v 2.3. Department of Geography. University of Cincinnati.
Brandeis, T.J. and Oswalt, S.N. 2004. The status of the U.S. Virgin Islands’ Forests. Resource Bulletin SRS-122, USDA Forest Service Southern Research Station, Asheville, NC.
Bruce, C. M. and Hibbert, D. W. 2004. Pre-processing Methodology for Application to Landsat TM/ETM+ Imagery of the Wet Tropics. CSIRO Tropical Forest Research Cnetre and Rainforest CRC. Research Report.
Bryant, R., Moran, S.M., McElroy, S.A., Holified, C., Thome, K.J., Miura, T. 2003. Data Continuity of Earth Observing 1 (EO-1) Advanced Land Imager (ALI) and Landsat TM and ETM+. IEEE Transactions on Geoscience and Remote Sensing. Vol. 41, No. 6.
Carrão, H., Capão L., Fernando Bação, F., Caetano M. 2006. MERIS Based Land Cover Classification Self-Organizing Maps: Preliminary Results. Proceedings of the 2nd Workshop of the EARSeL SIG on Land Use and Land Cover, 2006.
Cattrall, C. and Thome, K.J. 2002. Adjacency effects in High Resolution Imagery. Remote Sensing Group, Optical Sciences, University of Arizona. Poster presented at the USGS High Spatial Resolution Commercial Imagery Workshop. Nov 2004.
Chander, G. Meyer, D.J., Helder, D.L. 2004. Cross Calibration of the Landsat-7 ETM+ and EO-1 ALI Sensor. IEEE Transactions on Geoscience and Remote Sensing, vol. 42, No. 12, pp. 2821-2831.
Chavez Jr., P. S. 1996. Image-based atmospheric correction — Revisited and improved. Photogrammetric Engineering and Remote Sensing. Vol. 62, No.9, 1025−1036.
64
Conservation Data Center. 2000. U.S. Virgin Islands Rapid Ecological Assessment. The University of the Virgin Islands. ECC, Conservation Data Center. St. Thomas, VI. U.S.A.
Croven, J. 2008. Important Bird Areas in the Caribbean. Chapter U.S. Virgin Islands. 392-338
Daly, C. 2006. Production of Rainfall Frequency Grids for Puerto Rico and the U.S. Virgin Islands Using a Specifically Optimized PRISM System. National Weather Service Hydrometeorological Design Studies Center. NOAA Atlas 14. Vol 3.
Daley. B. 2009. Forest, Agriculture and Development in a Changing Landscape: Land- cover change analysis of St. Croix, USVI, using Landsat satellites from 1992-2002. University of the Virgin Islands Agricultural Experiment Station. St. Thomas, VI. U.S.A. Technical Bulletin #13.
Damman, A.E. and Nellis, D.W. 1992. A Natural History Atlas to the Cays of the U.S. Virgin Islands. Florida. Pinapple Press, INC.
Drake, Jason B.; Dubayah, Ralph O; Clark, David B.; Knox, Robert G.; Blair, J. Bryan; Hofton, Michelle A.; Chazdon, Robin L.; Weishampel, John F.; Prince, Stephen D. 2002. Estimation of tropical forest structural characteristics using large-footprint lidar. Remote Sensing of Environment 79: 305-319
Donegan S, J. and Flynn, L, P. 2004. Comparison of the response of the Landsat 7 Enhanced Thematic Mapper Plus and the Earth Observing-1 Advanced Land Imager over active lava flows. Journal of Volcanology and Geothermal Research. Vol 135, pp 105 - 126
ERDAS, 2010. ERDAS Field Guide. Atlanta, Georgia. United States. 480p
Elmore, A.J. and Mustard, J.F. 2003. Precision and Accuracy of EO-1 Advanced Land Imager (ALI) Data for Semiarid Vegetation Studies. IEEE Transactions on Geoscience and Remote Sensing. Vol. 41, No. 6. pp. 1311-1320.
65
Eshtehari, A. and Ebadi, H. 2008. Image Fusion of Landsat ETM+ and Spot Satellite Images Using IHS, Brovey and PCA. Toosi University of Technology.
Ewel, J. J. and Whitmore, J. L. 1973. The Ecological Life Zones of Puerto Rico and the U.S. Virgin Islands. USDA Forest Service. Institute of Tropical Forestry. Research Paper ITF-018.
FGDC (Federal Geographic Data Committee). 2008. National Vegetation Classification Standard, Version 2. FGDC-STD-005-2008. Federal Geographic Data Committee, FGDC Secretariat, U.S. Geological Survey, Reston, VA.
Gardner, L. 2008. A Strategy for Management of Ghuts in the U.S. Virgin Islands. Water Resources Research Institue. University of the Virgin Islands.
Gould, W.A., C. Alarcón, B. Fevold, M.E. Jiménez, S. Martinuzzi, G. Potts, M. Quiñones, M. Solórzano, E. Ventosa. 2008. The Puerto Rico Gap Analysis Project. Volume 1: Land cover, vertebrate species distributions, and land stewardship. Gen. Tech. Rep. IITF-GTR-39. Río Piedras, PR. U.S. Department of Agriculture, Forest Service, International Institute of Tropical Forestry, 165 p.
Hearn, D.R. 2002. Earth Observing-1 Advanced Land Imager: Imaging Performance On- Orbit. Lincoln Laboratory. NASA Project Report EO-1-12.
Helmer, E. H. and Ruefenacht, B. 2007. A Comparison of radiometric normalization methods when filling cloud gaps in Landsat imagery. Canadian Journal of Remote Sensing. Vol 33. No. 4.
Kennaway, T.A., E.H. Helmer, M.A. Lefsky, T.A. Brandeis, K.R. Sherrill. 2008. Mapping land cover and estimating forest structure using satellite imagery and coarse resolution lidar in the Virgin Islands. Journal of Applied Remote Sensing, Vol. 2, 023551.
Kohonen, T. 1990. The Self-Organizing Map. Proceedings of the IEEE. VOl 78, pp 1464-80.
66
Jensen, J. R. 1996. Introductory Digital Image Processing – a Remote Sensing Perspective, Prentice Hall, Englewood Cliffs, N.J.
Lencioni, D. E., Hearn, D. R., Digenis, C. J., Mendenhall, J.A., Bicknell, W.E. 2005. The EO-1 Advanced Land Imager: An Overview. Lincoln Laboratory Journal. Vol. 16, No. 2.
Landsat Project Science Office. 1998. Landsat 7 Science Data Users Handbook. Updated August 4th, 2009. http://landsathandbook.gsfc.nasa.gov/handbook/handbook_toc.html.
Liang, S., Fang, Hongliang, F., Chen, M. 2001. Atmospheric Correction of Landsat ETM+ Land Surface Imagery – Part 1: Methods. IEEE Transactions on Geoscience and Remote Sensing. Vol 39, No. 11.
Li Z Eastman JR. 2006. Commitment and Typicality Measurements for the Self- Organizing Map. UCGIS 2006 Summer Assembly, Vancouver, Washington, June 28 - July 1, 2006.
Liu, J. G. 2000. Evaluation of Landsat-7 ETM+ Panchromatic Band for Image Fusion with Multispectral Bands. Natural Resources Research. Vol 9, No. 4.
Marsett, R.C., Qi, J., Heilman, P., Biedenbender, S.H., Watson, M.C., Amer, S., Weltz, M., Goodrich, D., Marsett, R. 2006. Remote sensing for grassland management in the arid southwest. Rangeland Ecology and Management. Vol 59, pp 530-540.
Martinuzzi, S.; Gould, W.; Vierling, L.; Hudak, A.; Nelson, R.; Evans, J. in press. Classifying forest type and succession in a tropical dry forest: a LiDAR perspective. Biotropica
Martinuzzi, S., Gould, W.A., Ramos Gonzalez, O.G. 2007. Creating Cloud-Free Landsat ETM+ Data Sets in Tropical Landscapes: Cloud and Cloud-Shadow Removal. U.S. Department of Agriculture, Forest Service, International Institute of Tropical Forestry, Río Piedras, PR.
67
Memarsadeghi, N., Le Moigne, J., Mount, D. 2006. Image Fusion Using Cokriging. Proceedings of the IEEE International Symposium on Geoscience and Remote Sensing (IGARSS’06), Denver, CO, 2006, pp. 2518-2521.
McGaughey, R.J. 2009. FUSION/LDV: Software for LiDAR Data Analysis and Visualization. FUSION version 2.70. U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station.
NASA Jet Propulsion Laboratory, 2008. ASTER, ASTER User Advisory. Last accessed August 2011. asterweb.jpl.nasa.gov/swir-alert.asp.
NOAA. 1999. National Oceanographic and Atmospheric Administration (NOAA). National Oceans Service (NOS). National Center for Coastal Ocean Science (NCCOS). Center for Coastal Monitoring and Assessment (CCMA). Biogeography Program.
Reese, H. M., Lillesand, T. M., Nagel, D.E., Stewart, J.S., Goldmann, R.A., Simmons, T.E., Chipman, J.W., Tessar, P.A. 2002. Statewide land cover derived from multiseasonal Landsat TM data A retrospective of the WISCLAND project. Remote Sensing of Environment. 82, 224-237.
Riaño, D., Chuvieco, E., Salas, J., Aguado, I. 2003. Assessment of Different Topographic Corrections in Landsat-TM Data for Mapping Vegetation Types. IEEE Transcations on Geoscience and Remote Sensing. Vol 41, No. 5.
Renata J. Platenberg Ralf H. Boulon, Jr. Conservation status of reptiles and amphibians in the U.S. Virgin Islands.
Schroeder, T. A., Cohen, W. B., Song, C., Canty, M. J., Yang, Z. 2006. Radiometric correction of multi-temporal Landsat data for characterization of early successional forest patterns in western Oregon. Remote Sensing of the Environment. Vol 103, pp 16-26.
Thenkabail, P.S., Enclona E.A., Ashton, M.S., Legg, C., Minko, J.D. 2004 Hyperion, IKONOS, ALI, and ETM+ sensors in the study of African rainforests. Remote Sensing of Environment. 90, 23-43.
68
Thomas, T. and Devine, B. 2005. Island Peak to Coral Reef: A Field Guide to the Plant and Marine Communities of the Virgin Islands. 1st. ed. U.S. Virgin Islands: University of the Virgin Islands.
U.S. Census Bureau. 2010.. Census 2000 for the U.S. Virgin Islands and the 2010 Census for the U.S. Virgin Islands.
United States Army Corps of Engineers. 2004. 8-bit True Color 1 Meter Ortho-rectified Images of Puerto Rico and USVI. Peachtree City, GA.
United States Army Corps of Engineers. 2007. Digital Aerial Imagery for 2006-2007 PRVI Orthophotography. Peachtree City, GA.
USGS Global Visualization Viewer. 2011. Earth Resources Observation Science Center (EROS). Last accessed September 2011. http://glovis.usgs.gov/
USGS EarthExplorer. 2011. Earth Explorer. Last accessed September 2011. http://edcsns17.cr.usgs.gov/NewEarthExplorer/
USGS. 2010. Earth Observing 1 (EO-1). Last accessed September 2011. edcsns17.cr.usgs.gov/eo1/ (MOVED) UPDATE
Van Tu, N. 2005. Quality Assessment and Improvement of the Exist Land Cover Map by Using and ETM Image. MSc Thesis. International Institute for Geo-Information and Earth Observation.
Vijayaraj, V. 2004. A quantitative Analysis of Pansharpened Images. MSc Thesis. Mississippi State University. Department of Electrical & Computer Engineering.
Weaver, P.L. 2006a. Estate Thomas Experimental Forest: St. Croix, U.S. Virgin Islands: Research history and potential. General technical Report IITF-30, U.S. Department of Agriculture, Forest Service, International Institute of Tropical Forestry, Río Piedras, PR. 62 p.
69
Weaver, P.L. 2006b. A summary of 20 years of forest monitoring in Cinnamon Bay watershed, St. John, U.S. Virgin Islands. General Technical Report IITF-34, , U.S. Department of Agriculture, Forest Service, International Institute of Tropical Forestry, Río Piedras, PR. 47 p.
White, A.W. 2003. Evaluation of EO-1 ALI date Through an Analysis of Land Cover and Land Use and Local Impacts of Hurricane Iris in Belize, Central America.
3001 Inc. 2004. Puerto Rico Ground Control Report. Contract No: DACW43-03-0504
70
PREDICTED ANIMAL HABITAT DISTRIBUTIONS AND SPECIES RICHNESS
Authors:
Mariano C. Solórzanoa, Gary Pottsa, Lisa Yntemab, and William Goulda
a International Institute of Tropical Forestry, USDA Forest Service b University of Puerto Rico, Río Piedras c Unaffiliated
Recommended Citation: Introduction
All species range maps are predictions about the occurrence of those species within a particular area (Csuti 1994). Traditionally, the predicted occurrences of most species begin with samples from collections made at individual point locations. Most species range maps are small-scale (e.g., >1:10,000,000) and derived primarily from point data to construct field guides which are suitable, at best, for approximating distribution at the regional level or counties for example. The purpose of the GAP vertebrate species maps is to provide more precise information about the current predicted distribution of individual native species’ habitats according to actual habitat characteristics within their general distribution and to allow calculation of predicted area of distributions and associations to specific habitat characteristics.
GAP maps are produced at a nominal scale of 1:100,000 or better and are intended for applications at the landscape or "gamma" scale (heterogeneous areas generally covering 1,000 to 1,000,000 hectares and made up of more than one kind of natural community). Applications of these data to site- or stand-level analyses (site--a microhabitat, generally 10 to 100 square meters; stand--a single habitat type, generally 0.1 to 1,000 ha; Whittaker 1977, see also Stoms and Estes 1993) will likely reveal the limitations of this process to incorporate differences in habitat quality (e.g., understory condition) or necessary microhabitat features such as standing dead trees.
Gap analysis uses the predicted distributions of animal species habitat to evaluate their conservation status relative to existing land management (Scott et al. 1993). However, the maps of species distributions may be used to answer a wide variety of management, planning, and research questions relating to individual species or groups of species. In addition to the maps, great utility may be found in the literature that is assembled into databases used to produce the maps. Perhaps most importantly, as a first effort in developing such detailed distributions, they should be viewed as testable hypotheses to be
71
confirmed or refuted in the field. We encourage biologists and naturalists to conduct such tests and report their findings in the appropriate literature and to the Gap Analysis Program such that new data may improve future iterations.
Previous to this effort there were five individual state-based projects describing the likely present-day distribution of species by habitat type across their ranges. Ordinary species (i.e., those not threatened with extinction or not managed as game animals) are generally not given sufficient consideration in land-use decisions in the context of large geographic regions or in relation to their actual habitats. Their decline, because of incremental habitat loss can, and does, result in one threatened or endangered species "surprise" after another. The distribution and habitat information that exists for most of these species is also frequently truncated by state boundaries. Effective management of such wide ranging species requires a regional approach. Simply creating a consistent spatial framework for storing, retrieving, manipulating, analyzing, and updating the totality of our knowledge about the status of each animal species is one of the most necessary and basic elements for preventing further erosion of biological resources.
Spatial models are an important tool for understanding wildlife-habitat relationships and for guiding natural resource management decisions (Stoms et al 1992, Pearce and Ferrier 2000, Penhollow and Stauffer 2000, Wright et al. 2000, Brugnach et al. 2003). For predictive models to be useful tools in the decision making process, they must be accurate, general, and easy to apply (Van Horne and Weins 1991). Bolger et al. (1997) have suggested modeling wildlife-habitat relationships at the landscape scale because management decisions are often best applied at this broad level, and the Gap Analysis Program takes this approach to conservation.
This chapter is divided into five sections. The first section details the methods used to create animal habitat distribution models and to evaluate their completeness and accuracy. In the second section, we present results of the modeling process. The third section presents summary information on species richness, based on the animal-habitat distribution models. In the fourth section, we evaluate the accuracy of the models using standard GAP protocols. Finally we discuss the overall process, strengths and weaknesses of the models, potential uses of the models and associated data, and recommendations for further work. Methods
The vertebrate-habitat distribution modeling component of the PR-USVI Gap Analysis Project was under the responsibility of GIS and Remote Sensing Lab IITF USFS staff. Much of the structural frame of this project was inherited from previous Puerto Rico Gap Analysis Project developed by the same lab. We followed a standard process to model species distributions. First we compiled a list of vertebrate species to be modeled that consisted of breeding resident, non-breeding resident, and the more common non- breeding migratory species that visit the islands. We also included some of the exotic or introduced species of management concern. Second we identified, collected, and compiled, species occurrence records, published distribution maps and literature, habitat
72
association and environmental response variables from the literature for these species. We then developed a series of hexagon distribution maps, and a generalized distribution map, one for each of the species. These maps underwent an expert review process in which expert collaborators from the U.S. Virgin Islands would comment and propose changes. These changes were incorporated into the distribution maps. We generated a model script for each species that described what portion or classes of existing and newly developed raster files (representing the landscape e.g. Landcover) would better represent that species habitat distribution. During the modeling process all habitats deemed suitable or associated with a species where extracted within the known distribution of the species. The models were developed using ESRI ArcGIS 9.3 model builder.
U.S. Virgin Islands Vertebrate Species List
Gap Analysis projects generally focus on species that are known to breed or regularly breed in the project area and that are regularly occurring non-accidentals (Csuti and Crist 1998). Csuti and Crist (1998) suggest, as a general definition, that “regular breeders” are those species breeding in the state during at least 5 of the past 10 years. However, this is often difficult to document and published literature often provide contradictory information about it. Contrary to other gap analysis efforts, we decided to expand our efforts to include the more commonly seen non-breeding migratory species, which represent an important part of the islands avifauna. We developed a vertebrate species list of the U.S. Virgin Islands using various sources of information including:
• Raffaele, H.A., Wiley, J., Garrido, O., Keith, A., and Raffaele, J. 1998. A Guide to the Birds of the West Indies. Princeton Univ. Press. Princeton • Platenberg, R.J., F. E. Hayes, D. B. McNair and J. J. Pierce. 2005. A comprehensive wildlife conservation strategy for U.S Virgin Islands. Division of Fish and Wildlife, St. Thomas. 251pp. • InfoNatura: Animals and Ecosystems of Latin America [web application]. 2007. Version 5.0. Arlington, Virginia (U.S.A): NatureServe. Available: http://www.natureserve.org/infonatura. • Rodrigues, A.L. 2002. U.S. Virgin Islands Rapid Bird Assessment. Prepared for The Nature Conservancy. 292 pp. • Gannon, M.R., A. Kurta, A. Rodríguez-Durán, and M.R. Willig. 2005. Bats of Puerto Rico – An Island Focus and a Caribbean Perspective. Texas Tech University Press. 239 pp.
The task of preparing a species list and assigning a status to each of them was complicated by the fact that species occurrence in the USVI varies from island to island. While some species are considered common in one of the main islands they might be considered very rare, extirpated or even exotic in the other two. This is the case of the Puerto Rican Ground Lizard Ameiva exsul a lizard that is native to St. Thomas and St. John but is considered exotic in St. Croix where it has been observed in recent years (Renata Platenberg personal communication) most probably introduced accidentally through human intervention. Some of the collaborators and expert reviewers were concerned that using a single or general status for all the USVI was going to provide a
73
false representation and that the status of a particular species should be considered on an island to island basis. Another concern from stakeholders was the classification of field records based on date of observation. Assigning a “Historical” category to a record, for example, meant an observation was made prior to 1970. This date, however, was considered too inclusive and collaborators considered that a more realistic date such as 1989, year in which hurricane Hugo struck the islands, should be applied. We made an effort to take into account these comments. We found that the U.S. Virgin Islands terrestrial vertebrate biodiversity is composed of over 294 species. These include breeding and non-breeding residents, non-breeding and breeding migratory, vagrant or accidental species, and established exotic. We selected a group of 147 species of birds, mammals, reptiles and amphibians for the USVI GAP Analysis. We left out those species that are very rare non-breeding migrants and vagrants as well as extirpated species such as Eleutherodactylus schwartzi. In addition to the breeding species we decided to include a group of non-breeding residents, breeding and non-breeding migratory species of importance for which there was occurrence data available. We also included some non native species (established exotic species), which may have a potential impact on the native biota, such as the mongoose.
140 119 120
100
80
60
40 21 20 7 7 0 Birds Reptiles Amphibians Mammals
Figure 1. Number of species selected per taxonomic group
Establishing collaboration relationships
We carried out an initial outreach meeting in St. Thomas to present the project objectives. We focused in establishing and maintaining collaboration relationships with people from different government agencies, NGOs, the scientific community and individuals from all of 3 main islands. Among them Lisa Yntema, birder from St. Croix; environmental consultant Russell Slaton; environmentalist Mario Francis; Laurel Brannick, Christy McManus and Zandy Hillis-Starr from the National Park Service; Carol Cramer-Burke from St. Croix Environmental Association; the Island Resources Foundation; Renata Platenberg, Jennifer Valiulis, and Judy Pierce from the Department of Planning and
74
Natural Resources Division of Fish and Wildlife, Claudia Lombard U.S. Fish and Wildlife Service Sandy Point National Wildlife Refuge; James Byrne and Shawn Margles from the Nature Conservancy, and Magen's Bay Authority.
One of our projects collaborators, Lisa Yntema, made substantial contributions to this component and was included as a co-author. Species distribution mapping was carried out by our projects coordinator and cartographic technician as well as species habitat relationship models. To support collaborative efforts we travelled to the USVI on several occasions and held workshops, meetings, and communicated through email and phone.
Gathering the necessary data
Usually Gap analysis projects have relied on government agencies and Natural Heritage programs to obtain the species occurrence records necessary to map species distributions. We found that species occurrence data was limited for the USVI and that it was scattered among the different agencies databases or staff hard-drives. Of all the records we obtained, only a small volume had geographic coordinates. Most of the records we used for preparing distribution maps didn’t have spatial coordinates associated with them. Only 1507 records (5 percent) have GPS coordinates associated with them, out of a total of 29738 records (Including “Confirmed”, “Probable” and “Predicted” records). These records represent 31% of all the confirmed records which totaled 4795. The fact that a great majority of the “Confirmed” records don’t have GPS coordinates associated with them is explained by the fact that a great number of the records came from observations in the field by collaborators who worked on gathering information using field maps (Figure 2) that we prepared displaying the topography along with the hexagons.
Figure 2. Hexagon grid field map used by collaborators to gather species field observation records.
In these cases the collaborators provided field records and assigned a hexagon number to each of them rather providing coordinates.
75
25000
20000
15000 Confirmed Probable 10000
5000
0 Birds Mammals Reptiles Amphibians
Figure 3. Number and category of records of records per taxonomic group (Predicted records which account for 53 of the total were not included in the chart).
Species occurrence records were obtained from the following sources:
Agencies and Organizations
• U.S. Virgin Islands Department of Natural and Environmental Resources – Division of Fish and Wildlife • St. John Audubon Society Christmas Bird Counts • U.S. National Park Service
Individuals
• Lisa Yntema, Unaffiliated, St Croix Carried out field bird surveys in St Croix and shared the records with the VI Gap project. • Brittany Parker, Department of Biology University of New Mexico.
Other significant sources of distribution and habitat information:
• NOAA Environmental Sensitivity Index Maps - A series of maps with location of natural elements sensitive to oil environmental disasters. • Rodrigues, A.L. 2002. U.S. Virgin Islands Rapid Bird Assessment. Prepared for The Nature Conservancy. 292 pp. • Platenberg, R.J., F. E. Hayes, D. B. McNair and J. J. Pierce. 2005. A comprehensive wildlife conservation strategy for U.S Virgin Islands. Division of Fish and Wildlife, St. Thomas. 251pp.
76
• Wege, D.C., and Anadon-Irrizarry, V. 2008. Important Bird Areas in the Caribbean: Key sites for conservation. Cambridge, UK: BirdLife International. (Birdlife Conservation Series No. 115. • Audubon Christmas Bird Counts 1983 – 2005 and 2008 Bird field observation records during St John CBC counts. Laurel Brannick and Phyllis P. Benton. A map was provided of the areas visited during the count. For the 2008 count hexagon ID was provided for each observation. • Bat occurrence records 2006-2008 Provided by Renata Platenberg from the Division of Fish and Wildlife DPNR – Bat field observation records with spatial coordinates, from St Thomas and St John collected by various experts. • HerpWaypoints.xls Provided by Renata Platenberg from the Division of Fish and Wildlife DPNR represents amphibian and reptile field observation records with spatial coordinates. • Cays.xls Provided by Renata Platenberg from the Division of Fish and Wildlife DPNR Bird surveys in USVI cays. • U.S. Virgin Islands Rapid Bird Assessment The Nature Conservancy Documented Field Bird Surveys in St Croix. • USGS Herpetofaunal Inventories Vol 2 U.S. Geological Survey Maps of herp species survey routes.
242
Reptiles 142 Amphibian 3 Birds Mammals
1208
Figure 4. Number of records with spatial coordinates per study group.
GIS datasets to be used in modeling were identified by the entire project team, and were obtained or created by the GIS and Remote Sensing lab staff. Model iterations were run in multiple computers by the project staff. Literature reviews were carried out by staff and volunteers. Species hexagon distribution maps were submitted to expert reviewers, and were revised as needed, based on review corrections and comments.
We used cartographic modeling techniques previously described by Gap projects to represent species distributions. We made efforts to map species distributions using three
77
different mapping units: Small hexagons grid (with 2 km2 hexagons), Intermediate hexagon grid (22km2 hexagons same as the one used for Puerto Rico Gap Analysis) and sub-watersheds HUC-14. However, due to time constraints we ended up using only the small hexagon grid to represent species distributions.
Mapping Standards and Data Sources:
VIGAP mapped predicted species distributions in accordance with the standards of the GAP Handbook as of 13 January, 2000. All GIS modeling of species distributions was conducted on HP workstations running ERDAS Imagine and ArcINFO 9.0.
Minimum Mapping Unit (MMU)
The Virgin Islands Gap Analysis Project recognizes two mapping unit scales that are related to
1) defining a species geographic range determined by its probability of occurrence or
2) to a species predicted distribution based on associated range, habitat, and life history
variables.
PRGAP has adopted a hexagon grid network for use in mapping U.S. Virgin Islands’ biological diversity. This hexagon grid (VIGAP-HEX) provides a uniform unit of area that can be used to represent the range and occurrence of vertebrate species across a very heterogeneous landscape.
Figure 5. Dimensions of an individual hexagon in VIGAP-HEX grid.
2 Each hexagon (Fig. 5) has an area of about 2 km which is the MMU for species geographic distributions. VIGAP-HEX consists of 321 individual hexagons with 91 occurring only over land, 174 over coastal areas, and 57 over open marine areas with small reefs and cays. The hexagon shape is based on the U.S. Environmental Protection Agency’s (USEPA) Environmental Monitoring and Assessment Program (EMAP)
typically used in Gap Analysis. However, EMAP represents the conterminous U.S. only,
and in lieu of EMAP coverage in the Caribbean, the VIGAP-HEX grid (Map 1) was developed by the U.S. Forest Service by tessellating the larger hexagonal grid used in Caribbean Forest Inventory and Analysis (FIA). The hexagon size used for distribution mapping represents an improvement in precision over the larger sizes used in other Gap Analysis Projects in the continental U.S. It reduces commission errors and is more
78
harmonious with the reality of the USVI which is characterized by drastic changes in habitat types in short distances. It constitutes an effort to satisfy the pressing needs for accurate maps of the conservation efforts of this region. It also satisfies the concerns of stakeholders who were unanimously in favor of a smaller size hexagon as the larger size hexagons were thought to be unsuitable for such small islands.
The second mapping unit scale is derived from the 10 m pixel resolution reflected in the Virgin Islands GAP land cover layer (Map 2). VIGAP uses this 10 m pixel resolution as the MMU for mapping species predicted distributions.
GIS coverages and raster datasets used in the animal species modeling process
1. Vegetation and Landcover - Land Cover developed by the USVI Gap Analysis Project. 2. Topography and Landforms - PRUSVI_Landforms.img: Landforms of Puerto Rico and the U.S. Virgin Islands. Developed by the GIS and Remote Sensing Lab IITF Forest Service 3. Elevation - USGS Digital Elevation Model / NOAA Bathymetry 4. River guts and streams: - USGS National Hydrography Dataset.
Mapping Species Geographic Distribution:
In order to map a species geographic distribution, each hexagon was attributed with the species probability of occurrence according to one of eight categories (Table 7) depending on the likelihood that the species occurs in the hexagon. Occurrence records may by confirmed with a point location, a date, and a reliable observers name. Records may be probable based on published range maps, location descriptions, or expert opinion, or predicted based on the occurrence of habitat and expert opinion that species is likely to occur. These records are maintained in a Microsoft Access database (MS Access 2007) totaling over 30 thousand records.
Species occurrence records information was derived from published literature and range maps, published or unpublished datasets, museum records, and expert opinion. The availability of distribution and habitat data for vertebrate species varied among islands. Even though the jurisdiction of government agencies whose mission includes environmental conservation encompasses all tree islands, St Thomas, St John and St Croix, vertebrate surveys and research hasn’t been homogeneous across islands. So the availability of species occurrence data varied, for the different groups of species, among islands.
A species record of occurrence may be confirmed when associated with a credible observation, including a record of the location, the observation date, and the observer’s name. In other words, greater confidence was given to an occurrence record if it included
79 coordinates and it could be located within a particular hexagon. Records for locations that were able to be located within an individual hexagon were also classified as confirmed if they were from a recent observation and came from a published document or a confident source such as field surveys from recognized organizations. This was the case with many records attributed to a particular cay or pond which were small enough to fit within a hexagon. Records lacking coordinates, observer or date may lead to hexagons classified as probable, such records or distribution data were obtained from published range maps, location descriptions, or expert opinion. Similarly hexagons were classified as predicted based on the occurrence of habitat, presence in adjacent hexagons and expert opinion that the species is likely to occur. Figure 3 presents the record allotment per taxonomic group (amphibian, bird, mammal, and reptile) and each group’s total number of occurrence records represented in the database. Once we had a hexagon distribution for each species we proceeded to create species ranges which are simplified versions of the species distributions. The ranges were developed by merging hexagons were species occur (Only those hexagons classified as “Probable”, “Predicted” and/or “Confirmed”) and then simplifying the boundary of the resulting polygon by using a smoothing technique in ArcGIS.
80
Table 7. Confidence level or category assigned to hexagon records for each species.
Confidence Level – Taxon Description Category Confirmed All Confidently assumed or known to occur in the hexagon. Sources include species locality records and expert opinion.
Predicted All Predicted to occur based on a combination of presence of suitable habitat and historical record and/or presence in adjacent hexagon. Sources include expert opinion only.
Probable All Probable occurrence based on a strong likelihood. Sources include expert opinion and/or published range maps or range descriptions. Historical All Confidently assumed or known to have occurred in the hexagon prior to 1990 and considered as invalid for recent distribution. Sources include species locality records and expert opinion. Questionable All All Occurrences within hexagon was still in question after expert review. Hexagons coded as questionable are not included in species current range distribution. Sources include expert opinion only. Excluded All Documented occurrence was excluded by expert review after having been coded as confirmed, predicted, or probable. Sources include expert opinion only.
Unknown All Information indicating the presence or absence of the species in this hexagon wasn’t found prior
Expert Review Process
Volunteer expert reviewers were given distribution maps with a series of guidelines explaining the necessary steps to carry out the review. The review involved changes in the classification of individual hexagons based on their personal knowledge or other sources of information. It also included sections on recommendations about particular habitats types to include in the species habitat relationship modeling. The expert review comments and changes were later incorporated into the maps.
Wildlife Habitat Relationships:
Species-habitat models were based primarily on species affinity for the USVI-Gap Landcover classes but also, in a lesser extent, with elevation range described by the Digital Elevation Model (DEM) developed by U.S. Geological Survey (USGS), landforms of Puerto Rico and the U.S. Virgin Islands developed by the U.S. Forest
81
Service GIS and Remote Sensing Laboratory. We also made use of other ancillary layers including the USGS National Hydrography Dataset (NHD) river guts and streams of the U.S. Virgin Islands which helped us determine proximity to rivers, and the Federal Emergency Management Agency Flood Zones dataset which helped us identify certain habitats that were potentially prone to flooding. Species habitat affinities where based on species life-history information obtained from the literature, from expert review and our best professional judgment. This information has been documented in the USVIGAP- VERT Access database (MS Access 2007), where we compiled all available information on species taxonomic classification, conservation status, worldwide distribution, distribution in U.S. Virgin Islands, associations with vegetation, geologic substrates, soils, climate or rainfall patterns, elevation, topography, habitat structure, or other terrain features. We also documented information on species migratory, reproductive, and demographic patterns, dietary habits, habitat use, and activity patterns. This information, as well as a specific wildlife habitat model, is given in Appendix 5 for each species. The modeling was done in a similar manner for all 4 groups of species.
Predicted Habitat Distribution Modeling:
The modeling was done in an analogous way for all species groups. The modeling process could be described as identifying habitat associated with a species within its known distribution. In terms of species hexagon distribution only hexagons classified as “Confirmed”, “Probable” or “Predicted” were used during the modeling. These were extracted and then merged along with a buffer represented by the hexagon directly adjacent to them to create generalized local ranges. These files were later used as a mask to extract portions of raster datasets and in so doing limiting the modeling to areas of known species occurrence. The habitat relation component of modeling process consisted of the translation of species habitat associations or relations into the distribution of suitable habitat in the GIS realm. The habitat relationship database helped us establish a series of conditional operations that would match suitable habitat to certain categories within GIS thematic layers or datasets. Finally, areas of suitable habitat outside the species known occurrence were masked out using the generalized ranges. The modeling procedures were run in ArcGIS 9 or higher using different geoprocessing tools, the Spatial Analyst Extension and the Model Builder. This software is created and distributed by the Environmental Systems Research Institute (ESRI) of Redlands, CA.
There was very little variation in the development of the predicted habitat models among the different taxonomic groups, due to differences in life history and the habitat components that were generally most important. Some examples of these differences are noted below.
Methods Specific to Mammals:
The only native mammals in the U.S. Virgin Islands are bats. Other mammals were introduced or managed to colonize the islands. Bat occurrence records were not available for all the islands, unfortunately, so the mammal distributions developed by the project have limitations. This also happened for established exotic species such as the white-
82
tailed deer (Odocoileus virginianus), for which we only had occurrence records for St Croix even though it is generally known to occur in both St. Thomas and St. John. For this species and for the West Indian mongoose we selected all the islands where they were known to occur as their distribution. We used raw waypoint records from the Department of Planning and Natural Resources Division of Fish and Wildlife (DPNR - DFW) provided by Renata Platenberg. In addition we made use of reports generated by that same department and published journals.
Methods Specific to Birds:
Birds are probably the most comprehensibly studied group of species in the U.S. Virgin Islands and various organizations and individuals keep field observation records. When mapping the distribution of birds we considered them to have the ability to roam around islands looking for suitable habitat. What we mean with this is that areas of adjacent islands or cays that were overlapped by hexagons of known occurrence in the distribution maps, were included as part of the species distribution model if in fact they contained suitable habitat for that particular species.
Methods Specific to Reptiles:
Occurrence data sources for reptiles were not homogeneous across the USVI. St. Thomas and St. John benefit from a more comprehensive species monitoring than St. Croix. Nevertheless we managed to gather enough distribution data to prepare reasonably good distribution maps. We were more careful to identify and pick specific islands were a particular reptile species occur and not to extend their range across islands and cays with the exception of marine turtles which can potentially visit the islands and look for nesting sites available in other cays or areas in addition to those where they have been observed and reported.
Methods Specific to Amphibians:
As with reptiles, amphibians were regarded as having no ability to roam around between islands and this was taken into account when mapping their distribution. Care was taken not to extend amphibian ranges to islands where no observations have been made. Apart from this difference, amphibian hexagon distributions were developed pretty much the same way as the other groups of species. Results
Mammals:
We modeled the habitat distribution of 8 mammal species that occur in the USVI, including 6 species of bats, a deer and the mongoose. Tadarida brasilensis was the species with the most limited distribution covering the center east of St. John. We didn’t
83
find records for this species from St. Thomas and St. Croix and we could only speculate whether it may or may have occurred in the other islands. Noctilio leporinus occurs in all 3 main islands but its habitat distribution is limited by the presence of suitable waterbodies. On the other hand, Stenoderma rufum, although found in all 3 islands, appears to have a more localized distribution and this was reflected in its habitat distribution model. All other mammal species showed a wider habitat distribution across the USVI archipelago including the Jamaican Fruit-eating Bat (Figure 11). In terms of the habitat use, all mammal species studied where associated, at least to some extent, with forests, woodlands, shrublands, and seasonally flooded forests.
Birds:
We selected 118 birds to model their predicted habitat. When modeling bird species habitat distribution we followed the same methodology used to model the distribution of other species groups. Species hexagon distribution mapping in St. John and St. Croix benefited from a greater amount of occurrence records and expert review. We found the least amount of bird occurrence records sources for St. Thomas. The other two islands have been well studied and more continually maybe because of the presence of the Virgin Islands National Park in St. John and a greater community of birders and environmental NGOs in St. Croix. Some 51 species within our selection are associated with water- bodies, ponds, streams, coastal waters, mangroves or wetlands and immediately surrounding habitats. It is also interesting that some 108 bird species are associated with dry forests, including closed and open forests, woodlands, shrublands, and gallery forests, as is the case of the Lesser Antillean bullfinch (Figure 9). In contrast, about 70 bird species are associated with moist forests, moist woodlands, and moist gallery forests.
Reptiles:
We model the distribution of 21 reptiles. We included sea turtles because, although they spend most of their lifetime in the ocean, they do nest in sandy beaches. The eighteen terrestrial reptiles make use of the dry forest, at least to some extent. Epicrates monensis granti, for example, was associated with all dry forest habitats in our model. However, reptiles appear to occupy all kinds of habitats, from forests, shrubland, grasslands, wetlands, mangroves, beaches, disturbed habitats such as barrens, and urban areas and even freshwater habitats which may be used by the green iguana. The predicted habitat distribution map of Alsophis portoricensis is shown in figure 5.
Amphibians:
Amphibians where associated with a great variety of habitats in the USVI including dry and moist forest, woodlands, shrublands, grasslands, wetlands, agricultural lands, urban areas, natural and artificial barrens, and water-bodies. The yellow mottled coqui, for example, is associated with forests, woodlands, shrublands, agricultural areas and seasonally flooded forests and wetlands (Figure 6 and 7).
84
Figure 6. Alsophis portoricensis species occurrence map.
85
Figure 7. Alsophis portoricensis predicted habitat map.
86
Figure 8. Eleutherodactylus lentus species occurrence map.
87
Figure 9. Eleutherodactylus lentus predicted habitat map.
88
Figure 10. Loxigilla noctis species occurrence map.
89
Figure 11. Loxigilla noctis predicted habitat map.
90
Figure 12. Artibeus jamaicensis species occurrence map.
91
Figure 13. Artibeus jamaicensis predicted habitat map.
92
Species Richness
GAP has often been associated with the mapping of species-rich areas or "hotspots." Richness maps identify where the same numbers of elements co-occur in the same geographic locations. These are color coded or shaded in intensity from the highest numbers of co-occurrence (richness) to the lowest. Richness is only one of many pattern metrics that may be derived using the data. Richest areas may or may not indicate best conservation opportunities. They may occur in already protected areas or may represent mostly already protected species or those not at risk. Still, they are often a useful starting point to examine conservation opportunities in combination with other analyses described in this report's Introduction and in the Analysis section. We also feel they may be useful for other rewarding applications such as identifying places of interest for wildlife observation and study.
Accuracy Assessment
Assessing the accuracy of the predicted vertebrate distributions is subject to many of the same problems as assessing land cover maps, as well as a host of more serious challenges related to both the behavioral aspects of species and the logistics of detecting them. These are described further in the Background section of the GAP Handbook on the national GAP home page. It is, however, necessary to provide some measure of confidence in the results of the gap analysis for species collectively, if not individually or by taxonomic group (comparison to stewardship and management status), and to allow users to judge the suitability of the distribution maps for their own uses. We, therefore, feel it is important to provide users with a statement about the accuracy of GAP-predicted vertebrate distributions within the limitations of available resources and practicalities of such an endeavor. We acknowledge that distribution maps are never finished products but are continually updated as new information is gathered. This reflects not only an improvement over the modeling process, but also the opportunity to map true changes in species distributions over time. However, we feel that assessing the accuracy of the current maps provides useful information about their reliability to potential users.
Our goal was to produce maps that predict distribution of terrestrial vertebrates and from that, total species richness and species content with an accuracy of 80% or higher. Failure to achieve this accuracy indicates the need to refine the data sets and models used for predicting distribution. There is a conscious effort in the GAP process, however, to err on the side of commission; in other words, to attribute a species as possibly present when they are not. There are two primary reasons for doing so: first, few species have systematic, unbiased known ranges and we believe science is best served by identifying a greater potential for sampling and investigation than a conservative approach that may miss such opportunities; second, in conducting the analysis of conservation representation (see the Analysis section), we believe it most appropriate to identify a species that may need additional conservation attention that is then refuted by further investigation rather than identifying a species as sufficiently protected that is discovered not to be by its subsequent loss.
93
Limitations and Discussion
The task of mapping species distributions and predicted habitat in the U.S. Virgin islands presented numerous obstacles. First there was the limited amount of field observation records. There isn’t a natural heritage program that maintains these kinds of records. Occurrence records are gathered and maintained by government agencies and are managed independently and
Literature Cited
Bolger, D.T., T.A. Scott, and J.T. Rotenberry. 21997. Breeding bird abundance in an urbanizing landscape in coastal southern California. Conservation Biology 11:406-421.
Brugnach, M., J.P. Bolte and G.A. Bradshaw. 2003. Determining the significance of threshold values uncertainty in rule-based classification models. Ecological Modeling. 160:63-76.
Csuti, B. 1994. Methods for developing terrestrial vertebrate distribution maps for Gap Analysis (version 1). In J.M. Scott and M.D. Jennings, editors. A handbook for Gap Analysis. Idaho Cooperative Fish and Wildlife Research Unit, University of Idaho, Moscow.
Pearce, J., and S. Ferrier. 2000. An evaluation of alternative algorithms for fitting species distribution models using logistic regression. Ecological Modeling 128: 127–147.
Penhollow, M. E., and D. F. Stauffer. 2000. Large scale habitat relationships of Neotropical migratory birds in Virginia. Journal of Wildlife Management 64:362–373.
Scott, J.M., F. Davis, B. Csuti, R. Noss, B. Butterfield, C. Groves, H. Anderson, S. Caicco, F. D'Erchia, T.C. Edwards, Jr., J. Ulliman, and G. Wright. 1993. Gap analysis: A geographic approach to protection of biological diversity. Wildlife Monographs 123.
Stoms, D. M., F. W. Davis, P. A. Stine, and M. Borchert, 1992. Beyond the traditional vegetation map towards a biodiversity database, Proceedings of GISILIS'92, San Jose, CA, November 10- 12, 1992, pp. 718-726.
Van Horne, B. and JA Weins. 1991. Forest bird habitat suitability models and the development of general habitat models. U.S.DI Fish and Wildlife. Res. 8. Washington DC
Wright, A., Marcus, W.A., Aspinall, R.J., 2000.Applications and limitations of using multispectral digital imagery to map geo-morphic stream units in a lower order stream. Geomorphology 33 (1–2), 107–120.
94
LAND STEWARDSHIP AND PROTECTED AREAS
Jessica Castro-Prieto Doctoral Student Environmental Science Department University of Puerto Rico, Río Piedras
Mariano C. Solórzano Cartographic Technician USVI Gap Analysis Project
Introduction
To fulfill the analytical mission of the Gap Analysis, it is necessary to compare the distribution of elements of biodiversity (e.g., ecosystems, species) with layers of land ownership and management. In the Gap Analysis, attributes of land ownership and management are combined and represented in the land stewardship maps. Although land stewardship maps do not guarantees species and habitats protection, they are a start to assess the likelihood of future threat to biotic elements through habitat conversion - the primary cause of biodiversity decline. A scale from 1 to 4 is used by Gap to denote the relative degree of maintenance of natural conditions in each land stewardship unit. The purpose of this categorization is to identify the level of protection, and if necessary the need for change in management status, for individual elements or areas containing high degrees of diversity. This information can provide objective and scientific information to decision makers and managers in a way to support decisions regarding biodiversity. The four status categories can generally be defined as follows (Scott et al. 1993, Edwards et al. 1995, Crist et al. 1995): Gap status 1: An area having permanent protection from conversion of natural land cover and a mandated management plan in operation to maintain a natural state within which disturbance events (of natural type, frequency, and intensity) are allowed to proceed without interference or are mimicked through management. Gap status 2: An area having permanent protection from conversion of natural land cover and a mandated management plan in operation to maintain a primarily
95
natural state, but which may receive use or management practices that degrade the quality of existing natural communities. Gap status 3: An area having permanent protection from conversion of natural land cover for the majority of the area, but subject to extractive uses of either a broad, low-intensity type or localized intense type. It also confers protection to federally listed endangered and threatened species throughout the area. Gap status 4: Lack of irrevocable easement or mandate to prevent conversion of natural habitat types to anthropogenic habitat types which allows for intensive use throughout the tract. Also includes those tracts for which the existence of such restrictions or sufficient information to establish a higher status is unknown. The Protected Areas Database of the United States (PADUS) in the USGS Gap Analysis Program, only submit protected areas with Gap 1and 2 to UNEP-World Conservation Monitoring Center (WCMC) for the World Database for Protected Areas (WDPA) as only these two categories meet the definition of protected by IUCN as “A clearly defined geographical space, recognized, dedicated and managed, through legal or other effective means, to achieve the long-term conservation of nature with associated ecosystem services and cultural values” (http://gapanalysis.usgs.gov). Similarly, their program conducts a ‘gap analysis’ to assess the long-term protection of biodiversity only for protected areas with these two categories. But, those areas with Gap 1-3 are provided to the Commission for Environmental Cooperation (CEC) for the North American Terrestrial Protected Areas Database, as they have a broader definition of protection (Duarte pers. comm. 2012).
Methods
Land Stewardship Mapping, Categorization and Protected Areas Selection
The USVI Gap Land Stewardship file is a 1:100,000 scale version of land ownership and management of the United States Virgin Islands, distinguishing local, state, and federal jurisdictions from private lands and delineating areas managed for the long-term maintenance of natural ecological processes and biodiversity.
96
We used the Lambert Conformal Conic projection, the North American 1983 Grid Coordinate System and the NAD 1983 State Plane Puerto Rico Virgin Islands FIPS 5200 datum. These data are intended to aid in state level assessment of natural resources and are not intended for use at a scale greater than 1:100,000. Land stewardship data sets were obtained from federal agencies such as the National Park Service (U.S. Department of Interior), USDA Forest Service, U.S. Fish and Wildlife Service, local government agencies such as: the USVI Department of Planning and Natural Resources (DPNR), the USVI Department of Sports, Parks and Recreation (DSPR), the USVI Department of Agriculture, and non-government organizations such as The Nature Conservancy and the St Croix Environmental Organization, among others. In addition, private consulting firms also provided information about stewardship areas in the USVI. These data sets were provided via disks, email, downloaded from agency electronic bulletin boards, or as hard copy from management plans. Data sets were received in different formats: ArcView shapefiles, ArcInfo coverages, and Arc export files. The data were converted to common projections, edge matching and editing, and then were brought into the Land Stewardship Arc GIS geodatabase of the USVI Gap Analysis Project. The final mapping was conducted in Arc GIS 9.0. Data and maps presented in this section were provided by a variety of sources that are individually responsible for their accuracy. They were created solely with the purpose of conducting the analyses described in this report. On 2008, our staff visited several protected areas, to conduct interviews to scientists and official land managers. Also, an electronic survey (Appendix 1) was handed to some of the land managers of those not visited areas. For other areas the information was gathered from official government documents, official government webpages or other sources. Land managers provided invaluable information about critical ecological elements of some areas (e.g., endangered species), current threats (e.g., urban development, invasive species, exotic species) and the status of management plans (e.g., in use, in process of development, obsolete, inexistent). Both, interviews and electronic surveys were used in assigning the Gap conservation status for each land unit. Similarly to CEC, we selected as protected areas those land stewardship units with Gaps 1-3 that by some means were created or managed for biodiversity
97 conservation. In other words, we excluded from the analysis those areas under Gap 4, which although they were legally protected, they were not necessarily created for protecting wildlife and natural habitats. Under this category were historic sites, the University campus and military zones, among others.
Results
We identified a set of 89 land stewardship areas. However, only 69 were classified with Gaps 1, 2 and 3 (Protected areas) and mapped (Table 1; Figure 1). A set of land stewardship or protected areas recognized by the USVI Department of Planning and Natural Resources Division of Fish and Wildlife couldn’t be mapped because we couldn’t find a spatial boundary (Platenberg et al. 2005). The 69 protected areas represent 13 % (4559.8 hectares) of the total land area of St. Thomas (469.8 ha), St. John (2894.1 ha), St. Croix (1195.9 ha) and their associated cays (Fig. 1). Land protection varies greatly from island to island. While St John provides protection to 56.6% of its territory (2894.1 ha), the other two main islands protect less than 6% of their territory. From the 69 protected areas, 19 (27.5%) of them were classified in Gap Status 1; 23 areas (33.3%) in Gap 2 and 27 areas (39.1%) in Gap 3 (Fig. 2). However, protected areas in Gap 1 represented 75.3% of the total hectares under protection because of the large size of some of the areas including the VI National Park, Sandy Point Wildlife Refuge and Buck Island Reef National Monument. By the time we analyzed the data for this report, only five protected areas had management plans and two additional ones were in process of development. Land ownership was shared among 10 public and private institutions and individuals, with the federal government as the primary owner (71.2 %), followed by the USVI government (14.9 %), non-governmental organizations (7.1 %) and private (3.2 %) (Fig. 3) The federal and the U.S. Virgin Islands governments both own land in the Salt River Bay National Historic Park and Ecological Preserve representing 3.7% of the protected lands. Similarly, 71.2 % of the protected areas is managed by federal agencies, 14.8 % by USVI governmental agencies, 10.2% by non-governmental organizations and the remaining 3.7% is co-managed between the USVI government and federal agencies
98 as can be seen in figure 4. Protected areas management was primarily shared among 11 public and private institutions, where the U.S. National Park Service, the USVI DSPR, the USVI Department of Planning and Natural Resources (Division of Fish and Wildlife) and The Nature Conservancy represents the main protected area’s managers in the USVI (Fig. 5).
Discussion and Limitations
The U.S. Virgin Islands protects more than 13% of its territory which may be considered an acceptable amount of protection taking into account the world’s (12.7%) and the Caribbean (11.2%) percentages calculated by the United Nations for 2010 (IUCN and UNEP 2009). However, land protection is not balanced across the islands. Most protected lands correspond to the VI National Park in St. John. The latter is increasing in size over the years as new land tracts are added thanks to an active land acquisition program. The process of assigning a Gap Status category to protected areas was made difficult for various reasons. Most of the identified stewardship or protected areas did not have an active management plan at the time of this report, or land management staff on site or otherwise law enforcement agents to safeguard the areas from current threats such as littering or resource extraction. Hence, we found that in some cases land ownership and management were not a guarantee for biodiversity conservation. For example, some areas had high conservation values, but lack of any kind of protection. Another problem was to assign a conservation category based on land ownership and management. For example, during the writing of this report several areas were under consideration for either development, conservation, or a combination of these activities. Since the existence of current threats to wildlife and natural habitats was subtle to land ownership and management criteria, we did not consider current or potential threats as separate factors. One of the major limitations that we faced during data search and analysis was the lack of accurate spatial information for some of the areas. On the other hand, sometimes we had GIS data for the area boundary but no management information to assign an
99 accurate Gap status. Several times we found some discrepancies between areas boundaries extracted from published reports and those calculated by our team, the same with areas names found in different reports or in the web. Our collaborators from the USVI agreed with the protected areas lists presented in this report. Nevertheless, several of them disagreed with the Gap status that we assigned to some of the areas, since these categories were not always reflecting the most accurate situation. For example, some of the areas classified as 1, 2 or 3, are current targets of land development projects, where portions of some of them had already been illegally burned and illegally bull-dozed. All collaborators agreed that one of the major threats to wildlife and ecosystems in the USVI is land development which results in exposed ground and soil compaction decreasing soil permeability and increasing the water runoff and flooding, and affecting not only biodiversity, but also local human populations.
Acknowledgements Thanks to several collaborators in the USVI who helped to improve the information presented in this report.
References Division of Fish and Wildlife. 1993. Compass Point Salt Pond Declared Marine Reserve. Tropic News 5 (4). IUCN and UNEP. 2009. The World Database on Protected Areas (WDPA). UNEP- WCMC. Cambridge, UK. Kendall, M.S., L.T. Takata, O. Jensen, Z. Hillis-Starr, and M.E. Monaco 2005. An Ecological Characterization of Salt River Bay National Historical Park and Ecological Preserve, USVI. NOAA Tech. Mem. NOS NCCOS 14. 116 pp. Platenberg, R. J., F. E. Hayes, D. B. McNair, and J. J. Pierce. 2005. A Comprehensive Wildlife Conservation Strategy for the U.S. Virgin Islands. Division of Fish and Wildlife, St. Thomas. 251 pp. Weaver, Peter L. 2006. Estate Thomas Experimental Forest, St. Croix, U.S. Virgin Islands: Research history and potential. Gen. Tech. Rep. IITF-30. San Juan. PR: U.S.D.A. Forest Service, International Institute of Tropical Forestry. 62 p.
100
Datasets: - The Nature Conservancy. 2006. Caribbean Decision Support System. - VI National Park Service website. http://www.nps.gov/viis/index.htm
Maps: - Current and Proposed Conservation Lands St Croix, USVI Map from Geographic Consulting, Inc - Office of the Tax Assessor and Cadastral of the USVI.
101
Table 1. Land stewardship and protected areas in the USVI indicating the management class, and also percentage and hectares protected in each Gap category. Protected areas are indicated in bold letters. *These areas are not classified as protected and were not included in the map in figure 1. Status 1 Status 2 Status 3 Status 4 Total
Area name Map ID Management ha % ha % ha % ha % ha % # class Altona Lagoon Beach Recreation Area 61 VI Government 6.8 0.8 6.8 0.1 Booby Rock Wildlife Sanctuary 37 VI Government 0.3 0.1 0.3 >0.1 Bovoni Cay Wildlife Sanctuary 18 VI Government 22.3 2.8 22.3 0.3 Buck Island National Wildlife Refuge 20 Federal 18.7 0.5 18.7 0.3 Buck Island Reef National Monument 66 Federal 70.4 2.0 70.4 1.1 Butler Bay Conservation Easement 40 NGO 32.0 4.0 32.0 0.5 Butler Bay Nature Preserve 41 NGO 46.6 5.8 46.6 0.7 Caledonia Gut 43 VI Government 36.5 4.5 36.5 0.6 Capella Island Wildlife Sanctuary 21 VI Government 9.4 2.2 9.4 0.1 Carval Rock Wildlife Sanctuary 32 VI Government 0.3 0.1 0.3 >0.1 Cas Cay Wildlife Sanctuary 19 VI Government 5.8 0.2 5.8 0.1 Christiansted National Historic Site* - VI Government 2.3 0.1 2.3 >0.1 Cockroach Cay Wildlife Sanctuary 1 VI Government 8.3 0.2 8.3 0.1 Compass Point Pond Marine Reserve and Wildlife Sanctuary (Compass Point Salt Pond) 22 VI Government 3.8 0.1 3.8 0.1 Congo Cay Wildlife Sanctuary 26 VI Government 10.2 0.3 10.2 0.2 Coral Bay Preserve 36 NGO 10.1 1.2 10.1 0.2 Cramer's Park* - VI Government 33.7 2.0 33.7 0.5 Creque Dam 44 VI Government 3.1 0.4 3.1 >0.1 Cricket Rock Wildlife Sanctuary 3 VI Government 1.0 0.2 1.0 >0.1 Derick O. Steinmann Memorial Beach 45 NGO 0.8 0.2 0.8 >0.1 Dog Island Wildlife Sanctuary 30 VI Government 5.3 1.2 5.3 0.1
102
Dutchcap Cay Wildlife Sanctuary 4 VI Government 13.2 0.4 13.2 0.2 East Bay and Point Udall 69 VI Government 54.3 6.7 54.3 0.8 East End Marine Park (Great Pond) 64 VI Government 55.1 1.6 55.1 0.9 Estate Adventure Nature Trail 50 VI Civil Society 2.9 0.7 2.9 >0.1 Estate Barren Spot N/A NGO 0.2 0.0 0.2 >0.1 Estate Barren Spot Wetlands N/A VI Government 19.0 2.3 19.0 0.3 Estate Boetzberg N/A VI Government 7.7 1.0 7.7 0.1 Estate Clairmont Park 52 NGO 95.6 11.8 95.6 1.5 Estate Great Pond 65 VI Government 12.9 1.6 12.9 0.2 Estate Little La Grange 46 NGO 3.8 0.5 3.8 0.1 Estate Little Princess 58 NGO 19.8 2.4 19.8 0.3 Estate Mount Washington Bird Sanctuary 42 NGO 8.0 1.9 8.0 0.1 Estate Prosperity beachfront N/A NGO 0.8 0.1 0.8 >0.1 Estate Thomas 57 Federal 60.0 13.9 60.0 0.9 Estate Whim 48 NGO 5.0 0.6 5.0 0.1 Fairchild Park 16 VI Government 2.8 0.3 2.8 >0.1 Fairleigh Dickinson Territorial Park 68 VI Government 32.4 4.0 32.4 0.5 Flanagan Island Wildlife Sanctuary 39 VI Government 8.5 2.0 8.5 0.1 Flat Cay Wildlife Sanctuary 13 VI Government 1.1 0.0 1.1 >0.1 Frank Bay Marine Reserve and Wildlife Sanctuary 31 VI Government 0.9 0.0 0.9 >0.1 Frenchcap Cay 28 VI Government 6.3 0.2 6.3 0.1 Grass Cay Wildlife Sanctuary 25 VI Government 22.5 2.8 22.5 0.4 Green Cay National Wildlife Refuge 62 Federal 4.4 0.1 4.4 0.1 Henry E. Rohlsen Airport and Landfill* - VI Government 595.0 35.3 595.0 9.3 Herman Hill Pond 59 NGO 6.4 0.8 6.4 0.1 Howard M. Scout Wall Camp* - NGO 6.6 0.4 6.6 0.1
103
Jack and Isaac's Bays Preserve 67 NGO 121.0 28.1 121.0 1.9 Kalkun Cay Wildlife Sanctuary 8 VI Government 1.5 0.0 1.5 >0.1 Leduck Island Wildlife Sanctuary 38 VI Government 5.8 1.3 5.8 0.1 Little St. Thomas 10 NGO 0.3 0.1 0.3 >0.1 Long Point Bay 49 NGO 8.0 1.9 8.0 0.1 Longpoint & Cotton Garden N/A VI Government 69.6 8.6 69.6 1.1 VI Government Magen's Bay Preserve 17 / NGO 126.6 29.4 126.6 2.0 Manning Bay Wetlands 51 VI Government 30.0 3.1 30.0 0.5 Nancy Spire Nature Preserve N/A NGO 17.9 2.2 17.9 0.3 National Guard Armory* - Federal 21.8 1.3 21.8 0.3 National Guard Nazareth Facility* - VI Government 5.9 0.4 5.9 0.1 Outer Brass Island Wildlife Sanctuary 14 VI Government 43.8 5.4 43.8 0.7 Perkins Cay Wildlife Sanctuary 33 VI Government 0.2 0.0 0.2 >0.1 Protestant Cay Wildlife Sanctuary 60 VI Government 2.7 0.3 2.7 >0.1 Ruth Cay Wildlife Sanctuary 53 VI Government 10.2 1.3 10.2 0.2 Saba Island Wildlife Sanctuary 12 VI Government 12.7 3.0 12.7 0.2 Sail Rock Wildlife Sanctuary 7 VI Government 1.1 0.1 1.1 >0.1 Salt Cay Wildlife Sanctuary 5 VI Government 24.0 3.0 24.0 0.4 Salt River Bay National Historic Federal / VI Park and Ecological Preserve 54 Government 166.7 4.8 166.7 2.6 Salt River, Estate Montpellier N/A NGO 0.2 0.0 0.2 >0.1 Sandy Point National Wildlife Refuge 47 Federal 221.8 6.4 221.8 3.5 Savana Island Wildlife Sanctuary 6 VI Government 70.6 8.7 70.6 1.1 Shark Island Wildlife Sanctuary 24 VI Government 0.4 0.1 0.4 >0.1 Sion Ridge Area (Park Service Warehouse) 56 Federal 5.8 0.7 5.8 0.1 Smith Bay 23 NGO 8.6 2.0 8.6 0.1 Southgate Pond Coastal Preserve 63 NGO 42.0 9.8 42.0 0.7
104
Spratt Bay Estates 15 NGO 12.2 1.5 12.2 0.2 Spring Garden N/A VI Government 28.7 3.5 28.7 0.4 Steven Cay Wildlife Sanctuary 29 VI Government 2.1 0.1 2.1 >0.1 St. George Village Botanical Garden* - NGO 11.1 0.7 11.1 0.2 St. Peter Great House and Botanical Garden* - NGO 2.7 0.2 2.7 >0.1 Sula Cay Wildlife Sanctuary 2 VI Government 1.0 0.0 1.0 >0.1 Turtledove Cay Wildlife Sanctuary 11 Federal 1.7 0.0 1.7 >0.1 Two Brothers 27 VI Government 0.2 0.0 0.2 >0.1 University of the Virgin Islands* - VI Government 56.0 3.3 56.0 0.9 University of the Virgin Islands and Agriculture Experimental Station* - VI Government 89.5 5.3 89.5 1.4 U.S. Virgin Islands National Park (including cays) 35 Federal 2875.3 82.9 2875.3 45.0 USVI Department of Agriculture* - VI Government 840.8 49.9 840.8 13.2 UVI Wetlands 55 VI Government 35.0 4.3 35.0 0.5 VI National Guard (Hams Bluff)* - Federal 19.9 1.2 19.9 0.3 West Cay 9 VI Government 16.7 2.1 16.7 0.3 Whistling Cay 34 VI Government 8.3 1.9 8.3 0.1 Summary Total stewardship area (ha) 3468.3 430.4 809.1 1685.3 6393.1 100 Stewardship areas in the USVI (%) 9.9 1.2 2.3 4.8 18.3 Protected Areas (ha) 3468.3 430.4 809.1 - 4707.8 Protected Areas (%) 9.9 1.2 2.3 13.5 Protected Areas for each category (%) 73.7 9.1 17.2 - 100 USVI Total area (ha) 34972.0 100
105
Figure 1- Map of protected areas in the USVI with Gap status 1, 2 and 3. Some other protected areas may exist. Some of the protected areas that we identified were mapped.
106
Figure 2. Percentage of protected areas in Gap status 1, 2, 3 and 4.
107
Figure 3 – Map of protected areas ownership in the USVI.
108
Figure 4- Protected areas management in the USVI.
109
8000 Total hectares Hectares under Gap 1,2 and 3 7000
6000
5000
4000
Area (ha) 3000
2000
1000
0
USVI Port Authority U. S Forest Service USVI DSPR and DPNR U.S National Park Service The Nature Conservancy St. Croix HistoricalSt. Croix LandmarkSociety Society U.S. Fish and WildlifeUniversity Service of the Virgin Islands USVI Department of Agriculture
Figure 4. Major land managers in the USVI. Total hectares managed (dark green) and hectares (light green) managed for biodiversity conservation (Gap status 1-3).
110
Puerto Rico and U.S. Virgin Islands Gap Analysis Project – Land Stewardship Survey
Introduction
This survey was prepared by the PRUSVI Gap Analysis Project (PRUSVI GAP) to be used in the U.S. Virgin Islands land stewardship assessment. The information provided here will be integrated into the PRUSVI GAP database and will be used for classifying USVI protected areas into one of 4 categories based on the amount of protection they provide to biodiversity. Please answer the survey as best you can. You can always add comments to your answers. If there are documents that you could provide us or that are available on the web that can help us fill out specific topics (e.g. the description of the area) please let us know. Please contact PRUSVI GAP coordinator Mariano Solórzano at [email protected] or at 787-766-5335 x. 303 if you have any doubts.
Name of the Area:______
Name of Individual Completing Survey ______Date______Position: ______Contact info: ______Number of Years working at site______
Staff info: Staff Yes / No On Site Off Site Management Research Law enforcement Interpretive
Stewardship
Land Manager (If multiple managers please include names of the different agencies or organizations): Land Owner (If multiple owners please include names of the different agencies or organizations): Date of Establishment: Land Manager Official: Area Acres: Area Hectares: Short description of the area: Contact info of the area (Phone/email/mailing address): Q: Is the area subdivided into different management units under different management?
A:
Q: Are there areas such as wilderness areas that restrict human uses to a minimum?
111
A:
Q: Is there a Mandated management plan in operation to maintain a natural state? (Yes/No):
A:
Q: Is there Permanent protection from conversion of natural land cover?
A:
Q: Is there Statutory or legally enforceable protection from conversion to anthropogenic (related to or influenced by the impact of humans on nature) use of all or selected biological features? A:
If YES, please ‘X’ the applicable criteria: Type of protection Protection of ALL features Protection of SELECTED features State or Federal LEGISLATION State or Federal REGULATION Private deed restriction Conservation easement intended for
permanent status
Land Management
Please ‘X’ the applicable criteria Management Practice Yes No Comments (Optional) Wildlife management: Water monitoring: Manage threatened and endangered species: Fire monitoring: Wildlife monitoring: Disease monitoring: Vegetation monitoring: Soil monitoring: Exotic plants monitoring: Exotic plant control: Exotic animals monitoring: Exotic animal control: Water level control: Wetland restoration: Reforestation: Trail maintenance: Nest structures: Trash clean-up: Hunting:
112
Management Practice Yes No Comments (Optional) Grazing: Haying: Education: Community outreach: Law enforcement: Are natural disturbance events allowed to proceed without interference? Are natural disturbance events mimicked through management practices? Are there prescribed burns? Are natural resources removed from the area?
Natural Resources
If YES, please ‘X’ the applicable criteria Moist lowland forest: Coral reef: Dry forest: Lake: Grassland: Pond: Mature forest (>60 years): River: Secondary forest: Karst topography Wet land: Estuary: Other(s): Archeological site:
Facilities
If YES, please ‘X’ the applicable criteria Yes On site Off site None: Laboratory: Main Office: Visitors Center: Library: Plant Nursery: Staff Housing: Other:
Public Use Opportunities
If YES, please ‘X’ the applicable criteria Hiking Trails: Kayaking:
113
Picnic Shelters: Hunting: Camp Grounds: Diving: Biking: Wildlife observation: Boating: Volunteer program Fishing: Brochure available Other(s): Archeological site:
No. of Annual Visitors:______
Resources for Staff If YES, please ‘X’ the applicable criteria Computers: Vehicle: GIS: Maps: Plotters: Field Guide Books: Boat: Species Information: Computers: Other:
Threats to Biodiversity If YES, please ‘X’ the applicable criteria Habitat Loss: Urban Development: Fragmentation: Plant collection: Deforestation: Wild animal trade: Mineral extraction: Cutting down of vegetation: Oil spills: Illegal Dumping: Water contamination: Grazing without a permit Soil contamination: Clandestine construction: Air pollution: Trespassing: Diseases: Crab trapping: Fungal disease: Deep see diving: Insect infestations: Snorkeling: Flooding: ATVs: Hurricanes/Depressions: Over hunting: Sedimentation: Water sports: Erosion: Over fishing: Landslides: Camp Fires: Nonnative species: Boating: Feral cats and dogs: Over tourism: Mongoose: Other(s):
Research (Yes/No)
Q: Is a Bibliography of the research carried out in the area available?
114
A:
Q: Are there Electronic reprints of the research papers or reports available?
A:
Q: Is there a Species list of the area available?
A:
Thanks for completing this survey! If you have other comments, please use space below. You can send the survey to Mariano Solórzano via e-mail [email protected] or mail it to:
Mariano Solórzano International Institute of Tropical Forestry 1209 Calle Ceiba, Jardín Botánico Sur, San Juan, PR 00926-1119
115
ANALYSIS BASED ON STEWARDSHIP AND MANAGEMENT STATUS Mariano C. Solórzano Cartographic Technician USVI Gap Analysis Project Project Coordinator
Introduction This chapter describes the method and results of the gap analysis as used by the Gap Analysis Program. As described in the general introduction to this report, the primary objective of GAP is to provide information on the distribution and status of several elements of biological diversity. Although GAP "seeks to identify habitat types and species not adequately represented in the current network of biodiversity management areas" (GAP Handbook, Preface, Version 1, p. I), it is unrealistic to create a standard definition of "adequate representation" for either land cover types or individual species (Noss et al. 1995). A practical solution to this problem is to report both percentages and absolute area of each element in biodiversity management areas and allow the user to determine which classes are adequately represented in natural areas. There are many other factors that should be considered in such determinations such as: • Historic loss or gain in distribution, • Nature of the spatial distribution, • Immediate versus long term risk, and • Degree of local adaptation among populations of the biotic elements that are worthy of individual conservation consideration. Such analyses are beyond the scope of this project, but we encourage their application coupled with field confirmation of the mapped distributions.
Currently, land cover types and terrestrial vertebrates are the primary focus of GAP's mapping efforts. However, other components of biodiversity, such as aquatic organisms or selected groups of invertebrates may be incorporated into GAP distributional data sets. Where appropriate, GAP data may also be analyzed to identify the location of a set of areas in which most or all land cover types or species are predicted to be represented. The use of "complementarity" analysis, that is, an approach that additively identifies a selection of locations that may represent biodiversity rather than "hot spots of species richness" may prove most effective for guiding biodiversity maintenance efforts. Several quantitative techniques have been developed that facilitate this process (see Pressey et al. 1993, Williams et al. 1996, Csuti et al. 1997, for details). These areas become candidates for field validation and may be incorporated into a system of areas managed for the long-term maintenance of biological diversity.
The network of Conservation Data Centers (CDCs) and Natural Heritage Programs
116
(NHPs) established cooperatively by The Nature Conservancy and various state agencies maintain detailed databases on the locations of rare elements of biodiversity. GAP cooperatively uses these data to develop predicted distributions of potentially suitable habitat for these elements, which may be valuable for identifying research needs and preliminary considerations for restoration or reintroduction. Conservation of such elements, however, is best accomplished through the fine-filter approach of the above organizations as described in the introduction. It is not the role of GAP to duplicate or disseminate Heritage Program or CDC Element Occurrence Records. Users interested in more specific information about the location, status, and ecology of populations of such species, are directed to their state Heritage Program or CDC.
Methods The gap analysis is accomplished by first producing: Maps of land cover (Map 2), predicted habitat distributions for selected animal species (Appendix 7), and maps of the land stewardship and management status (Map 5). By extracting or intersecting the habitat distribution of the natural elements that lie within the protected areas we were able to create tables that summarize the area and percent of total mapped distribution of each element in different land stewardship and management categories. We created a raster image of the protected areas Gap status categories 1 through 4 using ESRI ArcGIS 9.3. Then we intersected this with our two land cover datasets (USVI Gap Landcover and USVI NVC Landcover) and with our individual species predicted habitat distributions. The resulting output includes the number of hectares of each species or land cover unit in each Gap status category.
Results The data are provided in a format that allows users to carry out inquiries about the representation of each element in different land stewardship and management categories as appropriate to their own management objectives. This forms the basis of Gap's mission to provide land owners and managers with the information necessary to conduct informed policy development, planning, and management for biodiversity maintenance.
Land Cover Analysis Since we classified the land cover classes in two different landcover raster datasets we developed two sets of tables summarizing the results: One set of tables summarizes the results for the U.S. Virgin Islands Gap Landcover (Figures 4 to 6) starting with table 11 through table 15; and one set of tables summarizes the results for the U.S National Vegetation Classification of the U.S. Virgin Islands Landcover (Figures 7 to 9) starting with table 16 through table 21 provides the area in hectares of each classes’ mapped distribution and the percent of the classes' total distribution in each category. For example, a typical entry may indicate that mangrove forests and shrublands cover 340.3 ha of the territory in total and that 102 ha are ranked Status 1 or 2, which represents 30% of that type's total distribution.
117
As a coarse indicator of the status of the elements, we provide a breakdown along five levels of representation (0-<1%, 1-<10%, 10-<20%, and 20-<50%; >=50%). The <1% level indicates those elements with essentially none of their distribution in a protected status while levels of 10%, 20%, and 50% have been recommended in the literature as necessary amounts of conservation (Noss and Cooperrider 1994; Noss 1991; Odum and Odum 1972; Specht et al. 1974; Ride 1975; Miller 1994). Summaries of the analysis according to the thresholds described above are shown in Table 12-16 below.
Predicted Animal Species Distributions Analysis Table 22 provides the area in hectares of each species mapped distribution and the percent of the species total distribution in each category. For example, a typical entry may indicate that the Antillean crested hummingbird, Orthorhyncus cristatus, habitat covers 19662.33 hectares of the islands in total (55.9% of the islands), and that 2251.2 hectares of that habitat is classified as Gap status 1 and 2, which represents 11.45 percent of that species total habitat distribution.
As a coarse indicator of the status of the elements, we provide a breakdown along five levels of representation (<1%, 1% to <10%, 10% to <20%, and 20% to <50%; and >=50%). The <1% level indicates those elements with essentially none of their distribution in a protected status while levels of 10%, 20%, and 50% have been recommended in the literature as necessary amounts of conservation (Noss and Cooperrider 1994; Noss 1991; Odum and Odum 1972; Specht et al. 1974; Ride 1975; Miller 1994). Summaries of the analysis according to the thresholds described above are shown in Table 18-22 below. Table 23 lists the locally or federally listed endangered, endemic, and vulnerable species.
118
Table 11. Area in hectares and percent for each USVI Gap land cover type for the whole island and by Gap Status categories 1-4.
Protected Total habitat Status 1 or 2 Status 1 Status 2 Status 3 Status 4 Landcover type % HA % HA % HA % HA % HA % HA U.S. Virgin Islands 100.00 35162.97 11.09 3898.57 9.85 3461.80 1.24 436.77 1.94 682.01 86.97 30582.39 Dry Alluvial Evergreen Gallery Forest 0.21 74.72 8.46 6.32 7.33 5.48 1.12 0.84 0.27 0.20 91.27 68.20 Dry Alluvial Semideciduous Forest 0.42 147.18 33.12 48.74 27.48 40.45 5.63 8.29 0.75 1.11 66.13 97.33 Dry Alluvial Shrubland 5.12 1801.24 6.91 124.45 5.96 107.41 0.95 17.04 1.53 27.58 91.56 1649.21 Dry Alluvial Open Shrubland 2.36 828.1 5.29 43.84 3.89 32.25 1.40 11.59 0.45 3.72 94.26 780.54 Dry Alluvial Woodland 1.44 506.71 9.49 48.10 8.19 41.49 1.30 6.61 0.83 4.23 89.67 454.38 Dry Limestone Evergreen Gallery Forest 0.33 117.26 28.36 33.26 12.58 14.75 15.79 18.51 1.24 1.45 70.40 82.55 Dry Limestone Semideciduous Forest 0.32 111.29 28.57 31.79 8.84 9.84 19.72 21.95 1.71 1.90 69.73 77.60 Dry Limestone Shrubland 4.62 1624.56 5.10 82.81 4.03 65.42 1.07 17.39 1.55 25.12 93.36 1516.63 Dry Limestone Open Shrubland 0.57 200.51 4.85 9.72 4.24 8.51 0.60 1.21 3.94 7.90 91.21 182.89 Dry Limestone Woodland 1.04 366.36 6.53 23.91 4.14 15.15 2.39 8.76 1.01 3.71 92.46 338.74 Dry Noncalcareous Evergreen Gallery Forest 2.42 851.11 7.80 66.37 7.75 66.00 0.04 0.37 3.42 29.09 88.78 755.65 Dry Noncalcareous Semideciduous Forest 11.50 4042.09 33.96 1372.70 32.55 1315.67 1.41 57.03 0.98 39.61 65.06 2629.78 Dry Noncalcareous Shrubland 17.97 6319.14 7.30 461.47 6.47 409.14 0.83 52.33 2.68 169.29 90.02 5688.38 Dry Noncalcareous Open Shrubland 2.66 934.4 10.07 94.13 6.48 60.56 3.59 33.57 4.36 40.71 85.57 799.56 Dry Noncalcareous Woodland 9.71 3415.45 9.56 326.41 8.84 301.91 0.72 24.50 2.39 81.71 88.05 3007.33 Lowland Moist Alluvial Evergreen Gallery Forest 0.24 84.49 52.74 44.56 48.35 40.85 4.39 3.71 0.00 0.00 47.26 39.93 Lowland Moist Noncalcareous Evergreen Forest 1.18 414.7 37.70 156.33 37.63 156.07 0.06 0.26 1.06 4.41 61.24 253.96 Lowland Moist Noncalcareous Evergreen Gallery Forest 2.58 905.53 32.44 293.71 30.98 280.55 1.45 13.16 2.81 25.49 64.75 586.33 Lowland Moist Noncalcareous Shrubland 0.25 86.51 0.55 0.48 0.55 0.48 0.00 0.00 0.05 0.04 99.40 85.99 Lowland Moist Noncalcareous Open Shrubland 0.01 2.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 2.03 Lowland Moist Noncalcareous Woodland 0.34 119.63 2.57 3.07 2.54 3.04 0.03 0.03 0.07 0.08 97.37 116.48 Seasonally Flooded Nonsaline Forest 0.36 125.61 45.50 57.15 39.61 49.76 5.88 7.39 0.18 0.23 54.32 68.23 Seasonally Flooded Nonsaline Shrubland 0.14 49.8 27.43 13.66 26.24 13.07 1.18 0.59 0.68 0.34 71.89 35.80 Seasonally Flooded Saline Forest 0.04 13.88 30.04 4.17 26.95 3.74 3.10 0.43 3.46 0.48 66.50 9.23 Seasonally Flooded Saline Shrubland 0.40 139.11 15.27 21.24 12.13 16.88 3.13 4.36 8.48 11.79 76.26 106.08 Mangrove Forest and Shrubland 0.97 340.29 29.95 101.90 25.07 85.32 4.87 16.58 10.14 34.52 59.91 203.87 Dry Grassland and Pastures 6.94 2441.69 4.24 103.54 1.85 45.24 2.39 58.30 2.70 65.92 93.06 2272.23 Moist Grassland and Pastures 0.00 1.15 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 1.15 Seasonally Flooded Herbaceous Nonsaline Wetlands 0.03 11.76 2.30 0.27 2.30 0.27 0.00 0.00 1.28 0.15 96.43 11.34 Seasonally Flooded Herbaceous Saline Wetlands 0.15 52.22 16.64 8.69 13.02 6.80 3.62 1.89 7.87 4.11 75.49 39.42 Emergent Herbaceous Saline Wetlands 0.00 0.3 40.00 0.12 3.33 0.01 36.67 0.11 0.00 0.00 60.00 0.18 Emergent Herbaceous Nonsaline Wetlands 0.03 11.03 7.52 0.83 7.52 0.83 0.00 0.00 3.99 0.44 88.49 9.76 Hay and Row Crops 0.61 215.03 0.06 0.13 0.00 0.00 0.06 0.13 0.00 0.00 99.94 214.90 Woody Agriculture and Plantations 0.02 8.75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 8.75 Artificial Barrens 0.57 202.17 0.27 0.55 0.24 0.49 0.03 0.06 0.08 0.16 99.65 201.46 Fine to Medium Grained Sandy Beaches 0.27 94.5 30.92 29.22 24.47 23.12 6.46 6.10 4.95 4.68 64.13 60.60 Gravel Beaches 0.10 36.84 29.07 10.71 26.19 9.65 2.88 1.06 5.46 2.01 65.47 24.12 Mixed Sand and Gravel Beaches 0.11 38.2 19.03 7.27 14.90 5.69 4.14 1.58 9.40 3.59 71.57 27.34 Riparian and Other Natural Barrens 0.01 4.48 8.26 0.37 7.37 0.33 0.89 0.04 3.35 0.15 88.39 3.96 119
Riprap 0.05 17.82 1.85 0.33 1.85 0.33 0.00 0.00 0.00 0.00 98.15 17.49 Rocky Cliffs and Shelves 1.05 368.35 23.60 86.92 18.52 68.23 5.07 18.69 13.33 49.11 63.07 232.32 Salt and Mudflats 0.30 104.88 20.79 21.80 18.57 19.48 2.21 2.32 4.70 4.93 74.51 78.15 Maintained Grassland 9.88 3475.27 0.61 21.22 0.53 18.36 0.08 2.86 0.48 16.72 98.91 3437.33 Low-density Urban Development 2.39 841.91 2.59 21.78 2.16 18.21 0.42 3.57 0.97 8.13 96.45 812.00 Medium-density Urban Development 6.02 2116.59 0.39 8.25 0.39 8.21 0.00 0.04 0.03 0.61 99.58 2107.73 High-density Urban Development 3.23 1134.88 0.01 0.09 0.01 0.09 0.00 0.00 0.00 0.00 99.99 1134.79 Aquaculture 0.00 0.75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 0.75 Fresh Water 0.11 40.43 1.26 0.51 1.21 0.49 0.05 0.02 0.52 0.21 98.22 39.71 Salt Water 0.92 322.27 32.79 105.68 28.60 92.18 4.19 13.50 1.98 6.38 65.23 210.21
120
Table 12. USVI Gap land cover classes with less than 1% of their total area protected, sorted by the amount of habitat. Total areas for U.S. Virgin Islands are given in the first row. Total area of these land cover classes are given in the final row. Percent of total of these classes on the island (first column) is followed by percent of each type in Status 1-4.
Protected Total habitat Status 1 or 2 Status 1 Status 2 Status 3 Status 4 Landcover type % HA % HA % HA % HA % HA % HA U.S. Virgin Islands 100.00 35162.97 11.09 3898.57 9.85 3461.80 1.24 436.77 1.94 682.01 86.97 30582.39 Lowland Moist Noncalcareous Shrubland 0.2460259 86.51 0.55 0.48 0.55 0.48 0.00 0.00 0.05 0.04 99.40 85.99 Lowland Moist Noncalcareous Open Shrubland 0.0057731 2.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 2.03 Moist Grassland and Pastures 0.0032705 1.15 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 1.15 Hay and Row Crops 0.611524 215.03 0.06 0.13 0.00 0.00 0.06 0.13 0.00 0.00 99.94 214.90 Woody Agriculture and Plantations 0.0248841 8.75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 8.75 Artificial Barrens 0.5749514 202.17 0.27 0.55 0.24 0.49 0.03 0.06 0.08 0.16 99.65 201.46 Maintained Grassland 9.8833233 3475.27 0.61 21.22 0.53 18.36 0.08 2.86 0.48 16.72 98.91 3437.33 Medium-density Urban Development 6.0193721 2116.59 0.39 8.25 0.39 8.21 0.00 0.04 0.03 0.61 99.58 2107.73 High-density Urban Development 3.2274862 1134.88 0.01 0.09 0.01 0.09 0.00 0.00 0.00 0.00 99.99 1134.79 Aquaculture 0.0021329 0.75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 0.75
121
Table 13. USVI Gap land cover classes with 1% to less than 10% of their total area protected, sorted by the amount of habitat. Total areas for U.S. Virgin Islands are given in the first row. Total area of these land cover classes are given in the final row. Percent of total of these classes on the island (first column) is followed by percent of each type in Status 1-4.
Protected Total habitat Status 1 or 2 Status 1 Status 2 Status 3 Status 4 Landcover type % HA % HA % HA % HA % HA % HA U.S. Virgin Islands 100.00 35162.97 11.09 3898.57 9.85 3461.80 1.24 436.77 1.94 682.01 86.97 30582.39 Dry Alluvial Evergreen Gallery Forest 0.21 74.72 8.46 6.32 7.33 5.48 1.12 0.84 0.27 0.20 91.27 68.20 Dry Alluvial Shrubland 5.12 1801.24 6.91 124.45 5.96 107.41 0.95 17.04 1.53 27.58 91.56 1649.21 Dry Alluvial Open Shrubland 2.36 828.1 5.29 43.84 3.89 32.25 1.40 11.59 0.45 3.72 94.26 780.54 Dry Alluvial Woodland 1.44 506.71 9.49 48.10 8.19 41.49 1.30 6.61 0.83 4.23 89.67 454.38 Dry Limestone Shrubland 4.62 1624.56 5.10 82.81 4.03 65.42 1.07 17.39 1.55 25.12 93.36 1516.63 Dry Limestone Open Shrubland 0.57 200.51 4.85 9.72 4.24 8.51 0.60 1.21 3.94 7.90 91.21 182.89 Dry Limestone Woodland 1.04 366.36 6.53 23.91 4.14 15.15 2.39 8.76 1.01 3.71 92.46 338.74 Dry Noncalcareous Evergreen Gallery Forest 2.42 851.11 7.80 66.37 7.75 66.00 0.04 0.37 3.42 29.09 88.78 755.65 Dry Noncalcareous Shrubland 17.97 6319.14 7.30 461.47 6.47 409.14 0.83 52.33 2.68 169.29 90.02 5688.38 Dry Noncalcareous Woodland 9.71 3415.45 9.56 326.41 8.84 301.91 0.72 24.50 2.39 81.71 88.05 3007.33 Dry Grassland and Pastures 6.94 2441.69 4.24 103.54 1.85 45.24 2.39 58.30 2.70 65.92 93.06 2272.23 Seasonally Flooded Herbaceous Nonsaline Wetlands 0.03 11.76 2.30 0.27 2.30 0.27 0.00 0.00 1.28 0.15 96.43 11.34 Emergent Herbaceous Nonsaline Wetlands 0.03 11.03 7.52 0.83 7.52 0.83 0.00 0.00 3.99 0.44 88.49 9.76 Riparian and Other Natural Barrens 0.01 4.48 8.26 0.37 7.37 0.33 0.89 0.04 3.35 0.15 88.39 3.96 Riprap 0.05 17.82 1.85 0.33 1.85 0.33 0.00 0.00 0.00 0.00 98.15 17.49 Low-density Urban Development 2.39 841.91 2.59 21.78 2.16 18.21 0.42 3.57 0.97 8.13 96.45 812.00 Fresh Water 0.11 40.43 1.26 0.51 1.21 0.49 0.05 0.02 0.52 0.21 98.22 39.71
122
Table 14. USVI Gap land cover classes with 10% to less than 20% of their total area protected, sorted by the amount of habitat. Total areas for U.S. Virgin Islands are given in the first row. Total area of these land cover classes are given in the final row. Percent of total of these classes on the island (first column) is followed by percent of each type in Status 1-4.
Protected Total habitat Status 1 or 2 Status 1 Status 2 Status 3 Status 4 Landcover type % HA % HA % HA % HA % HA % HA U.S. Virgin Islands 100.00 35162.97 11.09 3898.57 9.85 3461.80 1.24 436.77 1.94 682.01 86.97 30582.39 Dry Noncalcareous Open Shrubland 2.66 934.4 10.07 94.13 6.48 60.56 3.59 33.57 4.36 40.71 85.57 799.56 Seasonally Flooded Saline Shrubland 0.40 139.11 15.27 21.24 12.13 16.88 3.13 4.36 8.48 11.79 76.26 106.08 Seasonally Flooded Herbaceous Saline Wetlands 0.15 52.22 16.64 8.69 13.02 6.80 3.62 1.89 7.87 4.11 75.49 39.42 Mixed Sand and Gravel Beaches 0.11 38.2 19.03 7.27 14.90 5.69 4.14 1.58 9.40 3.59 71.57 27.34
123
Table 15. USVI Gap land cover classes with 20% to less than 50% of their total area protected, sorted by the amount of habitat. Total areas for the U.S. Virgin Islands are given in the first row. Total area of these land cover classes are given in the final row. Percent of total of these classes on the island (first column) is followed by percent of each type in Status 1-4.
Protected Total habitat Status 1 or 2 Status 1 Status 2 Status 3 Status 4 Landcover type % HA % HA % HA % HA % HA % HA U.S. Virgin Islands 100.00 35162.97 11.09 3898.57 9.85 3461.80 1.24 436.77 1.94 682.01 86.97 30582.39 Dry Alluvial Semideciduous Forest 0.42 147.18 33.12 48.74 27.48 40.45 5.63 8.29 0.75 1.11 66.13 97.33 Dry Limestone Evergreen Gallery Forest 0.33 117.26 28.36 33.26 12.58 14.75 15.79 18.51 1.24 1.45 70.40 82.55 Dry Limestone Semideciduous Forest 0.32 111.29 28.57 31.79 8.84 9.84 19.72 21.95 1.71 1.90 69.73 77.60 Dry Noncalcareous Semideciduous Forest 11.50 4042.09 33.96 1372.70 32.55 1315.67 1.41 57.03 0.98 39.61 65.06 2629.78 Lowland Moist Noncalcareous Evergreen Forest 1.18 414.70 37.70 156.33 37.63 156.07 0.06 0.26 1.06 4.41 61.24 253.96 Lowland Moist Noncalcareous Evergreen Gallery Forest 2.58 905.53 32.44 293.71 30.98 280.55 1.45 13.16 2.81 25.49 64.75 586.33 Seasonally Flooded Nonsaline Forest 0.36 125.61 45.50 57.15 39.61 49.76 5.88 7.39 0.18 0.23 54.32 68.23 Seasonally Flooded Nonsaline Shrubland 0.14 49.80 27.43 13.66 26.24 13.07 1.18 0.59 0.68 0.34 71.89 35.80 Seasonally Flooded Saline Forest 0.04 13.88 30.04 4.17 26.95 3.74 3.10 0.43 3.46 0.48 66.50 9.23 Mangrove Forest and Shrubland 0.97 340.29 29.95 101.90 25.07 85.32 4.87 16.58 10.14 34.52 59.91 203.87 Emergent Herbaceous Saline Wetlands 0.00 0.30 40.00 0.12 3.33 0.01 36.67 0.11 0.00 0.00 60.00 0.18 Fine to Medium Grained Sandy Beaches 0.27 94.50 30.92 29.22 24.47 23.12 6.46 6.10 4.95 4.68 64.13 60.60 Gravel Beaches 0.10 36.84 29.07 10.71 26.19 9.65 2.88 1.06 5.46 2.01 65.47 24.12 Rocky Cliffs and Shelves 1.05 368.35 23.60 86.92 18.52 68.23 5.07 18.69 13.33 49.11 63.07 232.32 Salt and Mudflats 0.30 104.88 20.79 21.80 18.57 19.48 2.21 2.32 4.70 4.93 74.51 78.15 Salt Water 0.92 322.27 32.79 105.68 28.60 92.18 4.19 13.50 1.98 6.38 65.23 210.21
124
125
As explained above, we provide results according to thresholds of representation advocated in the literature to conserve biodiversity. The values in the tables will allow users to set any desirable threshold to identify elements requiring more protection according to their criteria. The following summaries highlight potential gaps and conservation needs.
Land Cover with <1% representation in GAP status 1 and 2 (Table 12): Within these 10 land cover classes we find the smallest in extent: Moist grasslands and Aquaculture. All these classes put together make up 20.6 % of the islands. They are primarily subject to intensive human use such as agriculture, housing, and development. However, lowland shrublands may also represent an intermediate step in the transition from lands that were historically agricultural to forests. The class of moist grasslands and pastures covers very little ground compared with the same class in Puerto Rico where it nearly one quarter of the island and is primarily abandoned agricultural land. As a group, 0.4% of these classes is protected under Gap Status 1 or 2, or 30.7 ha out of a total of 7,243 ha.
Land Cover with 1% to <10% representation in GAP status 1 and 2 (Table 13): Seventeen land cover units fall in this category and they account for 55% of the island. They range from an extent of less than 1% to 18% of the island. They contain a number of forests, woodlands and shrublands land cover classes, as well as riparian and natural barrens, dry grasslands and pastures, some wetlands, low-density urban development, freshwater bodies. Dry grasslands and pastures represent the class with the highest area cover and amount to 18% or 6320 hectares This land cover classes has suffered major changes in the past decades (Daley 2005). As a group 6.8% of these units are protected under Gap Status 1 or 2.
Land Cover with l0% to <20% representation in GAP status 1 and 2 (Table 14): Four land unit classes fall in this category and they account for only 3.3% of the island. They include one dry shrubland, one seasonally flooded shrubland a saline wetland, and mixed sand and gravel beaches. As a group they are 11.3% protected under Gap Status 1 and 2.
Land Cover with 20% to <50% representation in GAP status 1 and 2 (Table 15): Sixteen land cover classes fall in this group and they account for 20.5% of the island. However, most of the classes in this range represent less than 1% of the islands territory. They include a number of ecologically important areas including some dry forests, seasonally flooded forests, mangroves, some wetlands, salt and mud flats, gravel beaches and fine sandy beaches salt ponds. As a group they are 32.9% protected under Gap Status 1 and 2.
Land Cover with at least 50% representation in GAP status 1 and 2: None of the USVI Gap land cover units are over 50% protected under Gap Status 1 and 2.
126
Table 16. Area in hectares and percent for each land cover type (USVI NVC Landcover) for the whole USVI territory and by Gap Status categories 1-4.
Protected Total habitat Status 1 or 2 Status 1 Status 2 Status 3 Status 4 U.S. National Vegetation Classification % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.00 35162.98 11.09 3898.58 9.85 3461.81 1.24 436.77 1.94 682.01 86.97 30582.39 G585 Caribbean Dry Broadleaf Forest 24.43 8589.08 21.56 1851.65 20.08 1724.51 1.48 127.14 1.54 132.27 76.90 6605.16 G455 Caribbean Seasonal Evergreen Moist Lowland Forest 3.31 1162.72 9.38 109.02 7.68 89.27 1.70 19.75 2.65 30.82 87.97 1022.88 G454 Caribbean Moist Lowland-Submontane Forest Group 3.99 1404.72 35.21 494.60 33.99 477.47 1.22 17.13 2.13 29.90 62.66 880.22 G002 Caribbean Hardwood Swamp Group 0.93 328.40 29.30 96.22 25.41 83.45 3.89 12.77 3.91 12.84 66.79 219.34 G004 Caribbean Mangrove Tidal Swamp Group 0.97 340.29 29.95 101.90 25.07 85.32 4.87 16.58 10.14 34.52 59.91 203.87 D094 Caribbean and Central American Lowland Shrubland, Grassland and Savanna 40.68 14303.32 6.50 929.41 5.15 736.09 1.35 193.32 2.41 344.54 91.09 13029.37 M041 Caribbean and Central American Freshwater Marsh 0.03 11.33 8.38 0.95 7.41 0.84 0.97 0.11 3.88 0.44 87.73 9.94 Agricultural Vegetation 0.64 223.78 0.06 0.13 0.00 0.00 0.06 0.13 0.00 0.00 99.94 223.65 Fine to Medium Grained Sandy Beaches 0.27 94.50 30.92 29.22 24.47 23.12 6.46 6.10 4.95 4.68 64.13 60.60 Gravel Beaches 0.10 36.84 29.07 10.71 26.19 9.65 2.88 1.06 5.46 2.01 65.47 24.12 Mixed Sand and Gravel Beaches 0.11 38.20 19.03 7.27 14.90 5.69 4.14 1.58 9.40 3.59 71.57 27.34 Riparian and other natural barrens 0.01 4.48 8.26 0.37 7.37 0.33 0.89 0.04 3.35 0.15 88.39 3.96 Riprap 0.05 17.82 1.85 0.33 1.85 0.33 0.00 0.00 0.00 0.00 98.15 17.49 Rock Cliffs and Shelves 1.05 368.35 23.60 86.92 18.52 68.23 5.07 18.69 13.33 49.11 63.07 232.32 Salt and Mudflats 0.30 104.88 20.79 21.80 18.57 19.48 2.21 2.32 4.70 4.93 74.51 78.15 Artificial Barrens 0.57 202.17 0.27 0.55 0.24 0.49 0.03 0.06 0.08 0.16 99.65 201.46 Maintained Grassland 9.88 3475.27 0.61 21.22 0.53 18.36 0.08 2.86 0.48 16.72 98.91 3437.33 Low-Density Urban Development 2.39 841.91 2.59 21.78 2.16 18.21 0.42 3.57 0.97 8.13 96.45 812.00 Medium-Density Urban Development 6.02 2116.59 0.39 8.25 0.39 8.21 0.00 0.04 0.03 0.61 99.58 2107.73 High-density Urban Development 3.23 1134.88 0.01 0.09 0.01 0.09 0.00 0.00 0.00 0.00 99.99 1134.79 Aquaculture 0.00 0.75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 0.75 Fresh water 0.11 40.43 1.26 0.51 1.21 0.49 0.05 0.02 0.52 0.21 98.22 39.71 Salt water 0.92 322.27 32.79 105.68 28.60 92.18 4.19 13.50 1.98 6.38 65.23 210.21
127
Table 17. Land cover classes (USVI NVC Landcover) with less than 1% of their total area protected, sorted by the amount of habitat. Total areas for the U.S. Virgin Islands are given in the first row. Total area of these land cover classes are given in the final row. Percent of total of these classes on the island (first column) is followed by percent of each type in Status 1-4.
Protected Total habitat Status 1 or 2 Status 1 Status 2 Status 3 Status 4 U.S. National Vegetation Classification % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.00 35162.98 11.09 3898.58 9.85 3461.81 1.24 436.77 1.94 682.01 86.97 30582.39 Agricultural Vegetation 0.64 223.78 0.06 0.13 0.00 0.00 0.06 0.13 0.00 0.00 99.94 223.65 Artificial Barrens 0.57 202.17 0.27 0.55 0.24 0.49 0.03 0.06 0.08 0.16 99.65 201.46 Maintained Grassland 9.88 3475.27 0.61 21.22 0.53 18.36 0.08 2.86 0.48 16.72 98.91 3437.33 Medium-Density Urban Development 6.02 2116.59 0.39 8.25 0.39 8.21 0.00 0.04 0.03 0.61 99.58 2107.73 High-density Urban Development 3.23 1134.88 0.01 0.09 0.01 0.09 0.00 0.00 0.00 0.00 99.99 1134.79 Aquaculture 0.00 0.75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 0.75
128
Table 18. USVI NVC land cover classes with 1% to less than 10% of their total area protected, sorted by the amount of habitat. Total areas for U.S. Virgin Islands are given in the first row. Total area of these land cover classes are given in the final row. Percent of total of these classes on the island (first column) is followed by percent of each type in Status 1-4.
Protected Total habitat Status 1 or 2 Status 1 Status 2 Status 3 Statu s 4 U.S. National Vegetation Classification % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.00 35162.98 11.09 3898.58 9.85 3461.81 1.24 436.77 1.94 682.01 86.97 30582.39 G455 Caribbean Seasonal Evergreen Moist Lowland Forest 3.31 1162.72 9.38 109.02 7.68 89.27 1.70 19.75 2.65 30.82 87.97 1022.88 D094 Caribbean and Central American Lowland Shrubland, Grassland and Savanna 40.68 14303.32 6.50 929.41 5.15 736.09 1.35 193.32 2.41 344.54 91.09 13029.37 M041 Caribbean and Central American Freshwater Marsh 0.03 11.33 8.38 0.95 7.41 0.84 0.97 0.11 3.88 0.44 87.73 9.94 Riparian and other natural barrens 0.01 4.48 8.26 0.37 7.37 0.33 0.89 0.04 3.35 0.15 88.39 3.96 Riprap 0.05 17.82 1.85 0.33 1.85 0.33 0.00 0.00 0.00 0.00 98.15 17.49 Low-Density Urban Development 2.39 841.91 2.59 21.78 2.16 18.21 0.42 3.57 0.97 8.13 96.45 812.00 Fresh water 0.11 40.43 1.26 0.51 1.21 0.49 0.05 0.02 0.52 0.21 98.22 39.71
129
Table 19. Land cover classes with 10% to less than 20% of their total area protected, sorted by the amount of habitat. Total areas for U.S. Virgin Islands are given in the first row. Total area of these land cover classes are given in the final row. Percent of total of these classes on the island (first column) is followed by percent of each type in Status 1-4.
Protected Total habitat Status 1 or 2 Status 1 Status 2 Status 3 Status 4 U.S. National Vegetation Classification % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.00 35162.98 11.09 3898.58 9.85 3461.81 1.24 436.77 1.94 682.01 86.97 30582.39 Mixed Sand and Gravel Beaches 0.11 38.20 19.03 7.27 14.90 5.69 4.14 1.58 9.40 3.59 71.57 27.34
130
Table 20. Land cover classes with 20% to less than 50% of their total area protected, sorted by the amount of habitat. Total areas for the U.S. Virgin Islands are given in the first row. Total area of these land cover classes are given in the final row. Percent of total of these classes on the island (first column) is followed by percent of each type in Status 1-4.
Protected Total habitat Status 1 or 2 Status 1 Status 2 Status 3 Status 4 U.S. National Vegetation Classification % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.00 35162.98 11.09 3898.58 9.85 3461.81 1.24 436.77 1.94 682.01 86.97 30582.39 G585 Caribbean Dry Broadleaf Forest 24.43 8589.08 21.56 1851.65 20.08 1724.51 1.48 127.14 1.54 132.27 76.90 6605.16 G454 Caribbean Moist Lowland-Submontane Forest Group 3.99 1404.72 35.21 494.60 33.99 477.47 1.22 17.13 2.13 29.90 62.66 880.22 G002 Caribbean Hardwood Swamp Group 0.93 328.40 29.30 96.22 25.41 83.45 3.89 12.77 3.91 12.84 66.79 219.34 G004 Caribbean Mangrove Tidal Swamp Group 0.97 340.29 29.95 101.90 25.07 85.32 4.87 16.58 10.14 34.52 59.91 203.87 Fine to Medium Grained Sandy Beaches 0.27 94.50 30.92 29.22 24.47 23.12 6.46 6.10 4.95 4.68 64.13 60.60 Gravel Beaches 0.10 36.84 29.07 10.71 26.19 9.65 2.88 1.06 5.46 2.01 65.47 24.12 Rock Cliffs and Shelves 1.05 368.35 23.60 86.92 18.52 68.23 5.07 18.69 13.33 49.11 63.07 232.32 Salt and Mudflats 0.30 104.88 20.79 21.80 18.57 19.48 2.21 2.32 4.70 4.93 74.51 78.15 Salt water 0.92 322.27 32.79 105.68 28.60 92.18 4.19 13.50 1.98 6.38 65.23 210.21
131
The following summary highlight potential gaps and conservation needs for the U.S. virgin Islands National Vegetation Classification Landcover.
Land Cover with <1% representation in GAP status 1 and 2 (Table 12): Within these land cover classes we find the smallest in extent: Aquaculture. All these classes put together make up 20.6 % of the islands. They are primarily subject to intensive human use such as agriculture, urban development, artificial barrens and maintained grasslands. As a group, 0.4% of these classes is protected under Gap Status 1 or 2, or 30.2 ha out of a total of 7,153.4 ha.
Land Cover with 1% to <10% representation in GAP status 1 and 2 (Table 13): Seven land cover units fall in this category and they account for 55% of the island. They range from an extent of less than 1% to 40.7% of the island. Class D094 Lowland shrublands and grasslands is the most represented in the islands with 14,303.3 ha (40.7%). Also in this range of protection we have very little representation freshwater marshes, natural barrens, riprap, and freshwater. Low-density urban development represents roads and built-up areas in rural areas or isolated. As a group 6.5% of these units are protected under Gap Status 1 or 2.
Land Cover with l0% to <20% representation in GAP status 1 and 2 (Table 14): Only one land unit class, mixed sand and gravel beaches, falls in this category and it accounts for only 0.1% of the island. This class is 7.3% protected under Gap Status 1 and 2.
Land Cover with 20% to <50% representation in GAP status 1 and 2 (Table 15): The class that stands out the most is dry broadleaf forest which covers 24.4% of the islands territory. Moist lowland-submontane forests are also in this range of protection covering 4% of the islands. Other classes in this range cover 1% or less of the islands including hardwood, and mangrove tidal swamp sandy and gravel beaches, cliffs and shelves. Also included in this range are salt and mud flats and salt ponds, important classes for many migratory bird species. As a group 24.1% of these units are protected under Gap Status 1 or 2.
Land Cover with at least 50% representation in GAP status 1 and 2: None of the USVI Gap land cover units are over 50% protected under Gap Status 1 and 2.
132
Table 21. Summary table identifying the amount of predicted distribution in hectares (ha) and percent of the total island for all resident, migratory, endemic, and endangered terrestrial vertebrate species in the U.S. Virgin Islands. Percent and area (ha) of protected habitat is given for Gap Status 1 and 2 combined and for Gap Status 1-4 individually.
Protected Total Habitat Status 1 and 2 Status 1 StatusS 2 Status 3 Status 4 Common Name Common Name Scientific Name (English) (Spanish) Species ID % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.0 35163.0 11.1 3898.6 9.8 3461.8 1.2 436.8 1.9 682.0 87.0 30582.4 Actitis macularia Spotted Sandpiper Playero Coleador GAP0020 6.4 2255.1 14.9 335.0 13.3 300.1 1.6 35.0 3.1 69.4 82.1 1850.6 Alsophis portoricensis Puerto Rican Racer Culebra Corredora GAP0036 2.4 838.2 15.3 128.3 11.5 96.5 3.8 31.8 6.9 57.7 77.8 652.3 Puerto Rican Ground Ameiva exsul Lizard Siguana común GAP0053 29.6 10425.5 17.6 1832.5 16.5 1720.1 1.1 112.4 1.5 154.8 80.9 8438.2 St. Croix Ground Ameiva polops Lizard Siguana de Santa Cruz GAP0054 0.2 82.3 87.4 71.9 83.0 68.3 4.4 3.6 12.6 10.4 0.0 0.0 Amphisbaena Virgin Islands Culebrita Ciega de fenestrata Amphisbaena Islas Vírgenes GAP0061 13.9 4889.6 40.7 1987.9 38.4 1879.0 2.2 108.9 1.0 47.0 58.4 2854.7 Pato Cabeciblanco Anas americana American Wigeon Americano GAP0066 0.8 296.8 8.6 25.6 8.5 25.1 0.2 0.5 2.6 7.6 88.8 263.6 White-cheeked Pato Quijada Anas bahamensis Pintail Colorada GAP0067 2.0 705.4 29.9 211.3 26.5 186.9 3.5 24.4 5.3 37.3 64.8 456.9 Anas crecca Green-winged Teal Pato Aliverde GAP0069 0.9 299.4 10.2 30.5 4.6 13.9 5.5 16.5 2.5 7.6 87.3 261.3 Anas discors Blue-winged Teal Pato Zarcel GAP0071 2.2 763.2 23.7 181.0 20.9 159.5 2.8 21.5 3.7 28.1 72.6 554.1 Anolis acutus St. Croix Anole GAP0089 47.2 16594.5 3.3 554.0 2.1 341.9 1.3 212.1 2.2 369.3 94.4 15671.2 Anolis cristatellus Crested Anole Lagartijo Común GAP0091 19.1 6714.0 12.7 850.9 11.6 779.6 1.1 71.3 1.6 107.3 85.7 5755.8 Anolis pulchellus Common Grass Anole Lagartijo Jardinero GAP0100 3.2 1137.1 6.9 78.8 6.6 75.4 0.3 3.4 2.8 31.7 90.3 1026.6 Anolis stratulus Barred Anole Lagartijo Manchado GAP0102 29.3 10296.1 28.4 2923.4 27.0 2777.2 1.4 146.2 0.6 62.7 71.0 7310.0 Anous stolidus Brown Noddy Cervera GAP0104 0.3 94.6 58.8 55.6 42.8 40.5 16.0 15.1 0.9 0.9 40.3 38.1 Ardea alba Great Egret Garzón Blanco GAP0136 4.3 1518.5 24.4 370.1 22.2 337.3 2.2 32.8 4.6 70.0 71.0 1078.5 Ardea herodias Great Blue Heron Garzón Cenizo GAP0137 2.8 982.8 31.8 312.8 27.7 272.7 4.1 40.1 5.8 57.2 62.4 612.8 Arenaria interpres Ruddy Turnstone Playero Turco GAP0139 1.0 361.0 43.2 155.8 38.4 138.8 4.7 17.0 6.6 23.8 50.3 181.4 Arrhyton exiguum Garden Snake Culebra de Jardín GAP0142 14.1 4957.4 3.5 175.3 0.9 46.1 2.6 129.2 0.3 14.5 96.2 4767.6 Jamaican Fruit-eating Artibeus jamaicensis Bat Muciérlago Frutero GAP0143 80.0 28125.1 12.1 3407.9 11.0 3090.9 1.1 317.0 1.5 411.8 86.4 24305.4 Aythya affinis Lesser Scaup Pato Pechiblanco GAP0158 0.4 156.8 8.0 12.6 0.0 0.0 8.0 12.6 2.1 3.3 89.9 140.9 Aythya collaris Ring-necked Duck Pato Acollarado GAP0159 0.3 106.9 12.6 13.4 0.8 0.8 11.8 12.6 3.1 3.3 84.4 90.2 Brachyphylla Antillean Fruit-eating Murciélago cavernarum Bat Cavernícola GAP0186 50.9 17881.8 16.9 3027.0 15.7 2814.1 1.2 212.9 1.6 281.5 81.5 14573.4 Bubulcus ibis Cattle Egret Garza del Ganado GAP0192 42.5 14932.3 2.5 372.1 2.3 349.5 0.2 22.6 0.8 112.6 96.8 14447.5 Bufo marinus Giant Toad Sapo Común GAP0195 23.6 8298.8 3.2 264.2 2.3 194.5 0.8 69.7 1.7 138.5 95.1 7896.1 Buteo jamaicensis Red-tailed Hawk Guaraguao Colirrojo GAP0197 79.9 28107.9 11.4 3194.7 10.5 2945.9 0.9 248.7 1.3 355.1 87.4 24558.1 Butorides virescens Green Heron Martinete Verde GAP0201 7.5 2645.7 26.1 690.4 24.7 653.3 1.4 37.1 3.9 103.3 70.0 1851.9 Calidris alba Sanderling Playero Arenero GAP0209 0.6 193.5 41.5 80.3 36.6 70.9 4.8 9.4 7.9 15.3 50.6 98.0 Calidris canutus Red Knot Playero Gordo GAP0212 0.4 127.9 9.4 12.0 3.3 4.2 6.1 7.8 4.3 5.4 86.4 110.5 White-rumped Playero de Rabadilla Calidris fuscicollis Sandpiper Blanca GAP0214 2.1 736.0 6.2 45.7 3.8 28.1 2.4 17.6 2.6 18.8 91.2 671.5 Calidris himantopus Stilt Sandpiper Playero Patilargo GAP0215 2.2 764.6 24.0 183.7 21.3 162.9 2.7 20.8 6.4 49.2 69.5 531.7 Calidris mauri Western Sandpiper Playero Occidental GAP0216 1.5 527.8 23.6 124.8 23.6 124.8 0.0 0.0 9.0 47.7 67.3 355.3 Calidris melanotos Pectoral Sandpiper Playero Pectoral GAP0217 0.8 295.1 6.1 18.1 4.7 13.9 1.4 4.1 2.4 7.0 91.5 270.1 Calidris minutilla Least Sandpiper Playerito Menudillo GAP0218 1.5 528.3 18.8 99.3 16.0 84.6 2.8 14.7 4.6 24.3 76.6 404.7 Calidris pusilla Semipalmated Playero Gracioso GAP0219 3.4 1190.1 16.4 195.5 12.9 153.9 3.5 41.6 4.9 58.7 78.6 935.9 133
Sandpiper Catoptrophorus semipalmatus Willet Playero Aliblanco GAP0260 1.0 344.1 20.2 69.6 16.9 58.2 3.3 11.4 6.0 20.5 73.8 253.9 Martín Pescador Ceryle alcyon Belted Kingfisher Norteño GAP0270 17.5 6160.4 22.2 1366.6 20.6 1266.2 1.6 100.4 2.8 172.8 75.0 4621.0 Charadrius semipalmatus Semipalmated Plover Playero acollarado GAP0284 1.2 436.9 26.9 117.7 23.2 101.6 3.7 16.1 4.5 19.6 68.6 299.6 Charadrius vociferus Killdeer Chorlo Sabanero GAP0285 0.8 267.0 5.8 15.4 4.2 11.2 1.6 4.2 3.3 8.7 91.0 242.9 Charadrius wilsonia Wilson's Plover Chorlo Marítimo GAP0286 1.5 517.7 33.1 171.1 29.3 151.7 3.8 19.4 4.7 24.4 62.2 322.1 Pejeblanco, Tortuga Chelonia mydas Green Sea Turtle Verde GAP0291 0.2 58.9 44.2 26.0 33.1 19.5 11.1 6.5 6.1 3.6 49.6 29.2 Chordeiles Querequequé gundlachii Antillean Nighthawk Antillano GAP0303 13.7 4833.1 0.1 3.1 0.0 0.0 0.1 3.1 0.8 38.2 99.1 4791.9 Coccyzus Pájaro Bobo americanus Yellow-billed Cuckoo Pechiblanco GAP0321 4.3 1523.8 3.6 55.3 3.0 45.2 0.7 10.1 5.9 90.6 90.4 1377.9 Coccyzus minor Mangrove Cuckoo Pájaro Bobo Menor GAP0323 49.1 17281.7 17.4 3000.1 16.4 2836.3 0.9 163.8 1.6 280.2 81.0 14001.3 Coereba flaveola Bananaquit Reinita Común GAP0324 67.5 23729.1 14.0 3326.4 12.7 3019.0 1.3 307.4 1.6 378.5 84.4 20024.2 Columba livia Rock Dove Paloma Doméstica GAP0327 9.6 3375.8 2.1 70.6 1.8 61.0 0.3 9.5 0.6 20.0 97.3 3285.3 Common Ground- Columbina passerina Dove Rolita GAP0328 51.3 18022.9 3.4 619.8 2.9 521.6 0.5 98.2 0.9 164.7 95.6 17238.4 Crotophaga ani Smooth-billed Ani Garrapatero GAP0354 53.2 18715.8 5.1 961.7 3.9 737.4 1.2 224.3 1.3 249.8 93.5 17504.4 Dendroica Black-throated Blue caerulescens Warbler Reinita Azul GAP0386 12.8 4508.6 52.4 2363.1 52.4 2362.3 0.0 0.8 0.2 9.9 47.4 2135.6 Yellow-rumped Dendroica coronata Warbler Reinita Coronada GAP0390 31.4 11026.9 25.5 2807.9 25.2 2777.4 0.3 30.5 1.9 205.9 72.7 8013.1 Dendroica discolor Prairie Warbler Reinita Galana GAP0391 42.6 14991.7 12.8 1919.7 11.9 1789.4 0.9 130.4 1.5 228.7 85.7 12843.3 Dendroica magnolia Magnolia Warbler Reinita Manchada GAP0394 3.8 1339.9 67.7 907.4 67.7 907.4 0.0 0.0 0.8 11.2 31.4 421.4 Dendroica petechia Yellow Warbler Reinita Amarilla GAP0397 2.8 1001.0 30.7 307.6 26.1 260.8 4.7 46.8 6.3 62.9 63.0 630.5 Dendroica striata Blackpoll Warbler Reinita Rayada GAP0399 35.7 12554.4 3.3 417.7 2.3 283.5 1.1 134.3 1.6 199.8 95.1 11936.8 Dendroica tigrina Cape May Warbler Reinita Tigre GAP0400 27.9 9806.3 17.7 1739.3 17.6 1727.5 0.1 11.8 1.7 164.9 80.6 7902.1 Black-throated Green Dendroica virens Warbler Reinita Verdosa GAP0401 10.5 3706.5 30.9 1144.0 30.8 1143.0 0.0 1.0 0.6 21.1 68.6 2541.3 Dermochelys Leatherback Sea coriacea Turtle Tinglar GAP0403 0.8 291.0 6.2 17.9 5.3 15.4 0.9 2.5 8.3 24.2 85.5 248.9 Egretta caerulea Little Blue Heron Garza Azul GAP0424 3.3 1146.5 25.1 287.8 21.7 248.9 3.4 38.9 4.9 55.9 70.0 802.7 Egretta thula Snowy Egret Garza Blanca GAP0428 4.0 1416.3 24.1 341.3 21.5 304.3 2.6 37.0 4.8 68.5 71.1 1006.4 Egretta tricolor Tricolored Heron Garza Pechiblanca GAP0429 2.8 977.8 28.0 274.0 24.9 243.8 3.1 30.2 6.4 62.9 65.5 641.0 Elaenia martinica Caribbean Elaenia Juí Blanco GAP0439 43.5 15280.4 15.1 2314.0 14.4 2193.3 0.8 120.7 1.4 217.3 83.4 12749.2 Eleutherodactylus antillensis Antillean Frog Churí GAP0443 41.0 14411.1 19.7 2840.8 18.8 2711.4 0.9 129.4 1.1 152.2 79.2 11418.2 Eleutherodactylus cochranae Whistling Frog Coquí Pitito GAP0445 24.6 8658.4 27.6 2393.4 26.5 2291.9 1.2 101.5 0.7 61.3 71.6 6203.7 Eleutherodactylus coqui Common Coqui Coquí de las Yerbas GAP0447 5.6 1952.7 2.4 47.6 2.4 47.5 0.0 0.1 0.1 2.8 97.4 1902.3 Eleutherodactylus Coquí de Islas lentus Yellow Mottled Coqui Vírgenes GAP0453 28.5 10024.9 2.4 244.9 1.6 165.4 0.8 79.5 1.2 124.7 96.3 9655.2 Epicrates monensis Virgin Islands Tree Culebrón de la Isla granti Boa Virgin GAP0472 8.4 2962.1 1.7 49.0 0.0 1.4 1.6 47.6 0.1 2.8 98.3 2910.3 Eretmochelys imbricata Hawksbill Turtle Tortuga Carey GAP0489 0.2 76.7 40.2 30.8 29.4 22.6 10.8 8.3 4.7 3.6 55.1 42.2 Eulampis Zumbador de Pecho holosericeus Green-throated Carib Azul GAP0507 16.6 5822.6 15.9 923.0 14.7 855.3 1.2 67.7 1.9 113.5 82.2 4786.1 134
Falco columbarius Merlin Halcón Migratorio GAP0517 51.4 18065.5 11.2 2028.4 10.6 1907.4 0.7 120.9 1.7 314.7 87.0 15722.4 Falco peregrinus Peregrine Falcon Halcón Peregrino GAP0518 36.4 12808.6 11.7 1498.7 11.2 1431.7 0.5 67.0 0.9 119.3 87.4 11190.6 Falco sparverius American Kestrel Halcón Común GAP0520 61.8 21733.8 12.6 2745.6 11.1 2403.4 1.6 342.3 1.7 369.9 85.7 18618.4 Magnificent Fregata magnificens Frigatebird Fragata Magnifica GAP0523 3.5 1221.3 35.9 437.9 32.3 394.4 3.6 43.5 4.5 55.4 59.6 728.0 Fulica americana American Coot Gallinazo Americano GAP0524 1.0 363.3 14.3 52.0 10.5 38.2 3.8 13.8 8.1 29.4 77.6 281.9 Fulica caribaea Caribbean Coot Gallinazo Caribeño GAP0525 0.6 195.7 7.7 15.2 5.4 10.5 2.4 4.7 6.6 12.9 85.7 167.7 Gallinago delicata Wilson's Snipe Agachona Común GAP0528 0.5 175.1 12.8 22.4 10.0 17.4 2.9 5.0 2.9 5.0 84.3 147.7 Gallinula chloropus Common Moorhen Gallareta Común GAP0530 2.1 738.5 21.3 157.0 17.2 127.1 4.0 29.9 6.2 45.8 72.5 535.7 Geochelone Tortuga de Patas carbonaria Red-footed Tortoise Rojas GAP0535 9.4 3315.5 37.4 1240.3 37.2 1234.5 0.2 5.8 2.1 71.0 60.4 2004.2 Common Geothlypis trichas Yellowthroat Reinita Pica Tierra GAP0536 1.3 461.0 18.1 83.6 12.4 57.2 5.7 26.4 1.1 5.1 80.8 372.3 Paloma Perdíz de Geotrygon mystacea Bridled Quail-Dove Martinica GAP0539 6.1 2138.1 45.2 966.9 45.0 962.5 0.2 4.4 2.8 60.3 52.0 1111.0 Haematopus American palliatus Oystercatcher Ostrero Americano GAP0575 0.8 266.1 35.6 94.8 31.1 82.8 4.5 12.0 2.4 6.3 62.0 165.0 Helmitheros Worm-eating vermivorus Warbler Reinita Gusanera GAP0600 15.1 5310.6 43.4 2302.8 43.4 2302.6 0.0 0.2 0.9 47.9 55.7 2960.0 Hemidactylus Cosmopolitan House mabouia Gecko Salamanquesa GAP0602 22.0 7747.5 4.2 322.1 3.7 285.2 0.5 36.9 0.5 42.0 95.3 7383.4 Small Indian Herpestes javanicus Mongoose Mangosta GAP0606 72.0 25325.5 13.4 3387.4 12.0 3036.1 1.4 351.4 1.6 403.7 85.0 21534.3 Himantopus mexicanus Black-necked Stilt Viuda Mexicana GAP0609 1.8 616.1 26.3 162.3 22.6 139.0 3.8 23.3 5.7 35.2 67.9 418.6 Hirundo rustica Barn Swallow Golondrina Horquilla GAP0616 24.5 8619.8 6.5 560.6 5.2 447.0 1.3 113.6 1.8 153.7 91.7 7905.5 Iguana iguana Green Iguana Iguana Común GAP0643 30.6 10760.8 13.9 1495.4 12.2 1318.1 1.6 177.3 2.0 216.6 84.1 9048.9 Larus atricilla Laughing Gull Gaviota Cabecinegra GAP0673 2.3 799.3 30.1 240.7 25.4 203.4 4.7 37.3 4.0 31.7 65.9 526.9 Leptodactylus Ranita de Labio albilabris White-lipped Frog Blanco GAP0689 33.4 11757.3 6.3 744.0 5.7 668.9 0.6 75.1 0.8 96.2 92.9 10917.1 Limnodromus Short-billed griseus Dowitcher Agujeta Pico Corto GAP0691 1.8 647.4 14.5 93.8 12.1 78.6 2.4 15.2 2.3 15.0 83.2 538.7 Lesser Antillean Come Ñame Antillano Loxigilla noctis Bullfinch Menor GAP0712 27.0 9499.6 23.5 2231.2 23.4 2218.5 0.1 12.7 2.0 190.2 74.5 7078.2 Mabuya mabouya Slippery-backed sloanei Mabuya Lucía GAP0728 1.0 357.7 22.9 81.8 12.3 44.0 10.6 37.8 35.2 125.9 41.9 150.0 Margarops fuscatus Pearly-eyed Thrasher Zorzal Pardo GAP0747 75.0 26379.5 11.9 3139.1 11.1 2918.8 0.8 220.3 1.2 314.1 86.9 22926.3 Northern Mimus polyglottos Mockingbird Ruiseñor GAP0773 31.3 11012.6 12.3 1350.6 11.3 1242.0 1.0 108.6 1.6 177.9 86.1 9484.1 Black-and-white Mniotilta varia Warbler Reinita Trepadora GAP0776 28.7 10104.9 27.6 2787.1 26.2 2649.7 1.4 137.4 1.9 193.7 70.5 7124.1 Velvety Free-tailed Molossus molossus Bat Murciélago Casero GAP0778 90.4 31794.3 10.8 3449.5 9.7 3097.6 1.1 352.0 1.5 461.7 87.7 27883.1 Molothrus bonariensis Shiny Cowbird Tordo Lustroso GAP0779 10.0 3500.1 29.6 1036.4 29.5 1033.7 0.1 2.7 2.7 95.0 67.7 2368.7 Myiarchus Puerto Rican antillarum Flycatcher Juí de Puerto Rico GAP0799 11.5 4034.6 60.7 2449.1 60.5 2441.9 0.2 7.1 0.2 8.2 39.1 1577.3 Noctilio leporinus Greater Bulldog Bat Murciélago Pescador GAP0814 2.9 1014.9 46.5 472.2 42.3 429.3 4.2 42.9 2.0 20.2 51.5 522.4 Numenius phaeopus Whimbrel Playero Trinador GAP0819 3.1 1082.6 10.9 118.3 9.1 98.8 1.8 19.4 3.1 33.8 86.0 930.6 Yellow-crowned Nyctanassa violacea Night-heron Yaboa Comun GAP0821 5.6 1972.2 34.8 685.7 32.1 632.7 2.7 53.0 4.6 91.0 60.6 1195.5
135
Nycticorax Black-crowned Night- nycticorax heron Yaboa Real GAP0823 4.7 1636.8 35.5 581.6 34.4 562.3 1.2 19.3 3.9 64.2 60.5 991.0 Odocoileus virginianus White-tailed Deer Venado Rabiblanco GAP0829 70.2 24674.9 13.3 3287.9 12.3 3045.5 1.0 242.4 1.5 370.7 85.2 21016.3 Orthorhyncus Antillean Crested cristatus Hummingbird Zumbador Crestado GAP0861 55.9 19662.3 11.4 2251.2 10.1 1985.8 1.3 265.4 1.6 318.3 86.9 17092.9 Osteopilus septentrionalis Cuban Treefrog Hila Platanera GAP0862 19.3 6784.5 9.8 667.8 8.7 589.0 1.2 78.8 0.3 19.0 89.9 6097.8 Oxyura jamaicensis Ruddy Duck Pato Chorizo GAP0865 0.1 38.7 60.1 23.3 30.9 12.0 29.3 11.3 4.5 1.7 35.4 13.7 Pandion haliaetus Osprey Aguila Pescadora GAP0869 2.9 1023.1 25.6 262.2 23.3 238.6 2.3 23.6 5.3 53.9 69.1 707.0 Parula americana Northern Parula Reinita Pechidorada GAP0886 39.9 14038.3 20.0 2806.2 18.9 2649.7 1.1 156.6 1.8 247.6 78.2 10984.4 Passer domesticus House Sparrow Gorrión Doméstico GAP0887 19.1 6698.6 2.8 185.4 2.5 170.1 0.2 15.2 0.5 34.1 96.7 6479.2 Passerina cyanea Indigo Bunting Gorrión Azul GAP0890 10.6 3711.1 5.6 208.9 5.3 197.1 0.3 11.7 0.5 17.7 93.9 3484.6 Patagioenas White-crowned leucocephala Pigeon Paloma Cabeciblanca GAP0892 51.0 17923.3 17.7 3176.8 16.1 2883.1 1.6 293.6 2.2 401.7 80.0 14344.9 Patagioenas squamosa Scaly-naped Pigeon Paloma Turca GAP0893 58.4 20539.9 15.3 3151.8 14.0 2873.6 1.4 278.2 1.8 368.1 82.9 17020.1 Pelecanus occidentalis Brown Pelican Pelícano Pardo GAP0895 4.9 1731.0 34.0 589.3 30.4 526.0 3.7 63.4 4.4 75.7 61.6 1066.0 Petrochelidon Golondrina de pyrrhonota Cliff Swallow Peñasco GAP0898 9.5 3330.2 1.2 39.6 1.1 37.4 0.1 2.2 0.2 6.4 98.6 3284.2 Phaethon aethereus Red-billed Tropicbird Rabijunco Piquirrojo GAP0903 0.9 309.0 36.7 113.3 28.4 87.7 8.3 25.6 21.8 67.2 41.6 128.5 White-tailed Phaethon lepturus Tropicbird Rabijunco Coliblanco GAP0904 0.9 309.8 34.0 105.2 29.8 92.5 4.1 12.8 17.7 54.8 48.3 149.7 American Golden- Pluvialis dominica plover Chorlo Dorado GAP0924 2.2 786.1 11.5 90.0 10.0 78.6 1.5 11.5 1.1 8.3 87.5 687.8 Pluvialis squatarola Black-bellied Plover Chorlo Cabezón GAP0925 3.5 1225.3 8.7 106.7 7.1 87.0 1.6 19.7 1.8 22.1 89.5 1096.5 Podilymbus podiceps Pied-billed Grebe Zaramago GAP0926 0.6 211.4 24.5 51.8 19.1 40.4 5.4 11.4 2.5 5.4 73.0 154.2 Porphyrio martinica Purple Gallinule Gallareta Azul GAP0936 0.1 19.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 100.0 19.9 Porzana carolina Sora Gallito Sora GAP0937 0.4 145.3 41.1 59.7 36.5 53.0 4.6 6.7 3.4 4.9 55.6 80.7 Progne dominicensis Caribbean Martin Golondrina de Iglesias GAP0951 17.4 6119.3 2.2 132.3 1.9 113.6 0.3 18.8 1.1 66.1 96.8 5920.9 Prothonotary Protonotaria citrea Warbler Reinita Protonotaria GAP0955 10.1 3567.4 33.1 1182.4 32.9 1174.9 0.2 7.5 0.9 31.0 66.0 2354.0 Audubon's Puffinus lherminieri Shearwater Pampero de Audubon GAP0976 0.1 33.5 99.2 33.2 84.8 28.4 14.5 4.8 0.0 0.0 0.8 0.3 Rallus longirostris Clapper Rail Pollo de Mangle GAP0985 0.2 64.5 6.4 4.1 6.4 4.1 0.0 0.0 29.1 18.8 64.5 41.6 Riparia riparia Bank Swallow Golondrina Parda GAP0998 12.0 4212.5 14.9 626.1 13.9 587.2 0.9 38.9 0.2 6.5 85.0 3580.0 Seiurus aurocapillus Ovenbird Pizpita Dorada GAP1041 14.7 5180.5 39.0 2018.1 38.9 2017.1 0.0 1.0 0.4 22.7 60.6 3139.6 Louisiana Seiurus motacilla Waterthrush Pizpita de Río GAP1042 1.6 566.2 56.9 322.3 56.5 320.1 0.4 2.2 4.5 25.3 38.6 218.6 Seiurus Northern noveboracensis Waterthrush Pizpita de Mangle GAP1043 5.8 2039.1 34.2 697.4 31.6 644.2 2.6 53.1 3.4 70.3 62.4 1271.4 Setophaga ruticilla American Redstart Reinita Candelita GAP1062 5.0 1764.6 34.2 603.3 32.8 578.7 1.4 24.6 1.5 26.7 64.3 1134.7 Sphaerodactylus beattyi St. Croix Dwarf Gecko GAP1073 14.4 5057.6 4.7 236.6 2.8 142.1 1.9 94.4 3.0 149.3 92.4 4671.7 Sphaerodactylus Common Dwarf macrolepis Gecko Salamanquita Común GAP1077 62.1 21841.4 14.6 3187.9 13.3 2912.7 1.3 275.3 2.0 434.4 83.4 18219.0 Desmarest's Fig- Murciélago Rojo Stenoderma rufum eating Bat Frutero GAP1116 14.4 5074.9 24.4 1239.0 24.3 1231.0 0.2 8.0 3.3 168.8 72.3 3667.1 Sterna anaethetus Bridled Tern Gaviota Monja GAP1124 0.2 74.3 76.3 56.7 50.6 37.6 25.7 19.1 1.2 0.9 22.5 16.7 Sterna antillarum Least Tern Gaviota Chica GAP1125 0.8 288.9 23.7 68.4 18.6 53.9 5.0 14.5 6.6 19.1 69.7 201.4 136
Sterna dougallii Roseate Tern Palometa GAP1127 0.2 65.2 54.1 35.3 37.5 24.4 16.6 10.9 0.1 0.0 45.8 29.9 Sterna fuscata Sooty Tern Gaviota Oscura GAP1129 0.0 11.5 76.9 8.8 20.8 2.4 56.1 6.4 0.0 0.0 23.1 2.7 Sterna hirundo Common Tern Gaviota Común GAP1130 1.4 487.3 30.8 150.0 23.8 115.9 7.0 34.1 6.0 29.1 63.3 308.3 Sterna maxima Royal Tern Gaviota Real GAP1131 0.8 285.9 41.8 119.5 35.7 102.1 6.1 17.4 4.4 12.6 53.8 153.8 Sterna sandvicensis Sandwich Tern Gaviota Piquiaguda GAP1134 0.6 196.7 29.7 58.4 17.7 34.8 12.0 23.7 5.9 11.6 64.4 126.7 Sula dactylatra Masked Booby Boba Enmascarada GAP1145 0.1 19.2 99.9 19.2 91.6 17.6 8.3 1.6 0.0 0.0 0.1 0.0 Sula leucogaster Brown Booby Boba Parda GAP1146 0.7 238.9 47.5 113.4 40.0 95.6 7.5 17.8 15.1 36.2 37.4 89.3 Sula sula Red-footed Booby Boba Patirroja GAP1147 0.2 74.6 65.9 49.1 48.6 36.2 17.3 12.9 32.3 24.1 1.8 1.3 Tachybaptus dominicus Least Grebe Tigua GAP1166 1.9 653.2 0.0 0.1 0.0 0.0 0.0 0.1 4.2 27.2 95.8 625.9 Brazilian Free-tailed Murciélago de Cola Tadarida brasiliensis Bat Libre GAP1168 3.7 1303.9 48.0 625.9 48.0 625.9 0.0 0.0 0.8 9.9 51.2 668.1 Thecadactylus rapicauda Fat-tailed Gecko GAP1172 0.7 246.0 0.0 0.1 0.0 0.0 0.0 0.1 1.7 4.2 98.3 241.7 Tiaris bicolor Black-faced Grassquit Gorrión Negro GAP1180 49.1 17263.8 3.4 589.9 3.1 526.7 0.4 63.2 0.9 155.1 95.7 16518.8 Playero Guineilla Tringa flavipes Lesser Yellowlegs Pequeño GAP1198 2.0 704.0 26.4 186.1 23.0 161.7 3.5 24.5 4.2 29.2 69.4 488.6 Playero Guineilla Tringa melanoleuca Greater Yellowlegs Grande GAP1199 0.9 300.4 19.3 58.1 17.5 52.5 1.9 5.6 4.3 12.9 76.4 229.4 Tringa solitaria Solitary Sandpiper Playero Solitario GAP1201 0.3 88.3 1.5 1.4 1.5 1.3 0.0 0.0 0.2 0.2 98.2 86.8 Typhlops richardii Richard's Blind Snake Vibora Común GAP1212 37.0 13027.3 16.5 2153.5 15.6 2026.4 1.0 127.1 1.0 136.4 82.4 10737.3 Tyrannus dominicensis Gray Kingbird Pitirre Gris GAP1215 76.1 26766.1 8.4 2239.4 7.2 1929.8 1.2 309.6 1.6 426.8 90.0 24099.9 Black-whiskered Vireo altiloquus Vireo Julian Chivi GAP1234 36.6 12870.8 20.5 2636.2 20.0 2578.5 0.4 57.6 2.3 294.3 77.2 9940.4 Wilsonia citrina Hooded Warbler Reinita Viuda GAP1240 16.5 5798.9 41.4 2401.5 41.4 2399.3 0.0 2.1 1.1 62.1 57.5 3335.4 Zenaida asiatica White-winged Dove Tórtola Aliblanca GAP1250 58.1 20420.2 8.9 1826.8 8.3 1688.6 0.7 138.2 1.2 251.6 89.8 18341.8 Zenaida aurita Zenaida Dove Tórtola Cardosantera GAP1251 82.2 28920.5 7.7 2227.9 6.6 1902.9 1.1 325.0 1.5 421.3 90.8 26271.4
137
Table 22. Species with less than 1% of their total distribution protected (4 species). Total areas for the USVI are given in the first row. Percent of total of these species on the island (first column) is followed by percent of each species in Status 1-4.
Protected Total Habitat Status 1 and 2 Status 1 Status2 Status 3 Status 4 Common Name Common Name Scientific Name (English) (Spanish) Species ID % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.00 35162.98 11.09 3898.58 9.85 3461.81 1.24 436.77 1.94 682.01 86.97 30582.39 Antillean Querequequé Chordeiles gundlachii Nighthawk Antillano GAP0303 13.74 4833.12 0.06 3.07 0.00 0.00 0.06 3.07 0.79 38.15 99.15 4791.90 Porphyrio martinica Purple Gallinule Gallareta Azul GAP0936 0.06 19.87 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00 19.87 Tachybaptus dominicus Least Grebe Tigua GAP1166 1.86 653.16 0.01 0.07 0.00 0.02 0.01 0.05 4.17 27.24 95.82 625.85 Thecadactylus rapicauda Fat-tailed Gecko GAP1172 0.70 245.96 0.03 0.08 0.00 0.00 0.03 0.08 1.69 4.15 98.28 241.73
138
Table 23. Species with 1% to less than 10% of their total area protected (37 species). Total areas for the U.S. Virgin Islands are given in the first row. Percent of total of these species on the island (first column) is followed by percent of each species in Status 1-4.
Protected Total Habitat Status 1 and 2 Status 1 Status 2 Status 3 Status 4 Common Name Common Name Species Scientific Name (English) (Spanish) ID % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.0 35163.0 11.1 3898.6 9.8 3461.8 1.2 436.8 1.9 682.0 87.0 30582.4 Pato Cabeciblanco Anas americana American Wigeon Americano GAP0066 0.8 296.8 8.6 25.6 8.5 25.1 0.2 0.5 2.6 7.6 88.8 263.6 Anolis acutus St. Croix Anole GAP0089 47.2 16594.5 3.3 554.0 2.1 341.9 1.3 212.1 2.2 369.3 94.4 15671.2 Common Grass Anolis pulchellus Anole Lagartijo Jardinero GAP0100 3.2 1137.1 6.9 78.8 6.6 75.4 0.3 3.4 2.8 31.7 90.3 1026.6 Arrhyton exiguum Garden Snake Culebra de Jardín GAP0142 14.1 4957.4 3.5 175.3 0.9 46.1 2.6 129.2 0.3 14.5 96.2 4767.6 Aythya affinis Lesser Scaup Pato Pechiblanco GAP0158 0.4 156.8 8.0 12.6 0.0 0.0 8.0 12.6 2.1 3.3 89.9 140.9 Bubulcus ibis Cattle Egret Garza del Ganado GAP0192 42.5 14932.3 2.5 372.1 2.3 349.5 0.2 22.6 0.8 112.6 96.8 14447.5 Bufo marinus Giant Toad Sapo Común GAP0195 23.6 8298.8 3.2 264.2 2.3 194.5 0.8 69.7 1.7 138.5 95.1 7896.1 Calidris canutus Red Knot Playero Gordo GAP0212 0.4 127.9 9.4 12.0 3.3 4.2 6.1 7.8 4.3 5.4 86.4 110.5 White-rumped Playero de Calidris fuscicollis Sandpiper Rabadilla Blanca GAP0214 2.1 736.0 6.2 45.7 3.8 28.1 2.4 17.6 2.6 18.8 91.2 671.5 Calidris melanotos Pectoral Sandpiper Playero Pectoral GAP0217 0.8 295.1 6.1 18.1 4.7 13.9 1.4 4.1 2.4 7.0 91.5 270.1 Charadrius vociferus Killdeer Chorlo Sabanero GAP0285 0.8 267.0 5.8 15.4 4.2 11.2 1.6 4.2 3.3 8.7 91.0 242.9 Yellow-billed Pájaro Bobo Coccyzus americanus Cuckoo Pechiblanco GAP0321 4.3 1523.8 3.6 55.3 3.0 45.2 0.7 10.1 5.9 90.6 90.4 1377.9 Columba livia Rock Dove Paloma Doméstica GAP0327 9.6 3375.8 2.1 70.6 1.8 61.0 0.3 9.5 0.6 20.0 97.3 3285.3 Common Ground- Columbina passerina Dove Rolita GAP0328 51.3 18022.9 3.4 619.8 2.9 521.6 0.5 98.2 0.9 164.7 95.6 17238.4 Crotophaga ani Smooth-billed Ani Garrapatero GAP0354 53.2 18715.8 5.1 961.7 3.9 737.4 1.2 224.3 1.3 249.8 93.5 17504.4 Dendroica striata Blackpoll Warbler Reinita Rayada GAP0399 35.7 12554.4 3.3 417.7 2.3 283.5 1.1 134.3 1.6 199.8 95.1 11936.8 Dermochelys Leatherback Sea coriacea Turtle Tinglar GAP0403 0.8 291.0 6.2 17.9 5.3 15.4 0.9 2.5 8.3 24.2 85.5 248.9 Eleutherodactylus Coquí de las coqui Common Coqui Yerbas GAP0447 5.6 1952.7 2.4 47.6 2.4 47.5 0.0 0.1 0.1 2.8 97.4 1902.3 Eleutherodactylus Yellow Mottled Coquí de Islas lentus Coqui Vírgenes GAP0453 28.5 10024.9 2.4 244.9 1.6 165.4 0.8 79.5 1.2 124.7 96.3 9655.2 139
Epicrates monensis Virgin Islands Tree Culebrón de la Isla granti Boa Virgin GAP0472 8.4 2962.1 1.7 49.0 0.0 1.4 1.6 47.6 0.1 2.8 98.3 2910.3 Fulica caribaea Caribbean Coot Gallinazo Caribeño GAP0525 0.6 195.7 7.7 15.2 5.4 10.5 2.4 4.7 6.6 12.9 85.7 167.7 Hemidactylus Cosmopolitan mabouia House Gecko Salamanquesa GAP0602 22.0 7747.5 4.2 322.1 3.7 285.2 0.5 36.9 0.5 42.0 95.3 7383.4 Golondrina Hirundo rustica Barn Swallow Horquilla GAP0616 24.5 8619.8 6.5 560.6 5.2 447.0 1.3 113.6 1.8 153.7 91.7 7905.5 Leptodactylus Ranita de Labio albilabris White-lipped Frog Blanco GAP0689 33.4 11757.3 6.3 744.0 5.7 668.9 0.6 75.1 0.8 96.2 92.9 10917.1 Osteopilus septentrionalis Cuban Treefrog Hila Platanera GAP0862 19.3 6784.5 9.8 667.8 8.7 589.0 1.2 78.8 0.3 19.0 89.9 6097.8 Passer domesticus House Sparrow Gorrión Doméstico GAP0887 19.1 6698.6 2.8 185.4 2.5 170.1 0.2 15.2 0.5 34.1 96.7 6479.2 Passerina cyanea Indigo Bunting Gorrión Azul GAP0890 10.6 3711.1 5.6 208.9 5.3 197.1 0.3 11.7 0.5 17.7 93.9 3484.6 Petrochelidon Golondrina de pyrrhonota Cliff Swallow Peñasco GAP0898 9.5 3330.2 1.2 39.6 1.1 37.4 0.1 2.2 0.2 6.4 98.6 3284.2 Black-bellied Pluvialis squatarola Plover Chorlo Cabezón GAP0925 3.5 1225.3 8.7 106.7 7.1 87.0 1.6 19.7 1.8 22.1 89.5 1096.5 Golondrina de Progne dominicensis Caribbean Martin Iglesias GAP0951 17.4 6119.3 2.2 132.3 1.9 113.6 0.3 18.8 1.1 66.1 96.8 5920.9 Rallus longirostris Clapper Rail Pollo de Mangle GAP0985 0.2 64.5 6.4 4.1 6.4 4.1 0.0 0.0 29.1 18.8 64.5 41.6 Sphaerodactylus St. Croix Dwarf beattyi Gecko GAP1073 14.4 5057.6 4.7 236.6 2.8 142.1 1.9 94.4 3.0 149.3 92.4 4671.7 Black-faced Tiaris bicolor Grassquit Gorrión Negro GAP1180 49.1 17263.8 3.4 589.9 3.1 526.7 0.4 63.2 0.9 155.1 95.7 16518.8 Tringa solitaria Solitary Sandpiper Playero Solitario GAP1201 0.3 88.3 1.5 1.4 1.5 1.3 0.0 0.0 0.2 0.2 98.2 86.8 Tyrannus dominicensis Gray Kingbird Pitirre Gris GAP1215 76.1 26766.1 8.4 2239.4 7.2 1929.8 1.2 309.6 1.6 426.8 90.0 24099.9 White-winged Zenaida asiatica Dove Tórtola Aliblanca GAP1250 58.1 20420.2 8.9 1826.8 8.3 1688.6 0.7 138.2 1.2 251.6 89.8 18341.8 Tórtola Zenaida aurita Zenaida Dove Cardosantera GAP1251 82.2 28920.5 7.7 2227.9 6.6 1902.9 1.1 325.0 1.5 421.3 90.8 26271.4
140
Table 24. Species with 10% to less than 20% of their total area protected (41 species). Total areas for the U.S. Virgin Islands are given in the first row. Percent of total of these species on the island (first column) is followed by percent of each species in Status 1-4.
Protected Total Habitat (Status 1 and 2) Status 1 Status 2 Status 3 Status 4 Common Name Common Name Species Scientific Name (English) (Spanish) ID % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.00 35162.98 11.09 3898.58 9.85 3461.81 1.24 436.77 1.94 682.01 86.97 30582.39 Actitis macularia Spotted Sandpiper Playero Coleador GAP0020 6.41 2255.07 14.86 335.04 13.31 300.08 1.55 34.96 3.08 69.42 82.06 1850.61 Alsophis portoricensis Puerto Rican Racer Culebra Corredora GAP0036 2.38 838.21 15.30 128.28 11.51 96.50 3.79 31.78 6.88 57.68 77.81 652.25 Puerto Rican Ground Ameiva exsul Lizard Siguana común GAP0053 29.65 10425.46 17.58 1832.45 16.50 1720.06 1.08 112.39 1.48 154.81 80.94 8438.20 Anas crecca Green-winged Teal Pato Aliverde GAP0069 0.85 299.36 10.18 30.46 4.65 13.91 5.53 16.55 2.54 7.60 87.29 261.30 Anolis cristatellus Crested Anole Lagartijo Común GAP0091 19.09 6714.02 12.67 850.93 11.61 779.63 1.06 71.30 1.60 107.34 85.73 5755.75 Jamaican Fruit-eating Muciérlago Artibeus jamaicensis Bat Frutero GAP0143 79.98 28125.07 12.12 3407.87 10.99 3090.91 1.13 316.96 1.46 411.82 86.42 24305.38 Aythya collaris Ring-necked Duck Pato Acollarado GAP0159 0.30 106.89 12.57 13.44 0.78 0.83 11.80 12.61 3.07 3.28 84.36 90.17 Antillean Fruit-eating Murciélago Brachyphylla cavernarum Bat Cavernícola GAP0186 50.85 17881.82 16.93 3026.97 15.74 2814.12 1.19 212.85 1.57 281.45 81.50 14573.40 Guaraguao Buteo jamaicensis Red-tailed Hawk Colirrojo GAP0197 79.94 28107.91 11.37 3194.67 10.48 2945.93 0.88 248.74 1.26 355.14 87.37 24558.10 Playerito Calidris minutilla Least Sandpiper Menudillo GAP0218 1.50 528.29 18.79 99.28 16.01 84.58 2.78 14.70 4.60 24.31 76.61 404.70 Semipalmated Calidris pusilla Sandpiper Playero Gracioso GAP0219 3.38 1190.15 16.42 195.48 12.93 153.92 3.49 41.56 4.93 58.72 78.64 935.95 Pájaro Bobo Coccyzus minor Mangrove Cuckoo Menor GAP0323 49.15 17281.69 17.36 3000.11 16.41 2836.29 0.95 163.82 1.62 280.24 81.02 14001.34 Coereba flaveola Bananaquit Reinita Común GAP0324 67.48 23729.05 14.02 3326.36 12.72 3019.01 1.30 307.35 1.59 378.47 84.39 20024.22 Dendroica discolor Prairie Warbler Reinita Galana GAP0391 42.63 14991.66 12.81 1919.71 11.94 1789.35 0.87 130.36 1.53 228.70 85.67 12843.25 Dendroica tigrina Cape May Warbler Reinita Tigre GAP0400 27.89 9806.26 17.74 1739.31 17.62 1727.47 0.12 11.84 1.68 164.90 80.58 7902.05 141
Elaenia martinica Caribbean Elaenia Juí Blanco GAP0439 43.46 15280.44 15.14 2313.96 14.35 2193.26 0.79 120.70 1.42 217.32 83.43 12749.16 Eleutherodactylus antillensis Antillean Frog Churí GAP0443 40.98 14411.12 19.71 2840.77 18.81 2711.39 0.90 129.38 1.06 152.20 79.23 11418.15 Zumbador de Eulampis holosericeus Green-throated Carib Pecho Azul GAP0507 16.56 5822.58 15.85 923.01 14.69 855.34 1.16 67.67 1.95 113.48 82.20 4786.09 Falco columbarius Merlin Halcón Migratorio GAP0517 51.38 18065.49 11.23 2028.37 10.56 1907.44 0.67 120.93 1.74 314.74 87.03 15722.38 Falco peregrinus Peregrine Falcon Halcón Peregrino GAP0518 36.43 12808.61 11.70 1498.74 11.18 1431.73 0.52 67.01 0.93 119.30 87.37 11190.57 Falco sparverius American Kestrel Halcón Común GAP0520 61.81 21733.82 12.63 2745.60 11.06 2403.35 1.57 342.25 1.70 369.87 85.67 18618.35 Gallinazo Fulica americana American Coot Americano GAP0524 1.03 363.28 14.32 52.02 10.52 38.21 3.80 13.81 8.08 29.36 77.60 281.90 Gallinago delicata Wilson's Snipe Agachona Común GAP0528 0.50 175.11 12.81 22.43 9.95 17.43 2.86 5.00 2.87 5.02 84.32 147.66 Geothlypis trichas Common Yellowthroat Reinita Pica Tierra GAP0536 1.31 461.05 18.14 83.63 12.41 57.21 5.73 26.42 1.11 5.12 80.75 372.30 Herpestes javanicus Small Indian Mongoose Mangosta GAP0606 72.02 25325.46 13.38 3387.44 11.99 3036.05 1.39 351.39 1.59 403.75 85.03 21534.27 Iguana iguana Green Iguana Iguana Común GAP0643 30.60 10760.84 13.90 1495.39 12.25 1318.10 1.65 177.29 2.01 216.59 84.09 9048.86 Limnodromus griseus Short-billed Dowitcher Agujeta Pico Corto GAP0691 1.84 647.44 14.49 93.79 12.13 78.56 2.35 15.23 2.32 14.99 83.20 538.66 Margarops fuscatus Pearly-eyed Thrasher Zorzal Pardo GAP0747 75.02 26379.46 11.90 3139.10 11.06 2918.76 0.84 220.34 1.19 314.05 86.91 22926.31 Mimus polyglottos Northern Mockingbird Ruiseñor GAP0773 31.32 11012.59 12.26 1350.61 11.28 1242.01 0.99 108.60 1.62 177.90 86.12 9484.08 Molossus molossus Velvety Free-tailed Bat Murciélago Casero GAP0778 90.42 31794.34 10.85 3449.51 9.74 3097.56 1.11 351.95 1.45 461.73 87.70 27883.10 Numenius phaeopus Whimbrel Playero Trinador GAP0819 3.08 1082.61 10.92 118.25 9.13 98.80 1.80 19.45 3.12 33.76 85.96 930.60 Venado Odocoileus virginianus White-tailed Deer Rabiblanco GAP0829 70.17 24674.90 13.32 3287.91 12.34 3045.53 0.98 242.38 1.50 370.65 85.17 21016.34 Antillean Crested Zumbador Orthorhyncus cristatus Hummingbird Crestado GAP0861 55.92 19662.33 11.45 2251.16 10.10 1985.75 1.35 265.41 1.62 318.31 86.93 17092.86 Reinita Parula americana Northern Parula Pechidorada GAP0886 39.92 14038.30 19.99 2806.24 18.87 2649.69 1.12 156.55 1.76 247.64 78.25 10984.42 Patagioenas Paloma leucocephala White-crowned Pigeon Cabeciblanca GAP0892 50.97 17923.33 17.72 3176.76 16.09 2883.12 1.64 293.64 2.24 401.69 80.03 14344.88 Patagioenas squamosa Scaly-naped Pigeon Paloma Turca GAP0893 58.41 20539.92 15.34 3151.77 13.99 2873.57 1.35 278.20 1.79 368.06 82.86 17020.09 American Golden- Pluvialis dominica plover Chorlo Dorado GAP0924 2.24 786.13 11.45 90.04 10.00 78.59 1.46 11.45 1.05 8.27 87.49 687.82 Riparia riparia Bank Swallow Golondrina Parda GAP0998 11.98 4212.54 14.86 626.07 13.94 587.16 0.92 38.91 0.15 6.51 84.98 3579.96 Sphaerodactylus Salamanquita macrolepis Common Dwarf Gecko Común GAP1077 62.11 21841.36 14.60 3187.93 13.34 2912.66 1.26 275.27 1.99 434.43 83.42 18219.00 Playero Guineilla Tringa melanoleuca Greater Yellowlegs Grande GAP1199 0.85 300.41 19.34 58.10 17.48 52.51 1.86 5.59 4.29 12.88 76.37 229.43 Typhlops richardii Richard's Blind Snake Vibora Común GAP1212 37.05 13027.26 16.53 2153.50 15.55 2026.37 0.98 127.13 1.05 136.43 82.42 10737.33
142
Table 25. Species with 20% to less than 50% of their total area protected (56 species). Total areas for the U.S. Virgin Islands are given in the first row. Percent of total of these species on the island (first column) is followed by percent of each species in Status 1-4.
Protected Total Habitat Status 1 and 2 Status 1 Status 2 Status 3 Status 4 Common Name Common Name Species Scientific Name (English) (Spanish) ID % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.00 35162.98 11.09 3898.58 9.85 3461.81 1.24 436.77 1.94 682.01 86.97 30582.39 Amphisbaena Virgin Islands Culebrita Ciega de fenestrata Amphisbaena Islas Vírgenes GAP0061 13.91 4889.59 40.66 1987.93 38.43 1879.04 2.23 108.89 0.96 47.01 58.38 2854.65 Anolis stratulus Barred Anole Lagartijo Manchado GAP0102 29.28 10296.12 28.39 2923.43 26.97 2777.24 1.42 146.19 0.61 62.71 71.00 7309.98 Ardea alba Great Egret Garzón Blanco GAP0136 4.32 1518.51 24.37 370.09 22.21 337.32 2.16 32.77 4.61 69.96 71.02 1078.46 Ardea herodias Great Blue Heron Garzón Cenizo GAP0137 2.80 982.81 31.83 312.78 27.75 272.70 4.08 40.08 5.82 57.23 62.35 612.80 Arenaria interpres Ruddy Turnstone Playero Turco GAP0139 1.03 360.96 43.15 155.77 38.44 138.75 4.72 17.02 6.58 23.75 50.27 181.44 Butorides virescens Green Heron Martinete Verde GAP0201 7.52 2645.67 26.10 690.44 24.69 653.29 1.40 37.15 3.91 103.33 70.00 1851.90 Calidris alba Sanderling Playero Arenero GAP0209 0.55 193.54 41.47 80.27 36.63 70.89 4.85 9.38 7.90 15.28 50.63 97.99 Calidris himantopus Stilt Sandpiper Playero Patilargo GAP0215 2.17 764.59 24.02 183.68 21.30 162.85 2.72 20.83 6.44 49.21 69.54 531.70 Calidris mauri Western Sandpiper Playero Occidental GAP0216 1.50 527.75 23.65 124.81 23.65 124.81 0.00 0.00 9.03 47.67 67.32 355.27 Catoptrophorus semipalmatus Willet Playero Aliblanco GAP0260 0.98 344.13 20.24 69.64 16.92 58.23 3.32 11.41 5.97 20.55 73.79 253.94 Martín Pescador Ceryle alcyon Belted Kingfisher Norteño GAP0270 17.52 6160.37 22.18 1366.56 20.55 1266.16 1.63 100.40 2.81 172.83 75.01 4620.98 Charadrius Semipalmated Plover Playero acollarado GAP0284 1.24 436.93 26.93 117.67 23.25 101.58 3.68 16.09 4.49 19.64 68.57 299.62 143
semipalmatus Charadrius wilsonia Wilson's Plover Chorlo Marítimo GAP0286 1.47 517.67 33.05 171.10 29.30 151.68 3.75 19.42 4.72 24.43 62.23 322.14 Pejeblanco, Tortuga Chelonia mydas Green Sea Turtle Verde GAP0291 0.17 58.90 44.21 26.04 33.09 19.49 11.12 6.55 6.15 3.62 49.64 29.24 Dendroica petechia Yellow Warbler Reinita Amarilla GAP0397 2.85 1000.96 30.73 307.64 26.05 260.80 4.68 46.84 6.28 62.86 62.99 630.46 Black-throated Green Dendroica virens Warbler Reinita Verdosa GAP0401 10.54 3706.45 30.86 1143.98 30.84 1142.97 0.03 1.01 0.57 21.14 68.57 2541.33 Egretta caerulea Little Blue Heron Garza Azul GAP0424 3.26 1146.46 25.10 287.80 21.71 248.87 3.40 38.93 4.88 55.94 70.02 802.72 Egretta thula Snowy Egret Garza Blanca GAP0428 4.03 1416.26 24.10 341.34 21.49 304.31 2.61 37.03 4.84 68.50 71.06 1006.42 Egretta tricolor Tricolored Heron Garza Pechiblanca GAP0429 2.78 977.85 28.02 274.01 24.93 243.79 3.09 30.22 6.43 62.87 65.55 640.97 Eleutherodactylus cochranae Whistling Frog Coquí Pitito GAP0445 24.62 8658.39 27.64 2393.42 26.47 2291.94 1.17 101.48 0.71 61.26 71.65 6203.71 Eretmochelys imbricata Hawksbill Turtle Tortuga Carey GAP0489 0.22 76.67 40.21 30.83 29.41 22.55 10.80 8.28 4.72 3.62 55.07 42.22 Magnificent Fregata magnificens Frigatebird Fragata Magnifica GAP0523 3.47 1221.33 35.86 437.93 32.29 394.41 3.56 43.52 4.54 55.43 59.60 727.97 Gallinula chloropus Common Moorhen Gallareta Común GAP0530 2.10 738.53 21.26 157.01 17.21 127.12 4.05 29.89 6.20 45.80 72.54 535.72 Geochelone Tortuga de Patas carbonaria Red-footed Tortoise Rojas GAP0535 9.43 3315.46 37.41 1240.28 37.24 1234.52 0.17 5.76 2.14 71.03 60.45 2004.15 Paloma Perdíz de Geotrygon mystacea Bridled Quail-Dove Martinica GAP0539 6.08 2138.14 45.22 966.88 45.01 962.45 0.21 4.43 2.82 60.29 51.96 1110.97 American Haematopus palliatus Oystercatcher Ostrero Americano GAP0575 0.76 266.12 35.62 94.79 31.11 82.79 4.51 12.00 2.39 6.35 61.99 164.98 Helmitheros Worm-eating vermivorus Warbler Reinita Gusanera GAP0600 15.10 5310.64 43.36 2302.80 43.36 2302.57 0.00 0.23 0.90 47.88 55.74 2959.96 Himantopus mexicanus Black-necked Stilt Viuda Mexicana GAP0609 1.75 616.08 26.34 162.28 22.57 139.02 3.78 23.26 5.71 35.20 67.95 418.60 Larus atricilla Laughing Gull Gaviota Cabecinegra GAP0673 2.27 799.31 30.11 240.69 25.45 203.42 4.66 37.27 3.96 31.67 65.93 526.95 Lesser Antillean Come Ñame Antillano Loxigilla noctis Bullfinch Menor GAP0712 27.02 9499.64 23.49 2231.23 23.35 2218.51 0.13 12.72 2.00 190.24 74.51 7078.17 Mabuya mabouya Slippery-backed sloanei Mabuya Lucía GAP0728 1.02 357.72 22.87 81.82 12.29 43.98 10.58 37.84 35.20 125.91 41.93 149.99 Black-and-white Mniotilta varia Warbler Reinita Trepadora GAP0776 28.74 10104.85 27.58 2787.06 26.22 2649.69 1.36 137.37 1.92 193.68 70.50 7124.11 Molothrus bonariensis Shiny Cowbird Tordo Lustroso GAP0779 9.95 3500.08 29.61 1036.38 29.53 1033.66 0.08 2.72 2.71 95.01 67.68 2368.69 Noctilio leporinus Greater Bulldog Bat Murciélago Pescador GAP0814 2.89 1014.85 46.53 472.21 42.30 429.28 4.23 42.93 1.99 20.24 51.48 522.40 Yellow-crowned Nyctanassa violacea Night-heron Yaboa Comun GAP0821 5.61 1972.19 34.77 685.71 32.08 632.73 2.69 52.98 4.61 90.98 60.62 1195.50 Black-crowned Night- Nycticorax nycticorax heron Yaboa Real GAP0823 4.65 1636.75 35.53 581.61 34.36 562.35 1.18 19.26 3.92 64.19 60.54 990.95 Pandion haliaetus Osprey Aguila Pescadora GAP0869 2.91 1023.13 25.63 262.18 23.32 238.58 2.31 23.60 5.27 53.93 69.10 707.02 Pelecanus occidentalis Brown Pelican Pelícano Pardo GAP0895 4.92 1730.97 34.05 589.33 30.38 525.95 3.66 63.38 4.37 75.66 61.58 1065.98 Phaethon aethereus Red-billed Tropicbird Rabijunco Piquirrojo GAP0903 0.88 308.96 36.66 113.27 28.39 87.72 8.27 25.55 21.76 67.23 41.58 128.46 White-tailed Phaethon lepturus Tropicbird Rabijunco Coliblanco GAP0904 0.88 309.78 33.97 105.23 29.84 92.45 4.13 12.78 17.70 54.84 48.33 149.71 Podilymbus podiceps Pied-billed Grebe Zaramago GAP0926 0.60 211.37 24.51 51.80 19.10 40.38 5.40 11.42 2.54 5.36 72.96 154.21 Porzana carolina Sora Gallito Sora GAP0937 0.41 145.25 41.07 59.65 36.48 52.99 4.59 6.66 3.37 4.89 55.57 80.71 Prothonotary Protonotaria citrea Warbler Reinita Protonotaria GAP0955 10.15 3567.43 33.15 1182.45 32.93 1174.92 0.21 7.53 0.87 31.03 65.98 2353.95 Seiurus aurocapillus Ovenbird Pizpita Dorada GAP1041 14.73 5180.45 38.96 2018.14 38.94 2017.14 0.02 1.00 0.44 22.73 60.60 3139.58 Seiurus Northern noveboracensis Waterthrush Pizpita de Mangle GAP1043 5.80 2039.12 34.20 697.36 31.59 644.22 2.61 53.14 3.45 70.32 62.35 1271.44 144
Setophaga ruticilla American Redstart Reinita Candelita GAP1062 5.02 1764.64 34.19 603.27 32.79 578.70 1.39 24.57 1.51 26.66 64.30 1134.71 Desmarest's Fig- Murciélago Rojo Stenoderma rufum eating Bat Frutero GAP1116 14.43 5074.90 24.41 1238.98 24.26 1230.98 0.16 8.00 3.33 168.83 72.26 3667.09 Sterna antillarum Least Tern Gaviota Chica GAP1125 0.82 288.90 23.68 68.40 18.64 53.86 5.03 14.54 6.60 19.06 69.73 201.44 Sterna hirundo Common Tern Gaviota Común GAP1130 1.39 487.32 30.77 149.96 23.78 115.88 6.99 34.08 5.97 29.08 63.26 308.28 Sterna maxima Royal Tern Gaviota Real GAP1131 0.81 285.88 41.80 119.51 35.72 102.12 6.08 17.39 4.41 12.61 53.78 153.76 Sterna sandvicensis Sandwich Tern Gaviota Piquiaguda GAP1134 0.56 196.67 29.71 58.43 17.67 34.75 12.04 23.68 5.88 11.56 64.41 126.68 Sula leucogaster Brown Booby Boba Parda GAP1146 0.68 238.90 47.48 113.43 40.02 95.60 7.46 17.83 15.14 36.16 37.38 89.31 Brazilian Free-tailed Murciélago de Cola Tadarida brasiliensis Bat Libre GAP1168 3.71 1303.87 48.00 625.85 48.00 625.85 0.00 0.00 0.76 9.95 51.24 668.07 Playero Guineilla Tringa flavipes Lesser Yellowlegs Pequeño GAP1198 2.00 703.95 26.44 186.13 22.97 161.67 3.47 24.46 4.15 29.24 69.41 488.58 Black-whiskered Vireo altiloquus Vireo Julian Chivi GAP1234 36.60 12870.83 20.48 2636.16 20.03 2578.54 0.45 57.62 2.29 294.29 77.23 9940.38 Wilsonia citrina Hooded Warbler Reinita Viuda GAP1240 16.49 5798.94 41.41 2401.48 41.38 2399.33 0.04 2.15 1.07 62.11 57.52 3335.35
145
Table 26. Species with more than 50% of their total predicted habitat protected (14 species). Total areas for the U.S. Virgin Islands are given in the first row. Percent of total of these species on the island (first column) is followed by percent of each species in Status 1-4.
Protected Total Habitat Status 1 and 2 Status 1 Status 2 Status 3 Status 4 Common Name Common Name Species Scientific Name (English) (Spanish) ID % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.00 35162.98 11.09 3898.58 9.85 3461.81 1.24 436.77 1.94 682.01 86.97 30582.39 Siguana de Santa Ameiva polops St. Croix Ground Lizard Cruz GAP0054 0.23 82.30 87.39 71.92 82.96 68.28 4.42 3.64 12.61 10.38 0.00 0.00 Anous stolidus Brown Noddy Cervera GAP0104 0.27 94.57 58.75 55.56 42.78 40.46 15.97 15.10 0.94 0.89 40.31 38.12 Dendroica Black-throated Blue caerulescens Warbler Reinita Azul GAP0386 12.82 4508.64 52.41 2363.10 52.39 2362.30 0.02 0.80 0.22 9.90 47.37 2135.64 Dendroica magnolia Magnolia Warbler Reinita Manchada GAP0394 3.81 1339.90 67.72 907.40 67.72 907.38 0.00 0.02 0.83 11.15 31.45 421.35 Myiarchus Puerto Rican antillarum Flycatcher Juí de Puerto Rico GAP0799 11.47 4034.62 60.70 2449.05 60.52 2441.91 0.18 7.14 0.20 8.25 39.09 1577.32 Oxyura jamaicensis Ruddy Duck Pato Chorizo GAP0865 0.11 38.74 60.14 23.30 30.87 11.96 29.27 11.34 4.47 1.73 35.39 13.71 Puffinus Pampero de lherminieri Audubon's Shearwater Audubon GAP0976 0.10 33.49 99.22 33.23 84.77 28.39 14.45 4.84 0.00 0.00 0.78 0.26 Seiurus motacilla Louisiana Waterthrush Pizpita de Río GAP1042 1.61 566.17 56.93 322.33 56.54 320.12 0.39 2.21 4.46 25.26 38.61 218.58 Sterna anaethetus Bridled Tern Gaviota Monja GAP1124 0.21 74.29 76.28 56.67 50.63 37.61 25.66 19.06 1.20 0.89 22.52 16.73 Sterna dougallii Roseate Tern Palometa GAP1127 0.19 65.21 54.10 35.28 37.46 24.43 16.64 10.85 0.06 0.04 45.84 29.89 Sterna fuscata Sooty Tern Gaviota Oscura GAP1129 0.03 11.50 76.87 8.84 20.78 2.39 56.09 6.45 0.00 0.00 23.13 2.66 Sula dactylatra Masked Booby Boba Enmascarada GAP1145 0.05 19.20 99.90 19.18 91.56 17.58 8.33 1.60 0.00 0.00 0.10 0.02 Sula sula Red-footed Booby Boba Patirroja GAP1147 0.21 74.56 65.88 49.12 48.55 36.20 17.33 12.92 32.35 24.12 1.77 1.32
146
Table 23. Federally listed endangered (FE), threatened (FT) and partial status (PS) species and those afforded legal protection by U.S. Virgin Islands Endangered and Indigenous Species Act as Locally Endangered (LE), Locally Threatened (LT), Local Special Concern (LSC), and finally the Management Concern the species represent for the Division of Fish and Wildlife of the Department of Planing Natural indicating their area protected (47 species). Total areas for U.S. Virgin Islands are given in the first row. Percent of total of these species on the island (first column) is followed by percent of each species in Status 1-4.
Status 1 and 2 Total Habitat Status 1 Status 2 Status 3 Status 4 (Protected) Common Name Common Name Species Scientific Name (English) (Spanish) ID % HA % HA % HA % HA % HA % HA U.S. Virgin Islands Territory 100.00 35162.98 11.09 3898.58 9.85 3461.81 1.24 436.77 1.94 682.01 86.97 30582.39 Alsophis portoricensis (LT)(SC) Puerto Rican Racer Culebra Corredora GAP0036 2.38 838.21 15.30 128.28 11.51 96.50 3.79 31.78 6.88 57.68 77.81 652.25 Ameiva polops St. Croix Ground Siguana de Santa (FE)(LE)(SGC) Lizard Cruz GAP0054 0.23 82.30 87.39 71.92 82.96 68.28 4.42 3.64 12.61 10.38 0.00 0.00 Amphisbaena fenestrata Virgin Islands Culebrita Ciega de (DD)(SGC) Amphisbaena Islas Vírgenes GAP0061 13.91 4889.59 40.66 1987.93 38.43 1879.04 2.23 108.89 0.96 47.01 58.38 2854.65 White-cheeked Pato Quijada Anas bahamensis (LE)(SC) Pintail Colorada GAP0067 2.01 705.43 29.95 211.26 26.49 186.87 3.46 24.39 5.29 37.29 64.77 456.88 Ardea alba (LE)(SLC) Great Egret Garzón Blanco GAP0136 4.32 1518.51 24.37 370.09 22.21 337.32 2.16 32.77 4.61 69.96 71.02 1078.46 Ardea herodias (LE)(SLC) Great Blue Heron Garzón Cenizo GAP0137 2.80 982.81 31.83 312.78 27.75 272.70 4.08 40.08 5.82 57.23 62.35 612.80 Brachyphylla cavernarum Antillean Fruit- Murciélago (DD)(SGC) eating Bat Cavernícola GAP0186 50.85 17881.82 16.93 3026.97 15.74 2814.12 1.19 212.85 1.57 281.45 81.50 14573.40 Calidris canutus (LE)(SGC) Red Knot Playero Gordo GAP0212 0.36 127.94 9.36 11.97 3.29 4.21 6.07 7.76 4.25 5.44 86.39 110.53 Playerito Calidris minutilla (LSC)(SC) Least Sandpiper Menudillo GAP0218 1.50 528.29 18.79 99.28 16.01 84.58 2.78 14.70 4.60 24.31 76.61 404.70 Catoptrophorus semipalmatus (LT)(SGC) Willet Playero Aliblanco GAP0260 0.98 344.13 20.24 69.64 16.92 58.23 3.32 11.41 5.97 20.55 73.79 253.94 Charadrius wilsonia (LSC)(SC) Wilson's Plover Chorlo Marítimo GAP0286 1.47 517.67 33.05 171.10 29.30 151.68 3.75 19.42 4.72 24.43 62.23 322.14 Chelonia mydas Pejeblanco, (FT)(LT)(SGC) Green Sea Turtle Tortuga Verde GAP0291 0.17 58.90 44.21 26.04 33.09 19.49 11.12 6.55 6.15 3.62 49.64 29.24 Chordeiles gundlachii Antillean Querequequé (LE)(SGC) Nighthawk Antillano GAP0303 13.74 4833.12 0.06 3.07 0.00 0.00 0.06 3.07 0.79 38.15 99.15 4791.90 Coccyzus americanus Yellow-billed Pájaro Bobo (PS)(LSC)(SC) Cuckoo Pechiblanco GAP0321 4.33 1523.82 3.63 55.28 2.97 45.21 0.66 10.07 5.95 90.60 90.43 1377.94 Dendroica coronata Yellow-rumped (LSC)(SC) Warbler Reinita Coronada GAP0390 31.36 11026.89 25.46 2807.88 25.19 2777.37 0.28 30.51 1.87 205.93 72.67 8013.08 Dermochelys coriacea Leatherback Sea (FE)(LE)(SGC) Turtle Tinglar GAP0403 0.83 291.00 6.16 17.92 5.29 15.38 0.87 2.54 8.32 24.21 85.52 248.87 Egretta thula (LSC)(SC) Snowy Egret Garza Blanca GAP0428 4.03 1416.26 24.10 341.34 21.49 304.31 2.61 37.03 4.84 68.50 71.06 1006.42
Egretta tricolor (LSC)(SC) Tricolored Heron Garza Pechiblanca GAP0429 2.78 977.85 28.02 274.01 24.93 243.79 3.09 30.22 6.43 62.87 65.55 640.97 147
Eleutherodactylus lentus Yellow Mottled Coquí de Islas (DD)(SC) Coqui Vírgenes GAP0453 28.51 10024.86 2.44 244.94 1.65 165.40 0.79 79.54 1.24 124.70 96.31 9655.22 Epicrates monensis granti Virgin Islands Tree Culebrón de la Isla (FE)(LE)(SGC) Boa Virgin GAP0472 8.42 2962.05 1.65 48.95 0.05 1.37 1.61 47.58 0.09 2.79 98.25 2910.31 Eretmochelys imbricata (FE)(LE)(SGC) Hawksbill Turtle Tortuga Carey GAP0489 0.22 76.67 40.21 30.83 29.41 22.55 10.80 8.28 4.72 3.62 55.07 42.22 Falco peregrinus (FE)(LSC)(SC) Peregrine Falcon Halcón Peregrino GAP0518 36.43 12808.61 11.70 1498.74 11.18 1431.73 0.52 67.01 0.93 119.30 87.37 11190.57 Fregata magnificens Magnificent (LE)(SGC) Frigatebird Fragata Magnifica GAP0523 3.47 1221.33 35.86 437.93 32.29 394.41 3.56 43.52 4.54 55.43 59.60 727.97 Gallinazo Fulica americana (LT)(SGC) American Coot Americano GAP0524 1.03 363.28 14.32 52.02 10.52 38.21 3.80 13.81 8.08 29.36 77.60 281.90
Fulica caribaea (LE)(SGC) Caribbean Coot Gallinazo Caribeño GAP0525 0.56 195.71 7.75 15.16 5.36 10.49 2.39 4.67 6.59 12.89 85.67 167.66 Gallinula chloropus (PS)(SLC) Common Moorhen Gallareta Común GAP0530 2.10 738.53 21.26 157.01 17.21 127.12 4.05 29.89 6.20 45.80 72.54 535.72 Geothlypis trichas Common (LSC)(SC) Yellowthroat Reinita Pica Tierra GAP0536 1.31 461.05 18.14 83.63 12.41 57.21 5.73 26.42 1.11 5.12 80.75 372.30 Geotrygon mystacea Paloma Perdíz de (LT)(SGC) Bridled Quail-Dove Martinica GAP0539 6.08 2138.14 45.22 966.88 45.01 962.45 0.21 4.43 2.82 60.29 51.96 1110.97 Haematopus palliatus American Ostrero (LT)(SGC) Oystercatcher Americano GAP0575 0.76 266.12 35.62 94.79 31.11 82.79 4.51 12.00 2.39 6.35 61.99 164.98 Helmitheros vermivorus Worm-eating (LSC)(SC) Warbler Reinita Gusanera GAP0600 15.10 5310.64 43.36 2302.80 43.36 2302.57 0.00 0.23 0.90 47.88 55.74 2959.96 Himantopus mexicanus (PS)(SLC) Black-necked Stilt Viuda Mexicana GAP0609 1.75 616.08 26.34 162.28 22.57 139.02 3.78 23.26 5.71 35.20 67.95 418.60 Limnodromus griseus Short-billed (LSC)(SC) Dowitcher Agujeta Pico Corto GAP0691 1.84 647.44 14.49 93.79 12.13 78.56 2.35 15.23 2.32 14.99 83.20 538.66 Lesser Antillean Come Ñame Loxigilla noctis (LSC)(SC) Bullfinch Antillano Menor GAP0712 27.02 9499.64 23.49 2231.23 23.35 2218.51 0.13 12.72 2.00 190.24 74.51 7078.17 Mabuya mabouya sloanei Slippery-backed (LT)(SGC) Mabuya Lucía GAP0728 1.02 357.72 22.87 81.82 12.29 43.98 10.58 37.84 35.20 125.91 41.93 149.99 Myiarchus antillarum Puerto Rican (LE)(SGC) Flycatcher Juí de Puerto Rico GAP0799 11.47 4034.62 60.70 2449.05 60.52 2441.91 0.18 7.14 0.20 8.25 39.09 1577.32 Noctilio leporinus Greater Bulldog Murciélago (DD)(SGC) Bat Pescador GAP0814 2.89 1014.85 46.53 472.21 42.30 429.28 4.23 42.93 1.99 20.24 51.48 522.40 Numenius phaeopus (LT)(SGC) Whimbrel Playero Trinador GAP0819 3.08 1082.61 10.92 118.25 9.13 98.80 1.80 19.45 3.12 33.76 85.96 930.60 Nycticorax nycticorax Black-crowned (LSC)(SC) Night-heron Yaboa Real GAP0823 4.65 1636.75 35.53 581.61 34.36 562.35 1.18 19.26 3.92 64.19 60.54 990.95 Oxyura jamaicensis (LSC)(SC) Ruddy Duck Pato Chorizo GAP0865 0.11 38.74 60.14 23.30 30.87 11.96 29.27 11.34 4.47 1.73 35.39 13.71 Patagioenas leucocephala White-crowned Paloma (LT)(SGC) Pigeon Cabeciblanca GAP0892 50.97 17923.33 17.72 3176.76 16.09 2883.12 1.64 293.64 2.24 401.69 80.03 14344.88
148
Pelecanus occidentalis (FE)(LSC)(SC) Brown Pelican Pelícano Pardo GAP0895 4.92 1730.97 34.05 589.33 30.38 525.95 3.66 63.38 4.37 75.66 61.58 1065.98 Phaethon aethereus Red-billed Rabijunco (LSC)(SC) Tropicbird Piquirrojo GAP0903 0.88 308.96 36.66 113.27 28.39 87.72 8.27 25.55 21.76 67.23 41.58 128.46 Phaethon lepturus White-tailed Rabijunco (LT)(SGC) Tropicbird Coliblanco GAP0904 0.88 309.78 33.97 105.23 29.84 92.45 4.13 12.78 17.70 54.84 48.33 149.71
Podilymbus podiceps (SC) Pied-billed Grebe Zaramago GAP0926 0.60 211.37 24.51 51.80 19.10 40.38 5.40 11.42 2.54 5.36 72.96 154.21 Progne dominicensis Golondrina de (LSC)(SGC) Caribbean Martin Iglesias GAP0951 17.40 6119.25 2.16 132.33 1.86 113.56 0.31 18.77 1.08 66.05 96.76 5920.87 Protonotaria citrea Prothonotary Reinita (LSC)(SC) Warbler Protonotaria GAP0955 10.15 3567.43 33.15 1182.45 32.93 1174.92 0.21 7.53 0.87 31.03 65.98 2353.95 Puffinus lherminieri Audubon's Pampero de (LE)(SGC) Shearwater Audubon GAP0976 0.10 33.49 99.22 33.23 84.77 28.39 14.45 4.84 0.00 0.00 0.78 0.26 Rallus longirostris (PS)(LE)(SGC) Clapper Rail Pollo de Mangle GAP0985 0.18 64.52 6.37 4.11 6.37 4.11 0.00 0.00 29.09 18.77 64.54 41.64 Louisiana Seiurus motacilla (LSC)(SC) Waterthrush Pizpita de Río GAP1042 1.61 566.17 56.93 322.33 56.54 320.12 0.39 2.21 4.46 25.26 38.61 218.58 Stenoderma rufum Desmarest's Fig- Murciélago Rojo (DD)(SGC) eating Bat Frutero GAP1116 14.43 5074.90 24.41 1238.98 24.26 1230.98 0.16 8.00 3.33 168.83 72.26 3667.09 Sterna antillarum (PS, FE)(LSC)(SC) Least Tern Gaviota Chica GAP1125 0.82 288.90 23.68 68.40 18.64 53.86 5.03 14.54 6.60 19.06 69.73 201.44 Sterna dougallii (FT)(LSC)(SLC) Roseate Tern Palometa GAP1127 0.19 65.21 54.10 35.28 37.46 24.43 16.64 10.85 0.06 0.04 45.84 29.89 Boba Sula dactylatra (LE)(SGC) Masked Booby Enmascarada GAP1145 0.05 19.20 99.90 19.18 91.56 17.58 8.33 1.60 0.00 0.00 0.10 0.02 Sula sula (LT)(SGC) Red-footed Booby Boba Patirroja GAP1147 0.21 74.56 65.88 49.12 48.55 36.20 17.33 12.92 32.35 24.12 1.77 1.32 Tachybaptus dominicus (LE)(SGC) Least Grebe Tigua GAP1166 1.86 653.16 0.01 0.07 0.00 0.02 0.01 0.05 4.17 27.24 95.82 625.85 Tadarida brasiliensis Brazilian Free- Murciélago de (DD)(SGC) tailed Bat Cola Libre GAP1168 3.71 1303.87 48.00 625.85 48.00 625.85 0.00 0.00 0.76 9.95 51.24 668.07 Richard's Blind Typhlops richardii (DD)(SC) Snake Vibora Común GAP1212 37.05 13027.26 16.53 2153.50 15.55 2026.37 0.98 127.13 1.05 136.43 82.42 10737.33 Wilsonia citrina (LSC)(SC) Hooded Warbler Reinita Viuda GAP1240 16.49 5798.94 41.41 2401.48 41.38 2399.33 0.04 2.15 1.07 62.11 57.52 3335.35
149
The following summaries and charts highlight potential gaps and conservation needs for vertebrate species.
Species with <1% of predicted distribution in status 1 or 2 (Table 18 and Figure ?): Four species have less than 1% of their habitat protected. In terms of the amount of protected habitat, these species appear to be some of the most vulnerable, although this is only one of many factors that affect the long-term survival of a species. Three birds and one reptile fall within this range of habitat protection. Two of the birds are associated with wetlands or water- bodies in general: The least grebe and the purple gallinule. Habitats such as wetlands and salt ponds in U.S. Virgin Islands are under great threat of degradation or destruction as has occurred in the past (McNair et al. 2006). The other bird, the Antillean nighthawk, is a breeding migrant whose habitat is more represented in the landscape than the other three species (13.7%) but is a species of greatest concern for the DPNR Division of Fish and Wildlife (Plantenberg et al. 2005). The last one, a reptile, the fat-tailed gecko is an introduced species and represents little concern to local environmental agencies.
Species with 1% to <10% of predicted distribution in status 1 or 2 (Table 19 and Figure ?): Thirty-seven species have 1% to less than 10% of their habitat protected. This group of species includes birds, reptiles and amphibians; and some of these species habitats cover a significant amount of land area. Such is the case of the zenaida dove (82.2% of the islands), the smooth- billed ani (53.2%) and the St. Croix anole (47.2%). It could be said that these species are relatively widespread and as a consequence they have very little protection of their habitat (5.1% and 3.3% respectively). There are some species in this range of protection, however, whose habitat covers very little of the territory, even less than 1%, including some sandpipers, the leatherback sea turtle, the clapper rail, the Caribbean coot and others. These latter species share in common certain habitats, such as sandy beaches or wetlands. In addition, the endangered Virgin Islands boa, Epicrates monensis granti, is also in this range, and its habitat (8.4% of the islands or 2962 ha) currently enjoys very little protection (1.7%).
Species with l0% to <20% of predicted distribution in status 1 or 2 (Table 20): Forty-one species have 10% to less than 20% of their habitat protected. Most of these species associated habitats cover a significant amount of land area, 24% in average, but up to 90.4% of the territory in the case of the Velvety free-tailed bat. In contrast, the least sandpiper habitat only covers 1.5% of the territory and some other four species habitats represent less than 1%.
Species with 20% to <50% of predicted distribution in status 1 or 2 (Table 21): Fifty-six species have 20% to less than 50% of their habitat protected. These species habitats cover between 0.2% (green sea turtle) and 36.6% (black-whiskered vireo) of the islands territory, 6.7% in average and a standard deviation of 6.4%. In other words, the amount of area covered by the habitats associated with these species shows much less variation when compared to the previous two range categories of habitat protection.
150
Species with at least 50% representation in GAP status 1 and 2 (Table 22 and Figure ?): Thirteen species have more than 50% of their habitat protected. Most of these species habitats cover less than 1% of the islands territory. This is the case of some marine birds - three terns, two boobies and the brown noddy – whose habitat is mostly located in off-shore cays which have been designated marine sanctuaries by the DPNR Division of Fish and Wildlife. Although this category encompasses mostly birds there is also one reptile: St. Croix Ground Lizard. This latter has been extirpated from St. Croix’s main island and is now relegated to the four cays surrounding St. Croix. It is listed as endangered under the U.S. Endangered Species Act. Two species in this protection range, the Puerto Rican Flycatcher and the Black-throated Blue Warbler, cover more territory than the rest, 11.5% and 12.8% respectively.
Endangered species representation in GAP status 1 and 2 (Table 23): Fifty-eight species are listed as either federally threatened or endangered or given partial status, are locally listed by the USVI Indigenous Species Act as locally threatened, locally endangered, data deficient or special concern, or are classified as species of concern or species of greatest concern by the DPNR Division of Fish and Wildlife. Most cover less than 10% of the islands territory and 18 cover less than 1%. Only two cover more than 50% of the territory: the Antillean fruit-eating bat and the white-crowned pigeon. However, 48 of these species have more than 10% of their habitat under protection. There are some results that stand out. For example, the Virgin Island tree boa only has 1.6% of its predicted habitat under protection, and the endemic yellow mottled coqui, only 2.4% protected.
% Island coverage % Habitat protection 0 5 10 15
Porphyrio martinica cuatro
Thecadactylus rapicauda tres
Tachybaptus dominicus dos
Chordeiles gundlachii uno
0.00 0.05 0.10 0.15 0.20 Figure ? Island coverage and percent protection for species with <1% of predicted distribution in status 1 or 2.
151
% Island coverage % Habitat protection 0 20 40 60 80 100
Rallus longirostris text37
Tringa solitaria text36 Calidris canutus text35 Aythya affinis text34 Fulica caribaea text33 Charadrius vociferus text32 Dermochelys coriacea text31 Calidris melanotos text30
Anas americana text29
Calidris fuscicollis text28 Anolis pulchellus text27 Pluvialis squatarola text26 Coccyzus americanus text25 Eleutherodactylus coqui text24 Epicrates monensis granti text23 Petrochelidon pyrrhonota text22 Columba livia text21
Passerina cyanea text20
Arrhyton exiguum text19 Sphaerodactylus beattyi text18 Progne dominicensis text17 Passer domesticus text16 Osteopilus septentrionalis text15 Hemidactylus mabouia text14 Bufo marinus text13
Hirundo rustica text12
Eleutherodactylus lentus text11
Leptodactylus albilabris text10 Dendroica striata text9 Bubulcus ibis text8 Anolis acutus text7 Tiaris bicolor text6 Columbina passerina text5 Crotophaga ani text4
Zenaida asiatica text3
Tyrannus dominicensis text2 Zenaida aurita text1 0 2 4 6 8 10
Figure ?. Island coverage and percent protection for species with 1% to <10% of predicted distribution in status 1 or 2.
152
% Island coverage % Habitat protected 0 25 50 75 100
Aythya collaris te… Gallinago delicata te… Anas crecca te… Tringa melanoleuca te… Fulica americana te… Geothlypis trichas te… Calidris minutilla te… Limnodromus griseus te… Pluvialis dominica te… Alsophis portoricensis te… Numenius phaeopus te… Calidris pusilla te… Actitis macularia te… Riparia riparia te… Eulampis holosericeus te… Anolis cristatellus te…
Dendroica tigrina te… Ameiva exsul te… Iguana iguana te… Mimus polyglottos te… Falco peregrinus te… Typhlops richardii te… Parula americana te… Eleutherodactylus antillensis te… Dendroica discolor te… Elaenia martinica te… Coccyzus minor te… Brachyphylla cavernarum te… Patagioenas leucocephala te… Falco columbarius te… Orthorhyncus cristatus te…
Patagioenas squamosa te… Falco sparverius te… Sphaerodactylus macrolepis te… Coereba flaveola te… Odocoileus virginianus te… Herpestes javanicus te… Margarops fuscatus te… Buteo jamaicensis te… Artibeus jamaicensis te… Molossus molossus te… 0 5 10 15 20
Figure ?. Island coverage and percent protection for species with l0% to <20% of predicted distribution in status 1 or 2
153
% Island coverage % Habitat protection 0 10 20 30 40
Chelonia mydas 56 Eretmochelys imbricata 55 Porzana carolina 54 Calidris alba 53 Sterna sandvicensis 52 Podilymbus podiceps 51 Sula leucogaster 50 Haematopus palliatus 49 Sterna maxima 48 Sterna antillarum 47 Phaethon aethereus 46 Phaethon lepturus 45 Catoptrophorus… 44 Mabuya mabouya sloanei 43 Arenaria interpres 42 Charadrius semipalmatus 41 Sterna hirundo 40 Charadrius wilsonia 39 Calidris mauri 38 Himantopus mexicanus 37 Tringa flavipes 36 Gallinula chloropus 35 Calidris himantopus 34 Larus atricilla 33 Egretta tricolor 32 Ardea herodias 31 Dendroica petechia 30 Noctilio leporinus 29 Pandion haliaetus 28 Egretta caerulea 27 Fregata magnificens 26 Tadarida brasiliensis 25 Egretta thula 24 Ardea alba 23 Nycticorax nycticorax 22 Pelecanus occidentalis 21 Setophaga ruticilla 20 Nyctanassa violacea 19 Seiurus noveboracensis 18 Geotrygon mystacea 17 Butorides virescens 16 Geochelone carbonaria 15 Molothrus bonariensis 14 Protonotaria citrea 13 Dendroica virens 12 Amphisbaena fenestrata 11 Stenoderma rufum 10 Seiurus aurocapillus 9 Helmitheros vermivorus 8 Wilsonia citrina 7 Ceryle alcyon 6 Eleutherodactylus… 5 Loxigilla noctis 4 Mniotilta varia 3 Anolis stratulus 2 Vireo altiloquus 1 0 10 20 30 40 50 Figure ?. Island coverage and percent protection for species with 20% to <50% of predicted distribution in status 1 or 2.
154
% Habitat protection % Island coverage 0 5 10 15
Sterna fuscata text11
Sula dactylatra text12 Puffinus lherminieri text7 Oxyura jamaicensis text6
Sterna dougallii text10
Sterna anaethetus text9 Sula sula text13 Ameiva polops text1
Anous stolidus text2
Seiurus motacilla text8 Dendroica magnolia text4
Myiarchus antillarum text5
Dendroica caerulescens text3 0 20 40 60 80 100
Figure ?. Island coverage and percent protection for species with at least 50% representation in GAP status 1 and 2.
155
Limitations and Discussion
When applying the results of our analyses, it is critical that the following limitations are considered: 1) the limitations described for each of the component parts (land cover mapping, animal species mapping, stewardship mapping) of the analyses, 2) the spatial and thematic map accuracy of the components, and 3) the suitability of the results for the intended application.
Limitations and Discussion for Land Cover Analysis
Assessing the conservation status of natural land cover is limited by several factors: GAP has typically found the accuracy of the mapped distributions of natural communities at the floristic (e.g., alliance) level to be substantially lower and more variable than that of animal distributions; any aggregation of biotic units (e.g., above species) is a surrogate for species or lower levels of biotic organization and will under represent conservation need (Pressey and Logan 1995); and for the most part we cannot distinguish the degree of natural condition or value of the mapped units due to management manipulation, exotic invasion, or spatial configuration. Considering an aggregation of species such as we have mapped to be sufficiently represented in existing conservation areas cannot be determined solely by the percentage of the community represented because the aggregation has unmapped variation in species composition that we could not measure. Until individual plant species distributions can be mapped, it is not possible to assure that the full range of vegetation biodiversity is represented, and surrogates must be used.
156
CONCLUSIONS AND MANAGEMENT IMPLICATIONS
The U.S. Virgin Islands have a wealth of rich coastal habitats, protected tropical dry forests, beaches and rocky cays that harbor a diversity of bird, reptile, amphibian, and mammal species in a very small area. Numbers can deceiving when land areas or numbers of species are compared between small tropical island systems and larger mainland habitat. The island systems are dwarfed by the size of continental counterparts. These islands do however offer unique and extremely important resources for resident and migratory species that depend on the mix of terrestrial and marine resources characteristic of the U.S. Virgin Islands. The islands are also important places for people as they foster cultural diversity among residents and serve as important destinations for tourists. Conserving these resources in a matrix of human and natural uses calls for ingenuity in planning and sustainable management. The U.S. Virgin Islands have a number of natural resources at risk due to sea level rise, land use changes, lack of monitoring support, and lack of coordination in conservation strategies among islands, government agencies, and conservation organizations. Currently, 13% of the terrestrial area is protected. This is greater than neighboring Puerto Rico at 8% but below regional and global averages closer to 15%. Additionally the protected areas are concentrated on St. John, while St. Thomas and St. Croix are less than 6 % protected.
157
PRODUCT USE AND AVAILABILITY
How to Obtain the Products It is the goal of the Gap Analysis Program and the USGS Biological Resources Division (BRD) to make the data and associated information as widely available as possible. Use of the data requires specialized geographic information systems (GIS) software and substantial computing power. Additional information on how to use the data or obtain GIS services is provided below and on the GAP home page (URL below). While a CDROM or DVD of the data will be the most convenient way to obtain the data, it may also be downloaded via the Internet from the national GAP home page at:
http://www.gap.uidaho.edu/
The home page will also provide, over the long term, the status of our state's project, future updates, data availability, and contacts. Within a few months of this project's completion, CD- ROMs or DVDs of the final report and data should be available at a nominal cost--the above home page will provide ordering information. To find information on this state GAP project's status and data, follow the links to "Current Projects" and then to the particular state of interest.
Disclaimer Following is the official Biological Resources Division (BRD) disclaimer as of 29 January 1996, followed by additional disclaimers from GAP. Prior to using the data, you should consult the GAP home page (see How to Obtain the Data, above) for the current disclaimer.
Although these data have been processed successfully on a computer system at the BRD, no warranty expressed or implied is made regarding the accuracy or utility of the data on any other system or for general or scientific purposes, nor shall the act of distribution constitute any such warranty. This disclaimer applies both to individual use of the data and aggregate use with other data. It is strongly recommended that these data are directly acquired from a BRD server [see above for approved data providers] and not indirectly through other sources which may have changed the data in some way. It is also strongly recommended that careful attention be paid to the content of the metadata file associated with these data. The Biological Resources Division shall not be held liable for improper or incorrect use of the data described and/or contained herein.
These data were compiled with regard to the following standards. Please be aware of the limitations of the data. These data are meant to be used at a scale of 1:100,000 or smaller (such as 1:250,000 or 1:500,000) for the purpose of assessing the conservation status of animals and vegetation types over large geographic regions. The data may or may not have been assessed for statistical accuracy. Data evaluation and improvement may be ongoing. The Biological Resources Division makes no claim as to the data's suitability for other purposes. This is writable data which may have been altered from the original product if not obtained from a designated data distributor identified above.
158
Metadata
Proper documentation of information sources and processes used to assemble GAP data layers is central to the successful application of GAP data.
Metadata is a description of the content, quality, lineage, contact, condition, and other characteristics of data. It is a valuable tool that preserves the usefulness of data over time by detailing methods for data collection and data set creation. It greatly minimizes duplication of effort in the collection of expensive digital data and fosters sharing of digital data resources. Metadata supports local data asset management such as local inventory and data catalogs, and external user communities such as Clearinghouses and websites. It provides adequate guidance for end-use application of data such as detailed lineage and context. Metadata makes it possible for data users to search, retrieve, and evaluate data set information by providing standardized descriptions of geospatial and biological data.
The Federal Geographic Data Committee approved the Content Standard for Digital Geospatial Metadata (FGDC-STD-001-1998) in June 1998 and NBII (
Each spatial data layer submitted must be accompanied by its metadata (*.xml or .sgml file) in the same directory. You must also include an additional directory (called "meta_master') which will include each metadata file in four forms (*.txt, *.xml, *.html, and *.sgml).
There are many tools available for metadata creation. For some examples, see http://www.nbii.gov/datainfo/metadata/tools/. Please note that some tools are free, and some are not. The redundancy in output format is to provide one file for error checking (*.txt), one for presentation on the Internet (*.html), and two for indexing elements for the spatial data clearinghouse (*xml, *.sgml). Remember, metadata describes the development of the spatial data set being documented. If there are companion files to the GIS data, use metadata to reference (reports, spreadsheet, another GIS layer).
USGS (NBII and FGDC) personnel conduct metadata training to meet FGDC standards and to include biological data. The metadata workshop provides an introduction to the metadata standard with hands-on practice producing documentation for a sample data set using appropriate software: Intergraph’s “Spatial Metadata Management System (SMMS)” and USDA Forest Service North Central Research Station’s “Metavist” are commonly used. The focus of the workshop is an understanding of the metadata standard, but other topics will include the metadata clearinghouse, metadata development tools, and strategies for metadata production. See
159
Appropriate and Inappropriate Use of These Data All information is created with a specific end use or uses in mind. This is especially true for GIS data, which is expensive to produce and must be directed to meet the immediate program needs. For GAP, minimum standards were set (see A Handbook for Gap Analysis, Scott et al. 1993) to meet program objectives.
Recognizing, however, that GAP would be the first, and for many years likely the only, source of statewide biological GIS maps, the data were created with the expectation that they would be used for other applications. Therefore, we list below both appropriate and inappropriate uses. This list is in no way exhaustive but should serve as a guide to assess whether a proposed use can or cannot be supported by GAP data. For most uses, it is unlikely that GAP will provide the only data needed, and for uses with a regulatory outcome, field surveys should verify the result. In the end, it will be the responsibility of each data user to determine if GAP data can answer the question being asked, and if they are the best tool to answer that question.
Scale: First we must address the issue of appropriate scale to which these data may be applied. The data were produced with an intended application at the ecoregion level, that is, geographic areas from several hundred thousand to millions of hectares in size. The data provide a coarse- filter approach to analysis, meaning that not every occurrence of every plant community or animal species habitat is mapped, only larger, more generalized distributions. The data are also based on the USGS 1:100,000 scale of mapping in both detail and precision. When determining whether to apply GAP data to a particular use, there are two primary questions: do you want to use the data as a map for the particular geographic area, or do you wish to use the data to provide context for a particular area? The distinction can be made with the following example: You could use GAP land cover to determine the approximate amount of oak woodland occurring in a county, or you could map oak woodland with aerial photography to determine the exact amount. You then could use GAP data to determine the approximate percentage of all oak woodland in the region or state that occurs in the county, and thus gain a sense of how important the county's distribution is to maintaining that plant community.
Appropriate Uses The above example illustrates two appropriate uses of the data: as a coarse map for a large area such as a county, and to provide context for finer-level maps. The following is a general list of applications: • Statewide biodiversity planning • Regional (Councils of Government) planning • Regional habitat conservation planning • County comprehensive planning • Large-area resource management planning • Coarse-filter evaluation of potential impacts or benefits of major projects or plan initiatives on biodiversity, such as utility or transportation corridors, wilderness proposals, regional open space and recreation proposals, etc. • Determining relative amounts of management responsibility for specific biological resources among land stewards to facilitate cooperative management and planning. • Basic research on regional distributions of plants and animals and to help target both specific species and geographic areas for needed research.
160
• Environmental impact assessment for large projects or military activities. • Estimation of potential economic impacts from loss of biological resource-based activities. • Education at all levels and for both students and citizens.
Inappropriate Uses It is far easier to identify appropriate uses than inappropriate ones, however, there is a "fuzzy line" that is eventually crossed when the differences in resolution of the data, size of geographic area being analyzed, and precision of the answer required for the question are no longer compatible. Examples include: • Using the data to map small areas (less than thousands of hectares), typically requiring mapping resolution at 1:24,000 scale and using aerial photographs or ground surveys. • Combining GAP data with other data finer than 1:100,000 scale to produce new hybrid maps or answer queries. • Generating specific areal measurements from the data finer than the nearest thousand hectares (minimum mapping unit size and accuracy affect this precision). • Establishing exact boundaries for regulation or acquisition. • Establishing definite occurrence or non-occurrence of any feature for an exact geographic area (for land cover, the percent accuracy will provide a measure of probability). • Determining abundance, health, or condition of any feature. • Establishing a measure of accuracy of any other data by comparison with GAP data. • Altering the data in any way and redistributing them as a GAP data product. • Using the data without acquiring and reviewing the metadata and this report.
161
GLOSSARY
algorithm - a procedure to solve a problem or model a solution (In GAP typically refers to a GIS procedure used to model animal distributions) alliance level - a land unit made up of a group of natural communities that have the same dominant or co-dominant plant species or, in the absence of vegetation, by the dominant land cover typically described according to the Anderson land cover classification (see "Natural Community Alliance" in Grossman et al. 1994) alpha diversity - a single within-habitat measure of species diversity regardless of internal pattern, generally over an area of 0.1 to 1,000 hectares (see Whittaker 1960, 1977) anthropogenic - caused by humans assemblages - a group of ecologically interrelated plant and animal species
band, spectral - a segment of the electromagnetic spectrum defined by a range of wavelengths (e.g. blue, green, red, near infrared, far infrared) that comprise the Satellite imagery e.g. Landsat ETM+ imagery
beta diversity - the change in species diversity among different natural communities of a landscape; an index of between-habitat diversity (see Whittaker 1960, 1977)
biodiversity - generally, the variety of life and its interrelated processes
biogeographic - relating to the geographical distribution of plants and animals biological diversity - see biodiversity
cartographic - pertaining to the art or technique of making maps or charts
classify - to assign objects, features, or areas on an image to spectral classes based upon their appearance as opposed to “classification” referring to a scheme for describing the hierarchies of vegetation or animal species for an area
coarse filter - the general conservation activities that conserve the common elements of the landscape matrix, as opposed to the "fine filter" conservation activities that are aimed at special cases such as rare elements (see Jenkins 1985)
community - a group of coexisting plants and animals cover type - a non-technical higher-level floristic and structural description of vegetation cover
162
cross-walking - matching equivalent land cover categories between two or more classification systems
delineate - identifying the boundaries between more or less homogenous areas on
remotely-sensed images as visible from differences in tone and texture
delta diversity - the change in species diversity between landscapes along major climatic or physiographic gradients (see Whittaker 1977)
digitization - entering spatial data digitally into a Geographic Information System
ecoregion - a large region, usually spanning several million hectares, characterized by having similar biota, climate, and physiography (topography, hydrology, etc.).
ecosystem - a biological community, its physical environment, and the processes through which matter and energy are transferred among the components
edge-matching - the process of connecting polygons at the boundary between two independently created maps, either between TM scenes or between state GAP data sets
element - a plant community or animal species mapped by GAP. May also be referred to as "element of biodiversity".
error of commission - the occurrence of a species (or other map category) is erroneously predicted in an area where it is in fact absent
error of omission - when a model fails to predict the occurrence of a species that is actually present in an area
extinction - disappearance of a species throughout its entire range extirpation - disappearance of a species from part of its range fine filter - see "coarse filter" floristic - pertaining to the plant species that make up the vegetation of a given area. formation level - the level of land cover categorization between Group and Alliance describing the structural attributes of a land unit, for example, "Evergreen Coniferous Woodlands with Rounded Crowns" (see Jennings 1993) gamma diversity - the species diversity of a landscape, generally covering 1,000 to 1,000,000 hectares, made up of more than one kind of natural community (see Whittaker 1977)
163
gap analysis - a comparison of the distribution of elements of biodiversity with that of areas managed for their long-term viability to identify elements with inadequate representation
Geographic Information Systems (GIS) - computer hardware and software for storing, retrieving, manipulating, and analyzing spatial data
Global Positioning System (GPS) - an instrument that utilizes satellite signals to pinpoint its location on the earth's surface
ground truthing - verifying maps by checking the actual occurrence of plant and animal species in the field at representative sample locations
habitat - the physical structure, vegetation composition, and physiognomy of an area, the characteristics of which determine its suitability for particular animal or plant species
hectare - a metric unit of area of 10,000 square meters and equal to 2.47 acres metadata - information about data, e.g., their source, lineage, content, structure, and availability minimum mapping unit - the smallest area that is depicted on a map neotropics - the biogeographic region stretching southward from the tropic of Cancer and including southern Mexico, Central and South America, and the West Indies phenology - the study of periodic biological phenomena, such as flowering, breeding, and migration, especially as related to climate phenotype - the environmentally and genetically determined observable appearance of an organism, especially as considered with respect to all possible genetically influenced expressions of one specific character physiognomic - based on physical features physiographic province - a region having a pattern of relief features or land forms that differ significantly from that of adjacent regions pixel - the smallest spatial unit in a raster data structure polygon - an area enclosed by lines in a vector-based Geographic Information System data layer or a region of contiguous homogeneous pixels in a raster system predicted habitat map – map that displays the distribution of the habitats associated with a particular species preprocessing - those operations that prepare data for subsequent analysis, usually by attempts to correct or compensate for systematic, radiometric, and geometric errors
164
pro-active - acting in anticipation of an event as opposed to reacting after the fact
range - the geographic limit of the species
range unit - a spatial, geographic unit to record and display species geographic range. reach - a stream or river segment between inflowing tributaries
registration, spatial - matching different images to each other by finding points on the images that can be matched to known points on the ground
remote sensing - deriving information about the earth's surface from images acquired at a
distance, usually relying on measurement of electromagnetic radiation reflected or emitted from the feature of interest
resolution, spatial - the ability of a remote sensing system to record and display fine detail in a distinguishable manner or the smallest feature that can be distinguished or resolved on a map or image, such as a TM pixel
scale, map - the ratio of distance on a map to distance in the real word, expressed as a fraction; the smaller the denominator, the larger the scale, e.g. 1:24,000 is larger than 1:100,000
species richness - the number of species of a particular interest group found in a given area
supervised classification - the process of classifying TM pixels of unknown identity by using samples of known identity (i.e., pixels already assigned to informational classes by ground truthing or registration with known land cover) as training data
synoptic - constituting a brief statement or outline of a subject; presenting a summary tessellation - the division of a map into areas of equal and uniform shape such as the EPA- EMAP hexagon
Thematic Mapper - a sensor on LANDSAT 4 and 5 satellites that records information in seven spectral bands, has a spatial resolution of about 30 m x 30 m, and represents digital values in 256 levels of brightness per band
Universal Transverse Mercator - one of several map projections or systems of transformations that enables locations on the spherical earth to be represented systematically on a flat map
Universal Transverse Mercator grid - a geographic reference system used as the basis for worldwide locational coding of information in a GIS or on a map
165
unsupervised classification - the definition, identification, labeling, and mapping of natural groups, or classes, of spectral values within a scene. These spectral classes are reasonably uniform in brightness in several spectral channels.
vector format - a data structure that uses polygons, arcs (lines), and points as fundamental units for analysis and manipulation in a Geographic Information System wildlife habitat relationship
model - a method of linking patterns of known habitat use by animal species with maps of existing vegetation, thereby identifying the spatial extent of important habitat features for use in conservation and management.
GLOSSARY OF ACRONYMS
ACSM American Congress on Surveying and Mapping ADAMAS Aquatic Database Management System ADEM Alabama Department of Environmental Management AML ARC/INFO Macro Language ASPRS American Society for Photogrammetry & Remote Sensing AVHRR Advanced Very High Resolution Radiometer (satellite system) BEST Biomonitoring of Environmental Status and Trends BLM Bureau of Land Management CAFF Conservation of Arctic Flora and Fauna C-CAP Coastwatch Change Analysis Program (NOAA) CDC Conservation Data Center CEC Council on Environmental Cooperation CENR Committee on Environment and Natural Resources CERES California Environmental Resources Evaluation System CIESIN Consortium for International Earth Science Information Network CODA Conservation Options and Decision Analysis (software) CRMP Coordinated Resource Management Plan CRUC Cooperative Research Unit Center DEM Digital Elevation Model DFW Division of Fish and Wildlife DLG-E Digital line graph - enhanced DNER Department of Natural and Environmental Resources DOI Department of the Interior DPNR Department of Planning and Natural Resources EDC EROS Data Center ECOMAP The National Hierarchical Framework of Ecological Units mapping project of the USDA Forest Service EMAP Environmental Monitoring & Assessment Program EMAP-LC EMAP-Landscape Characterization (USEPA) 149 EMSL Environmental Monitoring & Systems Laboratory (USEPA) EMTC Environmental Management Technical Center (NBS)
166
EOS Earth Observing System EOSAT Earth Observation Satellite Company (the commercial operator of the Landsat satellite system) EOSDIS EOS Data & Information System ERL Environmental Research Laboratory, Corvallis (U.S.EPA) EROS Earth Resources Observation Systems (USGS) ESRI Environmental Systems Research Institute ETM+ Enhanced Thematic Mapper plus FEMA Federal Emergency Management Agency FGDC Federal Geographic Data Committee FIA Forest Inventory and Analysis FTP file transfer protocol FY Fiscal Year GAO General Accounting Office (Congress) GAP Gap Analysis Program GCDIS Global Change Data and Information System GIS Geographic Information System GLIS Global Land Information System (USGS) GLOBE Global Learning and Observations to Benefit the Environment GPS Global Positioning System GRASS Geographic Resources Analysis Support System GRIS Geographic Resource Information Systems HRMSI High Resolution Multispectral Stereo Imager HUC Hydrologic Unit Code IALE International Association of Landscape Ecology IDRISI A GIS developed by Clark University IITF International Institute of Tropical Forestry IUCN International Union for Conservation of Nature LAPS Land Acquisition Priority System LC/LU Land Cover/Land Use (USGS) 150 MIPS Map and Image Processing System MOU Memorandum of Understanding MMU Minimum mapping unit MRLC Multi-Resolution Land Characteristics Consortium MSS Multi-Spectral Scanner MTPE Mission to Planet Earth NAFTA North American Free Trade Agreement NALC North American Landscape Characterization (USEPA, USGS) NAWQA National Water Quality Assessment (USGS) NBII National Biological Information Infrastructure NBS National Biological Service NCCP Natural Communities Conservation Planning program (in CA) NDCDB National Digital Cartographic Data Base NERC National Ecology Research Center (Ft. Collins, CO) NGO Nongovernment Organization NHD National Hydrography Dataset
167
NMD National Mapping Division NOAA National Oceanographic and Atmospheric Administration NPS National Park Service NRCS Natural Resources Conservation Service NSDI National Spatial Data Infrastructure NSTC National Science and Technology Council NWI National Wetlands Inventory (USFWS) OMB Office of Management and Budget (Administration) OSIS Oregon Species Information System PARC Public Access Resource Center PI Principal Investigator PRGAP Puerto Rico Gap Analysis Project PR-USVIGAP Puerto Rico and the United States Virgin Islands Gap Analysis Project SAB Science Advisory Board (USEPA) SCICOLL Scientific Collections Permit Database SEA St. Croix Environmental Association TNC The Nature Conservancy USEPA United States Environmental Protection Agency USFS United States Forest Service USFWS United States Fish and Wildlife Service USGS United States Geological Survey USVI United States Virgin Islands VIGAP Virgin Islands Gap Analysis Project
168