A Regional Analysis of Geographic Priorities for Biodiversity Conservation in Latin America and the Caribbean

CONTENTS

List of Figures

List of Tables

List of Acronyms

Acknowledgments

Preface

Executive Summary

Chapter One. Background and Purpose

Introduction
Definitions and Conceptual Issues
Data and Methodological Issues

Chapter Two. Assessing Biological Value

Introduction
Approach
Results
Data Limitations and Qualifiers on Use
Conclusions

Chapter Three. Assessing the Conservation Status of Ecoregions

Introduction
Approach
Results
Data Limitations and Qualifiers on Use
Conclusions

Chapter Four. Assessing Policy/Institutional Feasibility

Introduction
Approach
Results
Data Limitations and Qualifiers on Use
Conclusions

Chapter Five. Assessing Utility

Introduction
Approach
Results
Data Limitations and Qualifiers on Use
Conclusions

Chapter Six. Integration Results

Introduction
Approach
Integration Models
Integration Results
Data Limitations and Qualifiers on Use

Chapter Seven. Recommendations and Next Steps

How USAID Should Use These Recommendations
How Others Might Use This Report
Next Steps

Appendices

Appendix A: Maps
Appendix B: Integration: Regional Habitat Units Characterized by Biological Value and Conservation Status
Appendix C: Summary of Four Integration Models
Appendix D: Sample Data Forms Used by Workshop Participants to Assess Biological Value, Conservation Status and Policy/Institutional Feasibility
Appendix E: Using Policy/Institutional Data in Assessing Conservation Priorities
Appendix F: Alphabetical List of Workshop Participants
Appendix G: Workshop Agenda
Appendix H: Literature Cited and Bibliography of Other Selected Biodiversity Priority-Setting References

Document Information

LIST OF FIGURES

Figures in Main Text

Figure 1. Regional Habitat Units of Latin America and the Caribbean

Figure 2. Biological Value of Regional Habitat Units

Figure 3. Conservation Status of Regional Habitat Units

Figure 4. Priority Regional Habitat Units for Biodiversity Conservation

Figures in Map Appendix A

Figure A-1. Biological Value: Plants

Figure A-2. Biological Value: Mammals

Figure A-3. Biological Value: Birds

Figure A-4. Biological Value: Insects

Figure A-5. Biological Value: Herpetofauna

Figure A-6. Biological Value: Fish

Figure A-7. Conservation Status of Ecoregions of Latin America and the Caribbean: Current Conservation Status

Figure A-8. Conservation Status of Ecoregions of Latin America and the Caribbean: Modification of Current Status Based on Threat Analysis

LIST OF TABLES

LIST OF ACRONYMS

ACKNOWLEDGMENTS

The Biodiversity Support Program would like to fully acknowledge the collaborative nature of this exercise. The results reported here would not have been possible without the significant contributions of time and expertise from the following institutions and individuals who served on the NGO Working Group:

Conservation International (CI)-Russ Mittermeier, Silvio Olivieri, Chris Rodstrom
The Nature Conservancy (TNC)-Kent Redford, Roger Sayre
Wildlife Conservation Society (WCS)-John Robinson, Alejandro Grajal
World Resources Institute (WRI)-Dirk Bryant, Nels Johnson
World Wildlife Fund (WWF)-Eric Dinerstein, David Olson, Frances Seymour
Biodiversity Support Program (BSP)-Meg Symington, Kathryn Saterson

We thank Lisa Justice of TNC and Ilana Locker of BSP for their tireless efforts to organize participant invitations and logistics for the workshop.

The participants in the Miami workshop are the source of much of the data that the exercise required. They worked long hours and we thank them for their willingness to apply their detailed knowledge to a regional scale in order to assist policy makers at USAID.

For their long nights of map making, we thank the GIs specialists: Ludmilla Aguiar, Ricardo Machado and Chris Rodstrom (CI), Roger Sayre (TNC) and Avis Webster (WWF). Environmental Systems Research Institute (ESRI) contributed the time of Lance Shipman and Steve Vu, and provided logistical support with computer hardware and software in Miami. Elizabeth McCance did the lion's share of the data collection and synthesis for WCS's biological value section.

Data and analysis for the utility component were provided by Bruce Wilcox of the Institute for Sustainable Development (ISD). The World Conservation Monitoring Center (WCMC) provided digitized data on parks and protected areas and wetlands. Janet Abramovitz and WCMC provided data on biodiversity budgets and spending. The World Bank's LATEN Division's work with WWF provided the ecoregion classification that was the basis for the biogeographic units used in this exercise and moved this project forward faster than would have been otherwise possible.

Useful comments on various drafts of the priority-setting framework and approach were provided by Jim Adriance, Janis Alcorn, Walter Arensberg, Andrew Balmford, Garo Batmanian, Bruce Cabarle, Jason Clay, Steve Cornelius, Pat Foster-Turley, Peter Frumhoff, Doug Fuller, Rodrigo Gamez, Robert Goodland, Pamela Hathaway, Dan Janzen, Steve Kellert, Krishna Kumar, Michelle Lemay, Kathy MacKinnon, Jeff McNeely, Kate Newman, Marianne Schmink, Bill Siemer, Mike Smith, Michael Soulé, John Terborgh and Monique Zegarra.

And finally, we thank our colleagues at USAID for providing funding and technical input to this work. In particular, we thank Glenn Prickett, for initiating the project, and Mike Philley, Jeff Brokaw and Eric Fajer.

PREFACE

After many years of leadership by USAID and our partners, biodiversity conservation has truly arrived as a global priority. The Clinton Administration, for its part, has identified biodiversity as one of two global environmental priorities for USAID, along with the mitigation of global climate change.

This report marks an important step in our effort to turn that general commitment into a specific operational strategy. USAID asked the Biodiversity Support Program to work with leading conservation institutions-Conservation International, The Nature Conservancy, Wildlife Conservation Society, World Resources Institute, and World Wildlife Fund-to produce a framework for setting geographic priorities for biodiversity conservation. These organizations in turn called together leading experts from Latin America and the Caribbean to produce the findings you have in your hands.

Setting biodiversity priorities is a complex and often controversial task. And yet, conservation and development institutions set priorities every day through their budget decisions. USAID believes that establishing clear criteria and seeking advice from outside experts will help us set more effective priorities for our biodiversity programs.

USAID will consider the priorities identified in this report in making strategic decisions about our biodiversity programs. The report employs a sound set of criteria that USAID endorses: biological value, conservation status, biodiversity's usefulness to human communities, and the institutional and political feasibility for conservation. While we act on biodiversity priorities, we will remain mindful that every nation's biodiversity is critical to its own sustainable development. Thus, we will continue to support biodiversity conservation where appropriate on a country-by-country basis.

By supporting the development of this report, USAID also hopes to contribute to a broader debate in the international community on setting biodiversity priorities. We look forward to a continuing dialogue on these issues with our colleagues in biodiversity conservation-national governments, NGOs, the private sector, and other donors.

-GLENN T. PRICKETT CHIEF ENVIRONMENTAL ADVISOR US AGENCY FOR INTERNATIONAL DEVELOPMENT

EXECUTIVE SUMMARY

BACKGROUND

The US Agency for International Development (USAID) has identified biodiversity conservation as one of two global environmental priorities. As a result, the Agency hosted a meeting in December 1993 to explore criteria for setting geographic priorities for biodiversity conservation. The criteria agreed upon at this meeting included biological importance, threat, utility, and institutional feasibility. Following this meeting, USAID requested that the Biodiversity Support Program (BSP) lead an effort to develop a priority-setting framework, starting from the criteria listed above, to identify priority geographic areas for biodiversity conservation. BSP invited five leading conservation organizations-World Wildlife Fund, The Nature Conservancy, World Resources Institute, Conservation International and Wildlife Conservation Society-to serve on a Non-Governmental Organization (NGO) Working Group to assist in developing a framework for determining geographic conservation priorities and then applying it to Latin America and the Caribbean. The framework was developed from January - May 1994. Data were collected for the Latin American region from June - August, and a workshop with the participation of Latin American biodiversity experts was held in Miami in September 1994 to review the data and apply the framework to determine geographic conservation priorities.

PRINCIPLES

The principles underlying this geographic priority-setting approach follow:

Every nation's biodiversity is critical to its own sustainable development. Therefore, biodiversity conservation is important for every country. The recommendations from this exercise will help determine which areas should be priorities for biodiversity conservation at the regional level. The focus of this exercise is on where to conserve, not on what, how or why to conserve.

APPROACH

The approach used to determine investment priorities included four levels of analysis: 1) biological importance; 2) conservation threat and opportunity; 3) policy/institutional feasibility; and 4) human utility. All of the analyses, with the exception of policy/institutional feasibility, were based on biologically and ecologically distinct geographic units, not countries. These geographic units were called Regional Habitat Units (RHUs). The NGO Working Group collected preliminary data which were then critiqued, revised and supplemented at the Miami workshop by biological and social scientists with regional expertise.

The biological importance of areas was determined following a process that first defined priority areas for six taxonomic groups (plants, insects, birds, herpetofauna (reptiles and amphibians), mammals, and freshwater fish) and then integrated these taxonomic priority areas to define biological priority areas. To determine conservation threat and opportunity, data on five key landscape-level features were collected: the presence/absence of large blocks of original habitat, the percent of remaining original habitat, the rate of conversion, degree of degradation and fragmentation, and degree of protection. The policy/institutional (PI) component defined national institutional and policy characteristics which are favorable to the effectiveness of conservation-oriented investments. In the utility component of the analysis, genetic resources, productive and protective resources, and carbon sequestration were used to rank the potential utility of areas. All of the levels of analysis were then integrated by workshop participants to determine biodiversity conservation priorities at the regional level.

The regional geographic priorities identified by the NGO Working Group are based on an approach to integrating the four levels of analysis drawn from the integration approaches developed by workshop participants. Ranking of Regional Habitat Units within Major Habitat Types was based on a consideration and weighting of biological value and conservation status. Regional Habitat Units with the same rank based on these two values were differentiated based on the results of the utility analysis. Due to lack of satisfactory data, political/institutional criteria were not incorporated into the NGO Working Group's final ranking of priorities.

RESULTS

The results of the biological value and conservation status analyses, as well as the final ranking of priority areas are presented in Table 1. Seven Regional Habitat Units are recommended as highest priority for biodiversity conservation (one within each Major Habitat Type):

Another seven Regional Habitat Units are recommended as high priority for biodiversity conservation (also one from each Major Habitat Type):

Because of this exercise's emphasis on representation of different Major Habitat Types, and the integration of multiple criteria to assess biological value and conservation status, the list of priority Regional Habitat Units contains a number of areas that have not received significant conservation attention in the past. For example, temperate forest (pine- oak forests in Mexico and southern temperate forests in Chile), xeric (Caatinga in Brazil, deserts and other xeric formations in Mexico) and dry forest (the Chaco in Argentina, Paraguay and Bolivia, the Cerrado of Brazil) ecosystems were identified by the workshop participants as having high priority for biodiversity conservation efforts at the regional level.

CAVEATS AND LIMITATIONS

The workshop was extremely useful in refining the approach taken to the policy/institutional (PI) level of analysis. However, workshop participants agreed that the data collected prior to the workshop were insufficient to apply in the context of recommending biodiversity conservation priorities. Therefore, although PI criteria are not incorporated into the results presented above, examples of how PI data might be applied in future priority-setting exercises are provided in an appendix to this report. The PI criteria developed in the context of this exercise should be used by USAID to help determine where biodiversity conservation investments should be made within the large Regional Habitat Units identified above, as well as for determining what types of investments should be made in these areas.

The results of any attempt to determine geographic priorities for biodiversity conservation are obviously dependent on the scale at which the analysis is done. The scale of this analysis is region-wide, and thus it would be inappropriate to use the results presented here to make investment recommendations on a smaller, sub-regional (e.g., Central America) or country-level, scale. It is our hope that this priority-setting exercise is only the first step in an ongoing process, that would use continually finer scales to determine biodiversity conservation priorities. Indeed, an evaluation provided by workshop participants indicated that a significant number were so impressed with the overall approach that they wanted to try it at the national level within their own countries.

The exclusion of marine, freshwater, and mangrove ecoregions from this analysis limits our recommendations to terrestrial priorities. We are currently moving forward with an initiative to adapt and apply the framework to identify aquatic (including marine) biodiversity conservation priorities in Latin America and the Caribbean.

Neither biological value nor conservation status are static qualities. As new data are collected and new threats emerge, the results and rankings reported here could change. The data used in this exercise are all available in tabular format so that results can be amended and modified as new data become available.

TABLE 1. RANKING PRIORITY HABITAT UNITS

CHAPTER ONE
BACKGROUND AND PURPOSE

INTRODUCTION

In December 1993, the US Agency for International Development (USAID) convened an informal workshop for Agency staff and outside experts to explore criteria for setting geographic priorities for biodiversity conservation. USAID had recently identified biodiversity as one of two global environmental priorities and, consequently, had an interest in targeting its biodiversity programs more strategically. The consensus of the group assembled was that, given declining financial resources at USAID, geographic priorities would help ensure that USAID biodiversity investments remain focused and relevant. The group also agreed that priorities should not be based on species richness alone, but should consider multiple criteria such as phyletic and ecosystem diversity, threat, usefulness of biodiversity to humans and the policy and institutional factors that affect the probability of conservation investments being successful.

The Biodiversity Support Program (BSP) was then requested by USAID to lead an effort to develop a framework for setting geographic biodiversity conservation priorities and to then use the framework to generate geographic conservation priorities for USAID's use in Latin America and the Caribbean (LAC). BSP invited representatives of five major international NGOs to form a collaborative NGO Working Group to assist in this effort. The Working Group included representatives of Conservation International (CI), The Nature Conservancy (TNC), Wildlife Conservation Society (WCS), World Resources Institute (WRI), and World Wildlife Fund (WWF). Over a 9-month period, beginning in January 1994, the Working Group developed a framework and collected and synthesized data to identify geographic conservation priorities for USAID. In September 1994, a workshop, in which regional experts participated, was held in Miami, Florida to review the framework methodology, review and refine collected data, identify additional data sources, and apply the framework and data to identify preliminary regional geographic priorities for USAID in Latin America and the Caribbean.

The purpose of this priority-setting exercise was to identify areas of outstanding regional importance for biodiversity conservation. The approach taken can be characterized as "integrative" in that it integrates a number of criteria in determining priorities. We believe that the production and application of a logical, transparent and integrative geographic priority-setting framework represents a significant improvement over less systematic decision-making that has characterized biodiversity funding by most NGOs, government agencies, bilateral and multilateral funding agencies to date.

The approach we developed identifies geographic priorities based on a combination of four major categories of criteria: 1) the biological value of an area; 2) the conservation status of an area at the landscape level; 3) the policy and institutional characteristics that indicate whether a conservation investment in a given area is likely to have an impact; and 4) the potential human utility of an area's biodiversity. USAID and other donors are increasingly concerned with the role of biological diversity in sustainable development and with maximizing the cost effectiveness of their investment. The integrative approach resulting from the consideration of policy/institutional feasibility and human utility distinguishes this framework from previous priority-setting exercises. Another characteristic that distinguishes this exercise from many previous efforts is the emphasis on representation of different habitat types in the conservation priorities identified, and the use of biogeographic units, rather than countries as the basis for the analysis.

DEFINITIONS AND CONCEPTUAL ISSUES

Biodiversity is a term which refers to the variety and variability among living organisms, the ecological complexes in which they naturally occur, and the ways in which they interact with each other and the geosphere.

Priorities (and criteria for setting priorities) must be derived from clearly stated goals and objectives. The NGO Working Group recommends that USAID's support for conservation should help to conserve the broadest possible range of the Earth's biological diversity through conservation of priority ecoregions within the major habitat types of each biogeographic realm. Thus, the analysis presented here is based on ecologically distinct biogeographic units (ecoregions), not country units. The ecoregion classification scheme that is the basis for this approach is based on a system derived by WWF's Conservation Science Program (Dinerstein et al. 1995). An ecoregion represents a geographically distinct assemblage of communities that share a large majority of their species, ecological dynamics, and similar environmental conditions, and whose ecological interactions are critical for their long-term persistence. Ecoregions are grouped into Major Habitat Types (MHTs) based on similarities in terms of their general structure, climatic regimes, major ecological processes, level of species turnover with distance (beta diversity), and whose flora and fauna show similar guild structures and life histories.

For the purposes of the USAID priority-setting exercise, however, ecoregions were considered too fine a scale, so we used clusters of biogeographically similar ecoregions, which we called Regional Habitat Units (or RHUs), as the basic unit of analysis (Figure 1). It was felt that this scale of analysis was feasible given time and budget constraints without unduly sacrificing representation. A listing of the hierarchical classification for Major Habitat Types and Regional Habitat Units used in this priority-setting process can be found in Table 2. Using this hierarchical system, conservation priorities for each Major Habitat Type were identified. The approach calls for comparing biological value, conservation status, policy/institutional characteristics and utility of Regional Habitat Units only within the same Major Habitat Type, e.g., tropical dry forests in Mexico would be compared with tropical dry forests in Central America-both within the tropical dry forest major habitat-but not with an RHU such as Paramo, which is from the herbaceous montane major habitat.

The purpose of this priority-setting process was to determine where to undertake conservation initiatives, and it does not address what types of initiatives should be supported, although the form and quality of these conservation investments is obviously of critical importance. An important assumption of the exercise is that the list of priority areas will not be interpreted as an exclusive list of where USAID should invest its biodiversity conservation funds. It is recognized that biodiversity and biological resources are important to every country's sustainable development. USAID's March 1994 Strategies for Sustainable Development states that "USAID will...support conservation and sustainable use of biological resources where this is judged to be a priority for sustainable development at the country level."

This report addresses only biodiversity conservation priorities for terrestrial systems. The time and budget constraints of the exercise precluded consideration of aquatic/marine ecosystems at this time. We are currently moving forward with an initiative to adapt and apply the framework to aquatic and marine systems. Although several freshwater fish experts attended the workshop in Miami and identified preliminary biological priority areas based on fish faunas, they felt that freshwater aquatic systems should be addressed more fully in the context of a combined aquatic/marine priority-setting exercise for Latin America and the Caribbean. Mangrove ecoregions also were not addressed at the Miami workshop, but were the subject of a separate WWF workshop in November 1994 (Olson and Cintron, in prep.).

DATA AND METHODOLOGICAL ISSUES

This priority-setting effort collected existing data from a variety of sources, attempted to capture new data where possible, and provided this information to workshop participants in a spatially referenced form when possible. The biological value data were compiled by Conservation International and the Wildlife Conservation Society. Information on species identities, abundances and geographic distribution, and other relevant data were collected from primary, secondary and tertiary references, entered into tabular databases and mapped using geographical information systems (GIs).

The World Wildlife Fund Conservation Science Program provided an analysis of conservation status (threat and opportunity) at the landscape level by mapping the entire LAC region into 212 ecoregions, which, for the purposes of this exercise, were combined and consolidated to form Regional Habitat Units as described above.

The World Resources Institute compiled the policy/institutional (PI) data to be used in characterizing countries and sub-national units and worked with the World Wildlife Fund Policy Program to develop a PI assessment instrument. An analysis of the potential utility of the biodiversity found in the RHUs was prepared by the Institute for Sustainable Development (ISD).

Because most of these data sets are incomplete, and because some of the priority-setting criteria required subjective judgments, the knowledge and experience of individuals from the LAC region was incorporated into the priority-setting process during the workshop in Miami. The workshop involved 74 participants, including 40 regional experts on Latin America and the Caribbean (including both biologists and social scientists), 15 representatives of the NGO Working Group, 5 GIs/administrative staff, 11 USAID staff from Washington and Missions, and 3 observers. The workshop also provided an opportunity to incorporate the outlook and viewpoints of Latin Americans into the priority-setting process. A list of the workshop participants and invitees can be found in Appendix F.

Background information compiled by Working Group members and made available to the workshop participants included: a 1:10,000,000 map describing major ecoregions and regional habitat units for the region; a set of background maps describing the distribution of particular taxa in the region (in 8-1/2 x 11" format); and a set of maps describing the region's topography, hydrology, climate, vegetation types, soil types, and various ecological zonings (Holdridge, Bailey, Udvardy) to be used as references (at a scale of 1:10,000,000). Numerous maps of the region, including coastline, country boundaries, major rivers and lakes, were produced on transparent mylar at a scale of 1:10,000,000 and distributed to workshop participants to be used for overlaying on base maps and drawing the boundaries of areas that needed to be defined over the course of the workshop. Policy/institutional and utility data were provided to workshop participants in tabular form.

The workshop took place over four days. The first two days were spent in working groups characterizing the biological value of areas and considering how to assess policy and institutional characteristics. The third day of the workshop was devoted to assessing the conservation status (threat and potential) of the ecoregions and RHUs. On the final day of the workshop, participants were asked to develop a methodology for integrating the various levels of data to determine conservation priorities. The results of the four levels of analysis and the final regional geographic conservation priorities resulting from the integration of these layers are presented in this report. A detailed agenda of the Miami workshop can be found in Appendix G.

TABLE 2. HIERARCHICAL CLASSIFICATION OF MAJOR HABITAT TYPES AND REGIONAL HABITAT UNITS FOR THE LATIN AMERICAN AND CARIBBEAN REGION

CHAPTER TWO: ASSESSING BIOLOGICAL VALUE

INTRODUCTION

The methodology used to determine the biological value of Regional Habitat Units in Latin America and the Caribbean was developed jointly by Conservation International and the Wildlife Conservation Society. Methods developed for previous priority-setting exercises conducted by Conservation International were drawn upon to incorporate biological data and expert opinion in a workshop setting (IBAMA/INPA/CI, 1991; Conservation International, 1995). The methods were designed to achieve three major goals: 1) to provide a sound basis for understanding the biological importance of different areas in the region; 2) to provide thorough documentation of each phase of the process using both maps and tables, to allow USAID and others interested in the results to revisit the conclusions from the workshop; and 3) to identify biological priority areas within each Major Habitat Type.

APPROACH

Biological priority areas were determined following a process that first defined areas identified as priorities for the six taxa being analyzed (plants, insects, birds, herpetofauna, mammals, and freshwater fish) and then integrated these taxonomic priority areas to define integrated biological priority areas. This two-phase approach allowed for the experts to first review the data and define priorities for the taxonomic group with which they were most familiar. This provided a valuable first step before attempting an integrated assessment of biological value. Further, it allowed for a separate comparison of the individual taxonomic priorities and integrated biological priorities following the workshop, and highlights the strengths and weaknesses of each step.

Taxonomic priority areas were delineated concurrently by six taxonomic expert groups: plants; mammals; birds; insects; herpetofauna; and fish. The composition of these taxonomic working groups can be found in Appendix G. Each taxonomic priority area was characterized using a standard form (see Appendix D), and rated as being of "high," "medium" or "low" biological value based on a variety of criteria that differed slightly for each taxonomic group but included species richness, phyletic diversity, number of endemic species, beta diversity and presence of rare/endangered species. Subsequent to the workshop, those areas rated to be of "high" biological value were designated as regionally outstanding, areas of "medium" biological value as regionally significant, and areas of "low" biological value as locally important. The NGO Working Group felt that this revised terminology better reflected the workshop participants' conviction that biodiversity is valuable everywhere, and will minimize the potential for misunderstanding the results.

The second phase of the biological value analysis integrated the results from each taxonomic group. This phase of the analysis was conducted during a single plenary session with each of the taxonomic groups represented. First, a common characterization of Major Habitat Types and Regional Habitat Units was agreed upon to make comparisons of rankings of the taxonomic priority areas across taxonomic groups possible. The plenary session harmonized the different geographic delineations used by the different taxonomic groups, adjusting the boundaries to reflect the consensus of all groups. It was decided to group 35 Regional Habitat Units (RHUs) within seven Major Habitat Types (MHTs) as described in Table 2.

With the common base map set, the groups provided their scores, adjusting their taxonomic analyses to the newly defined areas. Each taxonomic group provided a biological value ranking for their taxonomic group for every RHU. The fish taxonomic group did not provide a numeric ranking for the integration process. It was decided to keep their results separate, since the base map decided upon by the other five taxonomic groups was not felt to be appropriate for freshwater organisms. The fish taxonomic group analysis (see Figure A-6) was used as a separate modifier in some models when biological value was integrated with conservation status (see Chapter Six, Integration Results). A final biological value rank was determined from the sum of the five taxonomic values (plants, mammals, birds, insects and herpetofauna) within each Major Habitat Type. RHUs were ranked for biological value within each MHT, thus ensuring that each of the seven major habitats had high priority areas.

RESULTS

The final scoring for each of the taxonomic groups and the integrated biological value ranking for each RHU is shown in Table 3. These results are illustrated in map form in Figure 2 and Figures A-1 to A-6. Some initial generalizations concerning the results obtained are apparent. Two RHUs, the tropical Andes and the Atlantic coastal forests of Brazil, consistently scored high and were identified as regionally outstanding across all taxonomic groups. There is no similar consensus for lower priority areas.

The results from the separate taxonomic analyses provide valuable insight into the perspective each biological discipline has on biological value. The different delineations and scoring by each taxonomic group reflect the way in which biologists group organisms within their discipline and how those organisms are distributed. There are also interesting results from the reports of the taxonomic working groups regarding the kind of scientific knowledge that exists for the region. According to taxonomic working group members, there is generally greater understanding of species composition than of ecosystem functioning. This may be partly due to the historical pattern of biological research which has often focused on species and populations, rather than complex interactions among groups of organisms.

Several important results emerge from the integrated biological priorities component of the workshop. In the context of Latin America and the Caribbean, numerous biological priority areas emerge that have not received significant attention in the past. Of particular note are the temperate forest and xeric formations. Temperate forests with regionally outstanding biological value include the pine-oak forests of Mexico and the southern temperate forests of Chile. Xeric formations and tropical dry forest, including the Caatinga of Brazil, the deserts and xeric formations of Mexico, the Chaco of Argentina, Paraguay and Bolivia, and the Cerrado of Brazil, were also identified as having regionally outstanding biological value.

Some of the areas identified in Miami as having regionally outstanding biological value were also identified in previous priority-setting efforts that focused solely on tropical moist forests (Myers 1988, Conservation International 1990). The Atlantic coastal forests of Brazil, the upper Amazon basin and the tropical Andes are thus reaffirmed from a biological perspective as critical areas for conservation within the tropical forest habitat type. Some areas that were identified as priorities in previous efforts, such as portions of the Central American lowland forest (Myers 1990) and many of the endemic bird areas in Latin America and the Caribbean (Bibby et al. 1992) were not identified in the current exercise, primarily as a result of the larger scale at which our analysis was conducted (see Data Limitations and Qualifiers on Use below).

DATA LIMITATIONS AND QUALIFIERS ON USE

Any attempt to assess biological value must address the fact that the definition of priority areas is dependent upon the scale of the analysis. The way in which habitats are grouped will affect any assessment of their biological value. Endemism is a particularly elusive factor to quantify, since the unit of analysis itself determines what is and is not endemic. Small patches of habitat containing a large number of endemic and rare species may be overlooked within a larger, less biologically rich region. Conversely, smaller, less biologically rich areas may be swept up within a larger, more valuable region and classified similarly.

The Regional Habitat Units used in this analysis reflect the consensus of experts from five separate biological disciplines. Each RHU was developed with an assumption that the biological value assigned by each taxonomic group would represent the average value across the entire unit. This assumption does not imply, however, that the entire area within an RHU is of completely uniform biological value. In fact, given the size of the units, and the fact that the RHUs are based on potential, rather than actual vegetation, heterogeneity within an RHU is inevitable and a finer scale analysis will be required to identify specific areas within each RHU that should be the focus of conservation efforts.

The RHUs and biological priority areas, as defined by workshop participants, vary widely in size. An examination of the spatial characteristics of the biological value map reveals two clear relationships. First, the average size of RHUs categorized as regionally outstanding is twice that of RHUs categorized as regionally significant or locally important. Second, the total area of regionally outstanding RHUs is more than twice that of the regionally significant and locally important areas combined, with two of the regionally outstanding RHUs-Upper Amazon and Cerrado-Pantanal-accounting for nearly 25% of the total land area in Latin America and the Caribbean. Obviously, a finer level of resolution is required to determine what specific sites within each of these large biologically valuable areas should be targeted for investment.

Although inclusion of broad regional and taxonomic expertise was the workshop goal, the composition of specific taxonomic working groups (see detailed workshop agenda, Appendix G) reflected inevitable occasional gaps in comprehensive knowledge. The insect group focused primarily on butterflies, since this was the primary expertise of its members, as well as being the most well known group in terms of its distribution and taxonomy. The insect working group also did not feel qualified to assess the biological value of large areas in northern Mexico and southern South America. Other groups had similar biases in expertise, such as for primates in the mammal group and tree frogs in the herpetofauna group. Neotropical migrants were not weighted heavily by the bird taxonomic group because migrant species constitute a small percentage of the total avifauna in the region as a whole and because the group focused on endemism as the most important feature determining biological value, and most migrants have wide winter ranges. We believe the specific emphases found within the taxonomic groups in Miami do not compromise the results obtained, but should be noted for purposes of transparency.

CONCLUSIONS

The workshop brought together leading biologists who are experts on the species and ecosystems of Latin America and the Caribbean. While each of these scientists has an individual expertise on some species or geographic areas, their collective knowledge and experience provide the best possible understanding of the region as a whole. The workshop approach and use of GIs allowed us to quickly capture the information these biologists have gained through decades of field work. Detailed biological information on species distribution, areas of endemism, and the physiography of the region, provided a basis from which to work. The result of this process was a broad consensus about which areas are of greatest importance biologically at a regional level.

TABLE 3.

FINAL RANKING OF REGIONAL HABITAT UNITS FOR BIOLOGICAL VALUE

CHAPTER THREE
ASSESSING THE CONSERVATION STATUS OF ECOREGIONS

INTRODUCTION

The second major step in determining which areas are highest priority for biodiversity conservation across Latin America and the Caribbean was to assess the landscape integrity and conservation status (threat and opportunity) of biogeographic units within the region. This step ensures that the regional conservation strategy: 1) intervenes quickly to ward off complete degradation and conversion in the most threatened ecoregions; and 2) creates programs to conserve the most intact examples of biologically valuable habitats with the best chances for long-term persistence.

The most robust method for assessing conservation status is to evaluate landscape condition, now a major focus in the field of conservation biology. An assessment of landscape-level features and dynamics provides powerful insights into the integrity of an ecoregion and helps to predict the long-term persistence of ecological processes that maintain biodiversity. Because the alteration of ecological processes is difficult to measure, biologists rely instead on specific landscape level features as indicators that prove easier to estimate. The analysis of these features, summarized in the section below, provides an objective and transparent method for generating one type of information needed by USAID and other donors to identify the most urgent biodiversity conservation needs in LAC.

APPROACH

For the benefit of this exercise (and its own project with the World Bank LATEN division and the Global Environmental Facility (GEF)), the Conservation Science Program of WWF developed an approach to assess the conservation status of ecoregions in the tradition of the IUCN Red Data Book for threatened and endangered species. This method was used successfully at the Miami workshop to clarify the conservation status of 148 LAC ecoregions. Workshop participants provided data for each ecoregion using a standard form (see Appendix D) to assess the current (or "snapshot") conservation status of an ecoregion based on current habitat configurations. Data on five key landscape-level features were collected: 1) the presence/absence of large blocks of original habitat; 2) the percent of remaining original habitat; 3) the rate of conversion; 4) degree of degradation and fragmentation; and 5) degree of protection.

Each variable considered received a numerical value that, when weighted and summed, provided a snapshot assessment of ecoregion conservation status. In weighting these variables, greater weight was given to total loss of original habitat (40%) and number of large blocks of intact habitat (20%) and degree of fragmentation and degradation (20%) since these variables are the best indicators of the probability of persistence of ecological processes within ecoregions. Rate of conversion of remaining habitat (10%) and degree of protection (10%) accounted for the remainder of the 100 point score. The presence of large blocks of original habitat, a high percentage of remaining habitat, and some degree of formal protection highlight opportunities for conservation within the ecoregion. Combined with degree of fragmentation and degradation, these variables also help predict the maintenance of ecological processes (e.g., predation, pollination and seed dispersal systems, nutrient cycling, migration, dispersal, and gene flow) that, ultimately, will determine how much biodiversity will persist over the long-term.

Based on the numerical value obtained, as explained above, ecoregions were classified as falling within one of six categories: Extinct (completely converted); Critical; Endangered; Vulnerable; Stable; and Relatively Intact. Brief definitions of these categories follow below.

EXTINCT: No natural communities resembling original ecosystems remain. Some of the original biota still present but persist only within highly modified communities and landscapes.

CRITICAL: Remaining intact habitat restricted to isolated small fragments with low probabilities of long-term persistence (e.g., < 5 years) without immediate or continuing protection and restoration.

ENDANGERED: Remaining intact habitat restricted to isolated fragments of varying size (some larger blocks still present) with medium to low probabilities of long-term persistence (e.g., 5-10 years) without immediate or continuing protection or restoration.

VULNERABLE: Remaining intact habitat occurs in habitat blocks ranging from large to small, but many intact clusters will likely persist over the next 10-20 years given adequate protection and moderate restoration.

RELATIVELY STABLE: Natural communities have been altered in certain areas, causing local declines in exploited populations and disruption of ecosystem processes. These disturbed areas can be extensive, but are still patchily distributed relative to the area of intact habitats.

RELATIVELY INTACT: Natural communities within an ecoregion are largely intact with species, populations, and ecosystem processes occurring within their natural ranges of variation.

The criteria for classifying ecoregions was tailored to reflect biological and ecological differences among Major Habitat Types. For example, large blocks in forested ecoregions were defined as greater than 250 km2 in extent, whereas large blocks in grasslands and xeric systems were defined as greater than 100 km2. Consequently, the snapshot assessment of an ecoregion was assessed only in comparison with other ecoregions in its same Major Habitat Type.

In the second step, the current conservation status was modified by workshop participants based on their best assessment of the type, intensity, and timeframe of severe threats to the biota and landscapes of an ecoregion (see Appendix D for sample threat analysis worksheet) to yield a modified conservation status. The next step was to scale up from the ecoregion level to the Regional Habitat Unit (RHU) level to allow comparisons with the other data layers. Because many of the RHUs cover large areas, it proved helpful to conduct the conservation status assessment at the ecoregion level first to ensure transparency of results and greater objectivity. Conservationists interested in how any RHU's conservation status was determined can refer to the status assigned to the ecoregions that compose that RHU and the values assigned to the landscape-level variables and threat indicators for each ecoregion.

RESULTS

Ecoregion Analysis
The most striking observation in comparing the snapshot conservation status map of ecoregions (Figure A-7) and the final conservation status map (Figure A-8) is how few ecoregions are designated as either STABLE or INTACT. Essentially, much of the area in these two categories is located in the Amazon basin, temperate forests of the southern cone, and western Mexican xeric systems. In contrast, many more ecoregions are identified as CRITICAL or ENDANGERED. Several regions have a high proportion of critical ecoregions: the northern Andes and lowland dry forests; the Atlantic coastal forests of Brazil and the Pampas of the southern cone. Endangered ecoregions are abundant in the northern and central Andes, most of Central America, the steppe and winter rainfall areas of the southern cone, the Cerrado and other dry forests south of the Amazon basin, and the Caribbean. Two ecoregions, both in Mexico (Tehuantepec savanna and Gulf of Mexico Palmar), were considered EXTINCT by workshop participants. This observation indicates the importance of conducting the conservation status assessment at the ecoregion level since the two extinct ecoregions would not have been classified as such under the more coarse-grained RHU classification scheme.

At the ecoregion level, final conservation status was estimated as follows:

Thirty-four ecoregions move up in category (i.e., were considered more threatened) after the assessment of threat was applied to the snapshot conservation status. In contrast, three ecoregions (Belizean swamp forests, pantepui, and subpolar Nothofagus forests), were downgraded to a less threatened category.

Among the Major Habitat Types, the only extinct ecoregions occurred in the herbaceous lowland grassland category. Numerically, the highest number of critical and endangered ecoregions occurred in tropical moist forests and tropical dry forests. However, only 7% of the tropical dry forest ecoregions were relatively stable or intact, whereas 25% of the tropical moist forests were stable or intact. From these data, we can conclude that tropical dry forests are on average more threatened than tropical moist forests in the LAC region. Most xeric ecoregions were either critical, endangered, or vulnerable. Grassland ecoregions illustrated the widest spread in conservation status.

Regional Habitat Unit (RHU) Analysis
The final conservation status for the RHUs was estimated as follows and is illustrated in Figure 3. The conservation status of each RHU is also listed in Table 1.

Almost half of all RHUs (16) were classified as vulnerable (46%), and only two RHUs (Pantepui and Guayana montane) were classified as relatively intact. Three out of three temperate forest RHUs (100%) and 5 out of 6 tropical dry forest RHUs (83%) were classified as critical or endangered. From these data, we can conclude that temperate forests and tropical dry forests are the most threatened Major Habitat Types in the LAC region.

Aquatic Biodiversity
The experts for freshwater fish who attended the workshop developed a preliminary method to assess the conservation status of aquatic ecoregions based on the following eight watershed-level parameters: 1) unimpacted basin area; 2) aquatic habitat loss; 3) barriers to movement; 4) conversion rate of aquatic and associated terrestrial systems; 5) degree of protection; 6) water quality; 7) riparian modification; and 8) introduced organisms. The aquatic biodiversity team delineated forty-two aquatic ecoregions for the LAC region. They urged the NGO Working Group to seek support for completing a full analysis of aquatic biodiversity priorities. Input from experts for different aquatic taxonomic groups (e.g., fish, insects, mollusks, crustaceans, plants) was strongly recommended.

DATA LIMITATIONS AND QUALIFIERS ON USE

The method developed by the fish taxonomic group represents an excellent start towards addressing the conservation status of aquatic ecosystems. The conservation community could greatly benefit by supporting a program to assess the conservation status of freshwater ecoregions in the near future. The inclusion of data on the conservation status of aquatic/marine and mangrove ecoregions would strengthen the analysis presented here.

It should be emphasized that the ecoregions and RHUs were all based on "potential vegetation" and not actual. Although good maps of existing vegetation were available for some areas, workshop participant expertise was used to estimate the percentage of original habitat remaining in most ecoregions and RHUs. In general, workshop participants found it easier to assess the conservation status of ecoregions within forested habitat types than non-forested habitat types, due to the relative ease with which habitat degradation can be judged in forested habitats. A further improvement on the process used to determine conservation status at the Miami workshop would involve providing workshop participants with detailed reference material-hard data, spatially referenced where possible-on threats: population growth rates; deforestation rates; land use; etc., which might guide them in their threat analysis assessment.

It should also be noted that the criteria used in this analysis of conservation threats do not reflect the higher order threats to the integrity of terrestrial ecosystems in Latin America and the Caribbean posed by changes in the biotic regulation of the flow of energy, water, carbon and nutrients through the biosphere (e.g., regional climate change, global climate change, soil loss and changes in net primary productivity). A priority-setting exercise that addressed the biotic life-support functions of ecosystems would require a closer collaboration between ecosystem ecologists and conservation biologists than has existed to date.

CONCLUSIONS

Critical vs. Intact Ecosystems
Conservationists correctly emphasize the importance of conserving the last remaining habitats in critical ecosystems. The complete loss of a habitat within a critical RHU, or ecoregion for that matter, would entail the total extinction of numerous species, particularly invertebrates and plants, as well as unique communities. Because the level of endemicity for invertebrates and plants is relatively high at the RHU level, even for RHUs that have been assessed as having locally important biological value, all critical RHUs (or ecoregions) should be viewed as highest priorities. The conservation community must act decisively to reverse the complete alteration of natural habitats within critical RHUs or ecoregions if we are to avoid losing significant biodiversity.

At the same time, the few RHUs that still contain relatively intact ecosystems deserve immediate conservation investment because it is within these now rare intact landscapes that ecosystem processes and species have the best chance for long-term persistence. Conservation investments within these areas are also the most cost-effective, and it would be unwise to miss the window of opportunity for strategic conservation planning. Therefore, a wise strategy would be to distribute conservation investments across the spectrum of conservation status categories from critical to relatively stable.

CHAPTER FOUR
ASSESSING POLICY/ INSTITUTIONAL FEASIBILITY

INTRODUCTION

The purpose of the Policy/Institutional (PI) component of the priority-setting analysis is to factor in features of the institutional and policy "landscape" favorable to the effectiveness of conservation-oriented investments. In the same way that certain physical characteristics (such as extent of habitat conversion) render some areas more amenable to the achievement of long-term conservation objectives than other areas, institutional and policy environments can also influence the likelihood of success.

Decision-makers involved in allocating scarce resources for conservation investments often do take policy/institutional factors into account, but usually in an intuitive, ad hoc fashion. The attempt to design an instrument to take these factors into account in a systematic, transparent manner is one of the key aspects of the framework that sets it apart from previous priority-setting methods, and is the one aspect of this approach that received the most favorable comment from external reviewers prior to the Miami workshop.

Because little work has been done on quantifying institutional and policy factors for biodiversity priority-setting or other purposes, a considerable amount of preparatory work and review had to take place before the Miami workshop. NGO Working Group members assembled a Washington-based advisory group familiar with conservation policy issues in Latin America. This advisory group met twice in mid-1994 to help develop and review an assessment methodology. WRI compiled data on 16 indicators of national level political commitment and institutional capacity that were available in existing databases for the LAC countries.

NGO Working Group members crafted a draft checklist assessment instrument (a questionnaire to capture expert opinion on unquantifiable variables favoring the success of conservation investments). The checklist assessment instrument was reviewed by several experts in public opinion and Delphi technique assessment methodologies. Members of the Washington-based advisory group were then asked to use the instrument to evaluate two or three countries with which they were most familiar, and provide feedback on the process. The survey instrument was then further revised, incorporating these comments, prior to being presented at the Miami workshop.

APPROACH

Nineteen experts in the area of policy and institutions in the LAC region (including 5 USAID Mission staff) comprised the PI working group at the Miami workshop. Members of the PI working group focused much of their discussion on the fact that the relevance of PI analysis to geographic priority-setting is directly linked with the nature of the proposed conservation investment. In other words, in order to answer the question of where to invest, one must specify what type of investment is being made.

As a partial resolution of this issue, members of the PI working group in Miami developed an "investment portfolio" model that would balance short and longer term payoffs and greater and lesser degrees of risk. The group defined two different PI vectors relevant to priority-setting:

1) existing policy and institutional capacity conducive to effective, on-the-ground conservation interventions in the short-term; and

2) policy and institutional environments conducive to productive investments in the development of such capacity (i.e., currently weak institutional capacity combined with strong political commitment and positive trend). In Table 4 below, areas with PI characteristics of Cell A would be favored for short-term, low-risk investments in immediate, on-the-ground conservation interventions, followed by those in Cell C. Areas with PI characteristics of Cell B would be favored for longer-term, higher risk investments in capacity-building. Areas with PI characteristics of Cell D would not be favored for conservation investments of either type.

RESULTS

Workshop participants in the PI working group reviewed and critiqued the PI data sets that had been compiled by WRI prior to the workshop. The group discussed the strengths and weaknesses of various indicators of political commitment and institutional capacity for conservation, and determined that, given the nature and quality of available data sets, an expert-opinion driven approach was a necessary alternative.

The PI working group then considered, critiqued, and revised a draft checklist assessment instrument (see Appendix D) to gather expert opinion on features of the policy/institutional landscape deemed relevant to the success of conservation investments. Utilizing their own regional expertise, PI working group members attempted to apply the expert opinion approach to countries in the region to generate a preliminary PI data set for use at the workshop. However, due to the limited number and country expertise of participants, an insufficient standardization of responses, and the difficulty of "grading" national commitment and capacity, they decided not to report a data set back to the full workshop.

DATA LIMITATIONS AND QUALIFIERS ON USE

PI working group members agreed that the data compiled by WRI prior to the workshop were not adequate to serve as PI indicators. The working group produced a "wish list" of data that would be useful in future PI decision-making processes (see Appendix E1).

For the purposes of the workshop, the group produced a checklist to assess PI characteristics at the country level; however, there was wide agreement that the instrument should be applied at the national level and then supplemented with sub-national level assessment as necessary. Participants also agreed that the checklist provided insufficient guidance to standardize scoring, and recommended that a further revision include explicit definitions of each possible score for each indicator. Appendix E2 notes other qualifiers on the use of PI data that were identified by the PI working group members in Miami.

CONCLUSIONS

The workshop resulted in an overall consensus, with some remaining reservations, that PI analysis can help answer the question-at least as a tie-breaker-of where scarce conservation resources should be invested, and can certainly help answer the question of what kind of investment should be made. A consensus statement (see Appendix E2) produced by the PI working group describes areas of agreement. It was also agreed that there was a need for further PI data collection and analysis in countries and, due to variation within countries, parts of countries where it might make a difference in priority rankings within and among priority RHUs. Data obtained from a mail survey administered by WRI subsequent to the workshop is used to provide examples of how PI data might be used in these contexts (see Appendix E3).

The Miami workshop also generated several "lessons learned" for structuring the PI component analysis in future priority-setting workshops. For example, bringing an expert opinion-generated data set to the workshop to be critiqued would likely accelerate the consensus-building process in the future.

Two other conclusions emerged from the PI working group sessions in Miami, with implications that reach beyond the immediate priority-setting goal of this exercise. First, working group members concluded that the checklist instrument could be useful to project managers within country missions to collect data relevant to decisions about the types of conservation-oriented activities USAID (or other donor organizations) should fund. Second, working group members concluded that while a considerable amount of uncaptured PI data exists (e.g., data collected by individual countries, or found within grey literature), very little of it is actually available for drawing sub-national, national, or regional, much less global, comparisons. This presents a challenge to international organizations that consider environmental data collection and dissemination to be part of their mandate, as socioeconomic and political data of this type are important for assessing national progress in meeting requirements as called for in both the Convention on Biological Diversity and Agenda 21.

CHAPTER FIVE
ASSESSING UTILITY

INTRODUCTION

Elements of biodiversity-specific and unique ecosystems/habitats, species/populations, gene pools/genetic material-which are perceived as useful or beneficial to humans are said to have utility value. This contrasts with "non-utility" value, which includes the intrinsic or aesthetic value humans attach to natural objects. The intrinsic values of biodiversity based on taxonomic diversity (e.g., number of species found in an ecoregion) or phylogenetic uniqueness are purely a matter of scientific data and judgment, while those based on utility ultimately reflect the normative values of society at large. Such normative values vary among the different groups benefiting from biodiversity. However, some utility benefits can be considered more or less universal. The purpose of the utility component of this priority-setting analysis is to consider high human utility value as a possible means of discriminating among otherwise equal regional habitat units.

APPROACH

Prior to the workshop in Miami, the Institute for Sustainable Development assembled data on five categories of biodiversity utility. Regional Habitat Units were characterized and ranked within Major Habitat Types based on these data. The five categories were carbon sequestration, genetic resources, productive and protective resources, hydrological resources and indigenous resources. The latter two were eliminated from further consideration in Miami because workshop participants thought the current data were incomplete or inappropriate. Descriptions of the categories which were used follow below.

Carbon Sequestration: This value attempts to measure the capacity of an ecosystem to contribute to the amelioration of global climate change through the storage of carbon. Biomass carbon content is the amount of carbon per area in live vegetation characteristic of a given Regional Habitat Unit. Its significance as an indicator of utility value is two-fold. Forests, in particular, can act either as sources or sinks of atmospheric carbon, and thus potentially exacerbate or ameliorate the potential for global warming resulting primarily from the combustion of fossil fuels. Considerable carbon is also stored in (and potentially lost to the atmosphere from) soil, and is measured as the rate of carbon accumulation per area.

Genetic Resources: A genetic resource is the genetic variability of a species (or variety or breed), useful or potentially useful to humans for the domestication or improvement of crop plants, livestock, poultry, fish, etc. The world food supply and all agricultural economies are dependent on a continual flow of genetic material which ultimately is derived from wild relatives and so-called land races (locally cultivated varieties or breeds). The centers of origin (and centers of diversity) of a number of the world's most important crops are in Latin America-cotton, tomato, potato, cocoa, sweet potato, pineapple, rubber, and others. Four separate categories were used to arrive at a value for genetic resources: Centers of Plant Diversity; Origins of Important Crop Species; Forest Tree Genetic Resources; and Domesticated Animal Origins.

Productive and Protective Resources: This value category combines a measure of an RHU's allochthonus resources (i.e., energy, biomass or nutrients that are exported beyond the boundaries of the RHU), the physical protection it provides, and its net primary productivity. Their measurement indicates relative importance of Regional Habitat Units in providing such benefits as being a source of nutrients or nursery site for fishery stocks, providing protection from coastal or riparian erosion, or assimilating pollutants.

More detail on how such utility data are compiled and weighted to produce a rank for each RHU for each of these categories can be found in Wilcox (1994).

RESULTS

Workshop participants agreed that human utility values were important in setting priorities for conservation investments, and stressed the need to consider local as well as global utility values. Participants recognized that measures of utility could potentially capture ecosystem function values of biodiversity not captured by biological values. The PI working group considered, critiqued, and agreed on the relative weighting of utility value categories presented at the workshop. They recommended a weighting that gives the highest value to genetic resources, followed by productive and protective resources, and carbon sequestration, in that order. The group agreed that "unique" utility values attributable to a specific ecoregion, such as wild relatives of important food grains, are more important than an ecoregion's incremental contribution to a "non-unique" value, such as carbon sequestration. Table 5 summarizes the final results of the utility analysis according to the weightings agreed upon by the working group participants.

While workshop participants found that human utility values are generally correlated with biological value, in some cases their explicit consideration would change the overall ranking of priority Regional Habitat Units depending on the particular integration model used (see Chapter Six, Integration Results, below). This was especially the case for RHUs with a high concentration of genetic resources, although the presence of wetlands, mangroves, or other "protective and productive resources" would also make a difference. Notable differences were found within Tropical Moist Lowland Forests and Tropical Dry Forests, in which genetic resources scores varied independently of biological value for some RHUs. Other differences were found in the Herbaceous Lowland Grasslands category.

DATA LIMITATIONS AND QUALIFIERS ON USE

Workshop participants were concerned about the lack of local data on human utility values, and indeed decided to discard the utility category that included hydrological values due to the lack of data. They recommended that additional investment be made in obtaining the local data, particularly data on hydrological resources such as watersheds.

The PI working group also discussed the appropriate use of available data on the presence of indigenous peoples. It was agreed that the measure should not be used as an overall indicator of utility value because it does not capture the dependence of non-indigenous or mixed cultural groups on biodiversity. However, it was agreed that data on indigenous populations would be valuable as an indicator of knowledge of biodiversity utility.

It should be noted that the three categories of utility chosen to generate the utility score (carbon sequestration, genetic resources and productive and protective resources) are by definition higher in forests than non-forests. Therefore, these data should only be used to make comparisons within Major Habitat Types, as they are in this report, and should not be used to make comparisons between areas from different Major Habitat Types.

CONCLUSIONS

There was general consensus at the workshop that human utility values are important to include in biodiversity priority-setting, although they would be greatly strengthened by adding local data on hydrological resources such as watersheds and other local utility values. In future priority-setting exercises, more time and thought should be devoted to consideration of how utility values should be integrated into the overall framework.

TABLE 5. FINAL RANKING OF REGIONAL HABITAT UNITS FOR UTILITY

CHAPTER SIX
INTEGRATION RESULTS

INTRODUCTION

The priority-setting framework produced by the NGO Working Group prior to the Miami workshop stated that a list of priority areas for biodiversity conservation would be produced by combining the results of the biological value, conservation status and policy/institution analyses, and incorporating considerations of human utility as a modifier for particular areas as appropriate. It was agreed that this integration process would occur in the context of the Miami workshop.

APPROACH

On the final day of the workshop, participants were divided into four groups, and each asked to devise a methodology to integrate the results obtained over the first three days of the workshop to arrive at a list of priority areas for biodiversity conservation. Each of the four groups contained a mixture of participants who had focused primarily on biological value, conservation status or policy/institutional feasibility. An illustrative model for integrating biological value and conservation status that had been developed by the NGO Working Group was presented to participants before breaking up into working groups. Each integration group developed an integration model and a resulting list of priority Regional Habitat Units, which were then presented in a plenary session.

INTEGRATION MODELS

The approaches taken by the four integration groups all produced remarkably similar lists of final recommendations. The four models were all based on the rankings of Regional Habitat Units according to biological value and conservation status (see matrices in Appendix B), but differed somewhat in their weightings of these two factors and their differential incorporation of policy/institutional (PI) and utility data. Summaries of all four integration models can be found in Appendix C. Three of these models were substantively the same (Models #1, 3, and 4 in Appendix C), in that they weighted biological value and conservation status equally, with regionally outstanding biological value > regionally significant > locally important; and Critical conservation status > Endangered > Vulnerable > Relatively Stable > Relatively Intact. Several of these integration groups explicitly discussed using utility as a "tie-breaker" among otherwise equivalent RHUs; others discussed using utility to "boost" the biological value of a particular RHU. Fish biodiversity was also used by one integration group to add "bonus points" to those regional habitat units identified as priorities using biological value and conservation status criteria. Model #4 (in Appendix C) was the most specific in its decision rules, and is thus considered representative of the other two, very similar, models.

In contrast, Model #2 (in Appendix C) considered biological value first in setting priorities. In considering conservation status, the model ranked Critical conservation status > Endangered > Relatively Intact > Relatively Stable > Vulnerable. This is different from the three other models which ranked RHUs within the same biological value group in order of increasing threat to habitat integrity. The integration group that developed this model, however, felt that relatively intact RHUs also deserve immediate conservation investment because it is within these now rare landscapes that ecosystem processes and species have the best chance for long-term persistence. Model #2 also explicitly considered PI criteria when prioritizing Regional Habitat Units within the medium biological value category. However, the highest priority RHUs as identified by Model #2 and listed in Table 6 can be identified without reference to PI characteristics since these all fall within the regionally outstanding biological value category. If adequate PI data were available, the order in which the six RHUs in the Regionally Significant category for biological value should be selected for investment could be affected.

Depending on whether these integration models are applied within Major Habitat Types or across Major Habitat Types, slightly different results are obtained. It was not completely clear to all workshop participants whether our goal of conservation in all the Major Habitat Types, and the fact that the biological and conservation threat analyses were done within Major Habitat Types, would require that integration rankings also be done within Major Habitat Types. Some groups made recommendations across Major Habitat Types (and there was even confusion within a single group on whether their results were across or within Major Habitat Types). Therefore, Table 6 reports the recommended biodiversity conservation priorities based on Model #4 and Model #2 applied both within Major Habitat Types and across Major Habitat Types.

INTEGRATION RESULTS

The NGO Working Group recommends geographic priorities based on an approach to integrating the four levels of analyses that is a hybrid of the two basic integration approaches developed by workshop participants. The NGO Working Group considered it important to recommend equal numbers of RHUs from all Major Habitat Types. Ranking of Regional Habitat Units within Major Habitat Types was based on a consideration and weighting of biological value and conservation status. Two RHUs with the same rank based on these two values were differentiated based on their rank for utility. Due to a lack of satisfactory data, political/institutional criteria were not incorporated into the NGO Working Group's ranking (see Data Limitations, below).

Based on the NGO Working Group's model, the following seven RHUs are identified as highest priority for biodiversity conservation (one within each Major Habitat Type): Atlantic Forest; Tropical Andes; Cerrado-Pantanal; Mexican Xerics; Patagonian Steppe; Puna; and Mexican Pine-Oak. Seven additional RHUs (also one from each MHT) that are recommended as high priority are: Upper Amazon; Caribbean Moist; Chaco; Caatinga; Pampas; Paramo; and Southern Temperate Forest. These recommendations are illustrated in Figure 4 and presented in tabular form in the final column of Table 6.

TABLE 6. INTEGRATION RANKING OF REGIONAL HABITAT UNITS FOR CONSERVATION PRIORITY

DATA LIMITATIONS AND QUALIFIERS ON USE

Lack of Satisfactory PI Data
Workshop participants discussed three ways to integrate PI analysis into the recommended geographic priorities:

1) Non-Integration: Some workshop participants argued that PI analysis should be used only to determine the kind of investment to be made after geographic priorities have been set based on other criteria.

2) Integration as a Tiebreaker: Many participants agreed that PI analysis should be used to choose between two otherwise equivalent regional habitat units (RHUs), or to prioritize among parts of single RHUs that straddle political boundaries.

3) Integration as a Decision Rule: Participants in one of the integration groups proposed that PI analysis could be used as a decision rule to prioritize among non-equivalent RHUs. For example, Model #2 (described above) would reject critically threatened RHUs of regionally significant or locally important biological value if PI values were low (i.e., fell into Cell D in Table 4) in favor of less threatened RHUs of similar biological value characterized by higher levels of political commitment or institutional capacity (Cells A, B, or C in Table 4).

Because a satisfactory data set was not available in Miami, PI criteria are not incorporated into the geographic priorities recommended by the NGO Working Group. However, two examples of how better PI data could be incorporated into the integration models described above can be found in Appendix E3.

Scale
As mentioned in Chapter Two, Assessing Biological Value, any attempt to assess biodiversity priorities must address the fact that the definition of priority areas is dependent on the scale of the analysis. Since the scale of this analysis was at the regional (LAC-wide) level, the resulting geographic priorities should only be used at the regional level, and should not be used to determine priorities at the sub-regional or country level. The geographic priorities identified in this report say very little about what biodiversity conservation priorities for the Caribbean, Central America or Bolivia should be. This exercise would have to be repeated at a sub-regional or country-wide scale in order to obtain geographic priorities relevant at the sub-regional or national level. Different criteria would undoubtedly be used at national and sub-regional scales than were used at the hemispheric regional scale.

Use of Expert Opinion
All four levels of analysis used in this priority-setting exercise, biological value, conservation status, policy and institutional feasibility, and utility, relied to some extent on expert opinion to confirm and supplement available published data. Although the workshop included broad regional and taxonomic expertise among participants, it is possible that a different set of experts might have generated a somewhat different list of biodiversity conservation priorities. The criteria and methods used by experts to rank areas for biological value and conservation status were standardized through the use of forms (see Appendix D) and the completed forms are available for review. We encourage all nations and donors to continue to support on-the-ground biological and conservation surveys to improve the availability of published data.

CHAPTER SEVEN
RECOMMENDATIONS AND NEXT STEPS

HOW USAID SHOULD USE THESE RECOMMENDATIONS

We recommend that USAID continue to play a leadership role within the donor community with regard to biodiversity conservation by reviewing current USAID, other donor, and national investments in the 14 RHUs identified as high priority by this exercise. USAID should then work to increase its and other donors' investments in those areas not currently receiving sufficient conservation investment. All remaining RHUs are considered appropriate for continued biodiversity conservation investments at the national and local levels. USAID should undertake the necessary policy/institutional feasibility analysis to determine which countries within the priority RHUs should receive increased investment. This does not mean that USAID must necessarily develop a new biodiversity project in each of the 14 RHUs identified as high priority for biodiversity conservation in this report. USAID could choose to increase investment in some of these areas by increasing awareness of their importance to biodiversity conservation on the part of other private, bilateral or multilateral donors making biodiversity investments in Latin America and the Caribbean. In an era of declining budgets and financial constraints, USAID must be creative to maintain its leadership in the field of biodiversity conservation.

USAID should not use the recommendations of this report to assess the appropriateness of biodiversity investments by USAID missions at the country level. The fact that Central American Lowland Forest was not identified in this exercise as a high priority, does not mean that USAID/Guatemala or USAID/Nicaragua should not support the conservation of this unique and important ecosystem. The conservation of this RHU at a Central America-wide scale, and certainly at a national scale, is undeniably important. There are many other similar examples throughout Latin America of the dangers of applying the results of this exercise at the wrong scale (see section on Scale, above). The approach used in this regional analysis could be applied to identify priorities at the national or sub-regional level, if criteria and data were adapted to the scale of the analysis.

USAID should use the results of this exercise to diversify its biodiversity investments into RHUs across the full spectrum of conservation status, from critical to relatively intact. At the regional scale, the "Red Data Book" approach used to assess conservation status can target those ecoregions in a critical state, where investments over the next five years will largely determine the fraction of species and habitats that might persist over the next few decades. The method also identifies those ecoregions where investments are needed to ensure that some of the most intact examples of each Major Habitat Type can be conserved for posterity and, for the present, can be kept from being degraded to vulnerable. We suggest that there is an inverse relationship between the conservation status of an ecoregion and the costs required to restore or protect its biota: the more threatened, the more costly it will be to maintain its integrity. Investing now in the more intact areas should also be part of any portfolio that emphasizes cost-effective investment.

It is also recommended that USAID continue to advocate the importance of geographic priority-setting, using the approach outlined in this report, in the Agency's dealings with other multilateral and bilateral donors and national governments. The Global Environmental Facility should be encouraged to undertake systematic geographic priority analyses using biological importance, conservation status and policy/institutional feasibility (as they have begun to do in the LAC region) to better justify the grants they make. USAID is supporting a number of national biodiversity conservation strategies around the globe in order to help countries meet their obligations under the biodiversity convention. Such strategies have, in the past, focused primarily on what types of conservation investments to make, with little attention to a systematic analysis of where in the country conservation action is most important. USAID has an opportunity to strengthen national conservation strategies by supporting cooperating country efforts to address both the "what" and the "where" simultaneously. Similar data are collected for both types of analyses, so inclusion of geographic considerations should not increase costs significantly.

HOW OTHERS MIGHT USE THIS REPORT

The Working Group feels that this priority-setting exercise has produced an innovative framework and set of useful indicators for conservation decision-making (verified and validated by a wide variety of regional experts) that should be useful to other organizations involved in priority-setting and conservation planning. By overlaying the conservation status of ecoregions with the data layer on biological value, conservationists create a powerful tool that guides investments in a more objective fashion than in past priority-setting strategies. This report should be a useful reference for countries and organizations seeking advice on what types of data are useful to guide decision-making and how these data might be effectively presented. It is hoped that this report will provide a baseline for future efforts.

The valuable biological, conservation status, policy/institutional and utility data collected for this priority-setting process will be made available to a wide audience of potential users. New data can be incorporated as they become available and the approach can be reapplied with new assumptions and weightings, as appropriate.

The approach and criteria developed for this exercise could also help guide countries in the LAC region to objectively prioritize biodiversity investments within their political boundaries. By applying biological value, conservation status and sub-national political and institutional feasibility criteria to the ecoregions within their national boundaries, as Argentina is now doing, they can better identify and protect the larger blocks of remaining habitat that are critical for biodiversity conservation, as well as the biologically rich but highly fragmented and altered ecoregions requiring urgent attention. One of the first things that countries are expected to do after ratifying the Biodiversity Convention is to prepare a National Biodiversity Conservation Strategy. As noted above, such strategies have tended to focus primarily on what types of conservation investments to make, with little attention to a systematic analysis of where in the country conservation action is most important. We hope that the geographic priority-setting framework presented here will be useful to many countries to assist them in identifying the "where" as well as the "what" in their national biodiversity conservation strategies.

NEXT STEPS

Marine
The fact that marine and freshwater aquatic biodiversity conservation priorities are not addressed in this report is a serious limitation. Recognizing this limitation, USAID has provided BSP with funds to carry out a parallel priority-setting exercise for marine and freshwater habitats. This exercise is now underway and a final report that identifies regional biodiversity conservation priorities in marine and freshwater habitats in Latin America and the Caribbean should be available in 1996.

Other Regions
The application of this geographic priority-setting framework to Latin America and the Caribbean was intended to serve as a pilot effort which would be followed by similar exercises for identifying biodiversity conservation priorities in other regions, such as Asia and Africa. Given the interest that many Miami workshop participants expressed in applying the framework at a national level, BSP is now exploring the possibility of refining the methodology so that it is useful at a variety of scales, including at the national level. As we learned from our experience in applying the framework in Latin America and the Caribbean, the question of where to conserve is directly linked to the question of what types of activities should be supported. We feel there would be many benefits to a country in identifying its geographic biodiversity conservation priorities at the same time that it is establishing its national biodiversity conservation priorities sectorally and programmatically.

Next

Document information

This publication was made possible through support provided to BSP by the Global Bureau of USAID, under the terms of Cooperative Agreement Number DHR-A-00-88-00044-00. The opinions expressed herein are those of the authors and do not necessarily reflect the views of USAID.

Prepared by: Biodiversity Support Program, Conservation International, The Nature Conservancy, Wildlife Conservation Society, World Resources Institute, and World Wildlife Fund.

ISBN 1-887531-17-3

© 1995 by The Biodiversity Support Program. All rights reserved. No part of this book may be reproduced without the permission of The Biodiversity Support Program, c/o World Wildlife Fund, 1250 24th Street, N.W., Washington, D.C. 20037, USA.

The Biodiversity Support Program (BSP)was established in 1988 with funding from the Research and Development Bureau of the U.S. Agency for International Development (USAID), under cooperative agreement number DHR-5554-A-00-8044. BSPs implemented by a consortium of World Wildlife Fund, The Nature Conservancy and World Resources Institute. BSP works to conserve biodiversity in developing countries by supporting innovative, on-the-ground projects that integrate conservation with social and economic development; research and analysis of conservation and development techniques; and information exchange and outreach.

Citation: Biodiversity Support Program, Conservation International, The Nature Conservancy, Wildlife Conservation Society, World Resources Institute, and World Wildlife Fund. 1995. A Regional Analysis of Geographic Priorities for Biodiversity Conservation in Latin America and the Caribbean. Biodiversity Support Program, Washington, DC, USA. 140 pp.

For more information and/or copies of this publication contact: The Biodiversity Support Program, c/o World Wildlife Fund, 1250 24th St., NW, Washington, DC 20037, USA. Tel. 202-293-4800, Fax 202-293-9211.