MAP OF THE ECOSYSTEMS OF CENTRAL AMERICA

From 1998 - 2000, the CCAD, World Bank and the Netherlands collaborated in the production of the Map of the Ecosystems of Central America. The project leader was Douglas Graham and the project coordinator Daan Vreugdenhil.

Full map and GIS title and authors:

Full final document title and authors:

Summery of a Biodiversity Mapping Project of Global Significance

Graphic impression of the map

All project files for downloading

The Central American Ecosystems Map is the culmination of decades of research by ecologists from across the region, many of whom worked directly or indirectly from national university positions. Production of the map was a team effort by the biodiversity and environmental conservation institutions of the Central American countries and their coordinating institution, the Comisión Centroamericana de Ambiente y Desarrollo (CCAD). The project team under the overall coordination of Daan Vreugdenhil of the World Institute for Conservation and Environment (WICE) and Douglas J. Graham of the World Bank is grateful for the vision and support of Mauricio Castro, Executive Director of CCAD, and Lorenzo Cardenal, Director of CCAD’s Mesoamerican Biological Corridor (MBC) project.  This effort was made possible by financing from a variety of sources: The Netherlands, the Global Environment Facility (through national MBC projects implemented by the World Bank and a regional MBC project implemented through the UNDP), the participating countries, and the World Bank. The initiative cost roughly $2 million and was carried out between early 1999 and mid-2001.

The primary objective of the mapping project is to map and describe the ecosystems of Mesoamerica (Belize, Guatemala, El Salvador, Honduras, Nicaragua, Costa
Rica, and Panama, using a comprehensive, regionally endorsed, classification system. Some of the key benefits:

• Input to the geographical delineation of the MBC;

• Inputs for national conservation and rural development strategies;

• Prioritization of protected areas (through presence/gaps analyses) given that our ecosystems are a proxy for unique sets of animal and plant communities and ecological processes;

• Assessment of the conservation value of protected areas;

• Creation of a baseline for further ecological stud-ies and biodiversity monitoring;
  Better information for environmental impact assessments;

• Enhanced understanding of the region’s ecology on the part of national and international scientists.
  

The mapping project was executed by national teams of lead scientists on biodiversity selected by the national authorities for biodiversity conservation in each of the participating countries, with coordination and assistance from an international team with extensive experience in vegetation mapping and use of geographic information system (GIS) applications and remote sensing. Production options were discussed with the national authorities, and in each country collaborating scientists from national universities or other institutions were contacted. A number of exchanges were needed between all par-ticipants to decide on the methodology and to exchange experiences during the course of the work. Once a method and approach had been decided, all the participating scientists needed to be trained in the methodology, interpretation, and handling of remotely sensed images, use of GIS applications, etc. In total, about 20 national scientists and government officials participated in intensive training sessions and many more in the various meetings and workshops that were organized. The Ecosystem Mapping Project involved five principal workshops/training sessions during the course of the mapping process. Regional harmonization and compatibility between the different national efforts has been attained through:

• Joint training;

• Coordination through frequent visits by the regional
  coordinator (Vreugdenhil) to each of the participating
  countries;

• Promotion by the World Bank and CCAD of intercountry
  coordination at various levels;

• Creation of a cross-country classification integration table;

• 
  Organization of a final integration workshop;

• Preparation of ecosystem descriptions.

The methodology used for the classification was the classification system developed under the auspices of the UNESCO in 1993 and further elaborated in 1974 by Möller-Dombois & Ellenberg: "A Tentative Physiognomic-Ecological Classification of Plant Formations of the Earth". It describes the above-ground or underwater vegetation structures and cover as observed in the field, described as plant life form. This classification is fundamentally a species-independent physiognomic, hierarchical vegetation classification system which also takes into account ecological factors such as climate, elevation, human influences such as grazing, hydric regimes, and survival strategies such as seasonality. The system was expanded with a basic classification for open water formations.

In Central America, the images of LANDSAT 5 Thematic Mapper are the most commonly used data sources for mapping of large areas. As a contribution to disaster relief, the USGS had made a set of images available for the territory covering Belize to the border between Nicaragua and Costa Rica. This considerably reduced the purchase costs of the imagery. For Costa Rica and Panama, some images were provided by the governments and some from the private collection of D. Muchoney. The provided imagery was considered for our use based on the following:

• Date of the image

• Spatial and spectral resolution

• Location accuracy

• Atmospheric conditions (cloud, haze)

• Sunlight angle 

• The countries covered by the USGS series

A few additional ones needed to be purchased For Belize, Guatemala, Honduras and Panama, the imagery was processed in bands 4, 5, 3, which allow for the strongest contrast between forested land (shades of brown) and non-forested land (shades of green). For Nicaragua and Costa Rica, mainly bands 5, 4, 3 (mostly shades of green) were processed on request of the lead scientists in these countries. For each country, the images were prepared in the projections of the national topographic maps. Georeferencing consisted basically of choosing reference points in both the image (cross roads, fixed river curves and coastal rocks, etc.) and defining a first grade polynomial regression line for optimizing the scene adjustment. For each scene 15 to 20 reference point were used, evenly distributed across the scene.

An additional set of imagery has been prepared in Projection UTM, Zone 15, Elipsoide Clarke 1866, Datum Nad 27, bands 4, 5, 3, so that users can project the entire map against a background of imagery used in the project. The regional map has been referenced as Projection Lambert-Azimuthal center -85, 13, north and false northing and easting of 1,000,000 (one million) m. The database originated from the "STEP" design of the University of Boston, (Muchoney et al. 1998) for many of the tracking and physical data. For biological data a new selection was composed after consultation of renowned external international scientists (Prof. Dr. Ir. R.A.A. Oldeman (University of Wageningen), Prof. Dr. A. Cleef (University of Amsterdam), and Dr. H. van Gils, ITC). Severe field testing by the national scientists lead to further development of the design. The resulting database allows for highly detailed description and characterisation of ecosystems and is probably suitable for worldwide use.

As fieldwork is the most expensive and the most difficult activity to organize in an ecosystem mapping project, all national teams but the Honduran team chose to carry our field sampling of a directed pre-selection of areas for field visits. In general relevé samples were selected on the following criteria:

• Insufficient expert knowledge of the region;

• Doubt about the classification on the image;

• Representation of the image class (each recognized
  class was preferably visited at 3 different locations).

• Regional consensus about methodology;

• Processed and georeferenced satellite images used in
  the course of the project;

• GIS vector files of the ecosystem maps of each of the
  seven countries with coordinated classification
  nomenclature;

• Ecosystem descriptions with good biological background
  information;

• MS Access-based database with the field data collected
  in the context of the project;

• The expansion of national herbariums with several
  thousands of herbarium specimens;

• Lead scientists of the scientific ecological community in
  each of the participating counties trained;

• A final report in Spanish and English and a final integrated
  map.

Collectively a total of 197 ecosystems have been recognized in the context of this study (including agriculture and urban). About 25 additional codes have been defined using modifiers for levels of human intervention but these do not represent distinct ecosystem classes per se.

The Global Significance of the Methodology for Biodiversity Identification by Proxy and Geographical Quantification

For decades, conservationists and biologists have been struggling with the determination of what needs to be protected, particularly in tropical countries.  While in the temperate climatic regions important habitats have been identified with a reasonable amount of accuracy to estimate to which extend most species and their habitats are protected.  Many factors have contributed to that situation:

• The presence of relatively large numbers of biologists and knowledgeable amateurs

• Almost a century of ever increasing systematic biodiversity inventory work

• Significant budgets from governments and large conservation organizations (Of course by no means sufficient)

• Manageable lists of national species

In the tropics, conditions are very different.  

• Although, some individual effort of biodiversity inventories often also date back to the beginning of last century, in many countries biodiversity inventories are just recently starting up. 

• Most tropical regions are located in developing countries and the available budgets for biodiversity inventories is minimal.  The few experience biologists are usually heavily restricted in their possibilities to carry out fieldwork.  Transportation, equipment and money for field expenses are almost always absent.  The low salaries of the biologists, usually makes it impossible for biologists to finance such costs by themselves.

• he number of available well-trained biologists who have actually found a job as a biologist that allows them to go out and study nature is extremely small, and the number of knowledgeable amateurs is extremely limited.

• But more striking than any of the aforementioned reasons is the simple fact, that biodiversity in the tropics is so diverse, that even an army of biologists would need to study for decades in any medium-sized tropical country to be able to identify and map out the national biodiversity. Any biologist who has worked for some time in the tropics gets this feeling of hopelessness. There are so many species! Where does one begin?!

Conservationist feel that it is important that as much as possible, all species of a country be conserved in a system of protected areas and national parks.  For many years,  highly skilled biologists would visit natural areas at a number of locations and register every species they could identify during the days they would visit each location.  But of course such well meant attempts never could do justice to the great biological wealth of these tropical habitats.  So what to do in stead?  In order to evaluate if a system of protected areas would host a representation of the natural heritage of a country, we really must have an idea of what is there.  Complete inventories of species in the tropics are impossible to make in the next 25 years, even if we would have enough money and trained personal to do it.  Many species simply are not yet described! So we had to find a method by approximation.  Something that could represent sets of species without knowing them and that would at least partially differentiate one set from another.

The method that has been used, the UNESCO Physiognomic–Ecological classification system, distinguishes and classifies vegetation types by their structure and different ecological criteria.  In our document, it is argued, that the criteria used by this system distinguish entire ecosystems; not merely plant communities.  Thus the system distinguishes between full sets of both plant and animal species present in each ecosystem.  In many cases it is possible to further differentiate certain vegetation structure types by a dominant species of set of species, thus deepening the biodiversity differentiation per UNESCO-class.

The classification system is suitable for all terrestrial ecosystems on earth, but it only partly succeeds in describing aquatic ecosystems.  As aquatic ecosystems are paramount in any protected area system, a basic descriptive set was added to also describe aquatic ecosystems.  Although, the aquatic ecosystem component needs further elaboration, the system thus created is suitable to describe all ecosystems on earth.

This is all great, but if it takes year