Prepared by Eleanor Sterling, American Museum of Natural History and Marsha Sitnik, Smithsonian Institution National Museum of Natural History
Background
Characterizing biological diversity becomes increasingly important as widespread habitat loss and extinctions expand across the globe. Accurate assessments of species richness and endemism, the degree of threat to biological diversity, and the area of habitat remaining are crucial underpinnings of any conservation management strategy. Rapid assessment techniques must be developed to gather this information in a comprehensive and cost-effective manner.
Current efforts to measure and monitor biodiversity at the regional and site levels tend to be undertaken by individual scientists or institutions, each with their own idea of how data should be gathered. It is often impossible to compare biodiversity data between sites because of incongruities in methodology and data bases. The need for standardized measures of biodiversity is now recognized, and researchers have made strides towards developing standardized protocols. These protocols and others being developed should be tested and then communicated broadly to researchers engaged in characterizing biodiversity distribution.
Traditional methods for monitoring changes in distribution and abundance of species are labor-intensive and realistically can only be applied to small geographic areas and selected taxa at a time. Remote-sensing data have the potential to be very useful in characterizing the distribution of biodiversity over a large geographic area. These data could potentially increase our ability to assess regional biodiversity rapidly and to monitor small- and large-scale changes in biodiversity over time. However, at the present time, thematic maps based on remote-sensing data do not adequately reflect patterns of biodiversity, i.e. different forest types or plant and animal communities.
Thus, the combination of remote-sensing data and field data on species distribution could provide a tool to: measure biological diversity more efficiently through improved thematic mapping algorithms; put ground-level surveys into a wider context; monitor particular areas in terms of disturbances (for instance the introduction of alien species); monitor changes over time (i.e. time series) in the most rapid, efficient way; input site/regional level information into a global surveillance system.
Time series (i.e. the comparison of two standardized surveys at the same locality at different times) of site-specific, detailed assessments of biodiversity can be used in designing monitoring programs and in evaluating the effectiveness of conservation programs.
Discussion of Issues
Measuring biodiversity on a regional or site level will entail multi-taxa, species-based assessments using standardized techniques. Classification and description of all the species living even in a limited area is impracticable. A more efficient and effective system involves sampling of "indicators" whose change in either time or space represents change in biological diversity as a whole. These indicators thus serve as measurable surrogates for the greater environment.
Selection of appropriate indicator species is a key element of this process. Most prior work with indicator species has focused on vertebrates, despite the fact that invertebrates are more numerous and major habitat engineers. Since the relationship between distribution of vertebrate species and other groups is not known, focusing on a sampling package of a variety of indicator taxa, including both faunal and floral, would be important. These sampling packages must be flexible in order to adapt to widely varying ecological conditions and national capacities.
Several criteria contribute to indicator taxa selection, including that they have:
- Relatively High Abundance
- High Species Richness
- Many Specialist Species
- Fine-Scale Measurability, Both Temporal And Geographical
- Reliability
- Ecologically Relevant Position In The Ecosystem
- Relevance To Policy Or Management Decisions
- Necessary Resources Available To Analyze (E.G. Systematic Framework, Trained Personnel, Etc.)
To obtain a critical mass of information on biodiversity distribution, several institutions and scientists need to collaborate on standardized data collection.
This assessment system can be applied to other regions to assess biodiversity cumulatively on larger geographic scales or it could be reapplied at the same site to contribute to studies of monitoring biodiversity over time. Monitoring data can contribute to evaluation of the impact of management strategies or of human activities on the conservation of biodiversity.
Data gathered for the assessments can contribute to national biodiversity information systems (as mandated in the Convention on Biological Diversity) that would facilitate strategy and policy development.
How to Address the Issue
Museum community (MC) and NGOs must select taxa for pool of potential "indicator" species based on estimates of each taxon's sensitivity to environmental change, applicability for generalizing across vegetation types (robusticity), and expertise in identification of the taxa within collaborating institutions.
Multi-taxa, multi-disciplinary data collection, including climatic variables, soil, sociological, ecological, remote sensing, estimates of threat and taxonomic data should be collected.
Indicator taxa should be tested for feasibility.
Algorithms could then be developed to combine remote-sensing data and geo-referenced data for a biodiversity thematic mapper.
Data can be applied to conservation needs.
In-country professionals should develop skills to apply data in national biodiversity information systems through this type of project.
NGOs and NASA need to choose pilot area(s) where information on biodiversity distribution and composition is well known, to more quickly and accurately test the efficacy of the model system.
Governmental and NGO organizations within the host country of the pilot area should be involved immediately after the workshop (preferably during!) on developing pilot project goals and objectives.
Recommended pilot project characteristics
Project must be multi-disciplinary, involving botanists, zoologists, space scientists and anthropologists where feasible.
Both earth- and space-based data will be collected using standardized methods and standardized databases.
Training in-country professionals in both space- and earth-based data collection will be an integral component of the pilot project.
The project should be developed around a current environmental problem in the pilot areas such as coral bleaching, rain forest conversion or forest fragmentation.
Project site should be in an area where the fauna and flora are well-inventoried to facilitate evaluation of efficacy of model system.
GIS analysis should include mapping data from satellite images, aerial surveys and existing maps. Layers could include information on soil characteristics, altitude, rainfall, percent canopy cover, forest structure, forest composition, and distribution of individual taxa. May be expanded to include land-use cover and land-use categories.
Species distributions in unsurveyed areas may be predicted through thematic maps produced with the developed new algorithm on the basis of congruence in environmental characteristics with areas known to contain species.
Evaluation of Results
Results from the pilot study can be evaluated in comparison with known species distribution and composition data to determine effectiveness of the various indicator species. Evaluation of the efficiency and applicability of this method, i.e., are shortcuts possible, or what needs to be modified for the method to be viable, is important. Evaluation via peer review would be recommended at the end of the first year.
back to main
