It was recognised from the outset of this project that gathering data for the different categories would present a challenge. This proved to be the case. A major issue in ensuring comparability between the species/habit groups came from the differences in data that were available for each group.
Although problems were encountered in gathering data for the various species/habit groups selected for this project, there was no indication that data would have been more readily available for other species/habit groups. The particular problem that arose with trees was due to including a group that was based on habit rather than taxonomy.
As anticipated, production of data on threatened species by country was straightforward as these figures could be generated directly from UNEP-WCMC's databases of threatened species. These data need to be kept under continual review and addition to allow future similar analyses to be made.
Similarly, identification of single country threatened species was a simple matter of identifying the subset that are only recorded in one country.
For the animal groups involved in this study, the information on threat status is reasonably comprehensive. The plant groups selected for the study were chosen on the basis of availability of data on threat status. In general, the data on palms are good and an expert consultant on ferns was available to provide input to the data on threatened ferns. (The taxonomy of ferns presents a particular problem, and the figures generated may vary considerably according to the taxonomy followed). The tree data that were available resulted from a recent three-year study that included regional workshops in Costa Rica, Zimbabwe and Vietnam. As a result the database has a regional bias with data being much more comprehensive for these regions. Data on temperate trees was notably less good.
In general, the issues discussed under 4.2.1 for trees apply equally to this category. Estimates of the number of forest occurring species are not available, either at the country level or by taxonomic group. The only species groups for which figures were obtained were the relatively small groups of ferns and palms.
A total of just under 7,000 threatened single country endemic species were evaluated in order to classify them as forest or non-forest species. A large proportion (81%) of these was considered to be forest occurring.
An effort was made to identify forest occurring, non-threatened single country endemic animal species. Information was gathered on c. 2,000 species, 50% of which were considered to be forest dwelling. However, the process was time-consuming because little information was readily available and, importantly, the quality of information was low. The results were considered to be of dubious conservation value, and we do not recommend that further work is done to assess the remaining 10,000 species.
Phase 2 of the project was designed to expand on the level of data for the two categories of data that proved to be obtainable for all species/habit groups: threatened single country endemic species and the subset of these that are forest occurring. It was clear that information on the names of the relevant species was necessary to support the figures presented.
This project comprised a desk study. A next step should involve circulating the list of species identified as threatened forest occurring single country endemics to each country. This would enable the data to be confirmed by each country and for more detailed distribution data to be obtained, ideally including point-location specimen data.
It would be interesting to analyse the information presented here on threatened forest occurring endemic species in terms of forest loss, area of forest remaining and area of protected forest. This would provide an indication of threat to the species listed in Annex 2. Mapping these species, particularly if detailed distribution data are available, against habitat and protected areas data is an obvious priority. This would indicate forest areas of importance for the survival of particular species. Conversely, by indicating species that occur in areas of high forest loss and low forest protection, the analyses would identify the forest occurring endemic species at greatest threat and hence a priority for protection.
The plant groups of ferns and palms show the greatest potential for further analyses, as these were the groups for which most data were available. Ferns have a greater relevance to a global study, as this group has a global distribution. The data on palms were the most comprehensive, but as palms are limited to the warmer regions of the world, they offer less scope as a global indicator of forest condition. A useful further activity would be to identify the names of the fern species involved. This would involve attention to fern taxonomy, which currently remains confusing, but would be an essential first step in establishing a definitive list of single country endemic ferns.
