"How global is the Global Biodiversity Information Facility?" ask the authors of the paper, published online by PLOS One (Yesson C, Brewer PW, Sutton T, Caithness N, Pahwa JS, et al. (2007) How Global Is the Global Biodiversity Information Facility?. PLoS ONE 2(11): e1124. doi:10.1371/journal.pone.0001124).

In order to examine the geographic and taxonomic coverage of the data available via the prototype GBIF data portal (as of October 2005), authors Chris Yesson, Alastair Culham and their colleagues compared data extracted from GBIF for the legume family of plants (peas, beans, etc.) with the International Legume Database & Information Service (ILDIS). They used Biogeomancer to add coordinates to some records that did not have them, and conducted spatial validation using the Taxonomic Database Working Group Geography Standard version 2.0 level 4 areas and the ILDIS distribution data.

Among other results, the authors found that the global species coverage for legumes, as represented in the GBIF data, was 27%, but that over 500 of the TDWG level 4 areas had less than 5% coverage; some of these areas are considered globally important areas for biodiversity. Thus, GBIF data for Legumes are a geographically biased sample.

Large parts of the globe, including hotspots of biodiversity in Africa and Asia, are GBIF-data deficient. These gaps could most readily be filled by existing databases being made available via GBIF by their owners, and by digitisation and sharing of the major museum collections via GBIF. The authors conclude that GBIF has taken substantial steps to achieve its goals for primary data provision, but support from the developed nations is needed to encourage more data providers to digitise and supply their records.

GBIF has recognised from its inception that for its full usefulness to the world to be realised, a large critical mass of data would have to be mobilised. Now that it has proven the concept that a worldwide, distributed data network is not only feasible but can succeed well, it has set a goal of mobilising 1 billion (1,000,000,000) records by the end of 2008 (see story on GBIF's 1 Billion Records goal).

The UK researchers found that at least 84% (a very high proportion) of current GBIF point data are correct, according to the stringent criteria set during their study. "We paid particular attention to the quality of location information stored with each specimen record, and the breadth of coverage for known species," said Chris Yesson, one of the lead scientists involved. GBIF provides tools for detecting those records that should not be used in a particular analysis.

However, current GBIF data coverage was found to be lacking in precisely those areas of the world that are known to be "biodiversity hotspots", meaning that the areas have exceptionally high numbers of species but these are not well-represented in GBIF data. "A focussed effort on these areas will create a truly global resource that creates unparalleled opportunities for understanding the world's biodiversity," concluded Yesson.

Alastair Culham, corresponding author of the study, said "GBIF provides a fantastic resource that may prove vital in understanding the impacts of global climate change at this critical time." The contribution that can be made will depend on increasing the numbers of digital records by orders of magnitude, and this increase must be made very soon.

The study used as an example the world's 11 largest botanical institutions. These had digitised and shared the data from only about 6% of their plant specimens as of October 2005. As of October 2007, that has increased to 9%. However, the need is for growth in digital data sharing that is at least 30 times faster than this.

Culham also emphasised that "governments of the world's developed nations should invest more money to support institutions and scientists around the world in their quest to digitise, publish and upload more of the data presently hidden in museums." GBIF calls for these and other data holders (for instance of monitoring and observational data) to help address the world's need to understand its biodiversity in order to promote sustainable development.

The BBRSC funded biologists at the University of Reading's School of Biological Sciences and computer scientists at the University of Cardiff to conduct the study. The team reviewed the quality and coverage of GBIF-mediated data by comparing those for the legume (beans, peas) family with information in the International Legume Diversity Information System (ILDIS).