BiodiversityWorld: a GRID-based distributed environment for studying global biodiversity patterns
Sutton, T.1, Bhagwat, S.3, Brewer, P.W.1, Bisby, F.A.1, Bromley, O.4, Caithness, N.1, Culham, A.1, Fiddian, N.J.2, Gray, W.A.2, Jones, A.C.2, Pittas, N.2, Scoble, M.J.3, Sutton, T.1, Valdes, P.J.5, White, R.J.2, Williams, P.3 & Xu, X21 Centre for Plant Diversity & Systematics, School of Plant Sciences, The University of Reading, READING RG6 6AS, UK
2 School of Computer Science, Object and Knowledge Based Systems Group, Cardiff University, CARDIFF CF24 3XF, UK
3 Department of Entomology, The Natural History Museum, Cromwell Road, LONDON SW7 5BD, UK
4 School of Biological Sciences, University of Southampton, SOUTHAMPTON SO16 7PX, UK
5 School of Geographical Sciences, University of Bristol, BRISTOL BS8 1SS, UK
BiodiversityWorld is an e-Science pilot project in the UK. Its aim is to implement a prototype problem solving environment for global biodiversity patterns within a GRID-based distributed system. After one year there is progress to report on both the underlying system and the potential for using it in biodiversity research.
A simple test version was implemented in the first six months and used to check the scoping of the system requirements. Then followed an intensive design phase, leading to our Version 2 architecture implemented from month eight and now available. Key features of the design are that modules are connected via a standardised API, and a layered architecture insulates this process from the underlying and changeable GRID communications layer. The design of variable and repeatable workflows, and the storage and management of data and metadata, are all central functions enabling the use of the system for biodiversity research.
The exemplar application areas have only just started, and in two it is too early to report; the work on biodiversity hotspots and optimal reserve selection at the NHM, and the work on phylogeny, climate and biogeography at Reading. The third, bioclimatic modelling of species distributions, has been used to test the prototypes. Lathyrus japonicus, the sea-pea found on shingle beaches across the coasts of N. America and Eurasia provides an illustrative example. Using current locations abstracted from databases using varied taxonomies, a bioclimatic envelope is developed, and then projected over the globe to locate other possible areas of suitability. Two methods in use differ interestingly in the details of the projection - one restricts sites accurately to its known distribution in the northern hemisphere, while the other predicts an area of possible suitability in Chile. Intriguingly we have now located reports of small populations in Chile. By projecting the model into past and future climates we hope to infer whether it must be a new arrival there, or whether it may be a relict from a formerly larger distribution, and whether it is likely to survive future climate change.