The use of the GARP genetic algorithm and Internet grid computing in the Lifemapper world atlas of species biodiversity

Lifemapper (http://www.lifemapper.org) is a predictive electronic atlas of the Earth's biological biodiversity. Using a screensaver version of the GARP genetic algorithm for modelling species distributions, Lifemapper harnesses vast computing resources through volunteers PCs similar to SETI@home, to develop models of the distribution of the world's fauna and flora. The Lifemapper project's primary goal is to provide an up-to-date and comprehensive database of species maps and predictive models (i.e. a fauna and flora of the world) using available data on species' locations. The models are developed using specimen data from distributed museum collections and an archive of geospatial environmental correlates. A central server maintains a dynamic archive of species maps and models for research, outreach to the general community and feedback to museum data providers. This paper is a case study in the role, use and justification of a genetic algorithm in development of large-scale environmental informatics infrastructure.