Advancing and Applying a Maximum Entropy Theory of Ecology

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

Understanding patterns in the spatial distribution, abundance, energetics, and network structure of species are central goals of ecology. Progress toward the goal of pattern prediction has been achieved recently for relatively pristine ecosystems with development of ecological theory based on a rigorous inference procedure, the method of maximum entropy, that originated from pioneering work in information theory half a century ago. This project will advance the existing Maximum Entropy Theory of Ecology by extending its taxonomic range and its ability to predict the regional spatial structure of populations, and by characterizing and understanding its predictive performance in highly disturbed ecosystems. The resulting predictive ability is key to formulating practical local, national, and global conservation and land use policies. Hence, a high priority of this research will be applying the theory to vexing conservation problems, such as improving estimation of species loss under habitat loss and climate change, improving the efficiency and accuracy of biodiversity census strategies, and improving estimation of the biological richness of biomes too large to census directly. Free, user-friendly software packages will be developed, allowing non-academic managers to apply maximum entropy methods to spatially explicit conservation questions. There will be training opportunities for women and underrepresented minorities in the quantitative sciences, and for students seeking to bridge the physical and biological sciences.