Speaker(s)
John Harte
John Harte is a physicist turned ecologist. His research interests span ecological field research, the theory of complex systems, and policy analysis. Current interests include applying insights from information theory to the analysis of complex ecosystems and empirical investigation of climate-ecosystem feedback dynamics. He is currently a Professor of the Graduate School in the Energy and Resources Group at UC Berkeley.
UC Berkeley Professor John Harte presented this talk as part of the BIDS Data Science Lectures series, at BIDS on Thursday, March 14, 2019.
Constrained maximization of information entropy yields least biased probability distributions and provides a framework for construction of complex systems theory. From physics to economics, from forensics to medicine, this powerful inference method has enriched science. Here I explain this method, apply it to ecology, and show that it predicts the detailed shapes of numerous patterns in nature that are of interest to ecologists. In relatively undisturbed ecosystems the theory works remarkably well, but a systematic pattern of failure is observed for ecosystems either losing species following anthropogenic disturbance or diversifying in historical time. An approach to extending the theory to rapidly changing systems will be sketched.
