Hierarchical Bayesian Methods in Ecology (10w2170)


Devin Goodsman (University of Alberta (Ph.D candidate))

(Paris Dauphine University)

Francois Teste (University of Alberta)


The "Hierarchical Bayesian Methods in Ecology" workshop will be hosted at The Banff International Research Station.

Ecological systems are extraordinarily complex. For example, if a researcher wishes to study lodgepole pine susceptibility to mountain pine beetle, they can look at the question on a stand level. The question would then be which pine stands are more susceptible to mountain pine beetle? However, the researcher could then ask whether certain trees within a pine stand are more resistant to mountain pine beetle attack. This is a tree-level question. Mountain pine beetles are analogous to ferries with passengers: they carry a variety of nematode worms, mites, fungi and bacteria that may help or hinder their colonization of host trees. Therefore a curious scientist could then ask how all of these organisms interact and how that affects beetle populations within trees and within stands. The answer to such questions requires the synthesis of beetle level processes, tree level factors and stand level variables. An elegant and adaptable numerical analysis is required to capture and accommodate such complexity. rnrnBayesian statistics are gaining in popularity amongst mathematical biologists and ecologists. The appeal of Bayesian methods lies in the simple interpretation of the results and their ability to model the uncertainty that arises when researchers collect, analyze, and interact with complex data. However, the great adaptability and complexity that is possible with Bayesian methods requires the researcher to have a deep understanding of models, and distributions. Prominent mathematical ecologists advocate the use of Bayesian statistics. However, Bayesian statistics are not commonly taught and are difficult to self-teach. Respected French Statistician, Christian P. Robert will guide an intensive workshop that allows local ecologist to apply Bayesian methods to their own datasets. Dr. Robert is the author of numerous books on Bayesian statistics and works extensively with environmental scientists internationally. rn

The Banff International Research Station for Mathematical Innovation and Discovery (BIRS) is a collaborative Canada-US-Mexico venture that provides an environment for creative interaction as well as the exchange of ideas, knowledge, and methods within the Mathematical Sciences, with related disciplines and with industry. The research station is located at The Banff Centre in Alberta and is supported by Canada's Natural Science and Engineering Research Council (NSERC), the US National Science Foundation (NSF), Alberta's Advanced Education and Technology, and Mexico's Consejo Nacional de Ciencia y Tecnolog�255a (CONACYT).