The Utility of Bayesian Decision Analysis and Environmental Problems
Organizer: Paul Black
(pblack@neptuneandco.com)
Neptune and Company, Inc.

Description:
Bayesian decision analysis is touted as a unified framework for coherent decision making. Two key components of Bayesian decision analysis are probability distributions and utility functions. Calculating posterior distributions and performing decision analysis can be computationally challenging, especially for complex environmental models. In addition, probability distributions and utility functions for environmental models must be specified through expert elicitation, stakeholder consensus, or data collection, all of which have their own set of technical and political challenges. Nevertheless, a grand appeal of the Bayesian approach for environmental decision making is the explicit treatment of uncertainty, including expert judgment. The impact of expert judgment on the environmental decision process, though integral, goes largely unassessed.

A traditional tool used to aid environmental decision-making is deterministic process modeling that attempts to reproduce the behavior of environmental systems. Expert judgment typically drives the level of model complexity chosen but model complexity choices are often not made within the context of the decision to be made. As the computational burden of environmental modeling is continually reduced probabilistic process modeling using Monte Carlo simulation is becoming routinely used to propagate uncertainty from model inputs through model predictions. However, the impact of model uncertainty is largely ignored due to conceptual and computational difficulties.

The next step in this evolution is placing this probabilistic modeling in a Bayesian framework that merges observational data and process modeling and allows model updating in a statistically rigorous manner as more data is collected. Ultimately the full power of the Bayesian approach in environmental decision making is realized as utility functions are explicitly merged with expert judgment, data, and probabilistic modeling.

Format:
This session will focus on the current state of the attempt to merge environmental process modeling and Bayesian decision analysis. Talks will highlight approaches for assessing different sources of decision uncertainty in process modeling emphasizing the interplay between model complexity and the decision process. Focus will also be placed on quantitatively incorporating expert judgment into the decision process.

Participants:
David Draper (presentation, Scenario and Parametric Uncertainty in GESAMAC: A Methodological Study in Nuclear Waste Disposal Risk Assessment)
David Draper is a Professor of Statistics in the School of Mathematical Sciences at the University of Bath in England. David did his Ph.D. work at the University of California, Berkley, finishing in 1981 and has since taught and done consulting and public policy research at the University of Chicago (1981-84), the RAND Corporation (1984-91), UCLA (1991-93), the University of Bath (1993-Present), with a sabbatical visit to the University of Washington in 1986. He is a fellow of the Royal Statistical Society and a member of both the IMSV and ASA. David has served as Associate Editor for the Journal of the American Statistical Association and the Journal of the Royal Statistical Society, and is the author or coauthor of four books and 46 articles and other substantial contributions to refereed journals.

Kenneth H. Reckhow and Mark E. Borsuk (presentation, A Probability Network for Water Quality Modeling and Decision Support)
Kenneth H. Reckhow received a B.S. in engineering physics from Cornell University in 1971 and a Ph.D. from Harvard University in environmental systems analysis in 1977. He is currently a professor at Duke University with faculty appointments in the School of the Environment, the Department of Civil and Environmental Engineering, and the Institute of Statistics and Decision Sciences. In addition, he is director of The University of North Carolina Water Resources Research Institute and an adjunct professor in the Department of Civil Engineering at North Carolina State University. He has published two books and over 80 papers, principally on water quality modeling, monitoring, and pollutant loading analysis.

Samantha Bates and Adrian Raftery (presentation, Bayesian Assessment of Uncertainty and Variability in Deterministic Environmental Exposure Models)
Samantha Bates graduated from the University of Auckland in Auckland, New Zealand with an Honors degree in Statistics in 1996 and arrived in Seattle, Washington that same year. She completed a M.S. in Statistics at the University of Washington (UW) in 1998, Currently she is a Ph.D. candidate in the Statistics department at the University of Washington working with Adrian Raftery.

Tom Stockton and Paul Black (presentation, Environmental Modeling and Bayesian Analysis for Assessing Human Health Impacts from Radioactive Contamination)
Tom Stockton graduated from the Duke University School of the Environment, with a Ph.D. in Environmental Modeling in 1998, and a M.S. in Environmental Management in 1985. He has performed water quality and water shed modeling for the States of North Carolina and Maryland since 1985. Currently, he is employed by Neptune and Company, Inc., where he is applying Bayesian techniques to modeling of environmental problems.