Bayesian Methods for Regional-Scale Stressor-Response Models

EPA Grant Number: R830887
Title: Bayesian Methods for Regional-Scale Stressor-Response Models
Investigators: Lamon, E. Conrad
Institution: Louisiana State University - Baton Rouge
EPA Project Officer: Hahn, Intaek
Project Period: May 14, 2003 through April 13, 2006 (Extended to April 13, 2008)
Project Amount: $389,168
RFA: Developing Regional-Scale Stressor-Response Models for Use in Environmental Decision-making (2002) RFA Text |  Recipients Lists
Research Category: Ecological Indicators/Assessment/Restoration , Ecosystems


This proposal seeks to use modern classification and regression trees and other Bayesian hierarchical techniques to link multiple environmental stressors to biological responses and quantify uncertainty in model predictions.


We propose a Bayesian Classification and Regression Tree (CART) approach that will: 1) report prediction uncertainty; 2) be consistent with the amount of data available; and 3) be flexible enough to permit updates and improvements. Composite data at national, regional and local scales will be used to develop a large, rich suite of environmental, chemical, physical, hydrologic and watershed characteristics to be used as potential predictor variables for eutrophication endpoints to identify and estimate regional stressor response models. Tree based methods are a flexible approach useful for variable subset selection and when the analyst suspects global nonlinearity and cannot (or does not want to) specify the functional form of possible interactions a priori. Model estimation will be accomplished using a training set of data randomly chosen from the full dataset assembled for this study. A validation, or test, dataset will be withheld from use in model estimation and used to evaluate "out of sample" predictive performance. Mean squared error (MSE), median absolute deviation (MAD) and cumulative log likelihood will be evaluated and reported for all models using both the training and test data.

Expected Results:

A systematic method for identification and estimation of regional scale stressor-response models in aquatic ecosystems will be useful in monitoring and assessment of aquatic resources, determination of TMDL's and for increased understanding of the differences between regions. The Bayesian approach offers many advantages from a decision theoretic point of view, including the estimation of the value of new information and proper probability distributions on the variable of interest as an output, which can be directly used in risk assessment or decision-making. Use of the decision theoretic framework to guide the modeling process will assure proper use of information regarding attributes that measure ecosystem value, and avoid the information loss associated with aggregation of attributes into indices.

Publications and Presentations:

Publications have been submitted on this project: View all 19 publications for this project

Journal Articles:

Journal Articles have been submitted on this project: View all 4 journal articles for this project

Supplemental Keywords:

ecological effects, EMAP, Markov Chain Monte Carlo methods, Metropolis-Hastings algorithm, NASQAN, National Eutrophication Survey, Nutrient Criteria Database, regionalization, statistics, water, watersheds., RFA, Economic, Social, & Behavioral Science Research Program, Scientific Discipline, Water, Ecosystem Protection/Environmental Exposure & Risk, Water & Watershed, Monitoring/Modeling, Regional/Scaling, decision-making, Ecology and Ecosystems, Watersheds, Economics & Decision Making, Social Science, risk assessment, ecosystem modeling, aquatic ecosystem, watershed, ecosystem assessment, Bayesian approach, decision analysis, decision making, environmental decision making, TMDL, ecological variation, regional scale impacts, water quality, assessment endpoint mechanistic research, ecology assessment models, ecosystem stress, watershed assessment, decision support tool, environmental risk assessment, Bayesian classifiers, water monitoring, adaptive implementation modeling

Progress and Final Reports:

  • 2003 Progress Report
  • 2004
  • 2005
  • 2006
  • Final Report