Assessing the Effects of Multiple Stressors in Environmental Monitoring Programs

EPA Grant Number: R827953
Title: Assessing the Effects of Multiple Stressors in Environmental Monitoring Programs
Investigators: Smith, Eric
Current Investigators: Smith, Eric , Ye, Keying
Institution: Virginia Polytechnic Institute and State University
EPA Project Officer: Louie, Nica
Project Period: October 1, 1999 through September 30, 2002
Project Amount: $266,388
RFA: Environmental Statistics (1999) RFA Text |  Recipients Lists
Research Category: Environmental Statistics , Health , Ecosystems


Major goals of environmental data analysis are the assessment of the condition of the environment and the relationship between condition and potential stressors. Environmental stress occurs at different temporal and spatial scales.


Methods used to relate measures of environmental health and environmental information are typically based on regression or correlation analysis and include methods such as multiple regression analysis, canonical correlation analysis and canonical correspondence analysis. Typical data sets are large in terms of the number of variables used in the modeling process and it may be quite difficult to select amongst different competing models. This research will investigate the use of Bayesian analysis as a set of methods for interpreting the importance of relationships between stressors and environmental variables and for prediction under this model uncertainty. The approaches that will be used is based on Bayesian model averaging and Bayesian hierarchical modeling. The research will use multiple regression analysis to evaluate relationships between biological measures of health and chemical, physical, habitat and land use variables. Bayesian model averaging will be used to assess variable importance and to make inferences and predictions. Models based on canonical correspondence analysis and canonical discriminant analysis are commonly used in the assessment of environmental data. The research will also extend Bayesian model averaging to applications involving canonical correlation analysis, canonical correspondence analysis and canonical discriminant analysis.

Expected Results:

The results of this research win aid ecological and environmental researchers trying to understand relationships between ecological and environmental measures of health and factors that influence these measures.

Supplemental Keywords:

principal components analysis, correspondence analysis, canonical variate analysis, canonical correlation analysis, redundancy analysis, canonical correspondence analysis, stressor effects, ecological statistics., RFA, Scientific Discipline, Economic, Social, & Behavioral Science Research Program, Ecosystem Protection/Environmental Exposure & Risk, Applied Math & Statistics, Ecology, Ecosystem/Assessment/Indicators, Ecosystem Protection, Mathematics, Ecological Effects - Environmental Exposure & Risk, Ecological Effects - Human Health, Ecological Risk Assessment, Ecology and Ecosystems, Environmental Statistics, Ecological Indicators, ecological exposure, risk assessment, co-pollutant effects, environmental stressor, aggregation, computer models, environmental risks, Bayesian method, correlative analysis, modeling, environmental monitoring networks, multiple stressors, ecological statistics, statistical models, regression analysis, data analysis, canonical correlative analysis, data models, innovative statistical models

Progress and Final Reports:

  • 2000 Progress Report
  • 2001
  • Final