Statistical Strategies for Environmental Modeling and Monitoring

Case Study

clouds

Challenges

With increased pressure from the political spectrum, the importance of statistics in environmental science is undisputed. The planning of experiments, sampling strategies and analysis of highly complex data are fundamental to carrying out environmental policies and research. There is increasing complexity of the processes and models being studied and a need to supply reliable assessments of risk and uncertainties.  There are many new issues and problems requiring innovative strategies and new methods. And the public needs to understand that there is a non-biased, third party who is analyzing the data. With environmental statistics becoming more visible in the news, it has become more politicized, which makes it harder for the public to both understand and interpret all the different results that are being published. The results of one's statistical analyses must be made understandable to scientists, policy makers and the public.

Outcomes & Results

NISS is able to provide a multi-disciplinary team together to look at these issues. NISS identifies and develops methodology for the long-run, complex cross-disciplinary issues. It also engages young researchers in the problems and communicated the results to scientists, policy-makers and the public.

NISS was hired by the U.S. Environmental Protection Agency to develop space-time models that would help estimate the detection and trend of exceedances in air pollution. NISS also looked at the health and pharmacokinetics using predictive models of dose and effect and helped with risk assessment of toxicants, looking at complex, multi-dimensional relations between concentrations of toxicants and outcomes.


Research Project

The Environmental Protection Agency and scientists from several universities across the United States and Canada worked on network design for environmental monitoring; space-time models, data fusion, health and pharmacokinetics; risk assesment of toxicants; and data representation and reporting.

Research Team: 

Principal Investigator(s): Peter Bloomfield, North Carolina State University; Doug Nychka,

Senior Investigators: Jerry Sacks, NISS; Ross Leadbetter and Richard Smith, University of North Carolina at Chapel Hill, Shao-Hang Chu, EPA; Thomas Curran, EPA, William Cox, EPA;  David Holland, EPA

Post Doctoral Fellow(s): Feng Gao, Patricia Styer