Anna Cressman - Final Oral Defense - 4/18/06

Decision support systems (DSS) are often employed in complex environmental problems to provide the user with an integrated solution approach and can range from purely qualitative to quantitative modeling to fully automated. This paper presents a semi-quantitative DSS approach for the problem of microbial loading in surface waters due to stormwater runoff. Six important stormwater variables are identified within the Neuse River Basin, N.C. including swine density, soil permeability, slope, highly developed land area, average annual precipitation, and average stream flow velocity. Linear regression is used to assess the correlation of each variable with measured fecal coliform levels in surface waters. The resultant DSS provides a relative ranking of each variable based on its predictive value and a scaling based on its dynamic range. The DSS presented provides a less complex, less resource intensive approach to the problem posed than other available stormwater modeling programs, but also provides a more reliable estimate of risk than a purely qualitative DSS. Policy makers can employ this method when structuring their approach to microbial loading decisions.

Committee:

Advisor: Douglas Crawford-Brown
Readers: Frederick K. Pfaender
Philip R. Berke (DCRP)

For further information please contact Rebecca Riggsbee Lloyd by email at Rebecca_Lloyd@unc.edu