Comparative Evaluation of Statistical and Mechanistic Models of Escherichia coli at Beaches in Southern Lake Michigan

• DOI: 10.1021/acs.est.5b05378
• Publication Date: 2016-01-30T01:28:52Z
• Authors: Ammar Safaie; Aaron Wendzel; Zhongfu Ge; Meredith B. Nevers; Richard L. Whitman; Steven R. Corsi; Mantha S. Phanikumar
• Keywords: Lakes; Michigan; Models, Statistical; Geography; 13. Climate action; Escherichia coli; 0207 environmental engineering; 02 engineering and technology; 14. Life underwater; Models, Theoretical; Water Microbiology; Bathing Beaches; 6. Clean water

Statistical and mechanistic models are popular tools for predicting the levels of indicator bacteria at recreational beaches. Researchers tend to use one class model or other, it is difficult generalize statements about their relative performance due differences in how developed, tested, used. We describe a cooperative modeling approach freshwater beaches impacted by point sources which insights derived from were used further improve statistical vice versa. The provided basis assessing improved using probability distributions generate high-resolution time series data source, long-term "tracer" transport based on observed electrical conductivity, better assimilation meteorological data, unstructured-grids resolve nearshore features. This resulted comparable both classes including parsimonious suitable real-time predictions an easily measurable environmental variable (turbidity). outlined here can be other sites has potential water quality resulting more accurate estimates beach closures.