The mixing of a river into coastal waters at two beaches: Environmental factors, E. coli contributions and applications for predictive models.

摘要

Beach closures and public health protection are confounded by analytical procedures that result in delays in notification of adverse water quality conditions and the lack of affordable analytical methods to identify pollutant sources. Attempts have been made to develop predictive frameworks using ancillary hydrometeorological data to statistically anticipate deteriorated water quality. Many urban coastal beaches are impacted by river runoff. In Kenosha Wisconsin, beach sanitary survey data from two beaches adjacent to the mouth of the Pike River were examined to ascertain whether simple river-lake mixing models identified river influence on coastal water quality and improved predictions of beach advisories.;Water samples (798 water samples) were collected from the Pike River (one location) and Lake Michigan beach locations to the north (three locations) and south (four locations) of the inflow during the summer months of 2012-2014. Specific conductivity was used as a conservative tracer for quantifying river-lake mixing. Mixing was dependent upon distance from the river mouth, river discharge, and wind and alongshore current directions (p<0.05). A two component mixing model quantified coastal E. coli concentrations when river waters were the dominant pollution source (n=9, R2= 0.5773-0.9282), except near the mouth where groundwater exfiltration confounded mixing calculations (n=8, R2=0.1704). An ensemble model (predictive model which estimated river influence on coastal waters) more accurately predicted exceedances of water quality standards compared to traditional multiple linear regression models as measured by sensitivity (fraction of exceedances accurately predicted; 0.419 vs. 0.194), but with more false positives. Given the importance of external river borne sources of E. coli to coastal beaches, models and data which address riverine mixing under a variety of hydrometeorological conditions have the potential to improve predictions of water quality in nearby waters and therefore protect public health.