Performance of Retention Ponds and Constructed Wetlands for Attenuating Bacterial Stressors

Microbial contamination from fecal origins in stormwater runoff poses a risk to human health through the consumption of drinking water and recreational and bathing contact with surface waters. Indicator bacteria serve as the regulatory meter by which water quality is measured and water quality standards (WQS) must be met. Research on constructed wetlands inactivation of fecal indicators in secondary and animal wastewater is well documented. Removals of fecal streptococci and coliforms generally exceeded 80% and 90%, respectively, in a review by Kadlec and Knight, 1996. Gersberg et al. and Garcia and Bécares concluded that extensively vegetated systems remove indicator bacteria at significantly higher rates from wastewater than unvegetated systems. However, because of the potentially high indicator bacteria concentrations in stormwater runoff, the untreated fraction in effluent from retention ponds and constructed wetlands may increase receiving water concentrations beyond WQS. This is in contrast to separate sanitary systems and combined stormwater and sanitary systems which, other than during sewer overflow conditions, chemically treat the wastewater routed to treatment plants. Experiments to evaluate the use of the first-order decay function for predicting indicator bacteria concentrations in effluent from best management practices (BMPs) were designed and completed by U.S. EPA's Urban Watershed Research Facility in Edison, NJ. Two studies, one at the bench-scale and the other at the pilot-scale, were completed to determine similarities and differences in inactivation rate constants, coefficients, and affects of environmental conditions on bacterial indicator concentrations. The focus of this paper is on the results of the pilot-scale studies which specifically explored the environmental factors that influence the rate of microbial inactivation as urban stormwater passes through retention ponds and constructed wetlands. The mesocosms designed and constructed for this project offered a unique setting allowing many characteristics associated with stormwater and flow to be held constant (i.e., influent characteristics, residence time, and pollutant loading). By varying testing dates with climatic conditions experienced throughout the year, an assessment of the impact of the environmental change on bacterial inactivation rates could be assessed. The results allowed comparison of rates of inactivation with seasonal wet-weather events and were used to determine new inactivation coefficients based on environmental variables. More information is needed to determine whether models that use first-order decay functions when predicting bacteria effluent concentrations from field BMPs (usually as a point source) are accurately providing effluent predictions and concomitant loads. Results suggest, depending on how bacterial loading is modeled during total maximum daily load (TMDL) development, that first-order decay may not adequately consider background concentrations when calculating dry-weather loading. Similarly, predicting loading over longer durations while ignoring seasonal changes may result in less accurate predictions of indicator bacteria loading. Both outcomes may impact actual bacterial loading and risk to human health with exposure to stormwater runoff contaminated water.