Storm Water Low-Impact Development, Conventional Structural, and Manufactured Treatment Strategies for Parking Lot Runoff

Abstract

Eleven storm water treatment strategies were evaluated for water quality performance and storm volume reduction during rainfall–runoff events between September 2004 and August 2005. Evaluated treatment strategies included structural best management practices (BMPs) (swales, retention ponds), low-impact development (LID) designs (treatment wetland, filtration and infiltration designs), and manufactured BMPs (filtration, infiltration, and hydrodynamic separators). Contaminant event mean concentration, performance efficiency, and mass-based first flush were evaluated for storms with varying rainfall–runoff characteristics. Previous research demonstrated that treatment performance of storm water control measures varies widely in response to site-specific contaminant loading functions. For that reason, the devices were tested in parallel, with a single influent source providing uniform loading to all devices. Treatment strategies were uniformly sized to target a rainfall–runoff depth equivalent to 90% of the annual volume of rainfall. Under the parallel and uniformly sized configuration, a normalized performance evaluation is possible because treatment strategies of the same scale receive runoff from events of the same duration, intensity, peak flow, volume, antecedent dry period, and watershed loading. Runoff constituent analyses included total suspended solids (TSS), total petroleum hydrocarbons-diesel, dissolved inorganic nitrogen, and total zinc. Several water quality parameters (temperature, dissolved oxygen, pH, conductivity) were monitored as real-time data. Performance evaluations indicate that several LID designs have removal efficiencies of 80% to 100%. In contrast, conventional structural BMPs perform poorly for most measures except for the pond with TSS. The manufactured systems tended to vary widely and were dependent on the design and contaminant of interest.