Mitigating the Effects of Bridge Deck Runoff: A Case Study Using Bioretention and a Bioswale

Stormwater runoff from roadways is a major source of surface water pollution in North Carolina. The North Carolina Department of Transportation (NCDOT) is tasked with implementing stormwater BMPs alongside linear highways. NCDOT has specific interest in runoff from bridge decks, which is often discharged through drainage holes in the deck directly to the stream below. NCDOT needs to determine optimal retrofit stormwater BMPs associated with bridge decks. Two bioretention cells and a bioswale were constructed in the easement of a bridge deck on I-540 at Mango Creek. One bioretention cell was adequately sized based on current North Carolina design guidance, while the other was undersized by one-half. Undersized bioretention cells could often be used in retrofit situations; therefore, it is important to understand how an undersized bioretention cell performs with respect to hydrology and water quality. Both bioretention cells employed 0.9 m (3 ft) of fill media, and had an internal water storage layer (IWS) of 0.6 m (2 ft). The swale was designed to convey the 2-year storm event without overtopping and had a surface area to length ratio of 130 m²/m. Runoff was piped from the northbound and southbound lanes to the bioretention cells and swale, respectively. Data collection began in October 2009. Weirs and stage recorders were used to monitor inflow to and outflow from each BMP. Flow-proportional, composite water quality samples were obtained at the inlet and outlet of each BMP. Monitored water quality parameters included TKN, NO_2–3-N, NH₄-N, TN, TP, and TSS. In terms of TN, TP, and TSS concentrations, bridge deck runoff quality at Mango Creek was at least 50% cleaner than at other bridge decks studied in North Carolina, Louisiana, and China. The undersized and standard bioretention cells produced similar effluent concentrations for nutrients and TSS; however, the standard bioretention cell reduced runoff volumes from events less than 2.5cm to a much greater extent (86% versus 49%). The bioswale did not reduce runoff volumes, and was ineffective in reducing pollutant concentrations.