THE ROLE OF INTERNAL WATER STORAGE DESIGN ON NITROGEN FATE

Abstract

Green stormwater infrastructure (GSI) is implemented in urban landscapes to manage stormwater quantity and quality. Bioretention is an infiltration-based GSI strategy and demonstrates variable performance for total nitrogen (TN) removal. Internal water storage (IWS) is a sub-grade design feature that uses an underdrain with an elevated outlet to force a submerged layer. When a carbon source is present, often woodchips, IWS facilitates denitrification—the microbial reduction of nitrate (NO3-) to nitrogen gas (N2). This work considers the impact of IWS underdrain configuration, geometry, and IWS media on hydraulics, TN, and NO3- removal to enhance IWS design. To explore the impact of underdrain height, three laboratory columns with underdrains located at the bottom (0 cm), middle (15 cm), and top (30 cm) of a gravel-woodchip IWS were coupled with USGS VS2DRTI simulations. For narrow IWS geometries, width to depth (w/d) ratio < 1, hydraulic efficiency (ev) decreased from 1.0 to 0.76 as underdrain height increased from the bottom (0 cm) to top (30 cm). Changes in ev were attributed to the presence of immobile (low flow) zones below raised underdrains that limited solute transport. The presence of immobile zones impacted NO3- removal efficiency which decreased from 63% (bottom underdrain) to 32% (top underdrain) for a hydraulic loading rate (HLR) of 2.5 cm/h. However, simulated scenarios beyond the lab scale revealed ev varied less than 10% for IWS w/d ratios > 1 and indicated flow dynamics observed for narrow columns do not always translate to wider field systems. Under transient flow conditions, minimizing effluent NO¬3- concentrations and loads ranked least to greatest in the order bottom > middle > top underdrain configurations and dual isotopes in NO3- confirmed the presence of denitrification in mobile zones. Laboratory columns with bottom underdrain configurations considered three IWS media compositions of gravel, gravel-woodchip, and gravel-woodchip-biochar. Synthetic stormwater was modified to include dissolved organic carbon (DOC), dissolved organic nitrogen (DON), and trace organic nitrogen compounds. Under continuous flow conditions, NO3- removal efficiency ranked in the order gravel-woodchip (78%) > gravel-woodchip-biochar (61%) > gravel (-10%) for a HLR of 2.5 cm/h. During antecedent dry periods, the gravel-woodchip-biochar and gravel-woodchip IWS removed NO3- within 18 hours following a transient event. However, the presence of biochar resulted in ammonium (NH4+) generation and effluent concentrations exceeded levels toxic to aquatic life. High-frequency field monitoring of an IWS with a raised underdrain was performed for eight storms over ten months. IWS nitrogen concentrations during storm events revealed that peak TN concentration generally occurred within the first hour during the rising limb of the IWS water level and that TN was likely exported from the system in the form of DON and NO3-. Additionally, NH4+ washout from unsaturated soil occurred during February through May and was attributed to sodium dispersion due to road salt application. This work coupled laboratory columns, modeling, and field studies to address the complexities of nitrogen management in bioretention as impacted by IWS underdrain height, geometry, ev, media selection, absorbent amendments, and seasonal patterns. When approaching IWS design for water quality enhancements, practitioners are encouraged to consider all these variables but recognize that the desired TN removal will not be achieved in some cases.