Evaluation of nitrogen removal and fate within a bioinfiltration stormwater control measure

The focus of this research is the nitrogen removal achieved in a bioinfiltration stormwater control measures (SCM). SCMs are engineered systems which use environmental approaches such as biological, chemical, and physical processes to treat stormwater for water quality while also reducing stormwater volumes and peak flows from surface water. A bioinfiltration SCM on Villanova University's campus in a residential student area was constructed in 2001 as a parking lot median retrofit to treat approximately a 2.54 cm storm event (1.27 cm as media storage, 1.27 cm as ponding). This research comprehensively examined nitrogen removal, which is one of the contaminants present in stormwater runoff. Both surface and subsurface stormwater samples collected throughout the site during the past nine years were tested for the following nitrogen species: Total Nitrogen (TN), Total Kljedahl Nitrogen (TKN), Nitrite (NO2 --N), Nitrate, (NO3--N), and NOx-N (NO2--N + NO3--N). The subsurface pore water samples were collected using lysimeters at the surface of the bioinfiltration media (LYS 0), at the bottom of the 1.2 m deep infiltration media (LYS 4), and an additional 1.2 m below the infiltration media (LYS 8). In addition to samples collected within the bioinfiltration system, surrounding in situ soil water and well water samples were tested for nitrogen to compare the engineered system to natural background conditions in removing nitrogen. Stormwater samples were analyzed from inflow to outflow for concentrations and loads. Discharge for the system occurred as infiltration into in situ soil; any overflow left as outflow from a converted storm drain.;From the 364 storms analyzed for hydrology, influent stormwater volume was significantly reduced through the bioinfiltration SCM. The bioinfiltration system captured 100% of storm events less than 1.27 cm, 97% of storms from 1.27 cm – 2.54 cm, and 50% of storms greater than 2.54 cm. Ponding durations measured from the initiation of rainfall averaged 32 hours that coupled with a recession soil rate of 0.5 cm/hr created longer saturated zones within the media.;TN, TKN, and NOx-N concentrations were significantly reduced from the top of the infiltration media to the bottom of the media (34%, 47%, and 53% for TN, TKN, and NOx-N, respectively), indicating that nitrogen removal occurred within the bioinfiltration media. No seasonal trends were observed for NO2--N or NO3 --N in the ponded sample and at the bottom of the media. Deeper lysimeters in native soil at 0.9 m and 1.8 m depths had comparable concentrations for TKN but significantly higher concentrations for NOx-N. NO 3--N comparisons between LYS 8 and surrounding groundwater wells indicated that the groundwater samples were significantly higher than the water that left the bottom of the bioinfiltration SCM and infiltrated 1.2 m below the bottom of the system. Nitrogen loads were statistically significantly reduced from surface water through the bioinfiltration media. Median load reductions between what entered the bioinfiltration SCM and what overflowed were 100% for TN, TKN, NOx-N, NO2--N, and NO3--N. Mass reductions via infiltration accounted for the main pollutant removal mechanism. Treatment for TKN, NO2 --N, and NO3--N loads occurred within the bioinfiltration media; therefore, further reducing pollutant loads from the infiltrating stormwater before reaching groundwater.;This SCM effectively mitigates stormwater runoff for volume and nitrogen removal. Overall, the stormwater that infiltrated through bioinfiltration SCM media had lower nitrogen concentrations than background lysimeter samples at deeper depths. Additional research is needed to determine the nitrogen removal mechanisms responsible for the loss of TKN and NOx-N by the bioinfiltration SCM outlined in this study. In summary, the bioinfiltration SCM provided significant volume reductions and water quality treatment.