Effects of storm size and frequency on nitrogen retention, denitrification, and N2O production in bioretention swale mesocosms

Abstract

Reported nitrogen (N) retention efficiencies for bioretention swales vary widely, but reasons for this are not well-understood, in part because almost no studies have measured (or characterized controls on) bioretention swale denitrification. Here, we apply a novel N2:Ar-based approach, in coordination with more established approaches, to estimate denitrification rates and compare bioretention N dynamics during artificial storms of two sizes (3.05 and 5.08 cm days−1) and following 4 inter-storm periods (initial storm with no prior storm, 1-, 7-, and 13-days). Denitrification rates during storms occurring after 7-days (520 ± 150 µmol N m−2 h−1) were significantly higher than those during an initialization storm (13 ± 34 µmol N m−2 h−1) or during a storm occurring one day after a previous storm (−63 ± 65 µmol N m−2 h−1). No significant differences in N processing were observed between 3.05 and 5.08 cm days−1 storms. Somewhat surprisingly, in all experiments [O2] remained near saturated, and N2O emissions were very low or undetectable. Mesocosms were largely a net sink for dissolved inorganic N (DIN) and a net source of dissolved organic N (DON). Denitrification was neither a dominant nor consistent pathway for N removal, accounting for a maximum of 23 ± 11% of DIN removal. Future research should continue to evaluate N assimilation as a N removal pathway in bioretention swales, as well as characterize N dynamics during unsaturated conditions associated with smaller rain events and during periods between the large storms examined here.