Influence of hydrologic pulses, flooding frequency, and vegetation on nitrous oxide emissions from created riparian marshes

The effects of hydrologic conditions, water quality gradients, and vegetation on nitrous oxide gaseous emissions were investigated in two identical 1-ha surface-flow created riverine wetlands in Columbus, Ohio, USA. For two years, both wetlands experienced seasonal (winter-spring) controlled hydrologic flood pulses followed by one year in which they received a steady flow rate of water. Nitrous oxide fluxes were quantified in a transverse gradient at different elevations (edge plots and high marsh plots with alternate wet and dry conditions, and low marsh plots and open water plots that were permanently flooded). The highest average of N2O fluxes was observed in high marsh plots (21.8 2.5 mu g-N m(-2) h(-1)), followed by edge plots (12.6 +/- 2.5 mu g-N m(-2) h(-1)), open water plots (9.9 +/- 2.1 mu g-N m(-2) h(-1)), and low marsh plots (7.0 +/- 4.8 mu g-N m(-2) h(-1)). Highest nitrous oxide fluxes were consistently observed in high marsh plots during summer when soil temperatures were >= 20 degrees C. In permanently flooded plots without vegetation, nitrous oxide fluxes were low, regardless of flood-pulse conditions. In high marsh plots, water table remained near the soil surface one week after flooding, causing an increase in N2O fluxes (25.9 +/- 13.9 mu g-N m(-2) h(-1)) compared with fluxes before (2.4 +/- 6.4 2.2 mu g-N m(-2) h(-1)) and during (6.9 +/- 2.2 mu g-N m(-2) h(-1)) flooding. In edge plots, nitrous oxide emissions increased during and after the flooding (11.3 +/- 3.2 and 7.3 +/- 3.3 mu g-N m(-2) h(-1)) compared with fluxes before the flood pulse (4.1 +/- 1.8 mu g-N m(-2) h(-1)). In low marsh and edge zones, no significant (P > 0.05) differences were observed in the seasonal N2O fluxes in the pulsing year versus steady-flow year. Spring N2O fluxes from high marsh plots were significantly (P=0.04) higher under steady-flow conditions (26.2 +/- 5.5 mu g-N m(-2) h(-1)) than under pulsing conditions (9.6 +/- 3.6 mu g-N m(-2) h(-1)), probably due to the water table near the surface that prevailed in those plots under steady flow condition. N2O fluxes were higher in plots with vegetation (39.6 +/- 13.7 mu g-N m(-2) h(-1)) than in plots without vegetation (-3.6 +/- 13.7 mu g-N m(-2) h(-1)) when plots were inundated; however, when no surface water was present, N2O fluxes were similar in plots with and without vegetation. Implications for large-scale wetland creation and restoration in the Mississippi River Basin and elsewhere for controlling nitrogen are discussed.