Production and transport of denitrification gases in shallow ground water

Abstract

The contribution of potentially intense denitrification in the saturated zone of hydromorphic soils to atmospheric N2O levels is poorly understood because few data exist on shallow ground water N2O production, consumption and transport to the atmosphere. The objective of the present study was to investigate the contribution of the saturated zone to surface N2O emission for two fen soils and a Gleyic Luvisol with ground water tables at the surface during the experimental period. Total denitrification, denitrifier N2O production, ground water dissolved N2 and/or N2O, and surface N2O emissions were measured in situ. Concentrations of dissolved gases and surface emissions were also simulated with a simple process-based model assuming measured rates of N2and N2O production or constant profiles of dissolved N2O concentration. NO3 − and N2O were abundant in all samples of the three sites. Substantial N2O surface emission originating from the saturated zone was measured at both fen sites. Total denitrification ranged from 0.9 to 1.58 mg N l−1 day−1 in the 15 – 35-cm layer of the fen soils and from 0.005 to 0.13 mg N l−1 day−1 in the 50 – 100-cm layer of the Gleyic Luvisol. The ratio of N2O production to total denitrification ranged from 0.07 to 0.32 in the fen soils and from 0.06 to 0.08 in the Gleyic Luvisol. The ratio of N2O production to total denitrification ranged from 0.07 to 0.32, and from 0.06 to 0.08, respectively. Concentrations of dissolved N2O measured at the fen sites were 1 to 2 orders of magnitude lower than simulated concentrations. In contrast, measured N2O emissions were within the order of magnitude of emissions simulated assuming measured N2 and N2O production rates. The agreement between measured and simulated N2O concentrations and surface emissions was satisfactory when gas diffusivity was multiplied by 10 and N2O reduction rate was multiplied by 20 in the simulation. Simulation of diffusive N2O emission assuming constant values of measured N2O concentration profiles resulted in emissions approximately one order of magnitude lower than the measured values. At the Lake Creek site, the measured peak concentration of dissolved N2 and the concentration simulated assuming measured values of N2 production were relatively close with values of 2.4 and 3.5 times the atmospheric equilibrium concentration, respectively. It was concluded that the disagreements between measured and simulated values of N2O concentrations and emissions resulted from the underestimation of model parameters for gas transfer and/or for N2O reduction to N2,which were not measured. Future modeling attempts should include use of measured values for all model parameters to obtain a more realistic description of the dynamics of N2O emission from the saturated zone.