DENITRIFICATION VARIABILITY AND CONTROL IN A RIPARIAN FEN IRRIGATED WITH AGRICULTURAL DRAINAGE WATER

Denitrification was measured by the C2H2 inhibition technique in a riparian fen irrigated with agricultural drainage water. 16 h after C2H2 treatment 88 +/- 14% of the total N2O contained in water-saturated cores could be accounted for by assuming equilibrium between the gas phase and the liquid phase. The denitrification activity averaged 2.8 and 8.8 mg N2O-N m-2 day-1 in the control plot and 1.6 and 21.9mg N2O-N m-2 day-1 in the irrigated plot during the dry and the runoff periods respectively. Four percent of the incoming NO3- was reduced to gaseous N. The spatial variability was often high, with coefficients of variation > 100% and was independent of seasonal changes in soil anaerobiosis. Soil NO3- and denitrification were poorly related, and bulk concentrations of NO3- below 200 mum suggested that the process was strongly limited by diffusion of NO3- into the soil during periods of flooding. Mean denitrification and water-filled pores correlated positively, r = 0.71*** for the control and r = 0.68*** for the irrigated plots. Water-soluble C was not related to denitrification. Multiple regression models including soil water, NO3, soluble C and temperature as independent variables, predicted between 21 and 55% of the denitrification, the highest value found when only mean data was considered. Water-filled pores was the most important variable. The observations on which 2 variables controlled denitrification were supported by laboratory experiments with manipulated cores. Water additions increased denitrification only in samples collected during the dry period. Anaerobic incubation of saturated cores did not affect the process. Restricted NO3- availability was clearly illustrated by the 25-41-fold increase obtained when NO3- was injected into cores at ambient and high carbon respectively. A response of up to 13-fold was observed when substrate-amended cores were made into slurries. Glucose did not increase denitrification by more than a factor of three.