Effect of antecedent soil moisture conditions on emissions and isotopologue distribution of N2O during denitrification

The present study determined the influence of initial moisture conditions on the production and consumption of nitrous oxide (N2O) during denitrification and on the isotopic fingerprint of soil-emitted N2O. Sieved arable soil was pre-incubated at two different moisture contents: pre-wet at 75% and pre-dry at 20% water-filled pore space. After wetting to 90% water-filled pore space the soils were amended with glucose (400 kg C ha(-1)) and KNO3 (80 kg N ha(-1)) and incubated for 10 days under a He/O-2-atmosphere. Antecedent moisture conditions affected denitrification. N-2 + N2O fluxes and the N2O-to-N-2 ratio were higher in soils which were pre-incubated under dry conditions, probably because mobilization of organic C during the pre-treatment enhanced denitrification. Gaseous N fluxes showed similar time patterns of production and reduction of N2O in both treatments, where N2O fluxes were initially increasing and maximised 3-4 days after fertilizer application, and N2 fluxes were delayed by 1-2 days. Time courses of delta N-15(bulk)-N2O and delta O-18-N2O exhibited in both treatments increasing trends until maximum N-2 fluxes occurred, reflecting isotope fractionation during intense NO3- reduction. Later this trend slowed down in the pre-dry treatment, while delta O-18-N2O was constant and delta N-15(bulk)-N2O decreased in the pre-wet treatment. We explain these time patterns by non-homogenous distribution of NO3- and denitrification activity, resulting from application of NO3- and glucose to the surface of the soil. We assume that several process zones were thus created, which affected differently the isotopic signature of N2O and the N2O and N-2 fluxes during the different stages of the process. We modelled the delta N-15(bulk)-N2O using process rates and associated fractionation factors for the pre-treated soils, which confirmed our hypothesis. The site preference (SP) initially decreased while N2O reduction was absent, which we could not explain by the N-flux pattern. During the subsequent increase in N-2 flux, SP and delta O-18-N2O increased concurrently, confirming that this isotope pattern is indicative for N2O reduction to N-2. The possible effect of the antecedent moisture conditions of the soil on N2O emissions was shown to be important. (C) 2010 Elsevier Ltd. All rights reserved.