Dissimilatory nitrate reduction to ammonium (DNRA), not denitrification dominates nitrate reduction in subtropical pasture soils upon rewetting

摘要

Soils under pasture are subjected to repeated wetting and drying cycles in response to rainfall, irrigation and evapotranspiration. The amplitude of these cycles is likely to increase under the predicted changes in rainfall variability, demanding a better quantitative understanding of the processes involved. The wetting of pasture soils triggers large pulses of N2O emissions, predominantly produced via denitrification. Under anaerobic conditions in the soil matrix, denitrification and dissimilatory nitrate reduction to ammonia (DNRA) are thought to compete for available NO3-. However, the relationship between gross NO3-production and consumption via denitrification (N-2 and N2O) and DNRA remains poorly understood. This study combines the direct quantification of N-2 and N2O with a numerical N-15 tracing model to establish the relationship between denitrification and DNRA in pasture soils after wetting. Soil microcosms were fertilised with NH4NO3 (35 mu g N g(-1) soil) using a triple N-15 labelling approach, wetted to four different water -filled pore space (WFPS) levels and incubated over two days. The abrupt increase in soil moisture triggered a burst of N-2 and N2O emissions, with peak fluxes of N-2 & 13.1 mu g N g(-1) soil day(-1) at high soil moisture levels. At 95% and 80% WFPS, denitrification was dominated by N2 emissions, with the N-2/(N-2+N2O) ratio ranging from 0.5 to 0.9. At 60% and 40% WFPS, the N-2/ (N-2 + N2O) ratio ranged from 0.2 to 0.3, showing N2O as the main product of denitrification. The wetting of dry pasture soils resulted in increased DNRA rates across soils and WFPS. Both denitrification and DNRA increased exponentially with WFPS and responded to NO3- availability, demonstrating both processes as N-substrate driven. The labile C/NO3- ratio was not correlated to DNRA rates and as such did not explain NO3- partitioning between denitrification and DNRA, likely due to the high C availability in the pasture soils. Increasing labile C availability stimulated heterotrophic soil respiration, which had no effect on denitrification rates, but increased DNRA. Increased soil respiration is likely to have lowered the soil redox potential, promoting a shift of NO3- consumption from denitrification to DNRA, which implies the soil redox potential rather than the C/NO3- ratio as the key factor for NO(3)(-)partitioning between denitrification and DNRA in C rich pasture soils. Our findings suggest that the high labile C availability under perennial pastures, together with the increase of labile C upon rewetting, drives heterotrophic soil respiration, reduces the soil redox potential and ultimately shifts NO(3)(-)consumption from denitrification to DNRA. This shift limits denitrification losses and is therefore critical for limiting N loss and increasing N retention in subtropical pasture soils.