Distinguishing sources of N2O in European grasslands by stable isotope analysis.

Nitrifiers and denitrifiers are the main producers of the greenhouse gas nitrous oxide (N(2)O). Knowledge of the respective contributions of each of these microbial groups to N(2)O production is a prerequisite for the development of effective mitigation strategies for N(2)O. Often, the differentiation is made by the use of inhibitors. Measurements of the natural abundance of the stable isotopes of N and O in N(2)O have been suggested as an alternative for the often unreliable inhibition studies. Here, we tested the natural abundance incubation method developed by Tilsner et al.1 with soils from four European grasslands differing in long-term management practices. Emission rates of N(2)O and stable isotope natural abundance of N(2)O and mineral N were measured in four different soil incubations: a control with 60% water-filled pore space (WFPS), a treatment with 60% WFPS and added ammonium (NH(4) (+)) to support nitrifiers, a control with 80% WFPS and a treatment with 80% WFPS and added nitrate (NO(3) (-)) to support denitrifiers. Decreases in NH(4) (+) concentrations, linked with relative (15)N-enrichment of residual NH(4) (+) and production of (15)N-depleted NO(3) (-), showed that nitrification was the main process for mineral N conversions. The N(2)O production, however, was generally dominated by reduction processes, as indicated by the up to 20 times larger N(2)O production under conditions favouring denitrification than under conditions favouring nitrification. Interestingly, the N(2)O concentration in the incubation atmospheres often levelled off or even decreased, accompanied by increases in delta(15)N and delta(18)O values of N(2)O. This points to uptake and further reduction of N(2)O to N(2), even under conditions with small concentrations of N(2)O in the atmosphere. The measurements of the natural abundances of (15)N and (18)O proved to be a valuable integral part of the natural abundance incubation method. Without these measurements, nitrification would not have been identified as essential for mineral N conversions and N(2)O consumption could not have been detected. Copyright 2004 John Wiley & Sons, Ltd.