N 2 O source partitioning in soils using 15 N site preference values corrected for the N 2 O reduction effect

Rationale The aim of this study was to determine the impact of isotope fractionation associated with N 2 O reduction during soil denitrification on N 2 O site preference (SP) values and hence quantify the potential bias on SP‐based N 2 O source partitioning. Methods The N 2 O SP values ( n  = 431) were derived from six soil incubation studies in N 2 ‐free atmosphere, and determined by isotope ratio mass spectrometry (IRMS). The N 2 and N 2 O concentrations were measured directly by gas chromatography. Net isotope effects (NIE) during N 2 O reduction to N 2 were compensated for using three different approaches: a closed‐system model, an open‐system model and a dynamic apparent NIE function. The resulting SP values were used for N 2 O source partitioning based on a two end‐member isotopic mass balance. Results The average SP 0 value, i.e. the average SP values of N 2 O prior to N 2 O reduction, was recalculated with the closed‐system model, resulting in −2.6 ‰ (±9.5), while the open‐system model and the dynamic apparent NIE model gave average SP 0 values of 2.9 ‰ (±6.3) and 1.7 ‰ (±6.3), respectively. The average source contribution of N 2 O from nitrification/fungal denitrification was 18.7% (±21.0) according to the closed‐system model, while the open‐system model and the dynamic apparent NIE function resulted in values of 31.0% (±14.0) and 28.3% (±14.0), respectively. Conclusions Using a closed‐system model with a fixed SP isotope effect may significantly overestimate the N 2 O reduction effect on SP values, especially when N 2 O reduction rates are high. This is probably due to soil inhomogeneity and can be compensated for by the application of a dynamic apparent NIE function, which takes the variable reduction rates in soil micropores into account. Copyright © 2016 John Wiley & Sons, Ltd.