Soils represent the major source of the atmospheric greenhouse gas nitrous oxide (N 2O) and there is a need to better constrain the total global flux and the relative contribution of the microbial source processes. The aim of our study was to evaluate isotopomer analysis of N 2O (intramolecular distribution of 15N) as well as conventional nitrogen and oxygen isotope ratios (i) as a tool to identify N 2O production processes in soils and (ii) to constrain the isotopic fingerprint of soil-derived N 2O. We conducted a microcosm study with arable loess soil fertilized with 20 mg N kg −1 of 15NO 3 −-labeled or non-labeled ammonium nitrate. Soils were incubated for 16 d at varying moisture (55%, 75% and 85% water-filled pore space (WFPS)) in order to establish different levels of nitrification and denitrification. Dual isotope and isotopomer ratios of emitted N 2O were determined by mass spectrometric analysis of δ 18O, average δ 15N ( δ 15N bulk) and 15N site preference (SP=difference in δ 15N between the central and peripheral N-positions of the asymmetric N 2O molecule). Total rates and N 2O emission of denitrification and nitrification were determined by 15N analysis of headspace gases and soil extracts of the 15NO 3 − treatment. N 2O emission and denitrification increased with moisture whereas gross nitrification was almost constant. In the 55% WFPS treatment, more than half of the N 2O flux was derived from nitrification, whereas denitrification was the dominant N 2O source in the 75% WFPS and 85% WFPS treatments. Moisture conditions were reflected by the isotopic signatures since highly significant differences were observed for average δ 15N bulk, SP and δ 18O. Experiment means of the 75% WFPS and 85% WFPS treatments gave negative δ 15N bulk (−18.0‰ and −34.8‰, respectively) and positive SP (8.6‰ and 15.3‰, respectively), which we explained by the fractionation during N 2O production and partial reduction to N 2. In the 55% WFPS treatment, mean SP was relatively low (1.9‰), which suggests that nitrification produced N 2O with low or negative SP. The observed influence of process condition on isotopomer signatures suggests that the isotopomer approach might be suitable for identifying N 2O source processes. However, more research is needed to determine the impact from process rates and microbial community structure. Isotopomer signatures were within the range reported from previous soil studies which supports the assumption that SP of soil-derived N 2O is lower than SP of tropospheric N 2O.