Few data are available to validate the Intergovernmental Panel on Climate Change's (IPCC) emission factors for indirect emissions of nitrous oxide (N2O). In particular the N2O emissions resulting from nitrogen leaching and the associated groundwater and surface drainage (EF5-g) are particularly poorly characterized. In situ push–pull methods have been used to identify the fate of NO3− in the groundwater. In this study, we adapted a previously published in situ denitrification push–pull method to examine the fate of 15N2O introduced into the subsoil–groundwater matrix. Enriched 15N2O was manufactured, added to groundwater via a closed system in the laboratory, and then introduced into the groundwater–subsoil matrix in an upland-marsh transition zone of a salt marsh and a forested alluvial riparian zone. Conservative tracers (SF6 and Br−) and 15N2O were injected into the groundwater and left for 1–4 h after which the groundwater was sampled. Added 15N2O behaved in a conservative manner at one site while the other site showed variability with some injections showing significant consumption (3–8 μg N2O-15N kg−1 soil day−1) of 15N2O. Our results show that the fate and dynamics of N2O in groundwater are complex and variable and that these dynamics should be considered in the development of improved IPCC inventory calculations.