A spatial inventory of N 2O emissions from agricultural and non-agricultural soils in Great Britain was prepared using a simple regression model within a GIS framework. The regression model was based on published N 2O data from soils of temperate climates. It describes emissions as a function of N input ( N), water filled pore space (WFPS), soil temperature ( T s) and land use ( A): ln (N 2 O) (kgN ha −1 y −1 )=−2.7+0.60ln N (kgN ha −1 y −1 )+0.61ln WFPS (%)+0.035 T s (°C)−0.99 A . The regression model predicted the highest fluxes of 6–21 kg N ha −1 y −1 from grazed grasslands. On tilled land, predicted N 2O emissions did not exceed 6 kg N ha −1 y −1, while fluxes below 0.1 kg N ha −1 y −1 were estimated for semi-natural land. N 2O emissions from soils in spring and summer were a factor of 2–3 higher than in the remaining part of the year. Total N 2O emissions for Great Britain were estimated at 127 kt N 2O-N y −1. Distribution maps of annual and seasonal N 2O emissions outlined the areas with the largest fluxes as those of intensive livestock farming in wet western regions of Britain. The annual emissions predicted by this study were much higher for agricultural soils than those suggested by the IPCC emission factor of 1.25% (0.25–2.25%) of N input, which range between 3 and 9 kg N ha −1 y −1 for grasslands and 2 and 3 kg N ha −1 y −1 for tillage crops and predict a total emission from agricultural sources of 56 kt N 2O-N y −1. Main uncertainty of the linear regression model is caused by scaling from published short-term N 2O emission data to annual averages.