Some key environmental variables controlling nitrous oxide emissions from agricultural and semi-natural soils in Scotland

Soil N2O emissions were measured from a range of mineral soils (22 sites) supporting woodlands, grassland and agriculture between 1991 and 1996 and were co-analysed in order to establish some overriding relationships between the N2O emission and soil temperature, water content, available NH+4 and NO−3, pH and N input by fertiliser, manure and atmospheric deposition. Fluxes were measured by chamber techniques. For individual sites, nitrogen, soil temperature and water content were the key variables controlling the emission rates. Nitrogen additions by fertilisation, manure and atmospheric deposition increased the emissions of N2O. From pasture grazed by sheep 1.7% of the N input from fertiliser and animal excreta was emitted as N2O. For an experimental sitka spruce plantation receiving acid mist and a mixed woodland in the immediate vicinity of a poultry farm losses of N2O were estimated at 3.7 and 0.8% of the atmospheric N deposited, respectively. In spite of the large variations in land use, soil physical and chemical characteristics of the 22 data sets collected at different years and at varying frequency, the mean flux from each site could be described by following multiple regression equation: log N2O (μg N m−2 h−1) = −1.04 + 0.165*soil temperature + 0.403*log N input (kg N ha−1 yr−1) −0.0145*gravimetric soil water content (% soil dry weight)