Denitrification loss from a grassland soil in the field receiving different rates of nitrogen as ammonium nitrate

Denitrification loss from a loam under a cut ryegrass sward receiving 0, 250 and 500 kg N ha^?1 a^?1 in four equal amounts was measured during 14 months using the acetylene-inhibition technique. The rate of denitrification responded rapidly to changes in soil water content as affected by rain. Mean rates of denitrification exceeded 0.2 kg N ha^?1 day^?1 only when the soil water content was >20% (w/w) and nitrate was >5μ N g^?1 in the upper 20 cm of the profile and when soil temperature at 2 cm was >5–8°C. When the soil dried to a water content <20%, denitrification decreased to <0.05 kg N ha^?1 day^?1. Highest rates (up to 2.0 kg N ha^?1 day^?1) were observed following application of fertilizer to soil at a water content of about 30% (w/w) in early spring. Denitrification in the control plot during this period was generally about a hundredth of that in plots treated with ammonium nitrate. High rates of N₂O loss (up to 0.30 kg N ha^?1 day-1) were invariably associated with high rates of denitrification (> 0.2 kg N ha^?1 day^?1). However, within 2–3 weeks following application of fertilizer to the plot receiving 250 kg N ha^?1 a^?1 the soil acted as a sink for atmospheric N₂O when its water content was >20% and its temperature >5–8°C. Annual N losses arising from denitrification were 1.6, 11.1 and 29.1 kg N ha^?1 for the plots receiving 0, 250 and 500 kg N ha^?1 a^?1, respectively. More than 60% of the annual loss occurred during a period of 8 weeks when fertilizer was applied to soil with a water content >20%.