Temporal variations in nitrous oxide fluxes from urine-affected grassland

Fluxes of nitrous oxide (N2O) from a permanent pasture were measured during a 1-y period using flux; chambers, Intact soil cores from the same pasture were also treated with cow urine (19 g N m(-2)) in a laboratory incubation experiment. Diurnal variations in flux rates were determined using a mobile laboratory for continuous field measurements of N2O fluxes at the field site. Nitrous oxide fluxes peaked after fertilizer (urea and NH4NO3) application, and winter fluxes were comparable to those measured during other parts of the year, with the annual N2O flux estimated at 3.2 kg N ha(-1) y(-1) or 1.3% of the fertilizer-N applied, Nitrous oxide fluxes followed the same diurnal variation as soil temperature, with N2O production peaking in late afternoon, Multiple linear regression analyses demonstrated weak relationships between N2O flux, rainfall and soil temperature (R-2 = 0.28). There was an immediate increase in N2O emission after urine application with rates reaching a peak of 89 mg N m(-2) d(-1) within 6 h, with an estimated 7% of the applied urine-N being lost as N2O over 42 d. Results indicate that the large spatial and temporal variability in N2O fluxes have to be assessed if accurate estimates of N2O losses are to be derived for gazed grasslands. (C) 1999 Elsevier Science Ltd. All rights reserved.