EVOLUTION OF NITROUS OXIDE FROM SOIL: II. EXPERIMENTAL RESULTS AND MODEL TESTING

The dynamic processes controlling N2O evolution in the field are complex; therefore, estimates of long-term evolution from periodic measurements are likely to be unreliable. Greater reliability in such estimations may be possible by reproducing these processes mathematically within a dynamic simulation model. A mathematical hypothesis for denitrification and N2O production was used within a larger ecosystem simulation model to study N2O evolution under laboratory and field conditions. Hourly N2O fluxes estimated by the model were compared with fluxes recorded over an inundated soil core during a 10-day laboratory incubation after a 100-g N Mg−1 amendment with 15N-KNO3. Estimated values were within 20% of those recorded for fluxes, ranging from 0 to 4 mg N m−2 h−1. Recovery of both mineral and labeled N from the amendment after 10 days was estimated within 2 g Mg−1 of recorded values. Hourly N2O fluxes estimated by the model were also compared with fluxes recorded during spring thaws during 2 years at a field site in central Alberta. Fluxes ranging from 0.1 to 10.0 mg N m−2 h−1 were reproduced by the model after different fertilizer treatments. Increased rates of N2O evolution followed periods of saturation in the upper profile caused by snowmelt. These increased rates continued for about 1 week in both years before declining as soil drying progressed. In the model, evolution arose from the volatilization, during drainage, of aqueous N2O generated from biological NO3- reduction that occurred during periods of low aqueous O2 concentrations caused by water accumulations in the thawing soil profile. The simulated evolution supports the hypothesis that denitrification accompanies soil thawing after snowmelt at this site.