Assessing Spring Thaw Nitrous Oxide Fluxes Simulated by the DNDC Model for Agricultural Soils

Large N2O emissions from agricultural soils have been reported during winter and spring thaw. The objective of this study was to assess the ability of the DNDC model to simulate N2O emissions resulting from freeze thaw cycles, particularly the timing of flux events. The DNDC model was tested against micrometeorological fluxes measured during 5 yr in Ontario, Canada. There was a very large discrepancy between simulated and observed fluxes in terms of magnitude and timing. The simulated event occurred, on average, 38 d later than observed, and N2O fluxes were up to 3.5 times larger than the highest measured flux. Examination of simulated soil conditions indicated that the mechanism underlying freeze-thaw-induced N2O flux in the DNDC model, release of ice-trapped N2O, was not correct. This misconception had not been identified before, possibly because cold conditions in previous studies were not as extreme as observed in our data set or because continuously measured N2O fluxes were not available for model assessment. As a result of this analysis, DNDC 9.1 was revised by removing the release of ice-trapped N2O and adding N2O newly produced by denitrification in the surface layer as the main mechanism for N2O production (DNDC 9.3). Comparison between simulated N2O fluxes using DNDC 9.3 and our data indicated improved timing to within 1 d of observed events. The magnitude of simulated flux differed from measurements by more than a factor of two, however, suggesting that an improved algorithm for N2O production and diffusion under soil freezing and thawing is needed.