General model for N 2 O and N 2 gas emissions from soils due to dentrification

• Published in: Global biogeochemical cycles 2000-12, Vol.14 (4), p.1045-1060
• Main Authors: Del Grosso, S. J., Parton, W. J., Mosier, A. R., Ojima, D. S., Kulmala, A. E., Phongpan, S.
• Format: Article
• Language: English
• ISSN: 0886-6236; 1944-9224
• DOI: 10.1029/1999GB001225

Observations of N gas loss from incubations of intact and disturbed soil cores were used to model N 2 O and N 2 emissions from soil as a result of denitrification. The model assumes that denitrification rates are controlled by the availability in soil of NO 3 ( e − acceptor), labile C compounds ( e − donor), and O 2 (competing e − acceptor). Heterotrophic soil respiration is used as a proxy for labile C availability while O 2 availability is a function of soil physical properties that influence gas diffusivity, soil WFPS, and O 2 demand. The potential for O 2 demand, as indicated by respiration rates, to contribute to soil anoxia varies inversely with a soil gas diffusivity coefficient which is regulated by soil porosity and pore size distribution. Model inputs include soil heterotrophic respiration rate, texture, NO 3 concentration, and WFPS. The model selects the minimum of the NO 3 and CO 2 functions to establish a maximum potential denitrification rate for particular levels of e − acceptor and C substrate and accounts for limitation of O 2 availability to estimate daily N 2 +N 2 O flux rates. The ratio of soil NO 3 concentration to CO 2 emission was found to reliably ( r 2 =0.5) model the ratio of N 2 to N 2 O gases emitted from the intact cores after accounting for differences in gas diffusivity among the soils. The output of the ratio function is combined with the estimate of total N gas flux rate to infer N 2 O emission. The model performed well when comparing observed and simulated values of N 2 O flux rates with the data used for model building ( r 2 =0.50) and when comparing observed and simulated N 2 O+N 2 gas emission rates from irrigated field soils used for model testing ( r 2 =0.47).