A microbial-explicit model with comprehensive nitrogen processes to quantify gaseous nitrogen production from agricultural soils

Abstract

Agricultural soils are a major source of anthropogenic N2O, but their N2O emission estimates are highly uncertain, mainly due to the complexity of nitrogen (N) processes. Most soil N models include only primary N processes such as nitrification and denitrification, which limits their ability to realistically simulate N transformations in soils and accurately estimate N2O emissions from soils. This study introduces a Microbial-Explicit Model incorporating Comprehensive Nitrogen processes (MEMCN) to evaluate and quantify the influences of various N processes on the production of N2O as well as NO and N2, including nitrification, denitrification, anaerobic ammonium oxidation (ANAMMOX), dissimilatory nitrate reduction to ammonium (DNRA), mineralization, and microbial assimilation (i.e., growth) and death. The MEMCN was evaluated in laboratory experiments using agricultural soils with different levels of N additions under anaerobic and aerobic conditions, and reproduced well the dynamics of NH4+, NO3−, NO2−, NO, N2O, and N2. After nitrification and denitrification, ANAMMOX and assimilation were found to be most important in controlling N transformations in agricultural soils. ANAMMOX directly increased N2 emissions by 139% at the beginning of the simulations (i.e., 48 h) under anaerobic conditions, while microbial assimilation indirectly reduced NO, N2O, and N2 emissions by 88%, 54%, and 58%, respectively, at the end of the simulations (i.e., 336 h) under aerobic conditions. Correspondingly, the biomass of ANAMMOX bacteria increased significantly at the beginning of the simulations under anaerobic conditions, while the biomass of nitrite oxidizing bacteria increased substantially under aerobic conditions. In contrast, DNRA, mineralization and microbial death had minor effect on soil N transformations, and the biomass of DNRA bacteria and heterotrophs did not change significantly during the simulations. Our study shows that it is necessary to include ANAMMOX and microbial assimilation in soil N models, while explicit simulation of microbial biomass dynamics may only be necessary if microbial biomass pools change significantly.