Simulated nitrous oxide emissions from multiple agroecosystems in the U.S. Corn Belt using the modified SWAT-C model

• Published in: Environmental pollution (1987) 2023-11, Vol.337, p.122537, Article 122537
• Main Authors: Liang, Kang, Qi, Junyu, Zhang, Xuesong, Emmett, Bryan, Johnson, Jane M.F., Malone, Robert W., Moglen, Glenn E., Venterea, Rodney T.
• Format: Article
• Language: English
• Subjects: Agroecosystem; Crop management; Greenhouse gases; Process-based modeling
• ISSN: 0269-7491; 1873-6424; 1873-6424
• DOI: 10.1016/j.envpol.2023.122537

Agriculture is a major source of nitrous oxide (N2O) emissions into the atmosphere. However, assessing the impacts of agricultural conservation practices, land use change, and climate adaptation measures on N2O emissions at a large scale is a challenge for process-based model applications. Here, we integrated six N2O emission algorithms for the nitrification processes and seven N2O emission algorithms for the denitrification process into the Soil and Water Assessment Tool-Carbon (SWAT-C). We evaluated the different combinations of methods in simulating N2O emissions under corn (Zea mays L.) production systems with various conservation practices, including fertilization, tillage, and crop rotation (represented by 14 experimental treatments and 83 treatment-years) at five experimental sites across the U.S. Midwest. The SWAT-C model exhibited wide variability in simulating daily average N2O emissions across treatment-years with different method configurations, as indicated by the ranges of R2, NSE, and BIAS (0.04–0.68, −1.78–0.60, and −0.94–0.001, respectively). Our results indicate that the denitrification process has a stronger impact on N2O emissions than the nitrification process. The best performing N2O emission algorithms are those rooted in the CENTURY model, which considers soil pH and respiration effects that were overlooked by other algorithms. The optimal N2O emission algorithm explained about 63% of the variability of annual average N2O emissions, with NSE and BIAS of 0.60 and −0.033, respectively. The model can reasonably represent the impacts of agricultural conservation practices on N2O emissions. We anticipate that the improved SWAT-C model, with its flexible configurations and robust modeling and assessment capabilities, will provide a valuable tool for studying and managing N2O emissions from agroecosystems. [Display omitted] •We integrated multiple N2O emission algorithms into the SWAT-C model.•Model performance was evaluated using data from the US Midwest.•The best-performing algorithm considered soil pH and respiration effects.•The model effectively captured the impacts of agricultural management practices on N2O emissions.•The model accurately represented the spatiotemporal variations in N2O emissions.