Field scale SWAT+ modeling of corn and soybean yields for the contiguous United States: National agroecosystem model development

Authors: CERKASOVA, NATALJA - Texas Agrilife Research; White, Michael; Arnold, Jeffrey; BIEGER, KATRIN - Aarhus University; ALLEN, PETER - Baylor University; GAO, JUNGANG - Texas Agrilife Research; Gambone, Marilyn; MEKI, MANYOWA - Agrilife Research; Kiniry, James; GASSMAN, PHILIP - Iowa State University

Submitted to: Agricultural Systems
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/1/2023
Publication Date: 8/1/2023
Citation: Cerkasova, N., White, M.J., Arnold, J.G., Bieger, K., Allen, P., Gao, J., Gambone, M.A., Meki, M., Kiniry, J.R., Gassman, P. 2023. Field scale SWAT+ modeling of corn and soybean yields for the contiguous United States: National agroecosystem model development. Agricultural Systems. 210. Article 103695. https://doi.org/10.1016/j.agsy.2023.103695.
DOI: https://doi.org/10.1016/j.agsy.2023.103695

Interpretive Summary: USDA is committed to ensuring sustainable growth of agriculture and food systems which includes productivity growth while reducing the environmental impact of agricultural production. In response to this goal, we developed a national agroecosystem model to provide a science-based, data-driven decision tool for developing conservation policy. The SWAT+ (Soil and Water Assessment Tool ) watershed model is the modeling engine for the national model. In this study, we examined the ability of the SWAT+ model to simulate corn and soybean production across the US. Results showed good agreement for both corn and soybean yields, with good model agreement in high-producing area of the US – the Corn Belt (Relative Error < ±5%). An analysis of possible crop production differences for corn and soybeans in irrigated, tiled, and non-irrigated-non-tiled fields is also discussed. This research assists in providing USDA policy makers with a national decision tool that can be used to assess the impact of management and climate on agricultural production in addition to the environmental sustainability of future changes.

Technical Abstract: In this paper we asses the capability of the updated SWAT+ model to accurately simulate crop yields across the contiguous United States, with the focus on Corn (Zea mays L.) and Soybean (Glycine max L. Merr.) production. We use available open-access data to setup a high-resolution modeling system, where every HUC8 (8-digit hydrologic unit) is represented as an individual SWAT+ simulation. All the 2201 HUC8 simulations in the contiguous U.S. are then interconnected from upstream to downstream and make up the National Agroecosystem Model (NAM). We used field boundary data to setup the NAM in such a way that every identified cultivated field is modeled as a unique HRU (Hydrologic Response Unit). Simulated corn and soybean yield from over 2.5 million field-type HRUs were compared with the reported average annual crop production of the respective area for the 2015-2020 period. Results show a good agreement for both crop yields (R2 = 0.999 for corn, R2=0.993 for soybeans), with a good model agreement in high-producing area of the US – the Corn Belt (Relative Error < ±5%). We also demonstrate the new soft-calibration module of SWAT+ and highlight major changes of the plant growth module and model parametrization. An analysis of possible crop production differences for corn and soybeans in irrigated, tiled, and non-irrigated-non-tiled fields is also discussed.