Spatial analysis and modeling the nitrogen flush after rainfall events at the field scale in SWAT

Authors: HANEY, ELIZABETH - Texas Agrilife Research; Haney, Richard; Arnold, Jeffrey; White, Michael; SRINIVASAN, RAGHAVAN - Texas A&M University; SENSEMAN, SCOTT - University Of Tennessee

Submitted to: American Journal of Environmental Sciences
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/22/2016
Publication Date: 5/3/2016
Publication URL: http://handle.nal.usda.gov/10113/5490057
Citation: Haney, E., Haney, R.L., Arnold, J.G., White, M.J., Srinivasan, R., Senseman, S.A. 2016. Spatial analysis and modeling the nitrogen flush after rainfall events at the field scale in SWAT. American Journal of Environmental Sciences. 12(2):102-121. doi:10.3844/ajessp.2016.102.121.

Interpretive Summary: Environmental models, such as the Soil Water Assessment Tool (SWAT) use a combination of scientific knowledge of environmental processes and spatial data to predict environmental outcomes under varying situations. The current nitrogen (N) model neglects the contribution of the microbial population to the plant available N pool, resulting in an underestimation of yield in low fertilizer situations. We used laboratory measurements of microbial activity and water extractable carbon (C) and N results to modify N cycling in the SWAT. Simulation results indicate that yearly yield values and the variability of these yield values were consistantly greater from the modified N model than from the SWAT model, as would be expected with the addition of N mineralization resulting from microbial activity. The equations used to model the complex biogeochemical N cycling relationships are elegant in their simplicity, yet capture the spatial complexity associated with their processes. The modified N model may be useful to regulators to help with the simulation of new conservation practices that include the effect of lower fertilizer inputs on nutrient runoff and pollution.

Technical Abstract: Current nitrogen (N) models tend to neglect the contribution of the microbial population to the plant available N pool, resulting in an underestimation of yield and possible over or underestimation of N runoff from natural and agricultural landscapes. We used the measurement of microbial activity coupled with the measurement of their food source, water extractable N and carbon (C), to provide initial N values and mineralization rates to modify N cycling in the Soil Water Assessment Tool (SWAT). Soil test data and spatial analysis of N mineralization values were used to: (1) quantify spatial variation of water extractable organic and inorganic N, soil inorganic N, and microbial activity; (2) develop a field scale model to determine N mineralization using updated soil testing methods for integration into the SWAT model; and (3) predict wheat yield. Simulation results indicate that yearly yield values and the variability of these yield values were consistantly greater from the modified N model than from the SWAT model, as would be expected with the addition of N mineralization resulting from microbial activity. The spatial variability in yield results by sample increased with the modified N model as compared to the SWAT model. The yield data resulting from the modified N model simulation were sensitive to soil nutrient values as well as variation in elevation. Temporal and climatic variability is accounted for by including a precipitation trigger for N mineralization. The equations used to model the complex biogeochemical N cycling relationships are elegant in their simplicity, yet capture the spatial complexity associated with their processes. The modified N model may be useful to regulators to help with the simulation of new conservation practices that include the effect of lower fertilizer inputs on nutrient runoff and pollution.