Coupling model of EPIC-Nitrogen2D and crop growth, soil water, nitrogen dynamics in winter wheat

Abstract: Quantitative estimation of soil moisture, nitrogen transformation and transport, and crop growth is important for improving nitrogen use efficiency and decreasing nitrogen pollution in agricultural areas. The objective of this study was to develop a transient soil water and nitrogen dynamics model with the consideration of crop growth. A crop growth module was developed based on Erosion/Productivity Impact Calculator (EPIC) model. It was then discretized by Galerkin finite element method to be integrated with the unsaturated-saturated soil water and nitrogen transformation and transport model Nitrogen2D. The integrated model was multi-functional in simulating soil water movements, soil nitrate and ammonium transformation and transport processes, and crop growth of specified type of crop. The crop growth module could calculate the root growing dynamics and its distribution in the soil, as well as the biomass in real time. And the module could also simulate the root uptake of water and nitrogen according to the real time root distribution. The yield of crop and nitrogen uptake mass would be output at the end of the crop growth period duration. The model was adopted to simulate the soil water and nitrogen distribution dynamics and winter wheat crop growth progresses in the irrigation and drainage experimental sites in Wuhan University, China. The experiments were implemented in 3 lysimeters, each with a size of 2 m × 2 m × 3 m. Winter wheat was planted and irrigated 3 times by the treated sewage water during the growing season. The nitrate and ammonium concentrations in the irrigation water were measured. The information on climatic factors such as temperature, precipitation was adopted. The reference evapotranspiration was calculated by Penman-Monteith equation. The temporal soil moisture content, soil temperature, ammonium and nitrate concentrations at different depths were measured, as well as the soil moisture, nitrate and ammonium concentration profiles before planting and after harvesting. The biomass and nitrogen uptake mass were measured after harvesting. The soil moisture and pressure head in the soil profile were measured simultaneously to obtain the soil water parameters of van Genutchen model. The model parameters including nitrogen transformation and transport parameters, and crop growth parameters were calibrated and validated by comparing the simulations with the measurements in test pit 8, 10 and 18. The Nash-Sutcliffe efficiency coefficient and agreement index of the simulated soil moisture content were 0.59-0.90, and 0.86-0.97 and these 2 statistical indexes for the simulated nitrogen concentration were 0.44-0.93, and 0.52-0.98. The simulation results showed good agreements with the measurements and indicated the model reliability for water and nitrogen simulation with crop growth. This study further compared the nitrogen root uptake results simulated by the proposed crop growth module and by a simple root uptake module. The results showed that the results from the crop growth module were in much agreement with the measurements with the relative root mean square error (RRMSE) 3.4%-46%, while the larger deviations between the simple model and the measurements were observed with the RRMSE 25%-176%, due to ignoring the root growing dynamics in real time and the effects of soil environments. Thus the integrated crop growth module and Nitrogen2D can be used for simulating soil water movements, the fate of nitrogen and crop growth in agricultural areas.