设施菜地 WHCNS_Veg 水氮管理模型.

设施 蔬菜 水分 氮素 WHCNS_Veg 模型评&#2 Excessive water and nitrogen (N) input and shallow root systems have led to serious N loss in greenhouse vegetable production system (GVPS), thereby threatening the soil and water-body environments. Identifying the fates of water and N is crucial to develop best management strategies in intensive GVPS. Soil-crop models are important as a water and N management tool to trade off crop yield and environmental cost. The objectives of this study were to develop a scientific water and N management tool for intensive GVPS in China, and evaluate the model performance in the simulating water dynamic, N fate and vegetable growth under different water and N management practices in China. In this study, based on the EU-Rotate_N model, a vegetable growth module was developed and fully incorporated into a soil water heat carbon nitrogen simulator (WHCNS). The coupled model, i.e., WHCNS_Veg, combined the soil module of the WHCNS model with the vegetable growth module of the EU-Rotate_N model. The key processes included soil movement, soil water evaporation, crop transpiration, soil N transport and transformation (net N mineralization, nitrification, ammonia volatilization, and denitrification) and vegetable growth. Two field experiments conducted in Shouguang city, Shandong province, and Tianjin city, Hebei province were used to test the coupled WHCNS_Veg model. Cucumber and tomato were planted in solar greenhouses at Shouguang and Tianjing experiments, respectively. Similar irrigation and fertilization management practices were both setup in the two experiments: 1) Furrow irrigation with conventional N fertilizer (farmer’s practice), FP; 2) Drip irrigation with optimal N fertilizer, OPT; 3) FP plus crop residues, FPR; and 4) OPT plus crop residues, DOR. Field experiment data including soil water (soil water content and matrix potential), soil N (nitrate concentration and nitrate leaching), plant N uptake, and marketable fresh yield under different water and N management practices were collected in the experiments. Results showed that the WHCNS_Veg model performed better in crop growth simulation than soil water and N dynamic simulation, with normalized root mean square error (nRMSE) not higher than 12.1%, index of agreement (IA) not less than 0.934 and Nash-Sutcliffe efficiency (NSE) not less than 0.829 for simulation of cucumber and tomato yield and plant N uptake. Soil water content had the highest simulation accuracy. The nRMSE, IA and NSE values for soil water content simulation were 6.2%-9.1%, 0.851-0.960 and 0.477-0.846, respectively. For soil nitrate concentration and nitrate leaching, the nRMSE values were 22.2%-40.1% and 4.6-26.0%, respectively, and the NSE values were −0.810-0.636 and 0.442-0.956, respectively. It indicated that the model simulation results are within the acceptable range. However, WHCNS_Veg model showed relative low simulation accuracy in soil matrix potential under drip irrigation condition, with nRMSE, IA and NSE values ranged from 22.9%-30.1%, 0.223-0.846, and -6.344-0.113, respectively. It indicated that the model has to be improved to simulate soil matrix potential under drip condition. In general, the coupled WHCNS_Veg model gives satisfactory results for soil water and N dynamics simulations as well as vegetable growth simulations. It has great potential to simulate and analyze water and N fates, and vegetable growth for the intensive greenhouse vegetable production in China. 0215; [ABSTRACT FROM AUTHOR]