基于APSIM模型的黄土旱塬区苜蓿——小麦轮作系统深层土壤水分及水分利用效率研究

黄土高原地区降水较少且季节性分配不均,苜蓿连续种植所导致的深层土壤干燥化问题已经引起普遍关注,苜蓿与粮食作物轮作是恢复苜蓿草地土壤水分、提高粮草种植系统可持续性的有效方式.但是长期轮作对土壤水分环境和水分利用效率的影响仍然缺乏研究,农业生产系统模拟模型(APSIM)具有广泛的适应性,可准确模拟耕作管理对作物系统资源利用...

Evaluating the deep-horizon soil water content and water use efficiency in the alfalfa-wheat rotation system on the dryland of Loess Plateau using APSIM

Rainfall on the Loess Plateau is limited and variable （350-650 mm·y^-1）. The problem of deep soil desiccation caused by continuous planting of alfalfa has aroused widespread attention. Alfalfa and food crop rotation is an effective way to restore soil moisture of alfalfa fields and improve the sustainability of forage cropping systems. However， there is still a lack of research on the effects of long-term rotation on the soil water environment and water use efficiency. The agricultural production systems simulator （APSIM） has extensive adaptability and can accurately simulate the impact of tillage management on the resource utilization of crop systems. Based on experimental data from Changwu， Zhenyuan， and Qingyang experimental stations in the dryland of Loess plateau for continuous alfalfa cropping and alfalfa based rotations， this study firstly verified the APSIM model in simulating deep soil moisture and alfalfa long-term production， and then simulated the effects of 81 different 38-year rotation patterns on farmland soil moisture， dry matter production， nitrogen absorption and water use efficiency. It was found that the APSIM model simulated the yield of alfalfa with high accuracy. The determination coefficient （R²）， mean square error （RMSE） and mean absolute error （MAE） for alfalfa yield were 0.65， 0.23 t·hm^-2 and 0.17 t·hm^-2， respectively， and the normalized mean square error （NRMSE） was 29.2%. The model could also accurately simulate the soil moisture content for the 0-1000 cm layer in long-term continuous alfalfa and alfalfa rotation cropping systems. The R²， RMSE， MAE and NRMSE were 0.73， 0.021 t·hm^-2， 0.017 t·hm^-2 and 11.7%， respectively. The determination coefficients of the simulation results in the rotation system was 0.83， RMSE， MAE and NRMSE were 0.024 t·hm^-2， 0.018 t·hm^-2 and 11.8%， respectively. The 81 rotation scenarios modelled comprised alfalfa and wheat rotation periods of 2， 4， 6， 8， 10， 12， 14， 16， 18 years respectively. The results showed that with increasing length of alfalfa rotation in the system， the soil water content became extremely low， falling to between 0.10-0.15 in the 0-1000 cm depth with rapid loss from the 400-600 cm soil horizon. When the alfalfa rotation period was more than 12 years， all rotation patterns experienced soil water deficit. Among the 81 scenarios， the total yields of 12-year alfalfa rotated with 14-， 16-， or 18-year wheat （L12W14， L12W16 and L8W16） were the highest. The system nitrogen uptake increased with increase in the number of the years planted with alfalfa， and decreased sharply after more than 10 years of wheat planting. After rotations of alfalfa exceeding 8 years， the water use efficiency of the system decreased with the increase in the number of years planted with alfalfa. Considering the soil water environment and water use efficiency， it is suggested that optimum length of alfalfa planting in an alfalfa-wheat rotation system in the study area should be 4-6 years， and the duration of the wheat rotation should be more than four years. The results provide guidance for dryland alfalfa management and forage-grain crop rotation practices on the Loess Plateau.