基于HYDRUS模型的华北平原小麦种植区水盐运移模拟

Water and salt transport simulation in the wheat growing area of the North China Plain based on HYDRUS model

  • 摘要: 土壤中水分和盐分是影响作物生长的两个关键因素,揭示水盐运移机制对阐明作物利用土壤水过程具有重要意义。本研究以华北平原典型农田——中国科学院禹城综合试验站为试验地,基于试验站内冬小麦种植地的长期土壤水分观测数据及室内土柱试验,应用HYDRUS-1D模型分别阐明土壤水分及盐分变化规律及分布特征,探究影响水盐运移的驱动因素,并评价HYDRUS-1D模型对研究区水盐运移模拟的适用性。水分运移模拟结果表明:浅层土壤水分运移模拟因受外界因素的剧烈影响而较深层土壤产生更大的误差,10 cm、20 cm、30 cm、40 cm和60 cm处水分运移模拟结果的均方根误差分别为0.0348 cm3·cm-3、0.0179 cm3·cm-3、0.0179 cm3·cm-3、0.0122 cm3·cm-3和0.0053 cm3·cm-3;水分运移模拟的纳什效率系数平均值为0.826,变异系数为0.0560,表明模拟结果与实测土壤水分变化过程一致性较好。土柱试验结果显示:灌水8 L,入渗12 h、24 h、40 h、45 h和48 h后,各时刻土壤盐分含量在垂向上整体呈现先增大后减少的分布规律,均方根误差分别为0.181 g·kg-1、0.131 g·kg-1、0.120 g·kg-1、0.034 g·kg-1和0.027 g·kg-1,平均误差的平均值为0.174 g·kg-1。受蒸发、耕作、根系等影响,理化性质变异性较大导致浅层土壤盐分运移模拟值与实测值偏差增大,纳什效率系数的变异系数达9.71。灌水8 L、16 L、24 L,入渗48 h后分别在土壤23 cm、26 cm、29 cm处出现盐分含量峰值,表明增加灌水量可加强盐分淋洗效果。此研究可为深入探究华北平原冬小麦土壤水盐运移规律、优化农田水资源管理、提高水资源利用效率提供一定理论基础。

     

    Abstract: Soil moisture and salinity are key factors that affect crop growth. Thus, it is important to investigate the mechanisms of water and salt transport to further clarify the process of soil water utilization in crops. The HYDRUS-1D model was applied to examine the spatial distribution and vertical variation in soil water and salinity and to explore the factors influencing water and salt transport. The model incorporated long-term soil moisture observation data and the results of indoor soil column experiments at the Yucheng Comprehensive Experimental Station, Chinese Academy of Sciences, a typical farmland in the North China Plain. Moreover, the applicability of the HYDRUS-1D model to the study area was evaluated. The results showed that the simulation of water transport in shallow soil had greater errors than that in deep soil, owing to the influence of external factors. The root mean square errors (RMSEs) of the water transport simulation results were 0.0348 cm3·cm-3, 0.0179 cm3·cm-3, 0.0179 cm3·cm-3, 0.0122 cm3·cm-3, and 0.0053 cm3·cm-3 at 10 cm, 20 cm, 30 cm, 40 cm, and 60 cm, respectively. The mean value of the Nash-Sutcliffe efficiency (NSE) coefficient was 0.826, and the coefficient of variation was 0.0560, indicating that the simulations of water transport were stable and consistent with the measured values. The soil column experiment results showed that after irrigation with 8 L water, salt salinity in the vertical direction first increased and then decreased; the RMSEs of the simulation results of salt transport were 0.181 g·kg-1, 0.131 g·kg-1, 0.120 g·kg-1, 0.034 g·kg-1, and 0.027 g·kg-1 after 12 h, 24 h, 40 h, 45 h, and 48 h, respectively. The mean error was 0.174 g·kg-1, which was in good agreement with the measured values, indicating that the model was suitable for the simulation of water and salt transport in the study area. However, owing to the influence of evaporation, tillage, and crop root system, large variations in physical and chemical properties resulted in large deviations between the simulated and measured values of salt transport in shallow soil, and the NSE coefficient reached 9.71. After 48 h of infiltration, the soil salinity peaked at 23 cm, 26 cm, and 29 cm depth with 8 L, 16 L, and 24 L irrigation quotas, respectively. These results showed that increased irrigation quotas can enhance salt leaching. This study confirmed that the HYDRUS-1D model could be used for theoretical studies of water and salt transport in the study area. This study also provides a theoretical basis for further exploration of water and salt transport in winter wheat, optimizing farmland water resource management, and improving the water resource utilization efficiency in the North China Plain.

     

/

返回文章
返回