基于混合产流模式的黄土高原流域水沙过程耦合模拟

为深入理解黄土高原干旱半干旱地区复杂地貌条件下流域水沙运移规律。基于垂向混合产流机理和运动波方程，构建分布式流域水文模型，耦合流域土壤侵蚀和泥沙输移过程模拟模块，并考虑梯田对水沙过程的影响，建立适用于黄土高原的分布式流域水沙过程模型。选取黄土高原延河支流西川河流域多年实测场次洪水过程的径流泥沙资料，对模型进行率定和验证。径流模拟的纳什效率系数在0.56以上，平均值超过0.70,模拟次洪过程的峰形、峰值、峰现时间与实测过程具有较好的一致性；侵蚀产输沙模拟精度较低，其纳什效率系数均值率定期为0.79,但验证期仅为0.45,模拟结果整体趋势与实测值较一致，但输沙量模拟峰值比实测值偏低。模型可以较精确地模拟黄土高原流域洪水产汇流过程，但输沙量模拟值偏低，一方面由于产汇流模块的误差传递；另一方面，对重力侵蚀考虑不足。因此，未来模型将考虑滑坡、崩塌等重力侵蚀过程，提升模拟精度和效率，为流域水沙过程模拟与流域综合治理提供有效工具。

Runoff and Sediment Processes Simulation Based on Mixed Runoff Generation Model in the Loess Plateau Watershed, China

In order to better understand the water and sediment transport characteristics in the arid and semi-arid region of the Loess Plateau under complex geomorphological conditions, a distributed watershed water and sediment process model suitable for the Loess Plateau was constructed based on the vertical mixing production mechanism and kinematic wave equation, coupled with watershed soil erosion and sediment transport modules, and the influence of terraces on water and sediment movement was considered. Measured runoff and sediment data of multiple years of flood events in the Xichuan River basin, a tributary of the Yanhe River in Loess Plateau, were used for calibration and validation of the model. The Nash-Sutcliffe efficiency coefficients of runoff simulation were higher than 0.56, with an average value greater than 0.70. The simulated flood peak shape, peak value, and peak time of the simulated floods were in good agreement with the measured floods. However, the model simulation accuracy of soil erosion and sediment transport was relatively low, with a mean Nash-Sutcliffe efficiency coefficient of 0.79 during the calibration period, but only 0.45 during the verification period. The overall trend of the simulation results was consistent with the measured values, but the peak sediment transport simulation values were lower than the measured values. The model can simulate the flood runoff process in watersheds in the Loess Plateau with relatively high accuracy, however the simulated sediment transport values were lower than the measured values, and this was due to both the error propagation of the production convergence flow module and the insufficient consideration of gravity erosion. Therefore, in the future, the model will consider gravity erosion processes such as landslides and rockfalls to improve simulation accuracy and efficiency, providing an effective tool for simulating watershed water and sediment processes and integrated watershed management.