华北地区分布式蓄超空间动态组合TOKASIDE-D模型研究

为探究我国华北地区由于下垫面和降水时空分布不均匀产生的地下水超采严重和产流机制混合多变(蓄满与超渗产流随着时间和空间发生交替主导)等问题的解决办法，在基于物理基础的分布式模型TOKASIDE的基础上，通过降雨和土壤含水量之间的判定关系，动态识别蓄满产流和超渗产流网格，构建分布式蓄超空间动态组合模型(TOKASIDE-D)，并选择华北地区的北辛店流域进行验证和分析。结果表明，与原模型相比，TOKASIDE-D模型在北辛店流域的洪水模拟精度有较为明显的提升，径流深误差合格率由44%提升至56%，洪峰误差合格率由44%提升至78%；TOKASIDE-D模型能够准确捕捉降雨、土壤含水量等动态因子对产流模式的控制效应。

TOKASIDE-D model of spatial dynamic combination of saturation-excess and infiltration-excess runoff for North China region

To investigate the solutions to the problems of severe groundwater over-extraction and mixed and variable runoff generation mechanisms(alternating dominance of the saturation-excess and infiltration-excess runoff generation with time and space)in North China, based on the distributed TOKASIDE model, a distributed spatial dynamic combination hydrological model(TOKASIDE-D)was established by dynamically identifying the saturation-excess and infiltration-excess runoff generation grids through the relationship determination between precipitation and soil water content. The Beixindian Watershed in North China is selected for the validation and the analysis. The results show that compared with the original model, the flood simulation accuracy of TOKASIDE-D model in the Beixindian Watershed is improved significantly, the qualified rate of runoff depth error is increased from 44% to 56%, and the qualified rate of flood peak error is increased from 44% to 78%. TOKASIDE-D model can accurately capture the control effect of rainfall, soil moisture and other dynamic factors on the runoff generation mode.