【摘要】 流域水文模型作为洪水预报和防灾减灾的核心技术，其汇流演算方法的选用对径流模拟效果有重要影响。Muskingum方法忽略回水效应，可能对水文模型在复杂地形区域的应用造成偏差，而WRF-Hydro(Weather Researchand Forecasting model(WRF) hydrological modeling system)模式不同汇流方法的对比研究仍显不足。针对清江流域水布垭水文站以上流域为研究区域，利用2016—2017年的7次径流过程，设计了扩散波方法采用默认CHANPARM.TBL参数表调参（WW）、Muskingum方法基于Route＿link.nc文件调参（MM）和扩散波方法采用修改后的CHANPARM.TBL参数表调参（WM）,3种试验方案进行模拟，对比扩散波、Muskingum两种河道汇流演算方法在WRF-Hydro模式中应用效果的差异，分析其影响因素。结果表明：使用两种河道汇流方法的WRF-Hydro模式均有良好的性能；MM试验方案情况下与WW/WM方案相比，Muskingum方法的平均NSE分别提高0.017、0.037，平均KGE分别降低0.012、0.021。在参数一致的情况下，扩散波方法因考虑回水效应，模拟径流流速较慢，峰值流量较小，峰现时间推迟，表明回水效应是影响汇流模拟的关键因素。更多还原

【Abstract】 Hydrological models are the core technology for flood forecasting, and the selection of channel routing algorithms has a significant impact on the effectiveness of runoff simulation. The Muskingum method ignores the effect of backflow, which may cause bias in the application of hydrological models in complex terrain areas. The comparative study of different channel routing algorithms in WRF Hydro mode is still insufficient. The basin above Shuibuya Hydrological Station in the Qingjiang River Basin is the research area. Simulate using 7 runoff processes from 2016 to 2017. Three experimental schemes were designed for simulation. The diffusion wave method which parameterized based on the default CHANPARM. TBL parameter table(WW), the Muskingum method which parameterized based on the Routenlink. nc file(MM), and the diffusion wave method which parameterized based on the modified CHANPARM. TBL parameter table(WM). Compare the differences in the application effects of diffusion wave and Muskingum river confluence calculation methods in the WRF Hydro model, and analyze their influencing factors. The results indicate that both WRF Hydro models using two river confluence methods have good performance. Compared with WW/WM, the average NSE of Muskingum method increased by0.017 and 0.037, and the average KGE decreased by 0.012 and 0.021. In the case of consistent parameters,the diffusion wave method simulates slower runoff velocity, smaller peak flow rate, and delayed peak appearance time due to considering the backflow effect, indicating that the backflow effect is a key factor affecting channel routing.更多还原