多源降水在门头沟山洪模拟中的应用及比较

Application and Comparison of Multi-Source Rainfall Data in the Simulation of Flash Flood in Mentougou of Beijing

2017年6月18日14时,北京门头沟区石羊沟流域爆发山洪泥石流,造成人员伤亡。使用站点降水、融合降水、雷达降水和卫星降水驱动WRF-Hydro水文模型运行,评估不同降水资料在本次山洪模拟中的效能。结果表明,各种降水的空间分布和时间演变较为相似,但除雷达降水外,其他降水均存在较大程度的低估。利用雷达降水模拟出的山洪与实际山洪在出现时间和洪峰流量上最为相近,利用其他降水模拟的山洪出现时间相近但低估了洪峰。目前常规台站观测降水暂不满足对小尺度突发山洪的研究和预警需求,亟需融合更多的自动站和雷达、卫星等实时高分辨率降水产品。

At 14:00 BT 18 June 2017,a flash flood happened in the Shiyanggou Watershed of Mentougou District,Beijing,causing casualties.The station precipitation,combined precipitation of station and satellite,radar precipitation and satellite precipitation are used to drive the WRF-Hydro hydrological model to evaluate the performance of these different precipitation data in the flash flood simulation.The results show that the spatial distribution and temporal variation of these data are similar.Except radar precipitation,other precipitation data underestimate the precipitation.In these flash flood simulation experiments,the flash flood simulated by radar precipitation is most similar to the actual flash flood in terms of occurrence time and peak flow.The flash floods simulated based on other precipitation data are close to the actual streamflow in terms of occurrence time but the peak flow is underestimated.At present,conventional precipitation products by routine observation stations can not meet the requirements of research and early warning of small-scale abrupt flash floods.So,it is urgent to integrate radar and satellite precipitation products with observation station precipitation to get real-time high-resolution precipitation products.