垂向混合模型在湫水河洪水预报预警中的应用

Application of Vertical Hybrid Model on Flood Forecast and Early-Warning of Qiushui River

针对半干旱半湿润地区的产汇流特点,采用垂向混合模型进行了洪水模拟与预报研究;面向洪水等级预警需求,提出了洪水等级与相对误差相结合的洪峰预报精度等级评价标准,并结合置信区间洪峰包含比这一评估指标,构建了洪水预报预警可靠性的综合评价方法。以黄河中游湫水河流域为例,选取流域出口林家坪水文站1980—2012年间洪峰大于300 m3/s的20场洪水进行模型率定与验证。结果表明:洪峰预报精度等级评价合格率在70%以上,可靠度为95%,优于新安江模型或陕北模型,在一定程度上能够满足当地洪水预报预警需求。

Aiming at the characteristics of runoff and routing in semi-arid and semi-humid areas,a vertical hybrid model was used to simulate and forecast floods. In order to meet the needs of early-warning of floods,a new evaluation standard of flood forecast accuracy based on the combination of flood peak level and relative error was put forward. Combined with the assessment index of flood peak coverage of confidence interval,a comprehensive evaluation method of reliability of flood forecast and early-warning was established. Taking the Qiushui River catchment in the middle Yellow River as an example,20 flood events with flood peaks greater than 300 m^3/s during the period from 1980 to 2012 at Linjiaping Station in the catchment outlet were selected to calibrate and validate the model. The results indicate that the flood forecast accuracy is above 70% and the coverage of confidence interval reaches 95%,which is better than the Xin＇anjiang model or the Shaanbei model.It can meet local flood forecasting and warning needs to some extent.