基于SWAT模型的里下河平原区水文模拟

Runoff simulation for the Lixiahe Plain based on SWAT model

本文以江苏里下河平原区为研究对象,结合该流域DEM与实测水系,运用凹陷化算法提取河网,并通过人工修正以实现河网的合理概化,使提取的河网水系等要素基本符合里下河区域实际水文状况。然后构建里下河平原区SWAT水文模型,以2006-2011年水文站汛期流量数据为基础,对该区域水文进行模拟,并对模型的适用性做了评估。研究结果表明:影响该区域径流模拟前三位敏感参数分别为径流曲线数(CN2)、土壤蒸发补偿系数(ESCO)和基流α系数(ALPHA_BF);经模型参数率定与验证检验,得出汛期内月模拟流量与实测流量的相关系数R2、Nash-Sutcliffe效率系数ENS均在0.80以上,相对误差|Re|均小于5.0%,表明SWAT模型在里下河平原区月径流模拟较好的适用性。该研究探讨了平原区分布式水文模型的应用,可为里下河平原区未来的水资源规划管理提供了科学参考依据。

The plain region of Lixiahe is characterized by flat topography,complicated river networks,and complex and uncertain hydrologic processes. Here,we extracted the sunken river network using the algorithm of ‘burn in' artificially modified to achieve a reasonable generalization through Arc GIS,so that the extraction of the reaches and the sub-basins in the Lixiahe area could agree with actual hydrologic processes in the area. We built SWAT hydrological models for the Lixiahe plain successfully,and simulated the hydrologic process of the Lixiahe plain based on hydrological stations from 2006 to 2011 traffic data in the flood season. The results showed that the impact of the top three regional runoff simulation parameters are sensitive to the runoff curve number(CN2),soil evaporation compensation factor(ESCO)and base flow coefficient α(ALPHA_BF). After model parameter calibration and validation tests,we found that monthly correlation coefficient during the flood season with flow simulation and measured flow simulation R2;Nash efficiency coefficients ENS were more than 0.80;and the relative error |Re|were less than 5.0%;therefore,the simulation of total monthly flow results have good applicability to simulate monthly runoff using the SWAT model in Lixiahe. This lays the foundation for the application of the area based on distributed hydrological modeling so as to provide a scientific reference for future water resource planning and management in the Lixiahe plain area.