华中多雨人口密集型流域洪水预报

Flooding prediction for a rainy,dense-population river basin of central China

建立一个集总式水文模型,即基于蓄满产流的三水源新安江模型,模拟抚河流域1981—1995年的18次暴雨洪水过程;采用马斯京根分段连续算法计算河流洪水演算和水流过程线;然后将所有水流过程线线性叠加,确定水流和洪水过程;利用日尺度数据和洪水频率数据对模型进行参数率定。结果表明:该模型在现场洪水模拟中的平均确定系数为0.911,模型校准的平均径流深度误差为4.73%,验证的平均径流深度误差为8.21%,三水源新安江模型可以被应用于抚河流域的洪水预报工作中。该研究为我国中部多雨地区的洪水预报研究提供了有益的参考。

Watershed runoff forecasting has been a research focus on hydrological model and flooding prediction.In China,Xinanjiang(XAJ)model is a widely used for flood forecasting,which is of significance to areas with booming economy,dense population,and high flooding risks,such as those located in central China.The Fu River basin located in Jiangxi Province is a rainy area with abundant precipitation(average annual rainfall is 1761 mm).A lumped hydrological model,three-source XAJ model based on excess storage runoff generation,was established to simulate 18 rainstorms and floods in Fu River basin from 1981 to 1995.Muskingen′s piecewise continuous algorithm was used to calculate river flood routing and the flow process line.The processes of the flow and flood were determined by linear superposition of all the flow processes.Parameters were calibrated by daily data and frequency data.Results showed that the average deterministic coefficient of the model in the field flood simulation was0.911,the average error of runoff depth for model calibration was 4.73%,and the average error of runoff depth for validation was 8.21%.Therefore,the three-source XAJ model could be used as a useful forecasting model in the flood forecasting system of Fu River basin.The useful references were provided for flood forecasting research in the rainy central China.