HSPF在热带沿海流域水文模拟中的应用

[目的] 探究HSPF模型在热带沿海流域的适用性和不确定性，为该模型在不同流域和地区的应用提供科学参考。[方法] 以位于热带沿海地区的三亚河流域为例，建立HSPF模型，选取2017-2019年径流量对模型进行率定和验证；通过Morris筛选法分析了水文过程中8个参数的敏感度，并与国内外研究不同流域的参数敏感度相对大小进行对比；同时利用MC-LHS方法对不同降雨量下模型的不确定性进行分析。[结果] HSPF模型能够很好地模拟研究区域实际的水文过程，率定期和验证期的NSE分别为0.93和0.98，相对误差（R_e）分别为0.87%和0.21%；地下水日消退系数为最敏感参数，而下层土壤蒸发系数和地下水出流中进入深层的比例对径流模拟的影响几乎可以忽略。[结论] 参数敏感度相对大小体现空间差异性。模型模拟的不确定性与降雨量之间的相关性明显。降雨量越大，模型模拟的不确定性和置信区间就越大，模型就越不稳定。

Aplication of HSPF in Hydrological Simulation in Tropical Coastal Watershed

[Objective] The applicability and uncertainty of HSPF model in tropical coastal basins was explored in order to provide scientific reference for the application of HSPF model in different basins and regions.[Methods] Taking the Sanya river basin located in the tropical coastal region as a study area, a HSPF model was established, and the flow from 2017 to 2019 was selected to calibrate and verify the model. The sensitivity of 8 parameters in the hydrological process was analyzed by Morris screening method, and compared with the relative sensitivity of parameters in different watersheds studied at home and abroad. Meanwhile, the uncertainty of the model under different rainfall conditions were analyzed by MC-LHS method.[Results] The HSPF model could simulate the actual hydrological process in the study area very well. The NSE of the periodic and validation period was 0.93 and 0.98 respectively, and the R_e was 0.87% and 0.21%, respectively. Among which, base groundwater recession was the most sensitive parameter, while the influence of lower zone evapotranspiration parameter and proportion of groundwater flow into deep was almost negligible.[Conclusion] The relative sensitivity of the parameters reflected the spatial differences. The uncertainty of model simulation is significantly correlated with rainfall, the greater the rainfall is, the greater the uncertainty and confidence interval of the model simulation are, and the more unstable the model is.