Modeling of large rainfall events plays an important role in water resources and floodplain management. Rainfall is resulted from complex interactions between climate factors (air moisture, temperature, wind speed, etc.) and land surface (topography, soil, land cover, etc.). Therefore, deriving accurate areal rainfall is not only relied on atmospheric boundary conditions, but also on the reliability and availability of soils, topography, and vegetation data. Consequently, uncertainties in both atmospheric and land surface conditions contributes to rainfall model errors. In this study, a blended technique combining dynamical and statistical downscaling has been explored. The proposed downscaling approach uses input provided from three different global reanalysis data sets including ERA-Interim, ERA20C, and CFSR. These reanalysis atmospheric data are hybridly downscaled by means of the Weather Research and Forecasting (WRF) model, which is followed by the application of an artificial neural network (ANN) model to further downscale the WRF output to a finer resolution over the studied region. The proposed technique has been applied to the third largest river basin in Vietnam, the Sai Gon–Dong Nai Rivers Basin; and the calibration and validation show the simulation results agreed well with observation data. Results of this study suggest that the proposed approach can improve the accuracy of simulated data, as it merges model simulations with observations over the modeled region. Another highlight of this approach is inexpensive computational demand on both computation times and output storage.

大降雨事件建模在水资源和洪泛区管理中发挥着重要作用。降雨是气候因素（空气湿度、温度、风速等）和地表（地形、土壤、土地覆盖等）之间复杂相互作用的结果。因此，推导出准确的区域降雨量不仅依赖于大气边界条件，还依赖于土壤、地形和植被数据的可靠性和可用性。因此，大气和地表条件的不确定性会导致降雨模型误差。在这项研究中，探索了一种结合动态和统计降尺度的混合技术。提议的降尺度方法使用三个不同的全球再分析数据集提供的输入，包括 ERA-Interim、ERA20C 和 CFSR。这些再分析大气数据通过天气研究和预报 (WRF) 模型混合缩小，然后应用人工神经网络 (ANN) 模型进一步缩小 WRF 输出到研究区域的更精细分辨率。该技术已应用于越南第三大河流域西贡-同奈河流域；校准和验证表明模拟结果与观测数据吻合良好。这项研究的结果表明，所提出的方法可以提高模拟数据的准确性，因为它将模型模拟与对建模区域的观察相结合。这种方法的另一个亮点是对计算时间和输出存储的廉价计算需求。