本项目以汉江流域为研究对象，以提高洪水预报精度和延长洪水预报预见期为研究目标，研究和实现基于概率预报的中尺度WRF模式和洪水预报模型耦合。针对耦合研究中存在的关键问题，项目将收集整理汉江流域1961-2011年典型暴雨洪水场次资料；分析研究不同水平分辨率对WRF模式降水预报影响，确定合适的水平分辨率，并建立VIC分布式模型和新安江模型；通过扰动物理方案，实现WRF降水集合预报；应用贝叶斯模型平均法实现耦合过程中降水概率预报和洪水概率预报。项目研究将促进和完善水文科学中洪水预报理论与方法的发展，为防洪决策提供科学可靠的技术支持。

To enhance the precision of flood forecasting and extend the lead-time of flood forecasting, a coupling system between a mesoscale numerical model (Weather Research Forecast, WRF) and flood forecasting models will be studied and realized based on the probabilistic forecasting method in the Hanjiang basin. To solve some key problems in the coupling process, the following issues will be proceed in next four years: (1) The representative storm and flood records will be collected and analyzed during the historical period of 1961-2011; (2)The impact of horizontal resolutions on predictions of precipitation will be analyzed by comparing three different horizontal resolutions; (3) A distributed hydrological model-Variable Infiltration Capacity and a concept hydrological model-Xin-anjiang hydrological model will be calibrated in Hanjiang basin; (4) To reduce the uncertainty of initial values and itself uncertainty of WRF, an ensemble forecasting for precipitation will be made by choosing different physics parameterization schemes; (5) Bayesian Model Averaging method will be used to transform the deterministic forecasting of precipitation and flood to the probabilistic forecasting in the coupling process. This project will promote the improvement of the flood forecasting theory and method in hydrological science, and also provide technique support to the flood prevention decision.