A Statistical Hydrological Model for Yangtze River Watershed Based on Stepwise Cluster Analysis

Streamflow prediction is one of the most important topic in operational hydrology. The responses of runoffs are different among watersheds due to the diversity of climatic conditions as well as watershed characteristics. In this study, a stepwise cluster analysis hydrological (SCAH) model is developed to reveal the nonlinear and dynamic rainfall-runoff relationship. The proposed approach is applied to predict the runoffs with regional climatic conditions in Yichang station, Hankou station and Datong station over the Yangtze River Watershed, China. The main conclusions are: (1) the performances of SCAH in both deterministic and probabilistic modeling are notable.; (2) the SCAH is insensitive to p levels in monthly runoff prediction with robust structural, (3) in term of the case study in Yangtze River watershed, it can be inferred that the water resource in the lower reaches of the Yangtze River is seriously affected by incoming water from the upper reaches according to the strong correlations. This study has indicated that the developed statistical hydrological model SCAH approach can characterize such hydrological process complicated with nonlinear and dynamic relationships and provide satisfactory predictions. Flexible data requirements, quick calibration, and reliable performances make SCAH be a more appealing tool in revealing rainfall-runoff relationships.