Zhou, Z.; Liu, S.; Hua H.; Chen, C.S.; Zhong G.; Lin, H., and Huang, C.W., 2014. Frequency analysis for predicting extreme precipitation in Changxing station of Taihu Basin, China.

Rainfall induced flooding is one of the most severe natural disasters in coastal regions. In recent years, along with global warming and sea level rising, extreme hydrological events, such as extreme precipitation, are of high occurrence. Meanwhile, rapid urbanization makes the urban environment transformed dramatically and results in additional flood ventures. Taihu Basin, located in Yangtze Delta, is the richest basin in China and flood control planning in this region is of high significance. Therefore, precipitation analysis, as a basic work of flood design, should be accurate and precise. In this study, based on precipitation data at Changxing Station of Taihu Basin, precipitation frequency analysis using mixed methods is performed. Two of the most applied distribution models, Pearson-III (P3) and Generalized Extreme Value (GEV), are investigated. For parameter estimation method of probability distribution functions, maximum likelihood estimation (MLE) and L-moments (LM) are used. In addition, seeking-matrix curve fitting based on conventional moments (CM) is also investigated to compare the calculation results. The performance of mixed methods is tested by two classical goodness-of-fit tests, Chi-Square test and K-S test. Consequently, GEV distribution model based on LM is evaluated to be the best fitting model for identifying and predicting future precipitation occurrence. So precipitation estimations from different return periods at Changxing Station are identified. This study is a new attempt to precipitation frequency analysis in the stations of Taihu Basin and the result can provide a reference for flood risk and water resource management in Taihu Basin and even in more other regions in China.