This study analyzed the variation of extreme precipitation by introducing Generalized Additive Models in Location, Scale, and Shape (GAMLSS) Model in the nonstationary frequency analysis based on the precipitation extreme series in the first rainy season (April, May, June) in the period of 1960-2012 derived from 79 meteorological stations in the southeast coastal region of China which covers Shanghai, Zhejiang, Fujian, Guangdong and Hainan. The greenhouse gas emissions index represented by human activities and comprehensive climatic index were built by principal component analysis (PCA). Simultaneously, contribution analysis method and GAMLSS model are applied to identify the impact of climate change and human activities on precipitation extremes considering both factors. The results show that: (1) The areas with significant increase and strong enhanced variation of precipitation extremes are mainly located in the northeast of Zhejiang, centralnorthern and western coastal Guangdong, and eastern Fujian. (2) There is a significant correlation between human activity increase and precipitation extreme variability, especially in the developed areas. The contribution of human activity to precipitation extremes in the stage of rapid socio- economic development (1986- 2012) is much greater than that in the stage of slow socio- economic development (1960- 1985), especially in Zhejiang and Guangdong provinces. (3) The northeast of Zhejiang, much of Fujian and the central-northern and western coastal Guangdong are high risk areas in terms of extreme precipitation. Thus, we should strengthen the risk prevention for extreme precipitation and its secondary disasters.