Abstract: The encounter of main stream and tributary floods has made the river's flood control situation more severe. It is of great significance to analyze the encounter situation of main and tributary floods in order to make full use of the flood control capacity of the cascade reservoirs in the river basin and reduce the flood loss in the downstream. At present, most of the flood encounter analysis is based on the measured data, and copula function has been applied to it because of its flexibility and effectiveness. The purpose of this study is to rationally formulate joint flood control dispatching methods in the various river basins of the Yangtze River and reduce the pressure of flood control in the Chuanjiang River In this paper, we used the multi-year actual measured data of Zhutuo Station, Cuntan Station on the main stream of Chuangjiang River, Beibei Station on the Jialing River and Wulong Station on the Wujiang River. Through Von Mises mixture distribution and P-Ⅲ distribution, this article aims to fit the annual maximum flood occurrence time and flood magnitude. In addition, this paper introduces Archimedean copula function to build joint distribution between main stream and tributary and between tributaries to analyze the possibility of flood encounter in the main and tributaries of the Yangtze River. The results show that: the probability of occurrence of once-in-a-thousand floods encounter in the main stream and tributaries of the Chuanjiang River is low, the probability of the occurrence of the floods encounter between the main stream of the Chuanjiang River and the Jialing River is relatively high and the magnitude is large, while the probability of the occurrence of the encounter between the Wujiang River, the main stream of the Chuanjiang River and the Jialing River is relatively low and the magnitude is small.