To identify the rainfall-flood relationship and reveal the probability of storm-flood encounters in the Yujiang river section of the upper Pinglu Canal, this study adopted a variety of Copula functions to construct a joint distribution model for joint storm-flood modeling. Meanwhile, the graphical evaluation method, the squared Euclidean distance, and the OLS criterion were applied to optimize the best Copula model. Finally, the probability of rainfall-flood encounters and the top likelihood of flooding under various rainfall scenarios were evaluated and investigated. The results show that the Clayton-based model is the best-fitting Copula function for the joint storm-flooding distribution, and the squared Euclidean distance and the OLS criterion reach 0.120 and 0.044, respectively; the risk of combined storm-flood is significantly, and the probability of joint flooding is 1.5 times higher than that of a single storm (or flooding) occurs;? the top probable flooding of the corresponding conditions under the typical rainfall scenarios of 0.5%, 1%, and 2% are 0.6%, 1.3%, and 2.6%, respectively. Flood control in the upper Yujiang River section of the Pinglu Canal is a difficult challenge.