To unveil the risk patterns of multi encountering compound hydrogeological disasters under changing environments, this study employed a bivariate Copula joint probability function and constructs a compound disaster encounter combination probability distribution model with a vine structure. The model underwent precise fitting using three fit quality test methods: log-likelihood (log-lik), Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC) for optimal selection. For various disaster occurrence scenarios, this research calculated the joint cumulative probability and Kendall joint recurrence period of compound disaster encounter combinations. Taking the compound disaster encounter events at Foshan station (rainfall-landslide (debris flow)-flash flood) and Denglongshan station (typhoon-storm surge-flood) in the Guangdong-Hong Kong-Macao Greater Bay Area as cases, probability distribution models for three types of disaster combinations were constructed. The results indicate that with the increase in the extremity of individual disaster scenarios, the joint cumulative probability and recurrence period of the compound disaster encounter combinations correspondingly rise, thereby revealing the risk patterns of compound disasters. This study can provide theoretical support for estimating the risk of encountering combinations of multi encountering compound disasters and has reference significance for the risk estimation of compound hydrogeological disasters.