The return period analysis of heavy rainfall disasters based on copula joint statistical modeling

This paper analyzes the multivariate and spatial distribution of heavy precipitation disasters and proposes a method for estimating disaster risk using a joint statistical model. We tested the model with hourly precipitation data from 122 meteorological stations in Shandong from 1990 to 2023. Different marginal distribution functions were used to fit precipitation duration and amount. A Copula joint distribution model established relationships between these variables to analyze heavy precipitation recurrence periods and disaster characteristics. Compared to univariate approaches, the Copula function more reasonably simulates natural disaster occurrence. The joint return period (JRP) estimated by the Copula function reveals that the JRP of 1-hour heavy rainfall is 89% higher than 6-hour rainfall, indicating significantly increased risk from short-term heavy rainfall in Shandong. This method provides a more scientific description of heavy precipitation disaster risk in different scenarios, particularly for short-term events, offering a robust foundation for disaster prevention planning and risk management.