Time of emergence of compound events: contribution of univariate and dependence properties

Abstract. Many climate-related disasters often result from a combination of several climate phenomena, also referred to as “compound events’’ (CEs). By interacting with each other, these phenomena can lead huge environmental and societal impacts, at scale potentially far greater than any events could have caused separately. Marginal dependence properties the forming CEs are key statistical characterising their probabilities occurrence. In this study, we propose new methodology assess time emergence CE probabilities, which is critical for mitigation strategies adaptation planning. Using copula theory, separate quantify contribution marginal overall probability changes multivariate hazards leading CEs. It provides better understanding how variables evolve contribute change in occurrences. For illustrative purposes, applied over 13-member multi-model ensemble (CMIP6) two case studies: compound wind precipitation extremes region Brittany (France), frost occurring during growing season preconditioned by warm temperatures (growing-period frost) central France. extremes, results show that emerge before end 21st century six models CMIP6 considered. growing-period frosts, significant detected 11 models. Yet, be very different one hazard another, model another. Depending on CE, some place strong importance both changes. These highlight considering properties, well inter-model variability, future risk assessments related