Abstract Climate change is likely to increase the frequency, severity, and duration of heat waves in many countries. To plan mitigation, adaptation, and resilience strategies, it is necessary to quantify heat wave risk at both the local level and the country level. A new, more granular methodology is proposed in order to integrate the impact of heat waves in hexagonal France on mortality with a short-term stress scenario. Based on open data and reproducible methodology, the approach can be used as a starting point to investigate other effects, such as urban heat islands. The present application is based on in situ observational weather data and environmental vulnerability data to construct adapted geographical clusters without relying on the administrative division of the territory. Excess mortality is modeled as a function of weather using machine learning. Using recent knowledge of climatology, we construct extreme weather scenarios to calculate a shock to mortality. Short-term shocks are compared, and their respective merits are discussed. The methodology has been shown to generate mortality shocks up to five times greater than those estimated by the French regulatory authority.
Raynal et al. (Fri,) studied this question.