Heat and Cold Mortality: Adaptation and Warnings

The aim of this study is twofold, first we estimate the effects of temperature on mortality distinguishing by several population groups. Second, we explore the mitigative effects of issuing a heat warning on the heat-mortality relationship. To this end, exploiting the errors in weather forecasts, we manage to compare days with similar meteorological conditions but different heat warning statuses. We employing a fixed-effects model applied to 1323 German sub-national areas between 2018 and 2023. We instrument the likelihood of issuing a warning with the error in wind forecast leveraging on the fact that underestimating wind leads to an underestimation of perceived temperatures (PT) in the hot rage and in turn to a lower probability of issuing heat warnings. We find an optimal temperature range of [10,15) PT C° with mortality increasing faster as temperature become hotter compared to colder. Nonetheless the effects are highly heterogeneous across age groups and gender, with older people more sensitive both to cold and hot temperatures and female more sensitive to cold ones. Moreover, we observe that the effects of a hot day least for at least 3 days and that consecutive hot days are up to 45% more lethal than isolated ones. Furthermore, areas that are on average warmer are better adapted to heat but did not improve their adaptation significantly between 2005 and 2023. Finally, heat impacts on mortality are moderated by the presence of heat warnings. The contribution of our results is threefold. First, we identify the most vulnerable groups both to heat and cold. Second, we reveal the crucial role of meteorological alerts in enhancing people’s adaptive capacity to climate change. Third, we underscore the importance of accurate weather forecasts for public safety.