Trends in temperature-mortality projections across urban and rural regions under different climate change scenarios in Switzerland

Abstract

Temperature vulnerability tends to be heterogenous between urban and rural regions, with usually larger risks in the former. However, it is unclear whether this pattern will persist under different climate change scenarios. Climate change combined with population growth, ageing and urbanization is expected to amplify the current temperature-related mortality burden, particularly since these drivers will follow diverse trajectories across urban and rural areas in the future. This study aimed to compare the impact of climate change regarding changes in heat- and cold-related mortality under various warming scenarios across urban/rural districts in Switzerland, while also accounting for changes in population demographics. We first estimated the district-age-specific (<75/>=75years) temperature-mortality association between 1990-2010 in Switzerland using a two-stage time series design with distributed-lag nonlinear models. We then projected heat- and cold-related mortality under 1.5C, 2.0C, 3.0C and 4.0C warming scenarios using downscaled temperature projections under RCP8.5 while accounting for the Shared Socio-economic pathway (SSP5) population development storyline. In Switzerland, heat- and cold-related mortality will increase by 2,420(95%CI:1654;3,884) and 993(82;1883) deaths /year, respectively, under 4.0C warming (vs. 1990-2010). Urban areas are more affected by both heat and cold and this pattern appears to be amplified under warmer scenarios. In particular, per 100 heat-related deaths in rural areas, there will be an increase from 362 to 391 deaths in urban areas, with similar results for cold (from 262 to 286 per 100 deaths in rural areas), indicating a slight increase in vulnerability to climate change in urban regions in the near future. Our preliminary findings suggest that climate change will increase current heat and cold-related mortality in Switzerland, and moreover, urban areas will continuously more be disproportionately affected. Further analyses will aim to account for urbanization projections and disentangle the contribution of the different drivers across urban and rural regions.