Urban multi-model climate projections of intense heat in Switzerland

This paper introduces a straightforward approach to generate multi-model climate projections of intense urban heat, based on an ensemble of state-of-the-art global and regional climate model simulations from EURO-CORDEX. The employed technique entails the empirical-statistical downscaling method quantile mapping (QM), which is applied in two different settings, first for bias correction and downscaling of raw climate model data to rural stations with long-term measurements and second for spatial transfer of bias-corrected and downscaled climate model data to the respective urban target site. The resulting products are daily minimum and maximum temperatures at five urban sites in Switzerland until the end of the 21st century under three emission scenarios (RCP2.6, RCP4.5, RCP8.5). We test the second-step QM approach in an extensive evaluation framework, using long-term observational data of two exemplary weather stations in Zurich. Results indicate remarkably good skill of QM in present-day climate. Comparing the generated urban climate projections with existing climate scenarios of adjacent rural sites allows us to represent the urban heat island (UHI) effect in future temperature-based heat indices, namely tropical nights, summer days and hot days. Urban areas will be more strongly affected by rising temperatures than rural sites in terms of fixed threshold exceedances, especially during nighttime. Projections for the end of the century for Zurich, for instance, suggest more than double the number of tropical nights (Tmin above 20 degrees C) at the urban site (45 nights per year, multi-model median) compared to the rural counterpart (20 nights) under RCP8.5.

Automated summary

This paper introduces a straightforward approach to generate multi-model climate projections of intense urban heat, based on an ensemble of state-of-the-art global and regional climate model simulations from EURO-CORDEX. The results suggest that urban areas will be more strongly affected by rising temperatures than rural sites, especially during nighttime.