Projections of heat waves with high impact on human health in Europe

•We study heat waves with high impact on human health in Europe.•Strong correlations are found between heat wave amplitude and mortality excess.•A quantile–quantile adjustment is applied to correct biases in RCMs.•Projections suggest increases in heat wave frequency, duration and amplitude.•Projections indicate a northward future extension of heat wave attributes.

Climate change will result in more intense, more frequent and longer lasting heat waves. The most hazardous conditions emerge when extreme daytime temperatures combine with warm night-time temperatures, high humidities and light winds for several consecutive days. Here, we assess present and future heat wave impacts on human health in Europe. Present daily physiologically equivalent temperatures (PET) are derived from the ERA-Interim reanalysis. PET allows to specifically focus on heat-related risks on humans. Regarding projections, a suite of high-resolution regional climate models – run under SRES A1B scenario – has been used. A quantile–quantile adjustment is applied to the daily simulated PET to correct biases in individual model climatologies and a multimodel ensemble strategy is adopted to encompass model errors. Two types of heat waves differently impacting human health – strong and extreme stress – are defined according to specified thresholds of thermal stress and duration. Heat wave number, frequency, duration and amplitude are derived for each type. Results reveal relatively strong correlations between the spatial distribution of strong and extreme heat wave amplitudes and mortality excess for the 2003 European summer. Projections suggest a steady increase and a northward extent of heat wave attributes in Europe. Strong stress heat wave frequencies could increase more than 40 days, lasting over 20 days more by 2075–2094. Amplitudes might augment up to 7 °C per heat wave day. Important increases in extreme stress heat wave attributes are also expected: up to 40 days in frequency, 30 days in duration and 4 °C in amplitude. We believe that with this information at hand policy makers and stakeholders on vulnerable populations to heat stress can respond more effectively to the future challenges imposed by climate warming.