What are the potential drivers of urban heat island changes in European cities in future regional climate simulations?

As a result of climate change, cities, like natural areas, are already warming and will continue to do so, even without taking urbanization into account. The urban heat island (UHI) effect, i.e. the difference in night-time air temperature between urban areas and their surroundings, poses major problems for inhabitants, for example in terms of heat stress.

Climate change projections are usually derived from global climate models (GCMs) downscaled to the local scale using statistical tools, or from regional climate models (RCMs). Due to their horizontal resolution, which is often too coarse, RCMs have historically been unable to adequately represent most urban areas, so little is known about projected changes in UHI in the future. A few studies have investigated the projected evolution of UHIs in the context of climate change, using different approaches, from GCMs and RCMs to high-resolution land surface models, but with little consistency between studies and great sensitivity to the city analyzed, its climate and especially the downscaling approach used.

Thanks to increasing computational resources and model development, RCMs are now able to achieve horizontal resolutions in the order of a few kilometers and can be coupled with various urban parametrization to improve the representation of cities in climate change projections.

Here we use GCM-driven RCM simulations from the EURO-CORDEX initiative (12.5 km) and the CORDEX Flagship Pilot Study on Convection (3 km) to study the future evolution of the UHIs of several European cities at the end of the century (2090-2099) under a scenario of very high greenhouse gas emissions (RCP8.5) and in which the cities are fixed in their historical states.

We study the potential atmospheric variables driving the UHI over the historical period and their evolution in the future, and compare these relationships between climate models of varying resolution and using different urban parametrization.