A statistical–dynamical downscaling methodology for the urban heat island applied to the EURO-CORDEX ensemble

Abstract Regional Climate Models (RCMs) are the primary climate information available to public stakeholders and city-planners support local adaptation policies. However, with resolution in order of ten kilometres, RCMs do not explicitly represent cities their influence on (e.g. Urban Heat Island; UHI). Downscaling methods required bridge gap between city scale. A statistical–dynamical downscaling methodology is developed quantify UHI Paris (France), based a Local Weather Types (LWTs) classification combined short-term high-resolution (1-km) urban simulations. The daily near-surface temperature amplitude, specific humidity, precipitation, wind speed direction simulated by used for LWTs attribution. time series associated randomly selected days mesoscale atmospheric model Meso-NH coupled canopy Town Energy Balance calculate corresponding successive LWTs. applied EURO-CORDEX ensemble driven ERA-Interim reanalysis, evaluated 2000–2008 period against station observations 2.5-km reanalysis. dynamical simulations slightly underestimate overestimate minimum maximum air respectively, but capture intensity biases tenth degree. show significant differences variables attribution, seasonal LWT frequencies captured. Consequently, reconstructed fields maintain small greatly improves compared raw data RCMs.