Methodology

Matching days come from ERA5 hourly fields for 2021-01-01 through 2025-12-31 (1,826 days, including 2024-02-29), aggregated to local solar days at 0.25° (~28 km). Local day for each cell runs midnight to midnight at UTC offset round(lon / 15) hours.

This is a recent-weather sample, not a 30-year climate normal. Microclimates (San Francisco fog, coastal breezes) average out at this grid scale. Antarctica (south of 60°S) is excluded.

Map criterion: daily maximum Steadman/BOM apparent temperature in shade, calculated hourly from 2 m air temperature and dewpoint plus 10 m wind speed. It estimates what the weather feels like to an appropriately dressed adult walking outdoors. It does not include direct sunlight, which can feel hotter. Optional criteria cap cloud cover, precipitation and wind gust.

Contains modified Copernicus Climate Change Service information (2026). Neither the European Commission nor ECMWF is responsible for any use of this information. ERA5 via NSF NCAR Curated ERA5 on AWS Open Data (DOI 10.5065/BH6N-5N20). Urban-centre boundaries, names and populations: European Commission GHS-WUP-MTUC R2025A ; other places: GeoNames (both CC BY 4.0). Basemap: OpenFreeMap / OpenMapTiles / OpenStreetMap contributors.

Rankings treat each GHSL urban centre as one place. Its population is the complete 2025 urban-centre population, and its weather score is an area-weighted average across every ERA5 land cell intersecting the urban boundary. Municipalities inside the same urban centre are not listed separately. Places outside GHSL urban centres remain standalone GeoNames entries only when their population is at least 10,000. Smaller municipalities still contribute to their urban centre.