Assessing Urban Heat Island in the European Union capitals by means of MODIS and Landsat data

The present work aims at estimating the Urban Heat Island effect in the capital cities of European Union (EU) countries. The main objectives of the research were: a) the estimation of the magnitude of UHI in the EU capitals and b) the evaluation of possible relationships with latitude and / or population density. To address these objectives two independent datasets were used that provide Land Surface Temperature estimates over the past eleven years (2013 to 2024), namely the MODIS Terra Land Surface Temperature product and the Landsat 8 Surface Temperature. These two LST products have differences in their spatial and temporal resolution, but were selected due to small difference in daytime acquisition time. The methodology focused on the construction of surface temperature time series from the above data sources for three distinct land cover classes, i.e. urban, forest, and cropland areas, retrieved from the CORINE 2018 land cover dataset. The time series from both MODIS and Landsat 8 were evaluated for their difference in their mean values in the three land cover types. All examined EU capitals were found to have statistically significant higher surface temperatures compared to their surrounding forest and cropland areas. MODIS and Landsat yielded comparable results, whereas the spatial pattern of UHI of EU capitals and its correlation with population density were evaluated highlighting the characteristics of increasing temperatures in large EU cities.