Differences in the Intensity of Surface and Canopy-Layer Urban Heat Islands in Europe

The satellite-derived surface urban heat island is often not properly distinguished from the in situ derived canopy-layer urban heat island in the field of remote sensing of urban climates. Yet, some studies have investigated their differences, focusing mainly on single urban agglomerations and limited periods. Moreover, a comparison of different urban heat island studies is impeded due to nonstandardized urban classifications. This study analyzes the impact of different definitions of urban and rural on the urban heat island intensity and investigates the differences between the surface and the canopy-layer urban heat island intensities for a long time period and a large number of urban agglomerations compared with most other urban heat island studies. We compare the urban heat island intensities for 91 European urban agglomerations using moderate-resolution imaging spectroradiometer land surface temperature and Netatmo-derived citizen weather station air temperature for the summers (June-August) of 2019-2021. The results show the following: First, the urban heat island types and their differences are significantly impacted by the underlying urban and rural classification. The findings underline that the studies based on different classification schemes should not be compared without considering the methodology. Second, the surface urban heat island is an all-day phenomenon, while the canopy-layer urban heat island is a nocturnal phenomenon and there is a substantial difference between both types during the diurnal cycle. Thus, land surface temperature should not be used as a proxy for the canopy-layer urban heat island. Third, the urban heat island does not indicate absolute temperatures. Intraseasonal changes in urban heat island intensities and urban temperatures can even have a different temporal evolution.