Challenges in the Use of Satellite-Derived Thermal Data in Urban Heat Island (UHI) Studies: A Case Study of Braganca, Portugal

The urbanization process causes various socio-environmental impacts, as it incorporates anthropogenic elements with low albedo, affecting the local microclimate and contributing to the formation of Urban Heat Islands (UHIs). The use of thermal data obtained from sensors onboard Earth Observation satellites is a methodology for studying UHI, with the most common satellites being Landsat (since Landsat 4, with Landsat-8 and 9 as the current operational models, equipped with Thermal Infrared Sensors (TIRS and TIRS-2), respectively); Aqua (Moderate Resolution Imaging Spectroradiometer (MODIS) sensor); and Terra (MODIS and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) sensors). Each sensor has its spatial, spectral, radiometric and temporal resolutions, which must be considered to ensure more accurate results. The objective of this study was to analyze the TIRS, TIRS-2, MODIS, and ASTER sensors in terms of their resolutions, limitations, challenges, and opportunities in Bragança (Portugal). We divided the data by seasonality and used Google Earth Engine to calculate the Land Surface Temperature (LST) for each image, comparing the results with in situ Air Temperature (Ta) data. As a result, all LSTs showed a strong/very strong correlation with Ta. Although the MODIS sensor has the higher temporal resolution, its low spatial resolution (1 km per pixel) was a limitation, considering the heterogeneity in Land Use and Land Cover (LULC) and Bragança´s area. ASTER provided a more detailed resolution (90m), but the limited number of images available for Bragança was challenging. TIR and TIR-2 provided better spatial resolution compared to MODIS (the data is collected at 100m and resampled to 30m), but it was also unable to detail LULC, especially in transition areas. For more robust studies, we suggest combining data from multiple sensors, taking advantage of the complementary benefits of their different resolutions and, eventually, incorporating higher-precision data, such as Unmanned Aerial Vehicles (UAVs).