Comparison between Air Temperature and different Land Surface Temperature satellite products for the City of Córdoba, Argentina

In the current context of climate change, the use of remote sensing in urban applications such as Urban Heat Islands risk mapping, provides valuable and timely information to address this challenge. By better understanding their environment and impacts, cities can work towards a more sustainable and resilient future. In recent years, platforms and code processors such as Google Earth Engine (GEE) have offered the possibility to visualize and work with different geoprocessed ready-to-use data. Land surface temperature (LST) is one of those products and it is a key parameter in the physics of land surface processes on regional and global scales. However, it is necessary to assess their performance before using them for studies such as social vulnerability, resilience, and risk management. In this study, different medium and high spatial resolution satellite products available in GEE were compared with 3-year data of 10 meteorological stations strategically located in the Metropolitan Area of Córdoba city, Argentina. The time series were compared using different statistical parameters, in which linear regressions were used and compared with a train and validated semi-empirical model. The results indicate that the predictive model outperforms the original LST data for all LST products analyzed. Thus, it is advisable to adjust the model before using the satellite-ready-to-use product.