Derivation of High-resolution Local Thermal Conditions from Satellite-based Data

The present study proposes a statistical approach for obtaining spatially detailed near-surface thermal information regarding the urban domain. This is accomplished by defining a workflow, which considers location-specific attributes of the built environment and which can be applied by utilizing publicly accessible data via remote sensing and GIS sources. To this end, we employ a large set of ground-based and remotely sensed observations from the years 2000 to 2015 to realize these objectives. The results reveal a relationship with an adjusted R-2 of 0.92 for day-time and 0.98 for night-time conditions. Air temperature estimates utilizing the proposed approach yield RMSEs of 2.26 K and 1.40 K for day-time and night-time inquiries respectively.