Urban Heat Islands (UHIs) are increasingly posing critical challenges to urban environments and human well-being. In response, we propose a novel methodology to identify Urban Heat Island Hotspots (UHIHs) to address the urgent need for effective management and mitigation strategies. Our research introduces the innovative concept of direct UHIHs detection and ranking, providing a framework for urban planning stakeholders to prioritise areas for regeneration based on UHIH severity.
In the second half of the 20th century, the implementation of the Soviet model in Eastern Europe induced substantial changes in the thermal environment of socialist cities.
Forced industrialization determined the urban expansion at a rate never seen before, with consequences on the quality of the environment, including the urban climate.
A sudden increase in heat load was experienced by Eastern European cities, with the construction of numerous high-density blocks of flats meant to sustain the massive urbanization campaigns initiated by the communist regimes.
Urban Heat Islands (UHIs) are increasingly posing critical challenges to urban environments and human well-being.
The research presents a comprehensive approach to the automated identification of UHI hotspots, addressing the pressing need for effective management and mitigation
strategies.
We propose:
• A new concept: hotspot ranking;
• A new algorithm for hotspot detection;
• A new tool for automatic detection and ranking hotspots.
Thermal infrared (TIR) satellite imagery is essential for mapping global land surface temperature (LST).
The spatial resolution of TIR bands is typically coarser than shorter wavelength multispectral bands, limiting the LST resolution to around 100 meters.
In complex urban environments, a finer LST spatial pattern is often needed. While NASA’s TIRS sensor onboard Landsat 8, Landsat 9, and ASTER onboard Terra, provide higher thermal spatial resolution, ESA’s Sentinel-2 offers finer spatial resolution in multispectral imagery (10 meters) but lacks TIR scanning.