Exploring the land-use urban heat island nexus under climate change conditions using machine learning approach: A spatio-temporal analysis of remotely sensed data.

Urbanization strongly correlates with land use land cover (LULC) dynamics, which further links to changes in land surface temperature (LST) & urban heat island (UHI) intensity. Each LULC type influences UHI differently with changing climate, therefore knowing this impact & connection is critical. To understand such relations, long temporal studies using remote sensing data play promising role by analysing the trend with continuity over vast area. Therefore, this study is aimed at machine learning centred spatio-temporal analysis of LST and land use indices to identify their intra-urban interaction during 1991-2021 (summer) in Imola city (specifically representing small urban environment) using Landsat-5/8 imageries. It was found that LST in 2021 increased by 38.36% from 1991, whereas average Normalised Difference Built-up Index (NDBI) increased by 43.75%, associating with increased thermal stress area evaluated using ecological evaluation index. Major LULC transformations included green area into agricultural arable-land and built-up. Finally, the modelled output shows that built-up & vegetation index have strongly impacted LST. This study, help to understand the relative impact of land-use dynamics on LST at intra-urban level specifically with respect to the small urban settings. Further assisting in designing and regenerating urban contexts with stable configuration, considering sustainability and liveable climate, for benefit of health of public and fragile population in particular.