Assessment of Outdoor Thermal Comfort using Landsat 8 Imageries with Machine learning tools over a Metropolitan City of India

Abstract Rapid urbanization and an enormous increase in the urban landscape have resulted in a drastic change in land surface temperature (LST) and outdoor thermal comfort in metropolitan cities and severely impact the health and well-being of residents. The thermal comfort of any region depends on various parameters such as atmospheric temperature, relative humidity, land use, and land cover classes (vegetation, water, built-up and barren). In the present study, an attempt has been made to understand the spatial variation of outdoor thermal comfort in a metropolitan city, Hyderabad (17° 23’ 13” N, 78° 29’ 30” E), India. High-resolution satellite imageries of Landsat 8 and available in situ meteorological observations are used for this purpose. Various environmental indices such as NDVI (Normalized difference vegetation index), NDWI (normalized difference water index), NBI (new built-up index), LST, brightness, greenness, and wetness are estimated using remote sensing techniques. The machine learning tool (SVM regression) was implemented to model the outdoor thermal comfort at a finer resolution. The spatial variation of outdoor thermal comfort was studied for the summer and winter seasons of 2018, 2019, and 2020 and analyzed to delineate the comfort and discomfort zones over the city. The results suggest that urban built-up and barren lands cause maximum discomfort to pedestrians, and vegetated areas and water bodies of urban spaces substantially decrease the thermal loads. Significant spatial variation of outdoor thermal conditions is noticed over different regions of the city, portraying the influence of the urban landscape.