Urban Overheating Assessment through Prediction of Surface Temperatures: A Case Study of Karachi, Pakistan

Global climate has been radically affected by the urbanization process in recent years. Karachi, Pakistan’s economic hub, is also showing signs of swift urbanization. Owing to construction infrastructure projects under China-Pakistan Economic Corridor (CPEC) and associated urbanization, Karachi’s significantly affected. The replacement natural surfaces anthropogenic materials results urban overheating increased local temperatures leading serious health issues higher air pollution. Thus, these temperature changes effects must be addressed minimize their impact on city’s population. For analyzing Karachi city, LST (land surface temperature) assessed current study, where data past 20 years (2000–2020) used. this purpose, remote sensing from Advanced Spaceborne Thermal Emission Reflection Radiometer Digital Elevation Model (ASTER GDEM) Moderate-Resolution Imaging Spectroradiometer (MODIS) sensors were utilized. long short-term memory (LSTM) model was utilized road density (RD), elevation, enhanced vegetation index (EVI) are used as input parameters. Upon comparing estimated measured LST, values mean absolute error (MAE), square (MSE), percentage (MAPE) 0.27 K, 0.237, 0.15% for January, 0.29 0.261, 0.13% May, respectively. low MAE, MSE, MAPE show a correlation between predicted observed values. Moreover, that more than 90% pixel falls least possible range −1 K +1 K. MSE Support Vector Regression (SVR) 0.52 0.453 0.18% 0.76 0.873, 0.26%. outperforms previous studies, shows accuracy, depicts greater reliability predict actual scenario. In future, based accurate model, city planners can propose mitigation strategies reduce harmful Urban Heat Island (UHI).