Modelling the impacts of land use/land cover changing pattern on urban thermal characteristics in Kuwait

Abstract

Rapid urbanization owing to population growth and economic development has made thermal environment-related studies increasingly prominent. This study aims to monitor and predict the changes in land use/land cover (LULC) and their impacts on land surface temperature (LST), urban heat island (UHI), and urban thermal field variance Index (UTFVI) in Kuwait from 1991 to 2021 for the very first time. Support Vector Machine (SVM) and Artificial Neural Network (ANN)machine learning algorithms were used to analyze and predict, respectively, using Landsat 4-5 and 8 images from 1991 to 2021 at 10 years interval. Results illustrated transformation of 27.24% bare land and 5.43% vegetation into the built-up area increased the mean LST by 5°C, resulting in an upsurge of UHI values by 0.861 and the strongest UFTVI effects by 108% from 1991 to2021. Predicted LULC, LST, UHI and UTFVI distribution modelled using the ANN-based Cellular Automata technique for 2031 shows the declination of vegetation cover (44%) and water bodies (57%) during 2021–2031, resulting upsurge in built-up areas (38%) and higher LST (10°C). Moreover, by 2031, 41% of the total study area will be covered by the strongest UTFVI affected zones. All the prediction was performed with a higher accuracy, and validated by RMSE and R2 values with predicted and real datasets. Assessing the predicted LULC, LST, UHI, and UTFVI distribution maps will help prospective policymakers and city planners to take the necessary actions by reducing heat stress impacts and make cities sustainable.