Modeling the impact of the future state of land use land cover change patterns on land surface temperatures beyond the frontiers of greater Kumasi: A coupled cellular automaton (CA) and Markov chains approaches

Abstract

There is ample evidence of the impact of land use/land cover (LULC) on land surface temperatures (LST) in Kumasi. However, the situation in the future is largely unknown. This study predicts the future state of LULC and LST in and beyond the frontiers of Greater Kumasi. The study used coupled Artificial Neural Networks (ANNs), Cellular-Automaton-Markov chains, and multiple regression to simulate the future state of LULC and LST for 2032 and 2042. Landsat imageries from 1986 to 2022 were acquired from USGS. Road networks, bus stops, and social centers were obtained from OpenStreetMap, and population density from the socioeconomic data and application center (SEDAC). Support vector machine (SVM) was used for image classification. Between 1986 and 2022, built-up areas increased by 18.63%; high-density forests decreased by 42.13%, whereas low-density forests increased by 20.21%. Between 1986 and 2042, the built-up area is expected to expand by 27%, high-density forests are likely to decline by 45%, whereas low-density forest are likely to increase by 19.097%. In contrast, water bodies and bare grounds are expected to decrease by 0.61% and 0.134%, respectively. Very high LSTs increased by 8.23% between 1986 and 2022. Between 1986 and 2042, LSTs are expected to increase spatially by 16.145%. Mean LSTs are expected to increase by 7.43 °C (36.03%) between 1986 and 2042. Urban expansion clearly drives the increase in LSTs, and this is more likely to persist in the future. These research findings therefore call for institutional and sector-based coordination between governmental institutions.