ASSESSING THE IMPACT OF URBANIZATION ON LAND SURFACE TEMPERATURE AND LAND USE/LAND COVER DYNAMICS IN ARKANSAS: A 20-YEAR SPATIOTEMPORAL ANALYSIS USING MACHINE LEARNING & GOOGLE EARTH ENGINE

Arkansas, known as the "Natural State", is experiencing significant urban expansion and land use changes, impacting environmental sustainability. This study investigates the relationship between land surface temperature (LST) and land use/land cover (LULC) changes in Arkansas, aiming to inform urban planning and sustainable development strategies. It evaluates the potential of machine learning techniques to accurately predict future LULC changes, aiding in land use planning and disaster management. Additionally, it examines the presence of urban heat islands in major Arkansas cities and how insights into LST and LULC changes can support long-term land use planning and climate resilience. Utilizing Landsat-5/8 data, the analysis focuses on time-series land cover indices and correlations between seasonal LST and LULC changes in Arkansas from 2001 to 2021, exploring how LULC affects LST dynamics, especially urban heat islands. Furthermore, it simulates future urbanization conditions using advanced machine learning algorithms, including the cellular automaton (CA) - Markov chain algorithm, to forecast future land use changes. This research sets the stage for future studies to enhance understanding and knowledge of urban sustainability and climate resilience.