Chikwawa District has experienced significant urban development, agricultural expansion, and population growth over the years. The study analyzed Landsat Satellite images from 1979, 1995, 2009, and 2023 to assess land use and land cover changes. The CA-Markov chain model, a spatially explicit model that integrates cellular automata and Markov chains, was applied to predict future LULC scenarios for 2035, 2045, and 2065. The Spearman rank correlation coefficient (?) was used to assess the statistical significance of changes in land cover classes. The results show that the Random Forest classification method yielded the highest accuracy, with Kappa values of 85%, 86%, 86%, and 90%, and overall accuracy of 88%, 90%, 90%, and 93% for the years 1979, 1995, 2009, and 2023, respectively. Over a 44-year period, land cover changes revealed a 21% decrease in forest, 3% in vegetation, and 0.3% in water bodies, while built-up areas and bare land increased by 22% and 16%. Predicted future changes indicate further decreases in forest cover (-28.6%, -33.8%, and -51%), vegetation (-12.9%, -19.8%, and -24.7%), and water (-28.2%, -12.3%, and -14%), with increases in built-up areas (+23.2%, +22.2%, and +22.8%) and bare land (+13%, +15.3%, and +37.1%) for the periods 2035-2023, 2045-2035, and 2065-2045, respectively. The CA-Markov model validated the predictive results, demonstrating a good too perfect agreement with the actual data for the years 2009, 2011, and 2023. The decline in vegetation and water bodies exacerbates the potential for environmental degradation, which could lead to greater vulnerability to climate change impacts, such as floods and droughts. This underscores the necessity for policy interventions to regulate land conversion, promote sustainable agricultural practices, and implement effective land-use planning frameworks.

Keywords: GIS, Remote sensing, CA-Markova Chain model, Random Forest, Land Use Land Cover