This study investigates the spatiotemporal dynamics of land use and land cover (LULC) in Islamabad, Pakistan, over three decades (1991–2021) and forecasts future scenarios up to 2051. The research integrates remote sensing, GIS data, and Landsat satellite imagery (5, 7, 8) to predict urban growth and support sustainable land management and urban planning. The study employs the Random Forest (RF) algorithm for LULC classification and the CA-Markov chain model for future LULC projections, generating transition probability matrices among different LULC classes. Results show significant changes in LULC, with vegetation cover decreasing from 49.21% to 25.81%, and forest cover increasing from 31.89% to 40.05%. Urban areas expanded from 7.55% to 25.59% of the total area, translating into an increase from 76.31 km² in 1991 to 258.61 km² in 2021. Forest area also expanded from 322.25 km² to 409.21 km². Projections indicate a further decline in vegetation cover and an increase in built-up areas to 371.44 km² by 2051, with a decrease in forest cover compared to its 2021 levels. The model's accuracy was confirmed with an overall accuracy exceeding 90% and a kappa coefficient around 0.88. The findings highlight the model's reliability and provide a theoretical framework integrating socio-economic development with environmental conservation. The results emphasize the need for a balanced approach to urban growth in Islamabad, underscoring the essential equilibrium between development and conservation for future urban planning and management. The study underscores the importance of using advanced predictive models in guiding sustainable urban development strategies. The study also demonstrates the effectiveness of the CA-Markov model in predicting LULC changes and the importance of incorporating socio-economic factors in urban growth projections. The study's results indicate a significant increase in built-up areas and a decrease in vegetation and bare land, highlighting the need for sustainable land management practices. The study's findings align with previous research, indicating that the model's predictions are reliable and accurate. The study's results also highlight the importance of using advanced techniques such as Random Forest and Markov chain models in understanding and predicting LULC changes. The study's findings provide valuable insights for future urban planning and sustainable development.This study investigates the spatiotemporal dynamics of land use and land cover (LULC) in Islamabad, Pakistan, over three decades (1991–2021) and forecasts future scenarios up to 2051. The research integrates remote sensing, GIS data, and Landsat satellite imagery (5, 7, 8) to predict urban growth and support sustainable land management and urban planning. The study employs the Random Forest (RF) algorithm for LULC classification and the CA-Markov chain model for future LULC projections, generating transition probability matrices among different LULC classes. Results show significant changes in LULC, with vegetation cover decreasing from 49.21% to 25.81%, and forest cover increasing from 31.89% to 40.05%. Urban areas expanded from 7.55% to 25.59% of the total area, translating into an increase from 76.31 km² in 1991 to 258.61 km² in 2021. Forest area also expanded from 322.25 km² to 409.21 km². Projections indicate a further decline in vegetation cover and an increase in built-up areas to 371.44 km² by 2051, with a decrease in forest cover compared to its 2021 levels. The model's accuracy was confirmed with an overall accuracy exceeding 90% and a kappa coefficient around 0.88. The findings highlight the model's reliability and provide a theoretical framework integrating socio-economic development with environmental conservation. The results emphasize the need for a balanced approach to urban growth in Islamabad, underscoring the essential equilibrium between development and conservation for future urban planning and management. The study underscores the importance of using advanced predictive models in guiding sustainable urban development strategies. The study also demonstrates the effectiveness of the CA-Markov model in predicting LULC changes and the importance of incorporating socio-economic factors in urban growth projections. The study's results indicate a significant increase in built-up areas and a decrease in vegetation and bare land, highlighting the need for sustainable land management practices. The study's findings align with previous research, indicating that the model's predictions are reliable and accurate. The study's results also highlight the importance of using advanced techniques such as Random Forest and Markov chain models in understanding and predicting LULC changes. The study's findings provide valuable insights for future urban planning and sustainable development.