Integrating Machine Learning, Cellular Automata-Artificial Neural Network Model for Projecting Future Land Use Patterns in Urban Landscape of Jaipur, India

Jaipur, India, is experiencing rapid urbanization that is significantly altering its land use and land cover (LULC) patterns, presenting both challenges and opportunities for sustainable development and socio-economic advancement. This study utilizes advanced geospatial and remote sensing technologies to assess these changes and project future scenarios. Specifically, satellite data were processed using Google Earth Engine, land cover was accurately classified using the Random Forest algorithm, and future projections were modeled through QGIS-MOLUSCE using a polynomial-based Cellular Automata–Artificial Neural Network (CA-ANN) approach. Analysis of Landsat imagery for the years 2000 and 2020 reveals a dramatic 188.59% increase in urban built-up areas and a 145.44% rise in vegetation cover, largely due to successful afforestation efforts. Meanwhile, barren land declined by 47.37%, and water bodies exhibited fluctuating trends, reflecting the intricate interplay between urban development and climatic variability. Looking ahead to 2045, model projections estimate that built-up areas will expand to approximately 1303.08 square kilometres, potentially threatening the integrity of vital green spaces and aquatic ecosystems. These findings highlight the urgent need for integrated policy interventions aimed at mitigating environmental risks such as urban heat island effects and biodiversity loss. By providing a detailed account of past and present LULC dynamics, this research delivers actionable, data-driven insights to support sustainable urban planning. Moreover, the integration of urban growth models with climate resilience strategies offers a replicable framework for managing urban expansion in other rapidly developing cities, particularly those situated in semi-arid regions.