Machine Learning and Remote Sensing-Based Assessment of Urban Land Use Dynamics and their Influence on the Thermal Environment of Jaipur City, India

Sumit Karagwal

Department of Geography, Mohanlal Sukhadia University, Udaipur, India.

Narendra Gupta

Commissionerate College of Education, Jaipur, India.

Amit Daiman *

GeoInsight Lab, Insight Development Consulting Group, New Delhi, India.

*Author to whom correspondence should be addressed.


Abstract

Urban expansion in semi-arid regions like Jaipur, India, has significantly altered land use patterns and intensified the Urban Heat Island (UHI) phenomenon over recent decades. This study presents a spatiotemporal assessment of land use and land cover (LULC) dynamics and their impact on the urban thermal environment of Jaipur from 1999 to 2024. Using multi-temporal Landsat satellite datasets and advanced geospatial analytics on the Google Earth Engine (GEE) platform, a Random Forest (RF) supervised classification approach was employed to map LULC changes across three timeframes (1999, 2011, and 2024). Simultaneously, Land Surface Temperature (LST) was derived from thermal bands to evaluate seasonal and spatial temperature variations. The findings reveal a substantial increase in built-up area, more than doubling over the study period, predominantly at the expense of agricultural and vegetative cover. This urban growth has directly contributed to elevated LST values, with newly urbanised zones exhibiting pronounced thermal anomalies. Seasonal LST analysis indicates consistently higher temperatures during summer, with spatial clustering of heat zones aligning with high-density built-up regions. The correlation between LULC transitions and thermal signatures was further validated using the NDVI, affirming vegetation loss as a major driver of surface heating. Integrating machine learning with remote sensing offers a scalable framework for urban environmental analysis. The study underscores the critical need for climate-sensitive urban planning, with recommendations for enhancing urban greening, preserving natural landscapes, and implementing heat-mitigation infrastructure in rapidly expanding cities. This work contributes to the growing field of urban climatology by providing actionable insights into how urbanisation patterns influence microclimatic conditions in semi-arid environments.

Keywords: Random forest, urbanisation, cloud computing, landscape dynamics, heat islands


How to Cite

Karagwal, Sumit, Narendra Gupta, and Amit Daiman. 2025. “Machine Learning and Remote Sensing-Based Assessment of Urban Land Use Dynamics and Their Influence on the Thermal Environment of Jaipur City, India”. International Journal of Environment and Climate Change 15 (8):444-56. https://doi.org/10.9734/ijecc/2025/v15i84987.

Downloads

Download data is not yet available.