Evaluating Land Use Transformation and Thermal Dynamics in Faridabad District Using Generative AI and GEE

Authors:

Abhishek Malik¹          Prof. Gaurav Kalotra²

1. Research Scholar, Department of Geography, Panjab University, Chandigarh
2. Professor, Department of Geography, Panjab University, Chandigarh

Abstract

Rapid urbanization in the National Capital Region (NCR) has significantly altered local microclimates, raising critical concerns regarding thermal stress and environmental sustainability. This study evaluates the spatiotemporal Land Use and Land Cover (LULC) transformations and their impact on Land Surface Temperature (LST) and Urban Heat Island (UHI) intensity in Faridabad District between 2017 and 2025. To overcome traditional computational bottlenecks, this research pioneers a novel GeoAI methodology, integrating Generative AI (Large Language Models) with the Google Earth Engine (GEE) cloud computing platform for automated spatial analysis. Utilizing the Dynamic World V1 dataset and Landsat 8 thermal imagery, the analysis reveals an 86.92 km² expansion in built-up urban infrastructure, primarily consuming open and barren lands. Consequently, the maximum UHI intensity within dense urban clusters amplified dramatically from +8.75°C in 2017 to +11.63°C in 2025. The findings confirm that unabated industrial densification is creating severe localized "heat domes." Furthermore, the study demonstrates that AI-assisted cloud scripting reduces multi-decadal processing times to mere seconds, offering a highly scalable, prompt-based framework for global urban climate monitoring.

Keywords: GeoAI, Google Earth Engine, Urban Heat Island, Land Use Transformation, Faridabad, Generative AI, Spatial Analysis.