Data‐Driven Analysis of Urban Heat Island Effect and Economic Effect Based on Machine Learning and Wavelet Coherence

ABSTRACT The configuration of urban infrastructure, green spaces, and vegetation is crucial in alleviating the impacts of the surface urban heat island (SUHI) effect. This study examined the spatiotemporal dynamics of the SUHI impact in Sialkot from 2001 to 2023, employing Landsat data. The Modified Mann–Kendall trend test indicated a stable but somewhat declining SUHI trend, along with a reduction in the IBI and an elevation in the NDVI. Wavelet coherence and rolling correlation analyses demonstrated complex short‐term and seasonal effects of vegetation on SUHI intensity. The XGBoost model exhibited substantial predictive accuracy ( R 2 = 0.86, RMSE = 0.18), with Land Surface Temperature recognized as the primary determinant. The findings highlight the essential function of green infrastructure in mitigating SUHI effects, indicating the efficacy of Pakistan's green programs. This research highlights the importance of machine learning in environmental studies and provides critical insights for sustainable urban design and policy formulation.