Decadal Evolution and Multi‐Scale Climate Teleconnections of Heatwaves Over Tamil Nadu, India (1981–2020): Trends, Spatial Expansion and Atmospheric Drivers

This study investigates the spatiotemporal evolution, climatic drivers, and predictive modelling of heatwaves (HWs) over Tamil Nadu (TN), India, during 1981–2030, using India Meteorological Department (IMD) gridded temperature and reanalysis datasets. The analysis reveals a progressive intensification and inland expansion of HW frequency and duration, with hotspots in Chennai, Cuddalore, Trichy, Madurai and Vellore during 2011–2020. El Niño–Southern Oscillation (ENSO)– Indian Ocean Dipole (IOD) coupling emerged as the dominant large‐scale driver of HW variability. The concurrence of El Niño and nIOD (Negative IOD) phases significantly enhanced HW frequency and duration through large‐scale subsidence, reduced convective activity and suppressed cloud cover. Conversely, La Niña years exhibited enhanced moisture advection, increased cloudiness and weaker heat extremes. ENSO‐driven anomalies in relative humidity (RH), total cloud cover (TCC) and wind circulation clearly demonstrated that El Niño‐induced dryness and weakened low‐level winds favor heatwave amplification, while La Niña fosters convective moderation. In addition, positive Atlantic Multidecadal Oscillation (AMO) and negative Arctic Oscillation (AO) phases during the recent decade strengthened subsidence and stagnant warm‐air conditions, jointly contributing to intensified HWs, suggesting an increasing influence of extra‐tropical oscillations under a warming climate. A hybrid Artificial Neural Network–Long Short‐term Bi‐directional Memory (ANN–LSBM) model integrating dynamic predictors (T max , RH, TCC) and climate indices (ENSO, IOD, AO, AMO) achieved a Root Mean Square Error (RMSE) of 19.97 and R 2 of 0.48, demonstrating stable convergence and moderate predictive skill. The findings emphasize the growing importance of both tropical and extra‐tropical climate drivers in modulating regional heat extremes and underline the urgent need for state‐level planner awareness, adaptive heat action plans, and integration of Artificial Intelligence (AI)‐based predictive tools to enhance TN's heatwave preparedness and resilience. In addition, land use and land cover (LULC) change detection between 2011 and 2020 revealed significant land transformation across TN, characterized by an increase in built‐up area and a decline in barren and floodplain regions. These transitions indicate rapid urbanization, surface modification and reduced vegetative cover, which have collectively contributed to enhanced surface heating and intensified local heatwave conditions. The LULC findings reinforce that regional warming and urban expansion are critical non‐climatic drivers influencing the observed escalation in heatwave frequency and intensity.