Impact analysis of urban green-blue-gray space interaction on heat Island intensity in Hyderabad Metropolitan Region using ECOSTRESS data

Urban Heat Island Intensity (UHII) results from rapid changes in urban green-blue-gray spaces. Present research evaluates seasonal UHII conditions in the Hyderabad Metropolitan Region by utilizing ECOSTRESS Land Surface Temperature (LST) measurements across the summer (March) and winter (December–January) periods from 2019 to 2025. The combination of the Normalized Difference Vegetation Index (NDVI), Normalized Difference Moisture Index (NDMI), Modified Normalized Difference Water Index (MNDWI), Impervious Surface (IMPER), and VIIRS Nighttime Light (NTL) represents urban green-blue-gray spaces. Peak LST values increases from 44.41 °C (Summer 2019) to 49.28 °C (Summer 2025) and from 38.91 °C (Winter 2019) to 41.24 °C (Winter 2025). The central urban area shared with industrial zones in the northeastern direction possess higher LST. Besides Multiple linear regression (MLR), machine learning models including Random Forest (RF), Extreme Gradient Boosting (XGBoost) and SHAP Analysis were applied to capture nonlinear environmental interactions influencing urban thermal variability. Compared to statical approaches, machine learning models own improved predictive accuracy across seasons. XGBoost achieves best performance during summer 2019 (R² = 0.72, RMSE = 1.69, MAE = 1.37), winter 2019 (R² = 0.39, RMSE = 1.27, MAE = 0.98) and for summer 2025 (R² = 0.56, RMSE = 2.15, MAE = 1.69), winter 2025 (R² = 0.54, RMSE = 1.95, MAE = 1.52). Results verify that green and blue space act as the mitigating factor of LST, which can be integrated into sustainable city development plans through forest planting alongside river basin as heat management strategy.