Urban green spaces play a crucial role in mitigating the Urban Heat Island (UHI) effect by cooling surface temperatures. This study used Landsat 8 data and urban green space surveys, applying Radiative Transfer Equation (RTE) and Boosted Regression Tree (BRT) models to analyze how green space morphology and vegetation composition influence land surface temperature (LST) across seasons. Results show that the percentage of green space area (PLAND) has the greatest impact on LST year-round. In summer and winter, additional morphological factors, such as Largest Patch Index (LPI), Mean Patch Size (MPS), and Edge Density (ED), also play important roles. A cooling effect is observed when PLAND exceeds thresholds of 34.5 % in spring, 44.5 % in summer, 39.6 % in autumn, and 37.4 % in winter, though the effect diminishes with further increases. Trees provide the highest contribution to LST reduction in all seasons, with particularly strong effects in summer (66.8 %) and autumn (65.7 %). Optimizing urban green space design through strategic planning and a balance of vegetation types can significantly enhance temperature regulation, reduce UHI effects, and improve urban ecological quality.