Greening and Land Surface Temperature in Newly Expanded and Old Built-Up Areas Across Urban Development Stages in the Yangtze River Delta.

Vegetation greenness helps cool cities, but how its cooling effect changes over time in newly expanded and existing old built-up areas (BUAs) across different urban development stages remains poorly understood. This study investigated the spatiotemporal trends of the Normalized Difference Vegetation Index (NDVI) and land surface temperature (LST) within newly expanded and old BUAs across cities of different development stages from 1992 to 2020. Utilizing the ESA CCI dataset for BUA delineation and Landsat-derived NDVI and LST, we applied the Mann-Kendall test and Sen's slope estimator to analyze 26 core cities. City sizes were classified into five categories from small cities to super megacities based on their status, and then assigned to urban development stages ranging from Stage 4 (least developed) to Stage 1 (most developed). The results revealed a significant decrease in NDVI across all areas in both new BUA (average significant slope = -0.0009, p < 0.05) and old BUA (average significant slope = -0.0002, p < 0.05), and a significant increase in LST in both new BUA (average significant slope = 0.0818, p < 0.05) and old BUA (average significant slope = 0.0245, p < 0.05). In addition, NDVI trends in BUA expansion varied significantly across urban development stages (p for trend = 0.016). Excluding Stage 1 (Shanghai only), vegetation loss attenuated from the sharpest decline in Stage 4 (average significant slope = -0.0021, 36.60% significant, p < 0.05) to a moderate decline in Stage 3 (-0.0011, 26.03%), and reversed to a slight vegetation gain in Stage 2 (0.0031, 19.14%), whereas LST showed no significant stage-dependent pattern. These findings underscore the widespread decreasing NDVI and increasing LST in both newly expanded and old BUAs, with the stage-wise vegetation differences potentially reflecting varying policy and economic contexts.