Abstract
This study assesses ecological change in two 100 resilient cities (100RC) (Pune, Surat) and two non-100RC cities (Chattogram, Nagpur) in South Asia using an improved remote sensing ecological index (RSEI). The enhanced model incorporates six indicators composite vegetation index (CVI), wetness, normalized difference build-up and bare soil index (NDBSI), impervious surface index (ISI), urban index (UI), and land surface temperature (LST)-with entropy-based weighting and a moving window approach to overcome traditional PCA limitations. The results reveal apparent differences between 100RC and non-100RC cities. Non-100RC cities exhibited high volatility, characterized by sharp ecological declines, partial recoveries, and subsequent collapses. In contrast, 100RC cities followed steadier but persistent downward trajectories: Surat exhibited a monotonic decline, while Pune improved until 2022 before declining in 2024. Hotspot analysis revealed that cold spots expanded dynamically in non-100RC cities, whereas they spread gradually in 100RC cities. Moran's I confirmed strong spatial clustering (0.89-0.97) across all cities. A two-way ANOVA revealed significant group, year, and interaction effects (p < 0.001). These findings suggest that resilience frameworks reduce ecological instability but are insufficient to halt structural degradation. Strengthening urban resilience requires more rigorous policy enforcement, continuous monitoring, and ecological restoration to achieve sustainable ecological outcomes.