Abstract
The remote sensing ecological index (RSEI) is an important tool for assessing ecosystem quality. However, its land surface temperature (LST) component poses challenges due to complex calculations and mismatched spatial resolution with other indicators. This study proposed an improved remote sensing ecological index (DRSEI). By replacing the LST component in RSEI with the difference index (DI) (representing PM(2.5) concentration), the new index better reflects air pollution's impact on ecosystem quality. The results demonstrated that DRSEI outperformed the RSEI in assessing ecosystem quality in Chongqing's urban area. It exhibited three advantages: stronger correlation with the ecological index (EI), standard deviation values closer to EI's baseline, and lower root mean square error. The applicability of the DRSEI and RSEI varied across different regions: the DRSEI proved to be more suitable for highly urbanized areas, whereas the RSEI performed better in suburban regions. Further analysis revealed that the spatial variability of indicators influenced their loadings in principal component analysis, thereby affecting ecosystem quality assessment results. This study emphasizes the importance of considering the spatial distribution of indicators when constructing ecological indices. The findings suggest DRSEI could effectively assess ecosystem quality in urbanized areas. This approach provides new insights for urban ecological monitoring and environmental management.