[Analysis of Spatial and Temporal Evolution of Ecosystem Health and Its Driving Factors on the Loess Plateau]

The clarification of health status of a regional ecosystems and their driving factors is of great significance for promoting sustainable development. Based on the "pressure-state-response （PSR）" framework， the ecosystem health levels of the Loess Plateau were evaluated for the years 2000， 2005， 2010， 2015， and 2020， and the main driving factors affecting ecosystem health levels were further identified by using the spatial-temporal geographically weighted regression （GTWR） model. The following results were obtained： ① During the study period， the ecosystem health level of the Loess Plateau fluctuated upward， with the value increasing from 0.452 in 2000 to 0.464 in 2020. Areas with good ecosystem health levels were mostly distributed in the central and northeastern regions， with values greater than 0.45. Areas with poor ecosystem health levels were mostly distributed in the northwest and southeast regions， with values lower than 0.40. ② From the perspective of spatial aggregation characteristics， the ecosystem health level of the Loess Plateau exhibited significant spatial aggregation."H-H aggregations"were mostly located in the central， western， and northeastern parts of the Loess Plateau， such as Xining， Yan'an， Taiyuan， and Tianshui. ③ The analysis of driving factors showed that the slope， annual average temperature， proportion of construction land， and land use intensity were the dominant factors affecting the ecosystem health of the Loess Plateau. Slope mainly drove positively， whereas annual average temperature， proportion of construction land， and land use intensity mainly drove negatively. Our results can provide a theoretical basis for the formulation of ecological protection and sustainable development policies on the Loess Plateau.