Abstract
To reveal the changes and future states of ecosystem service value (ESV) and landscape ecological risk index (LERI) in typical dry valley ecosystems in the 21st century, this study utilized high-precision land use data combined with ESV assessment, LERI assessment, bivariate spatial autocorrelation, Geodetector, CA-Markov model, and standard deviation ellipse methods. The spatio - temporal evolution, spatial associations and differentiation characteristics of ESV and LERI in the Anning River Basin from 2000 to 2020 were analyzed, and made predictions for 2030. The results indicated: (1) The ESV in the study area exhibited a decreasing trend from 2000 to 2030, with a cumulative reduction of 1.444 billion CNY, where forest land contributed over 78.46% of the total ESV. The basin presented a spatial pattern of lower ESV in the valley and higher in mountainous areas, dominated by high-value areas (accounting for approximately 41.54% of the total area). (2) LERI showed an overall declining trend during the study period, with a spatial pattern similar to ESV, characterized by low values in valleys and higher values in mountainous regions. The extremely low-risk areas were predominant (approximately 37.00%). (3) The spatial differentiation of ESV and LERI was jointly influenced by natural and economic factors, with agricultural production potential and annual evaporation identified as the dominant factors (q values of 0.18 and 0.28, respectively). The center of gravity of ESV distribution moved towards the northeast, whereas LERI shifted to the southeast. (4) ESV and LERI exhibited positive spatial correlations (Moran's I > 0), with high ESV-low LERI as the main LISA cluster (approximately 20.92%). These findings highlight the necessity for adopting tailored ecological protection measures based on the spatiotemporal distribution and evolution characteristics of ESV and LERI to promote sustainable regional ecological development.