Objective This study investigates the spatio-temporal dynamics of ecosystem services (ES) in Lhasa, a high-altitude city on the Tibetan Plateau, where fragile ecosystems face escalating pressures from urbanization. Ecosystem services, defined as benefits provided by natural ecosystems to sustain human well-being and ecological balance, are critical for biodiversity conservation and socio-economic development. While ES valuation (ESV) has been widely studied in low-altitude regions, research on high-altitude ecosystems remains limited despite their heightened vulnerability due to unique geographic and climatic conditions. The study aims to: 1) quantify urbanization impacts on ESV from 2007 to 2022; 2) evaluate the effectiveness of ecological restoration initiatives such as the Lhasa River Comprehensive Governance Project; 3) identify spatial thresholds in the ESV-human activity intensity (HAI) relationship to inform sustainable urban planning.

Methods A high-resolution spatial analysis framework (0.5 km × 0.5 km grid) was applied to assess ESV and HAI dynamics over 15 years (2007–2022). ESV was calculated using a dynamic model based on the improved equivalence factor method developed by Xie Gaodi and colleagues in 2015, integrating net primary productivity (NPP), precipitation, land cover, vegetation type, and environmental productivity. HAI was quantified through a composite index reflecting land-use change, population density, transportation infrastructure, and industrial development, with weights assigned based on ecological impacts. Spatial autocorrelation (Moran’s I ) and local cluster analysis were employed to identify interactions between ESV and HAI, accounting for the Tibetan Plateau’s topography.

Results The results of the study highlight significant spatio-temporal variations in ESV and human activity intensity in Lhasa. From 2007 to 2022, the total ESV decreased by 24.3%, amounting to a loss of 462.8 million CNY, with a brief recovery observed between 2012 and 2017. This recovery was largely driven by the Lhasa River Comprehensive Governance Project, which focused on wetland restoration and the improvement of hydrological services. The analysis of spatial patterns indicated that areas with high ESV values were predominantly concentrated around the Lhasa River Basin and the Lalu Wetland, while areas with low ESV values exhibited expansion in newly urbanized zones. A bivariate spatial autocorrelation analysis indicated that the positive spatial correlation between ESV and HAI weakened over the study period, with Moran's I decreasing from 0.249 in 2007 to 0.083 in 2022. Local cluster analysis further identified shifts in spatial patterns, with a 38.6% reduction in high-high clusters (areas with both high ESV and low HAI) and a 217% expansion in low-high clusters (areas with low ESV and high HAI), predominantly in newly urbanized areas. Moreover, the study identified a significant negative correlation(p < 0.05) between ESV and HAI in 23.5% of the grid cells by 2022, indicating a decoupling of the relationship between human activity and ecosystem service provision. The research identifies a threshold effect in the ESV − HAI relationship, where the decline in ESV accelerates when HAI exceeds a value of 0.65, with a regression coefficient of −1.32 and a p-value of less than 0.001. This threshold suggests that beyond a certain level of HAI, the negative impact on ecosystem services becomes more pronounced. between ESV and HAI in 23.5% of the grid cells by 2022, indicating a decoupling of the relationship between human activity and ecosystem service provision. The study identified a threshold effect in the ESV-HAI relationship, where the decline in ESV accelerated when HAI exceeded a value of 0.65, with a regression coefficient of −1.32 and a p-value of less than 0.001. This threshold suggests that beyond a certain level of human activity intensity, the negative impact on ecosystem services becomes more pronounced.

Conclusion From 2007 to 2022, Lhasa’s central urban area experienced significant land-use shifts, marked by cropland and grassland conversion to urban construction land. Ecosystem service value (ESV) initially decreased before a partial recovery, with an overall decline of 46.28 million CNY. Water bodies and wetlands contributed most to ESV, while low-value zones expanded in rapidly urbanizing areas. Human activity intensity (HAI) intensified radially, concentrating in older urban cores and expanding post-2012.Study limitations include sensitivity of valuation methods and data resolution constraints. Future work should integrate high-resolution models and climate scenarios to refine ESV-HAI interactions. Policy priorities include enforcing ecological redlines, protecting critical wetlands, and balancing urban growth with ecological resilience.