Comprehensive assessment of land degradation in the Sanjiangyuan region based on multi-source indicators.
Changxian Qi, Haiyue Peng, Maqiang Xu, Zeyu Li, Yanhe Wang, Qiong Li, Baozu Cui, Baohong Wang, Zongbao Wang, Jianlong Zhu, Junjun Qin, Linxing Zhao
Abstract
Open AccessAs a typical high-altitude ecologically fragile zone, the Sanjiangyuan region is facing severe ecological challenges, including grassland degradation, land desertification, and declining ecosystem productivity, posing a significant threat to regional ecological security. This study integrates vegetation and soil factors to establish the Optimal Land Degradation Index (OLDI) based on a constrained optimization algorithm as a novel indicator for monitoring land degradation. Trend analysis and driving factor analysis were employed to analyze the spatiotemporal patterns of land degradation in the Sanjiangyuan region, identifying the influencing factors and potential future developments. The results indicate that the land degradation risk pattern exhibits a stepwise distribution from southeast to northwest, with high-risk degraded land accounting for 41% of the total area, moderate-risk degraded land covering 39%, and low-risk degraded land comprising 20%. Over the past 23 years, land improvement has been the dominant trend, with a total improved area reaching 98,700 km2, representing 27% of the Sanjiangyuan region, while degraded land covers 7,700 km2, accounting for 2%. The driving factor analysis reveals that land degradation is influenced by multiple dimensions, including topography, climate factors, and human activities, exhibiting significant spatial heterogeneity. However, future trends suggest a potential shift toward worsening land degradation, highlighting the urgent need for prioritized ecological conservation measures to effectively address the issue. Overall, the findings provide scientific evidence to support regional ecological restoration strategies and offer a valuable reference for sustainable management of high-altitude fragile ecosystems.