<p>To address the contradiction among “high-intensity development, high ecological value, and high geological risk” in the Sichuan–Yunnan Ecological Barrier Zone, this study constructs an integrated assessment framework that incorporates geological, ecological, and human activity dimensions, in response to limitations of existing evaluation systems such as subjective weighting, inadequate characterization of ecological functions, and weak spatial integration. This framework consolidates nearly 25 years of data into 13 indicators covering geological disaster susceptibility, ecosystem service functions, and human engineering activity intensity. By coupling a Random Forest-weighted information value model with the InVEST model, it achieves dynamic weighting of multi-source data and analysis of nonlinear relationships. Assessment results show that the regional eco-geological environment exhibits a spatial pattern of “three belts and two cores,” with overall quality at an upper-medium level. Based on this, the study proposes a coordinated governance pathway of “geological risk mitigation, ecological value enhancement, and industrial structure optimization,” providing systematic decision-making support for regional ecological security maintenance and sustainable spatial utilization.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Comprehensive eco-geo-environmental assessment of the Sichuan–Yunnan ecological barrier zone using a random forest model

  • Xiaolin Yang,
  • Pengwei Wang,
  • Shucheng Tan,
  • Churan Feng,
  • Rongxue Fan,
  • Hao Tang

摘要

To address the contradiction among “high-intensity development, high ecological value, and high geological risk” in the Sichuan–Yunnan Ecological Barrier Zone, this study constructs an integrated assessment framework that incorporates geological, ecological, and human activity dimensions, in response to limitations of existing evaluation systems such as subjective weighting, inadequate characterization of ecological functions, and weak spatial integration. This framework consolidates nearly 25 years of data into 13 indicators covering geological disaster susceptibility, ecosystem service functions, and human engineering activity intensity. By coupling a Random Forest-weighted information value model with the InVEST model, it achieves dynamic weighting of multi-source data and analysis of nonlinear relationships. Assessment results show that the regional eco-geological environment exhibits a spatial pattern of “three belts and two cores,” with overall quality at an upper-medium level. Based on this, the study proposes a coordinated governance pathway of “geological risk mitigation, ecological value enhancement, and industrial structure optimization,” providing systematic decision-making support for regional ecological security maintenance and sustainable spatial utilization.