Full-chain comprehensive assessment and multi-scenario simulation of geological disaster vulnerability based on the VSD framework: a case study of Yunnan province in China

• A full-chain dynamic analysis framework was developed to assess past, present, and future GDV. • Simulates GDV under three development preferences: environment oriented, status quo, and economically oriented. • Reveals the spatiotemporal heterogeneity of exposure and GDV. • Suggests adaptive strategies to address trade-offs between economic growth and sustainable disaster mitigation efforts. • Offers insights for balancing disaster management and regional economic growth in geologically vulnerable areas globally. With increasing global climate change and human-induced disturbances to geological environments, the frequency and destructiveness of geological disasters are intensifying. Thus, scientifically assessing and predicting geological disasters and formulating disaster prevention and mitigation policies are of critical importance for promoting regional sustainable development. However, current evaluations of geological disasters often overlook the adaptive capacity of the disaster-bearing body and focus primarily on cross-sectional comparisons across different regions at specific time points, with insufficient attention to spatiotemporal dynamics. This study selects Yunnan Province, one of the most severely affected regions by geological disasters in China, as a case study. Based on the Vulnerability Scoping Diagram (VSD) framework, a full-chain dynamic analysis framework was developed to assess past and present geological disaster vulnerability (GDV) and simulate future scenarios. This framework integrates exposure, sensitivity, and adaptive capacity to establish an indicator system that reflects the unique economic and ecological characteristics of the region. Using various mathematical models, the spatial differentiation and evolutionary trends of GDV in Yunnan Province from 2000 to 2020 were quantitatively analyzed. Furthermore, the Ordered Weighted Averaging (OWA) algorithm and the Partical Swarm Optimization-Support Vector Machine (PSO-SVM) model were combined to simulate future GDV scenarios for 2030–2050 under three development preferences: environment oriented, status quo, and economically oriented. The results show that from 2030 to 2050, GDV in Yunnan Province is generally at a medium–high level, with a spatial distribution pattern of “centered on Kunming, with a radial increase extending outward.” Temporally, GDV exhibits an initial increase followed by a decline, with regions clustering primarily as high-high and low-low GDV areas. Additionally, the predicted GDV outcomes vary across different development preferences. In the economically oriented scenario, GDV predominantly stabilizes at moderate levels, while in the environment oriented scenario, GDV improves the fastest but subsequently shows a slight rebound. These findings suggest that single-focus decisions prioritizing either economic development or environmental protection result in unsustainable GDV improvements. Therefore, decision-makers should consider GDV fluctuations under various scenarios when setting strategic development goals for different regions and phases, ensuring localized and adaptive adjustments. This study aims to provide insights from the Yunnan case for balancing disaster governance and regional economic development in global regions prone to geological disasters.