Abstract:
Objective Debris-flow susceptibility and hazard assessment are important foundations for debris-flow prevention and risk management. To address insufficient representation of rock-mass fragmentation and ecological root-zone effects, strong subjectivity in index weighting, and inadequate uncertainty characterization in traditional methods, this study proposes a comprehensive debris-flow assessment method that integrates mechanism interpretation with uncertainty expression, taking the Xi'an section of the northern foothills of the Qinling Mountains as the study area.
Methods Nine evaluation factors were selected: relative relief, average slope, watershed area, distance to faults, landslide-collapse density, weathering-layer thickness, NDVI, vegetation type, and rock vulnerability index. The frequency ratio method was used for factor quantification, and factor weights were determined using a CRITIC-AUC hybrid weighting method. A multidimensional cloud model was used to characterize the randomness and fuzziness of susceptibility grades, and rainfall-triggering factors were incorporated to complete the hazard assessment.
Results The susceptibility assessment achieved an AUC value of 0.8652, indicating good discriminative capability. Extremely high-hazard zones covered approximately 542 km2, accounting for 9.88% of the study area, and were mainly distributed in piedmont steep-slope zones and deeply incised gullies. Hazard-site density increased markedly with hazard level, reaching 0.0592 in extremely high-hazard zones and 0.0138 in high-hazard zones, while values in low- and extremely low-hazard zones were close to zero. The rock vulnerability index, weathering-layer thickness, NDVI, and vegetation type contributed strongly, indicating that rock weathering and fragmentation, loose material supply, and ecological root-zone differences jointly control spatial differentiation of debris-flow hazard.
Conclusion The CRITIC-AUC-multidimensional cloud model considering rock-mass structure and ecological root-zone effects can effectively identify high-hazard debris-flow watersheds and improve the mechanistic interpretability of the assessment results.
下载: