摘要
[研究目的]会泽县地处云贵高原半山区,地质灾害频发.以滑坡灾害为研究对象,采用耦合模型开展易发性评价,以期探索适合云贵高原半山区地理环境特点的评价模型.[研究方法]选取地貌、坡度及坡向等12个评价因子,应用确定性系数(CF)与Logistic耦合模型计算滑坡灾害发生概率,基于GIS的自然间断法评价结果等级.通过合理性检验与敏感性检验验证模型.[研究结果](1)高程、距断层距离、距道路距离、坡度对会泽县滑坡易发性影响较为显著.(2)稳定、低、中和高的面积分别为46.25 km2(0.61%)、3 095.41 km2(52.77%)、1 296.64 km2(21.94%)和1 446.53 km2(21.94%).(3)确定性系数(CF)与Logistic耦合模型通过合理性检验,AUC值为0.927.[结论](1)滑坡中-高易发区位于会泽县西南部及东北部部分地区,低易发区及稳定区位于西北部、中部和南部地区.(2)采用CF-Logistic耦合模型开展滑坡灾害易发性评价具有较高合理性和准确性.
Abstract
This paper is the research on the geological disasters.[Objective]Located in the semi-mountainous region of the Yungui Plateau,Huize County is frequently affected by geological disasters,particularly landslides.This study focused on landslide hazards,with a coupled model being employed for vulnerability assessments,aiming to develop an evaluation model tailored to the unique geographical and environmental characteristics of the semi-mountainous area of the Yungui Plateau.[Methods]Twelve evaluation factors,including topography,slope,and aspect,were carefully selected.The probability of landslide occurrence was calculated using the Coefficient of Determination(CF)and Logistic coupling model.The results were graded using the GIS-based Natural Breaks method.The model underwent rigorous validation through rationality and sensitivity tests.[Results](1)Elevation,distance to faults,distance to roads,and slope were identified as significant influencers on landslide susceptibility in Huize County.(2)The areas categorized as stable,low,moderate,and high susceptibility cover 46.25 km2(0.61%),3 095.41 km2(52.77%),1 296.64 km2(21.94%),and 1 446.53 km2(21.94%),respectively.(3)The Coefficient of Determination(CF)and Logistic coupling model demonstrated high rationality,with an AUC value of 0.927.[Conclusions](1)Moderate to high susceptibility landslide areas are concentrated in the southwest and northeast regions of Huize County,while low susceptibility and stable areas are prevalent in the northwest,central,and southern regions.(2)The application of the CF-Logistic coupling model for landslide vulnerability assessment exhibits superior rationality and accuracy.
基金项目
云南省教育厅科学研究基金项目(2023J0157)
自然资源部重点实验室2022年开放基金项目(20220103)
万方数据