| 石文君,王宇栋,解晋航,李章杰,梁形形.基于多种模型对比的寻甸县地质灾害易发性分析[J].矿产勘查,2024,15(6):1092-1102 |
| 基于多种模型对比的寻甸县地质灾害易发性分析 |
| Analysis of geological hazard susceptibility in Xundian County based on the comparison of multiple models |
| 投稿时间：2024-01-09 修订日期：2024-04-07 |
| DOI：10.20008/j.kckc.202406020 |
| 中文关键词: 地质灾害 寻甸县 证据权法 组合赋权 耦合模型 易发性评价 |
| 英文关键词: geological hazards Xundian County weight evidence method portfolio empowerment coupled model predisposition assessment |
| 基金项目:本文受云南省科技厅基础研究计划项目（202401CF070138）资助。 |
|
| 摘要点击次数: 457 |
| 全文下载次数: 202 |
| 中文摘要: |
| 寻甸县地质条件复杂，地质灾害频发，而区域性易发性评价是预防地质灾害亟待解决的问题。本文以寻甸县斜坡类地质灾害为研究对象，采用证据权法、组合赋权法以及证据权-组合赋权耦合模型3种模型方式，基于GIS平台与SPSS 26.0软件，将选取的高程、坡度、土地利用等7项证据因素/诱发因子进行研究区易发性评价，划分出低、中、中高、高4类易发区。研究结果表明：（1）高、中高易发性分区内地质特质与研究得出的地质灾害影响因素相吻合，预测结果符合地质灾害空间分布特征；（2）落在高及中高分区内的灾害点占比分别为83.33%，84.68%，85.14%；（3）3种模型精确率均超过70%，其中，证据权-组合赋权耦合模型的特征曲线（ROC）的线下面积（AUC）精度检验值最大，为0.801，证据权法次之，为0.758，组合赋权法第三，为0.752。相对而言，证据权-组合赋权耦合模型更适合应用于地质灾害风险评价，为寻甸县灾害的分析提供方法思路，为区域的城镇化建设及发展提供参考，为防灾减灾和地质灾害预防提供科学依据。 |
| 英文摘要: |
| In order to reduce the frequency of geological disasters in Xundian County， taking the slope geological disasters such as landslides， collapses and debris flows in Xundian County as the research object， using the evidence right method， the combined weighting method and the coupling of the two methods， based on the GIS platform and SPSS 26.0 software， seven evidence factors/inducing factors such as elevation， slope and land use were selected to evaluate the susceptibility of the study area， and four susceptibility areas were divided： low， low， medium and high. The results show that： （1）The geological characteristics of the high， medium and high susceptibility zones are consistent with the influencing factors of geological disasters obtained from the study， and the prediction results are consistent with the spatial distribution characteristics of geological disasters. （2）The proportion of disaster points in the high and middle high sub-districts was 83.33%， 84.68% and 85.14%， respectively. （3）The accuracy of the three models is more than 70%， among which the accuracy of the area under the line （AUC） of the characteristic curve （ROC） of the coupled model is the largest， which is 0.801， followed by the evidence weight method （0.758）， and the combination weighting method is the third， which is 0.752. Comparatively speaking， the coupled model of evidence weight-combination weighting is more suitable for geological disaster risk assessment， which provides a scientific basis for disaster prevention and mitigation and geological disaster prevention in Xundian County. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |