逻辑回归和信息量模型在山区滑坡易发性评价中适宜性的分析

邹凤钗; 冷洋洋; 陶小郎; 尹彦迪

邹凤钗,冷洋洋,陶小郎,尹彦迪.逻辑回归和信息量模型在山区滑坡易发性评价中适宜性的分析[J].贵州地质,2022,39(3):280-286

逻辑回归和信息量模型在山区滑坡易发性评价中适宜性的分析

Suitability Analysis for Landslide Susceptibility Assessment in the Mountain Area by Logistic Regression and Information Model

投稿时间：2022-02-03

DOI：

中文关键词: 滑坡地质灾害易发性评价逻辑回归模型信息量模型万山区

英文关键词:Landslide Susceptibility evaluation Logistic regression model Information model Wanshan District

基金项目:贵州省地质灾害详细调查及风险评价成果集成(中央自然灾害防治体系建设补助资金Z195110010003)；2021年度中共贵州省委全面深化改革重大调研课题(贵州山区地质灾害成因机理及防治对策研究)。

作者	单位
邹凤钗	贵州省地质环境监测院，贵州贵阳550081
冷洋洋	贵州省地质环境监测院，贵州贵阳550081 成都理工大学，四川成都610059
陶小郎	贵州省地质环境监测院，贵州贵阳550081
尹彦迪	贵州省地质环境监测院，贵州贵阳550081

摘要点击次数: 448

全文下载次数: 564

中文摘要:

以万山区为例，在区域滑坡孕灾条件的基础上，筛选工程地质岩组、斜坡结构、平均坡度、地貌、距构造距离及距河流距离共6个易发条件因子，选取逻辑回归模型和信息量模型对山区滑坡进行易发性评价。结果显示逻辑回归模型中中高易发区面积占比分别为1578%和1970%，82%的地质灾害点落在该区域内；信息量模型中中高易发区面积占比为1241%、2519%，包含了区域88%的滑坡灾害点。最后通过实际发生的灾害点在各易发区的分布情况进行检验，逻辑回归模型中灾害点落在高易发区的比例远小于信息量模型，且高易发等级中灾害点实际发生的比值较小，说明针对山区区域滑坡地质灾害易发性评价结果预测上，信息量模型的评价结果更为客观准确。

英文摘要:

Taking Wanshan district as an example，six factors including engineering geological rock group，slope structure，average slope，landform,distance from structure and distance from water system were selected based on the study of landslide hazard condition in Wanshan DistrictLogistic regression model and information model were used to evaluate the landslide susceptibility，and the results showed that the proportion of medium-high susceptibility area is 15.78% and 19.70% respectively，and 82% of geological disasters fall in this regionIn the information model，the area of medium-high susceptibility area accounted for 12.41% and 25.19%，including 88% of the landslide disastersFinally，the distribution of actual disaster points in each prone area was testedThe proportion of disaster points in the high-prone area in the logistic regression model was much smaller than that in the information model，and the actual ratio of disaster points in the high-prone level was smallIt shows that the evaluation result of information quantity model is more objective and accurate for the regional landslide geological hazard susceptibility evaluation.

查看全文查看/发表评论下载PDF阅读器