摘要
以湖北省十堰市郧阳区为研究区,将坡度、高程、距道路的距离、距断层的距离、距水系的距离、植被指数、工程岩组作为该区域地质灾害影响因子。根据868个地质灾害隐患点,采用信息量方法计算每个影响因子的信息量值,作为逻辑回归的分类数据,建立基于信息量法+逻辑回归模型的易发性评估模型,同时与单独信息量模型的评价结果进行对比分析。结果表明,利用ROC曲线对该模型进行检验,信息量模型和信息量法+逻辑回归模型的评价精度分别为75.9%和76.4%,两者差异不大,但是工程岩组影响因子在信息量法+逻辑回归模型中无法通过显著性检验;同时考虑到,低易发区和极低易发区中,信息量模型地质灾害隐患点占比只有6.9%,与实际情况更为接近,故选用信息量模型的评价值作为十堰市郧阳区地质灾害易发性评价结果。由此说明,在地质灾害评价中数学模型的选择不一定两种方法模型叠加使用较一种模型效果好,需要结合项目实际情况来选择合适的评价模型。
This paper takes Yunyang District,Shiyan City,Hubei Province as the research area,and takes slope,elevation,distance from roads,distance from faults,distance from water systems,vegetation index,and engineering rock groups as the influencing factors of geological disasters in this area.According to 868 geological disasters hidden points,the information value method is used to calculate the information value of each impact factor,which is used as the classification data of logistic regression,and the susceptibility assessment model based on the information volume+logistic regression model is established,and it is compared and analyzed with the evaluation results of the individual information model.The results show that the evaluation accuracy of the information volume model and the information model+logistic regression model are 75.9%and 76.4%,respectively,using the ROC curve to test the model.There is not much difference between the two,but the engineering rock group influence factor cannot be passed the test in the information volume model+the logistic regression model.At the same time,considering that in the low-prone areas and extremely low-prone areas,the information volume model accounts for only 6.9%of the hidden danger points of geological disasters,which is closer to the actual situation.Therefore,the evaluation accuracy of information model is selected as the evaluation result of geological disaster susceptibility in Yunyang District of Shiyan City.This shows that the choice of mathematical model in geological hazard evaluation is not necessarily the effect of superimposing the two method models than one model.It is necessary to select the appropriate mathematical model based on the actual situation of the project.
作者
金朝
费雯丽
丁卫
陈曦
杜翌超
Jin Zhao;Fei Wenli;Ding Wei;Chen Xi;Du Yichao(Hubei Geological Survey,Wuhan,Hubei 430034;Eighth Geological Brigade of Hubei Geological Bureau,Xiangyang,Hubei 441000)
出处
《资源环境与工程》
2021年第6期845-850,886,共7页
Resources Environment & Engineering
基金
湖北省地质局部门预算项目“十堰市郧阳区地质灾害风险调查评价”和“热红外遥感地热资源潜力预测遥感研究”(MSDZ202114)。
关键词
地质灾害
信息量法
Logistic回归法
易发性评价
郧阳区
geological disasters
information quantity method
Logistic regression method
susceptibility evaluation
Yunyang District
