基于逻辑回归模型的来宾市岩溶塌陷敏感性评价

Study on Karst Collapse Susceptibility Assessment Based on Logistic Regression Model— A Case Study of Laibin City

岩溶塌陷是长期困扰广西的地质灾害现象。本文针对广西来宾市岩溶塌陷地质灾害,采用二分类变量非线性逻辑回归模型作为岩溶塌陷敏感性评价模型,借助于ArcGIS和SPSS软件工具,选取地形、地层岩性、断裂带、地下水位变幅、地下水位埋深、土壤渗透率、覆盖层厚度7个影响因子,将各因子归一化处理后借助于SPSS统计软件进行逻辑回归方法运算,计算每个影响因子的逻辑回归系数值,并根据逻辑回归模型得到研究区各单元的岩溶塌陷灾害敏感性概率,在ArcGIS中绘制研究区岩溶塌陷地质灾害敏感性分区图。经过模型验证和实地调查,表明逻辑回归模型模拟效果较好,可以用于指导研究区岩溶塌陷地质灾害防治。

Karst collapse is a long-term geological disaster in Guangxi.This paper adopts the binary classification variable nonlinear logistic regression model to evaluate the susceptibility of karst collapse in the city of Laibin,Guangxi.Moreover,7impact factors such as terrain,formation lithology,faults,amplitude of groundwater level,groundwater depth,percolation rate of soil,and overburden thickness are selected and the analysis tools are ArcGIS and SPSS.After being normalized,the coefficient of each impact factor is calculated by logistic regression method with SPSS.The susceptibility probability of each unit in the study area is obtained with the logistic regression model and the collapse susceptibility zoning is mapped with ArcGis.Through assessment model validation and field survey,the simulation effect of the model is proved to be good,which can be used to guide the prevention and control of collapse disasters in the study area.