摘要
白龙江流域是我国滑坡泥石流灾害四大高发区之一,进行该区域滑坡敏感性评价,能够为决策者在灾害管理和设施建设规划方面提供帮助,对区域防灾减灾具有重要指导意义。本研究采用边坡单元为基本研究单元,在野外调查及前人研究基础上,选择控制该区域滑坡发育的19个要素作为影响因子;经过主成分分析和独立性检验得到该区域对滑坡形成贡献最大的6个因子:高程、坡度、坡向、岩性、断裂距离和人口密度;分别使用二元逻辑回归模型(LR)和支持向量机模型(SVM)对该区域进行滑坡敏感性评价;最后,采用ROC曲线对模型精度进行验证。研究结果表明,两模型各能将38.76%、14.48%、9.40%、11.28%、26.07%和13.49%、21.61%、8.17%、26.70%、30.04%的边坡单元分别预测为极高危险区、高危险区、中度危险区、低危险区和极低危险区;精度验证结果表明两种模型均能有效地进行该区域滑坡敏感性评价,并且支持向量机模型具有更好的分类能力、预测精度和稳定性。
Bailong river basin is one of the four regions with high incidences of landslide and debris flow in China. Thus it is of vital importance to early out hazard mapping of the landslides in this region to provide references for disaster management and construction planning. Using slope units as the basic assessment units, this research firstly gets the 6most contributing factors of landslides by means of principal component analysis and independence test. Then, the methods of Logistic Regression（LR） and Support Vector Machine （SVM） are conducted for landslide hazard mapping. Results show that （1） both LR and SVM can effectively evaluate the hazards of landslides in the region; （2）the SVM has a better ability in classification, predicting accuracy and model stability. According to the results of the two models, the study area are classified into five categories, i.e., very high dangerous zone, high dangerous zone, moderate dangerous zone, low dangerous zone and very low dangerous zone, taking an area proportion of 38.76%, 14.48%, 9.40%, 11.28%, 26.07% and 13.49%, 21.61%, 8. 17%, 26.70%, 30. 04%, respectively.
出处
《工程地质学报》
CSCD
北大核心
2014年第1期56-63,共8页
Journal of Engineering Geology
基金
国家科技支撑计划项目(2011BAK12B06)
甘肃省科技重大专项(1102FKDA007)资助
关键词
白龙江流域
逻辑回归
支持向量机
敏感性制图
滑坡
Bsilong river, Logistic Regression, Support Vector Machine, Susceptibility mapping, landslide
