摘要
以丽江市古城区为研究区,选取高程、坡度、坡向、距道路距离、距水系距离、降雨量、距断层距离、归一化植被指数8个致灾因子,以区内164处滑坡为基础依据,利用ArcGIS数据分析平台,分别应用信息量和确定性系数模型对研究区滑坡易发性进行评价,并通过自然间断法进行危险等级划分。根据评价结果把该地区分为低、较低、中等、较高、高5个级别,结果显示:在易发性评价较高及以上的分级面积上,信息量模型和确定性系数模型分别占比27.11%和19.88%。利用ROC(Receiver Operating Characteristic)曲线对比分析两种模型评价结果的精度,确定性系数模型评价精度(AUC=0.868)略优于信息量模型评价精度(AUC=0.862),更适用丽江市古城区的地质灾害易发性评价。
Taking the Gucheng District as the study area, eight disaster-causing factors of elevation, slope, slope direction, distance from road, distance from water system, rainfall, distance from fault, and normalized vegetation index were selected, and 164 landslides in the area were used as the basis to evaluate the landslide susceptibility of the study area by applying the informativeness and deterministic coefficient models respectively with ArcGIS data analysis platform, and the hazard level was classified by the natural interruption method.According to the evaluation results, the area was divided into five levels: lower, low, medium, high and higher susceptibility areas, and the results showed that the informativeness model and deterministic coefficient model accounted for 27.11% and 19.88%,respectively, in the area of the partition with higher susceptibility evaluation and high susceptibility.The ROC(Receiver Operating Characteristic)curve was used to compare and analyze the accuracy of the evaluation results of the two models.The evaluation accuracy of the deterministic coefficient model(AUC=0.868)is slightly better than that of the informative model(AUC=0.862),which is more suitable for the evaluation of geohazard susceptibility in Gucheng District.
作者
葛宗旭
刘汉湖
陈行
GE Zong-xu;LIU Han-hu;CHEN Hang(Earth Sciences Academy,Chengdu University of Technology,Chengdu 615001,Sichuan,China)
出处
《云南地理环境研究》
2022年第3期69-76,共8页
Yunnan Geographic Environment Research
基金
云南省自然资源厅“2021年度云南省重点区域地质灾害精细化调查与风险评价”(YNGH[2021]-055)。
关键词
滑坡
地理信息系统
地质灾害
丽江市古城区
landslide
GIS
geological disaster
Gucheng District of Lijiang City