{"title":"考虑到条件因子的空间异质性,用于滑坡易发性评估的改进型缓冲控制取样策略","authors":"Lei-Lei Liu, Hao Xiao, Yi-Li Zhang, Can Yang","doi":"10.1007/s10064-024-04008-x","DOIUrl":null,"url":null,"abstract":"<div><p>The selection of landslide and non-landslide samples significantly influences the performance of machine learning (ML)-based landslide susceptibility assessment (LSA). The commonly used buffer-controlled sampling (BCS) strategy for selecting non-landslide samples overlooks the spatial heterogeneity of the geological environment and lacks a standardized method for determining buffer radius. As a result, the sampling process introduces significant uncertainty to ML models. This paper proposes an improved BCS strategy that incorporates the spatial heterogeneity of conditioning factors to address this issue. The proposed strategy generates a buffer zone for each landslide by merging all neighboring areas with the same attributes as the landslide and then calculates the average equivalent radius of those zones for comparative analysis. The random forest (RF) and the support vector machine (SVM) models are employed to predict the landslide susceptibility of Taojiang County, China, using both the improved and the traditional BCS strategy. Furthermore, the impact of different buffer radii on the model prediction is thoroughly investigated to provide guidance for the selection of buffer radius. The results demonstrate that a buffer radius of less than 3,000 m is optimal in Taojiang County. Compared with the traditional RF and SVM model, the corresponding improved models exhibit superior performance, with higher AUC values and increased peak frequency ratios in areas of very high susceptibility. These findings confirm the effectiveness of the proposed strategy, offering valuable guidance for buffer radius selection and improving the ML-based LSA.</p></div>","PeriodicalId":500,"journal":{"name":"Bulletin of Engineering Geology and the Environment","volume":"83 12","pages":""},"PeriodicalIF":3.7000,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"An improved buffer-controlled sampling strategy for landslide susceptibility assessment considering the spatial heterogeneity of conditioning factors\",\"authors\":\"Lei-Lei Liu, Hao Xiao, Yi-Li Zhang, Can Yang\",\"doi\":\"10.1007/s10064-024-04008-x\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>The selection of landslide and non-landslide samples significantly influences the performance of machine learning (ML)-based landslide susceptibility assessment (LSA). The commonly used buffer-controlled sampling (BCS) strategy for selecting non-landslide samples overlooks the spatial heterogeneity of the geological environment and lacks a standardized method for determining buffer radius. As a result, the sampling process introduces significant uncertainty to ML models. This paper proposes an improved BCS strategy that incorporates the spatial heterogeneity of conditioning factors to address this issue. The proposed strategy generates a buffer zone for each landslide by merging all neighboring areas with the same attributes as the landslide and then calculates the average equivalent radius of those zones for comparative analysis. The random forest (RF) and the support vector machine (SVM) models are employed to predict the landslide susceptibility of Taojiang County, China, using both the improved and the traditional BCS strategy. Furthermore, the impact of different buffer radii on the model prediction is thoroughly investigated to provide guidance for the selection of buffer radius. The results demonstrate that a buffer radius of less than 3,000 m is optimal in Taojiang County. Compared with the traditional RF and SVM model, the corresponding improved models exhibit superior performance, with higher AUC values and increased peak frequency ratios in areas of very high susceptibility. These findings confirm the effectiveness of the proposed strategy, offering valuable guidance for buffer radius selection and improving the ML-based LSA.</p></div>\",\"PeriodicalId\":500,\"journal\":{\"name\":\"Bulletin of Engineering Geology and the Environment\",\"volume\":\"83 12\",\"pages\":\"\"},\"PeriodicalIF\":3.7000,\"publicationDate\":\"2024-11-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Bulletin of Engineering Geology and the Environment\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://link.springer.com/article/10.1007/s10064-024-04008-x\",\"RegionNum\":2,\"RegionCategory\":\"工程技术\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENGINEERING, ENVIRONMENTAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Bulletin of Engineering Geology and the Environment","FirstCategoryId":"5","ListUrlMain":"https://link.springer.com/article/10.1007/s10064-024-04008-x","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ENVIRONMENTAL","Score":null,"Total":0}
引用次数: 0
摘要
滑坡和非滑坡样本的选择对基于机器学习(ML)的滑坡易感性评估(LSA)的性能有很大影响。用于选择非滑坡样本的常用缓冲区控制取样(BCS)策略忽视了地质环境的空间异质性,并且缺乏确定缓冲区半径的标准化方法。因此,取样过程会给 ML 模型带来很大的不确定性。本文针对这一问题提出了一种改进的 BCS 策略,该策略结合了条件因子的空间异质性。建议的策略通过合并与滑坡具有相同属性的所有相邻区域,为每个滑坡生成一个缓冲区,然后计算这些区域的平均等效半径,以便进行比较分析。利用随机森林(RF)和支持向量机(SVM)模型,采用改进的 BCS 策略和传统的 BCS 策略预测中国桃江县的滑坡易发性。此外,还深入研究了不同缓冲半径对模型预测的影响,为缓冲半径的选择提供指导。结果表明,在桃江县,小于 3,000 米的缓冲半径是最佳的。与传统的射频和 SVM 模型相比,相应的改进模型表现出更优越的性能,AUC 值更高,在极高易感地区的峰频比也有所提高。这些发现证实了所提策略的有效性,为缓冲区半径选择和改进基于 ML 的 LSA 提供了宝贵的指导。
An improved buffer-controlled sampling strategy for landslide susceptibility assessment considering the spatial heterogeneity of conditioning factors
The selection of landslide and non-landslide samples significantly influences the performance of machine learning (ML)-based landslide susceptibility assessment (LSA). The commonly used buffer-controlled sampling (BCS) strategy for selecting non-landslide samples overlooks the spatial heterogeneity of the geological environment and lacks a standardized method for determining buffer radius. As a result, the sampling process introduces significant uncertainty to ML models. This paper proposes an improved BCS strategy that incorporates the spatial heterogeneity of conditioning factors to address this issue. The proposed strategy generates a buffer zone for each landslide by merging all neighboring areas with the same attributes as the landslide and then calculates the average equivalent radius of those zones for comparative analysis. The random forest (RF) and the support vector machine (SVM) models are employed to predict the landslide susceptibility of Taojiang County, China, using both the improved and the traditional BCS strategy. Furthermore, the impact of different buffer radii on the model prediction is thoroughly investigated to provide guidance for the selection of buffer radius. The results demonstrate that a buffer radius of less than 3,000 m is optimal in Taojiang County. Compared with the traditional RF and SVM model, the corresponding improved models exhibit superior performance, with higher AUC values and increased peak frequency ratios in areas of very high susceptibility. These findings confirm the effectiveness of the proposed strategy, offering valuable guidance for buffer radius selection and improving the ML-based LSA.
期刊介绍:
Engineering geology is defined in the statutes of the IAEG as the science devoted to the investigation, study and solution of engineering and environmental problems which may arise as the result of the interaction between geology and the works or activities of man, as well as of the prediction of and development of measures for the prevention or remediation of geological hazards. Engineering geology embraces:
• the applications/implications of the geomorphology, structural geology, and hydrogeological conditions of geological formations;
• the characterisation of the mineralogical, physico-geomechanical, chemical and hydraulic properties of all earth materials involved in construction, resource recovery and environmental change;
• the assessment of the mechanical and hydrological behaviour of soil and rock masses;
• the prediction of changes to the above properties with time;
• the determination of the parameters to be considered in the stability analysis of engineering works and earth masses.