| 冯京辉,孙舒轻,赵鹏林,张文龙,杨帅,杨涛,朱楠男,张巧玲.融入 InSAR形变因子的典型黄土覆被区地质灾害危险性评价[J].矿产勘查,2025,16(6):1515-1526 |
| 融入 InSAR形变因子的典型黄土覆被区地质灾害危险性评价 |
| Incorporating InSAR-derived deformation insights into geological hazard risk assessment of loess areash |
| 投稿时间：2024-06-03 |
| DOI：10.20008/j.kckc.202506020 |
| 中文关键词: 典型黄土覆被区 地质灾害 InSAR 耦合模型 危险性评价 |
| 英文关键词: typical loess coverage area geological hazards InSAR coupling model risk assessment |
| 基金项目:本文受陕西省地质灾害隐患点及风险区域识别技术研究项目(陕自然资勘发[2022]70号)、陕西省地质灾害隐患动态识别项目(陕自然资勘发[2023]23号)和矿山地质环境遥感广域动态监测技术研究项目(202412)联合资助。 |
|
| 摘要点击次数: 874 |
| 全文下载次数: 1314 |
| 中文摘要: |
| 遥感技术提供全面、实时的地表信息，在地质灾害防治中具有重要作用，但针对特定地质背景下的典型黄土地区融入 InSAR形变因子的危险性评价存在不足。针对传统的评价因子和评价方法已不能满足提高当前黄土地区地质灾害评价准确性的需求，本文融入 InSAR形变因子构建地质灾害危险性评价指标体系，采用耦合随机森林(RF)—层次分析(AHP)赋权信息量(IV)模型方法对研究区开展地质灾害危险性评价，并使用遥感解译建立的地质灾害隐患数据库对其进行验证。结果表明:基于典型黄土地区地质灾害危险的 9个评价因子按重要性依次为高程、累积降雨量、道路距离、水系距离、InSAR、岩性、坡向、土地利用、坡度；评价分区结果显示高危险区主要分布在土质易湿陷、InSAR形变明显、水系发育密集和降雨量较大的山地沟谷和黄土梁峁区域；使用地质灾害解译数据库隐患点验证精度可得中危险区及以上隐患数量占比高达到 95.61%，高危险区平均隐患点数量可达 4个，低危险区隐患数分布不超过 1个，且通过 AUC值(0.832)计算结果表明模型的性能较好。因此本研究构建的融入 InSAR形变因子的 RF-AHP-IV耦合模型评价体系合理且预测准确性较高，为典型黄土地区地质灾害防治提供了科学决策基础，具有一定的实际应用价值。 |
| 英文摘要: |
| Remote sensing technology provides comprehensive and real-time surface information and plays acrucial role in geological hazard prevention and control. However, there are deficiencies in the risk assessment oftypical loess areas under specific geological backgrounds by integrating InSAR deformation factors. In view of thefact that traditional evaluation factors and evaluation methods can no longer meet the needs of improving the accu.racy of geological disaster assessment in the current loess areas, this paper integrates InSAR deformation factors toconstruct a geological disaster risk assessment index system, adopts the coupled random forest (RF) -hierarchyanalysis (AHP) weighted information volume (IV) model method to carry out geological disaster risk assessment inthe study area, and verifies it using the geological disaster hazard database established by remote sensing interpreta.tion. The results show that the nine evaluation factors of geological hazards in typical loess areas are elevation, cu.mulative rainfall, road distance, water system distance, InSAR, lithology, slope aspect, land use, and slope in orderof importance; the evaluation zoning results show that high-risk areas are mainly distributed in mountain valleysand loess ridges where soil is prone to collapse, InSAR deformation is obvious, water system development is dense,and rainfall is large; the accuracy of the hidden danger points in the geological hazard interpretation database is veri.fied, and the proportion of hidden dangers in medium-risk areas and above is as high as 95.61%, the average num.ber of hidden danger points in high-risk areas can reach 4, and the number of hidden dangers in low-risk areasdoes not exceed 1, and the calculation results of the AUC value (0.832) show that the performance of the model ishighly effective in its predictions.Therefore, the evaluation system of the RF-AHP-IV coupling model incorporatingInSAR deformation factors constructed in this study is reasonable and has high prediction accuracy, which providesa scientific decision-making basis for the prevention and control of geological hazards in typical loess areas and hascertain practical application value. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|