物探化探计算技术2025，Vol.47Issue(6)：931-940,10.DOI:10.12474/wthtjs.20241104-0001
顾及InSAR形变的区域滑坡易发性评价
Evaluation of regional landslide susceptibility considering InSAR deformation
赵芹 1李清泉1
作者信息
- 1. 重庆交通大学智慧城市学院,重庆 402247
- 折叠
摘要
Abstract
Aiming at the problem of late update of landslide data sets and the lack of dynamic evaluation factors to represent surface deformation in slope susceptibility evaluation,this paper took the Wanzhou District of Chongqing as the research area,interpreted new slip slope through SBAS-InSAR results,and introduced InSAR deformation rate into the evaluation factors.Particle swarm optimization(PSO)was used to optimize the hyperparameters of random forest(RF)and extreme gradient lift(XGBoost)models,and landslide susceptibility mapping was completed.The results showed that the deformation rate in the study area ranged from-170.441 to 127.406 mm/a,and 58 new landslides were identified.Both models show good training and test accuracy,among which XGBoost is more prominent.After adding InSAR deformation rate factor,the AUC values of RF and XGBoost on the test set are increased to 0.8431 and 0.8801,respectively.The areas with high landslide susceptibility in the study area are mainly concentrated along the Yangtze River and its tributaries.The results of landslide susceptibility evaluation considering InSAR deformation can provide theoretical support and scientific guidance for landslide prevention and control along the Yangtze River in Wanzhou District.关键词
滑坡易发性评价/SBAS-InSAR/粒子群优化/随机森林/极端梯度提升/万州区Key words
landslide susceptibility evaluation/SBAS-InSAR/particle swarm optimization/random forest/extreme gradient boosting/Wanzhou District分类
天文与地球科学引用本文复制引用
赵芹,李清泉..顾及InSAR形变的区域滑坡易发性评价[J].物探化探计算技术,2025,47(6):931-940,10.
