Deformation Monitoring and Prediction of Riverbank Slopes Based on InSAR and LSTM-RF: Mainstream Bank Slopes in the Middle-western Section of the Three Gorges Reservoir Area
Monitoring and predicting reservoir slope deformation, which is influenced by water level fluctuations and rainfall infiltration, is essential for slope stability assessment and early warning. Long short-term memory (LSTM) networks capture long-term dependency features in time-series data, while random forest (RF) models address complex nonlinear relationships among multiple variables. A method for slope deformation prediction, integrating LSTM and RF, was proposed. Small baseline subset-interferometric synthetic aperture radar (SBAS-InSAR) technology was used to process Sentinel-1 imagery from 2018 to 2022, generating a time-series deformation field for the mainstream bank slopes in the middle-western section of the Three Gorges Reservoir Area. The reliability of the InSAR inversion results was validated with global navigation satellite system (GNSS) monitoring data. Precipitation and water level data, along with deformation data, were coupled as inputs for the prediction model. The annual average deformation rate in the study area ranges from -59.31 to 36.4 mm/year, with acceleration during periods of concentrated rainfall and water level fluctuations. Predictions show that cumulative deformation will peak at -177.2 mm between June and September 2022. The root mean squared error (RMSE) of the test set ranges from 3.410 to 3.968 mm, and the mean absolute error (MAE) ranges from 2.522 to 3.088 mm. The model exhibits good prediction accuracy and stability, confirming the applicability of the method for predicting reservoir bank slope deformation.
DENG Yi-fei, YANG Chun-xia, HUANG Wen-yi, ZHOU Ying-tong, DENG Qi.
Deformation Monitoring and Prediction of Riverbank Slopes Based on InSAR and LSTM-RF: Mainstream Bank Slopes in the Middle-western Section of the Three Gorges Reservoir Area[J].
Science Technology and Engineering, 2026, 26(28): 12065-12074 DOI:10.12404/j.issn.1671-1815.2600197
点 A 与点B 的变化趋势较一致,分别由-147.2、-150.1 mm 逐步发展至 9 月 18 日的-165.0、-171.3 mm,之后波动幅度较小。点 D 在预测期内表现出更明显的阶段性起伏,整体仍呈下沉发展趋势。结合研究区水文背景,5—9月为降雨集中期且库水位存在季节性涨落,预测期内6—9月形变加速与降雨入渗及水位波动过程相对应,9月中旬后随降雨减弱形变变化幅度减弱,模型形变预测结果与降雨过程及库水位涨落阶段具有较好的对应关系。
ZhouChuangbin, JiangQinghui, YaoChi, et al. Research progress and challenges in deformation and stability of high-steep slopes in water conservancy and hydropower projects[J]. Journal of Hydropower Engineering, 2025, 44(1): 1-17.
CaoHai, LiHaifeng, XuTao, et al. Research on safety monitoring and risk early warning of slopes near dam reservoir shores[J]. Water Resources and Hydropower Engineering, 2025, 56(S2): 561-567.
[5]
LiuD, CuiY, WangH, et al. Assessment of local outburst flood risk from successive landslides: case study of Baige landslide-dammed lake, upper Jinsha River, eastern Tibet[J]. Journal of Hydrology, 2021, 599: 126294.
ZhouJian, TangMinggao, PeiFangge, et al. Short-term prediction of deformation in bank slides based on machine learning[J]. Mountain Research, 2023, 41(6): 891-903.
DengLizheng, YuanHongyong, ZhangMingzhi, et al. Research progress on landslide deformation monitoring and early warning technology[J]. Journal of Tsinghua University (Science and Technology), 2023, 63(6): 849-864.
DaiKeren, WuMingtang, ZhuoGuanchen, et al. Review on InSAR early identification and monitoring of reservoir landslides for large hydropower engineering projects in southwest mountainous area of China[J]. Journal of Earth Sciences and Environment, 2023, 45(3): 559-577.
TaoChangxu, ZhangYongshuang, RenSanshao. Research progress on InSAR technology application in landslide identification and monitoring[J]. Chinese Geology, 2025, 52(2): 513-526.
XuDongli, WangTao, DawaSangye, et al. Application of SBAS-InSAR technology in monitoring deformation of extra-large landslides: a case study of the Pangcun Landslide in Xizang Autonomous Region[J]. Science Technology and Engineering, 2021, 21(35): 14927-14935.
DongJianjun, MeiYuan, LiXin, et al. SBAS-InSAR monitoring of safety and stability of high-altitude waste dump slopes[J]. China Safety Science Journal, 2022, 32(1): 92-101.
JuShujun, GaoZhiliang, JiangMingcheng, et al. Time-series InSAR monitoring of geological hazards in the Dadu River Basin[J]. Science Technology and Engineering, 2023, 23(31): 13234-13242.
[20]
NiuW, HuX, LinB, et al. Detection and monitoring of potential geological disaster using SBAS-InSAR technology[J]. KSCE Journal of Civil Engineering, 2023, 27(11): 4884-4896.
ZhangQiaofeng, TangMinggao, ZhouJian, et al. Sensitivity analysis of landslide deformation and reservoir water level changes at baihetan reservoir[J]. Hydrogeology and Engineering Geology, 2026, 53(2): 213-222.
ZhangPan, LiWeile, LuHuayan, et al. Relationship between deformation characteristics of active landslides in reservoir areas and reservoir water level changes based on SBAS-InSAR: a case study of the Maoergaishan Hydropower Station in Heishui County, Sichuan Province[J]. Science Technology Engineering, 2024, 24(28): 11991-12002.
GePengfei, ChenMi, ZhangLingqi, et al. Analysis of rainfall impact on the Jinjiling Landslide in the Three Gorges Reservoir Area using time-series InSAR technology[J]. Science Technology and Engineering, 2025, 25(24): 10161-10174.
[27]
MengS, ShiZ, LiG, et al. Machine learning-based data mining of reservoir landslide triggering mechanisms and failure time prediction for displacement sudden state[J]. International Journal of Rock Mechanics and Mining Sciences, 2025, 194: 106234.
PanJianping, CaiZhuoyan, ZhaoRuiqi, et al. Prediction of surface deformation on unstable slopes using InSAR and GRU neural networks[J]. Bulletin of Surveying and Mapping, 2023(3): 33-38.
PengYibo, YangWeifang, YanXiangrong, et al. Three-dimensional surface deformation prediction in mining areas based on a combined SMA-CNN-GRU-Attention model[J]. Bulletin of Surveying and Mapping, 2025(3): 8-14.
WenLinhai, LiuPing, HuangXinkang, et al. Prediction of surface subsidence in mining areas based on the combined ARIMA-DBO-LSTM model: a case study of the Yangshui mining area in Kaiyang County, Guizhou Province[J]. Science Technology and Engineering, 2025, 25(24): 10193-10203.
JiangJinjin, LiuJia, JiangShan, et al. Prediction of soft soil surface subsidence based on a multivariate regression LSTM model: a case study in Zhuhai, Guangdong[J]. Chinese Journal of Geological Hazards and Prevention, 2025, 36(6): 119-126.
LiShuaifei, LiuChangyi, HuXiasong, et al. Deformation monitoring and prediction of the Xijitan giant landslide based on SBAS-InSAR technology and LSTM neural networks[J]. Arid Zone Research, 2025, 42(6): 1126-1137.
ChenYuanyuan, ZhaoBingkun, WangHui, et al. Time-series InSAR surface deformation prediction based on LSTM models[J]. Yangtze River, 2024, 55(3): 146-152.
HuangHuibao, JiangDejun, LiuHeng, et al. Deformation monitoring and prediction of the Dadu River Baifugou Hydropower Station based on multi-temporal InSAR technology[J]. Science Technology and Engineering, 2023, 23(30): 13112-13120.
WangBenhao, WangYanxia, XiangXueyong, et al. Monitoring and prediction of ground subsidence in mining areas using DS-InSAR and LSTM[J]. Bulletin of Surveying and Mapping, 2024(11): 44-48.
XuWenzheng, LuShuqiang, LinZhen, et al. Deformation monitoring and prediction of the Fanjiaping landslide using combined InSAR and neural network techniques[J]. Hydrogeology and Engineering Geology, 2025, 52(2): 150-163.
ZhuTongtong, ShiXuguo, ZhouChao, et al. Monitoring and analysis of the stability of the shuping landslide in the three gorges reservoir area using Sentinel-1 data from 2016 to 2020[J]. Geomatics and Information Science of Wuhan University, 2021, 46(10): 1560-1568.
YuXianyu, TangLi. A coupled SMOTE-Tomek and CNN model for landslide susceptibility assessment and its application: a case study of the Zigui—Badong section in the Three Gorges Reservoir Area[J]. The Chinese Journal of Geological Hazard and Control, 2024, 35(3): 141-151.
[56]
LiS, XuW, LiZ. Review of the SBAS InSAR time-series algorithms, applications, and challenges[J]. Geodesy and Geodynamics, 2022, 13(2): 114-126.
ChenXueye, QiXiaoshuai, PengBo, et al. Early identification of landslides using SBAS-InSAR technology combined with deep neural networks[J]. Geomatics and Information Science of Wuhan University, 2025, 50(6): 1210-1224.
FeiXinfeng, TianYe, ZhaoChaoying, et al. Identification and deformation monitoring of unstable slopes in the Longyangxia Re-servoir Area of the upper Yellow River based on multi-temporal InSAR technology[J]. Journal of Earth Sciences and Environment, 2023, 45(3): 578-589.
SunXiwen, HeXiaoxing, LuTieding, et al. Prediction of reservoir surface displacement deformation using a hybrid variational modal long short-term memory network[J]. Journal of National University of Defense Technology, 2025, 47(3): 151-161.
ZhangTengjun, YangHonglei, LiuYoufeng. Research on landslide deformation monitoring and prediction using combined time-series InSAR and Transformer[J]. Journal of Geodesy and Geodynamics, 2026, 46(5): 628-636.
TangFeifei, TangTianjun, ZhuHongzhou, et al. Prediction of rainfall-induced landslide displacement Using attention mechanism and Bi-LSTM[J]. Bulletin of Surveying and Mapping, 2022(9): 74-79.
GuiXiaoli, YangBin, SongYuxin, et al. Detailed evaluation of landslide susceptibility using SBAS-InSAR combined with machine learning[J]. Science of Surveying and Mapping, 2025, 50(11): 124-132.