白鹤滩水电站为世界第二大水电站，对其周边潜在失稳边坡开展长期变形监测和时空变形规律分析具重要意义。本文将InSAR技术用于白鹤滩水电站库岸潜在失稳边坡时序形变估计，以探究其在库水位和降水变化联合作用下的时空变形规律。针对传统InSAR线性形变模型中较少顾及环境因素影响的局限，本文提出改进的时序InSAR库岸区形变估计方法。这一方法在形变建模环节以周期模型为基础模型，融入降水因子以充分顾及库岸区域所受库水位变化和降水影响；在形变估计环节通过组建时序相位方程组解算形变速率、高程改正值和降水因子参数，进而实现潜在失稳边坡时序形变估计。实验以白鹤滩水电站库岸区域大弯子-骑骡沟段为例，获取了为期31个月的时序形变结果。研究发现此区域形变以线性趋势为主导，伴随库水位变化体现周期变化特征，并相对旱雨季过渡月有2个月的滞后效应；近江区域坡体形变大于远江区域，最大累积形变达到-155 mm；隧道下游出口段和中段特征点上的形变差异最高达98 mm，是大弯子隧道洞壁产生裂缝的主要原因。利用残余高通形变验证改进模型的建模精度，结果显示改进模型相比传统InSAR模型精度提升12.5%；利用现场GNSS监测结果验证形变的外部精度为±1.7 mm。研究结果可为白鹤滩水电站库岸潜在失稳边坡长期变形监测预警提供参考。

The Baihetan Hydropower Station is the world"s second-largest hydropower station, with a large number of potential unstable slopes distributed on both sides of its reservoir bank. It is of great significance to carry out long-term and effective deformation monitoring for it. Conventional ground contact monitoring methods are mostly based on discrete point monitoring, which makes it difficult to generate surface deformation results and timely extract boundary information of unstable zones. This article applies InSAR technology to monitor the temporal deformation of the unstable slope of the Baihetan hydropower station reservoir bank, in order to explore its spatiotemporal deformation regularity under the combined effect of water level and precipitation changes. Taking the Dawanzi-Qiluogou section of the Baihetan hydropower station reservoir area as an example, an improved time-series InSAR method is proposed to address the limitations of traditional InSAR linear deformation models that do not take into account the impact of climate and environment. This method is based on a periodic model in the deformation modeling process, and incorporates precipitation factors considering the impact of precipitation changes in the reservoir bank area. Then, the deformation rate, elevation correction value, and precipitation factor parameters are calculated using temporal phase observations to achieve the calculation of temporal deformation results. The experiment obtained the temporal deformation results of the Dawanzi-Qiluogou section for a period of 31 months. The research results showed that the slope deformation in the near river area of the Dawanzi-Qiluogou section was greater than that in the far river area, and the maximum cumulative deformation reached -155 mm. The deformation in this area is dominated by a linear The deformation in this area is dominated by a linear trend, accompanied by periodic changes in reservoir water level, and has a lag effect of 2 months relative to the peak precipitation. The cumulative deformation difference on both sides of the slope caused by water storage reaches 16.9 mm, which is the main cause of cracks in the tunnel wall of the downstream exit section of the Dawanzi Tunnel. Using residual high pass deformation to verify the modeling accuracy of the improved model, it was found that the improved model has an accuracy increase of 12.5% compared to traditional InSAR model; The accuracy of deformation obtained by the method proposed in this paper is ± 1.7 mm based on on-site GNSS monitoring results. The research results can provide reference for long-term deformation monitoring and landslide hazard identification of unstable slopes on the reservoir bank of Baihetan Hydropower Station.