基于InSAR 3-D观测的溪洛渡库区持续缓动型滑坡四维时空演化研究

A large amount of longtime slow-moving landslides are located in Reservoirs. InSAR technology has a ability on detecting and calculating three-dimensional deformation of landslide's surface. This project intends to explore the relationship of surface displacements and underground deformation of slow-moving landslide and their space-time evolution mechanism. A large slow-moving landslide in Huanghua town, which is in deformation before and after impoundment in Xiluodu Reservoir, is chosen as the research object. Firstly, the 3D deformation of the surface are calculated by regional coverage, high accuracy, multiple period deformation observation using InSAR technology, and shape and position of the sliding surface are estimated by some experiments. Secondly, based on the geotechnical properties of the sliding body and sliding zone, combined with the reservoir level, rainfall and field investigation, we analyze main influencing factors of landslide deformation and landslide "time - space" four-dimensional evolution rule. At last, the relationship between surface displacement and deep deformation is explored by means of numerical simulation. This project's main means of work is deep space observation. Based on the study of landslide three-dimensional surface displacement, aiming to break through the research problem of the inversion of landslide deep deformation characteristics, we are expecting to explore the four- dimensions of the longtime slow-moving landslide evolution mechanism. The implementation of this project will be not only to promote the application of InSAR technology in the landslide prevention in Xiluodu and to promote understanding of the formation mechanism and evolution process of slow-moving landslide, but also be of great significance in revealing the law of landslide deep deformation.

库区存在大量持续缓动型滑坡，利用InSAR技术可获取并计算滑坡表面的三维变形，本项目拟探索持续缓动型滑坡地表变形与深部变形的关系和时空演化机制。以溪洛渡库区蓄水前、后均有变形的黄华镇某大型滑坡为研究对象，通过改进的InSAR技术进行面覆盖、高精度、多期次的变形观测，计算地表三维变形，估算滑面形态与位置；根据滑体、滑带的岩土力学性质，结合库水位、降雨和现场调查等，分析确定滑坡变形主要影响因素并分析滑坡体地表"时间-空间"四维演化规律；结合数值模拟手段，探索地表变形与深部变形的关系。本项目以深空观测为主要工作手段，以滑坡地表三维变形为研究基础，以滑坡深部变形特征反演为目标，探索缓动型滑坡体的四维演化机制。项目不仅对于促进InSAR技术的应用和溪洛渡库区滑坡防治具有直接作用，更促进认识持续缓动型滑坡的形成机理及演化过程，且对于创新滑坡灾害调查工作模式和揭示滑坡深部普遍变形规律具有重要意义。

库区边坡变形一直是世界各国非常关注的问题。溪洛渡库区2013年5月首次蓄水后，库区边坡稳定性普遍恶化。为了了解水库蓄水后斜坡变形的空间和时间特征，研究采用InSAR技术，利用多源SAR数据（ENVISAT ASAR、ALOS PALSAR-1、ALOS-PALSAR-2、TerraSAR-X和Sentinel-1)对水库首次蓄水前、蓄水后和正常运行期间水位变化周期的变形斜坡进行了解译与分析。区域上，首次蓄水前有40个边坡处于变形状态，其中有36个在首次蓄水后继续变形。2014年10月至2021年2月，共解译152个变形边坡。整体来说，变形斜坡数量逐年减少。从2016年10月到2021年2月，29.6%的变形边坡在水位下降期间变形更显著，4.6%的变形边坡在水位上升期间变形更显著，其余的变形边坡(65.8%)在水位上升和下降期间均变形较大。在45个基岩变形边坡中，碳酸盐岩组有16个，碎屑岩岩组28个，变质岩(板岩)1个。统计结果表明，变形边坡中70%为土质岸坡，22%为顺层岩质岸坡。研究还发现，顺层岩质岸坡和崩坡积物组成的岸坡，其变形对水位上升很敏感，而由地表堆积物组成的土质边坡的变形对水位下降很敏感。结合岸坡结构类型、岩性等因素，进一步探讨了导致变形边坡破坏模式的可能原因，并详细分析了翌子村滑坡、沙依戈滑坡、岩湾变形体、星光三组变形体和作坊洗滑坡。在上述基础上，总结分析了多源SAR数据在库区的适用性。