耦合高精度InSAR监测的区域性地面沉降模拟研究

Ground subsidence has become the most serious geological hazard in China, mainly caused by over exploitations of underground resources. It has been reported that a total area of 79,000 square kilometers in national wide had subsided over 200 millimeters till 2009, and more than 50 cities encountered very serious ground subsidence. In order to tackle the increasing threats of ground subsidence, the State Council of China government approved a ten-year Plan in 2012 to call for measures to monitor, control and prevent land subsidence. Therefore, to conduct a thorough research on accurately monitoring and modeling ground subsidence is of great significance at this time. .Although in the past, many researches on ground subsidence monitoring and subsidence modeling respectively have been undertaken, they were usually conducted independently and in an isolate way. Especially, researches about subsidence monitoring concern only how to develop techniques for accurately mapping the subsidence, such as the time-series interferometric SAR (InSAR) techniques represented by PS-InSAR (permanent scatterer InSAR) and SBAS (small baseline subset) InSAR, but pay little attention on developing subsidence models by taking into account the InSAR measurements. On the other side, researches on building ground subsidence models usually rely on some ground-based measurements of land subsidence, such as those made through GPS and leveling. Those observations are usually very sparse in space. Subsequently, the subsidence models cannot be very accurate because they are based on incomplete observations of ground subsidence. .In this research, we seek to develop a novel and integrated solution for accurate monitoring and modeling of ground subsidence over large areas. In particular, firstly, we will focus on developing a new set of InSAR algorithms to process the phase information and retrieve the deformation over distributed scatterers (DSs), and to adopt piecewise linear deformation models for accurate retrieval of accumulative subsidence. This research represents an update to the classical time series InSAR techniques, which have been put into operational use to monitor ground subsidence over urban areas, but usually fails to achieve reliable subsidence estimation over rural areas where point-like scatterers are very sparse. Only after this issue is solved, the ground subsidence information over large areas composed of both urban and rural environments can be derived accurately. Secondly, we seek to develop a 3D numerical model fully coupling groundwater flow with ground subsidence. This model will be constrained by and consistent with the highly-dense ground subsidence data derived by the improved time series InSAR technique. Based on this robust 3D numerical model, a more reliable prediction of ground subsidence can be accordingly achieved. The two components of this research are inter-linked, and together could form the basis to build operational solutions against ground subsidence in China in the future.

地面沉降是影响我国生态文明建设和可持续发展的主要地质灾害。现有研究缺乏地面沉降监测与数值模拟的深度结合，因而难以准确揭示地面沉降的演化机理及发展趋势。本项目拟将地面沉降InSAR监测和数值模拟协同考虑，重点研究城区与非城区一体化的区域性地面沉降InSAR高频高密度监测、InSAR监测结果校正的地面沉降三维流固全耦合数值模拟两大关键难题，利用InSAR获取的高精度沉降监测，为高分辨率的地面沉降数值模拟提供校正和验证数据，建立更可靠的地面沉降数值模型，并开展由建筑荷载、地下水开采及二者叠加作用引起的地面沉降预测分析。研究成果将为我国地面沉降灾害的精细化防控提供重要科学基础。

本项目主要研究如何利用时间序列InSAR高精度监测地面沉降以及融合InSAR监测结果的地面沉降数值模拟和预测分析，是国内将地面沉降监测与数值模拟深度结合方面的首次探索，具有明显的创新性。. 项目取得了四方面的研究成果。1）提出了一种新的分布式目标的提取方法和相位估计方法，使得分布式目标也能进行较高精度的时序分析和形变提取，解决了点目标缺乏的非城区的InSAR测量点密度不足的难题。沧州地区的实验表明，引入分布式目标后，监测点密度提高了近5倍。2）提出了二阶连续分段线性形变模型。与标准的线性形变模型相比，二阶连续分段线性模型的监测精度提高了14.9%，该模型对于地表形变呈现较强非线性情况下的形变监测具有重要价值。3）结合沧州地区第四纪松散沉积层应力-应变特征室内试验结果与InSAR高分辨率地面沉降监测结果，建立了沧州地区高分辨率计算规模的地下水开采引发地面沉降的三维流固全耦合数值模型。采用InSAR监测数据对模型进行验证，81.82%的拟合点误差小于10mm/a。4）完成了沧州地区地面沉降模拟预测。根据沧州市建筑荷载的分布特征及地下水开采现状，假设保持2016年的地下水开采布局，以构建的地面沉降数值模型为基础，模拟预测了沧州市2017年1月~2020年12月的地面沉降量。. 项目研究成果在我国地面沉降的精细化防控、地下水资源的科学开采等方面具有重要的科学意义和实用价值。