地壳垂向形变的GNSS影像时空特征分析研究

GNSS coordinate time series are widely used to derive crustal vertical deformation, which can provide fundamental data for fields of geodynamics, geological hazards monitoring, it is of important scientific and practical significance. However, existing methods for vertical deformation monitoring with GNSS coordinate time series are confronted with the following challenges: increasing number of stations invalidates the isotropic assumption in a large area, making it difficult to obtain spatial patterns of vertical deformation; on the other side, in the longer time series, comprehensive effects of skewed distribution and seasonal signal aliasing bias the derived velocity and their uncertainties. To solve above problems, this proposal is planned to propose a novel method to handle skewed distribution of GNSS time series, to establish an optimal vertical velocity model that makes allowance for inter-annual and seasonal variations, improving velocity accuracy; it also aims to construct empirical spatial structure function and to refine spatial interpolation methods to break through spatial resolution constraints to generate crustal vertical deformation GNSS images; thereafter, vertical crustal deformation characteristics with sufficient both spatial and temporal resolution will be revealed. A software system with independent intellectual property rights will be developed and applied to typical areas in China and western U.S to validate above fundamental theories and key techniques. As a result, this proposal would obtain kinematic and "regional model" GNSS images transited from static and "single-point" model, revealing the spatiotemporal evolution characteristics of specified regional crustal vertical deformation. These evolvement characteristics of vertical crustal deformation under different time-space scales would provide technical support for understanding the interaction mechanism inside the earth.

利用GNSS坐标时间序列可获取地壳垂向形变信息，能为地球动力学、地质灾害监测等地学研究提供基础数据，具有重要的科学和现实意义。然而，随着观测站的数目增多和数据积累，现有GNSS监测垂向形变方法面临着不少挑战：日益增多的测站使得大范围内形变的各向同性假设不再适用，空间模式难以获取；更长的时间序列中，偏态分布与季节性信号的混叠效应，导致速度模型不够准确。为此，本项目拟建立顾及偏态分布的速度建模方法，实现无偏稳健的速度估计；兼顾时间和空间分辨率，构建经验时空结构函数，精化时空插值方法，以生成精确的地壳垂向形变GNSS影像（垂向形变图）；开发GNSS影像处理软件系统，并在华北、美国西部等典型垂向形变区域进行应用验证，获取从单点到区域、从静态到动态的GNSS影像，揭示区域内垂向形变的时空演变特征。研究成果将提供典型区域不同时空尺度下地壳垂向形变的演变特性，为理解地球内部相互作用机制提供技术支撑。

长期积累的GNSS坐标时间序列蕴含丰富的地壳形变信息，能为地球动力学、地质灾害监测等地学研究提供基础数据，具有重要的科学和现实意义。本项目利用GNSS垂向坐标时间序列开展了三方面的研究，主要集中在时间序列非线性变化成因方面、时间序列速度模型建模优化方面和GNSS影像构建方面。对于单独测站而言，定量分析了测站坐标阶跃对建立的GNSS参考框架稳定性、测站速度及周年振幅的影响；提出了一种针对阶跃不连续性的自动稳健探测改进方法（MSTARS），提升了阶跃探测的准确性，有效削弱了速度建模样本的偏态分布，改善了速度估计的可靠性；评估分析了不同GNSS卫星天线相位中心改正模型对尺度参数序列的影响，为已有GNSS速度成果的有效归化提供了基础；对于区域测站网而言，定量评价了中、大尺度下、典型空间滤波方法的效果，有效削弱了偏态分布与季节性信号的混叠效应，改善了速度模型的准确性；提出了一种联合时域与频域的GNSS基准站坐标时间序列周期信号建模方法，克服了基准站坐标时间序列的复杂周期性变化趋势及显著的非线性特征带来的速度估计不准确问题；提出了一种顾及周期信号的测站速度稳健估计方法，使速度样本分布更接近正态分布，避免速度、周期信号和瞬态异常信号的影响，并避免方差估计中经验尺度因子的使用；兼顾时间和空间分辨率，构建经验时空结构函数，精化时空插值方法，以生成精确的 地壳垂向形变GNSS影像（垂向形变图）；开发GNSS影像处理软件系统，并在华北、美国西部等典型垂向形变区域进行应用验证，获取从单点到区域、从静态到动态的GNSS影像，揭示区域内垂向形变的时空演变特征，为理解地球内部相互作用机制提供技术支撑。

内容获取失败，请点击重试