Comprehensive Analysis of Traveltime Data Collected Through GPS-Acoustic Observation of Seafloor Crustal Movements
Comprehensive Analysis of Traveltime Data Collected Through GPS-Acoustic Observation of Seafloor Crustal Movements
复制标题
通过 GPS 声学观测海底地壳运动收集的走时数据的综合分析
DOI：
10.1002/2017jb014733
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
and M. Kido
中科院分区：
文献类型：
--
作者：
Honsho;C.;and M. Kido
我们开发了一个全面的反演方案来分析通过GPS -声学观测收集的运动时数据。我们的方法使用一个称为“最低点总延迟”（NTD）的量，它类似于全球导航卫星系统分析中的天顶总延迟，来表示声速的时间变化。使用NTD的观测方程是一个近似公式。我们用数值实验检验了它的适用范围，并证明了它在实践中是很好的。我们将所有观测活动的旅行时数据一起用于反演计算，以确定以下内容：(1)特定参考活动时单个精密声学应答器的位置，(2)每次活动时刚性阵列从这些位置的位移，以及(3)每次活动期间NTD的时间变化。我们将此方法应用于实际数据，以更精确地固定阵列的几何形状。此外，我们已经在检测垂直运动和水平运动方面迈出了第一步，尽管需要进一步提高精度。在对反演得到的阵列位移进行线性回归时，其误差协方差对得到的速度有显著影响。在考虑协方差的情况下，无论在反演中采用何种运动作为参考，都能得到几乎相同的速度。
We have developed a comprehensive inversion scheme to analyze traveltime data collected through GPS‐acoustic observations on a campaign basis. Our method uses a quantity called the “nadir total delay” (NTD), which is analogous to the zenith total delay in Global Navigation Satellite Systems analyses, to represent the time variation of the sound speed. The observation equation using the NTD is an approximated formulation. We examine its applicable scope using numerical experiments and demonstrate that the approximation holds well enough in practice. Traveltime data from all our observation campaigns were utilized together for inversion calculation to determine the following: (1) the positions of individual precision acoustic transponders at the time of a particular reference campaign, (2) the displacement from these positions of a rigid array at the time of each campaign, and (3) the time variation of the NTD during each campaign. We applied this method to actual data to fix the array geometry more precisely. Also, we have made a first step in detecting vertical motions as well as horizontal motions, though further enhancement of the accuracy will be required. In performing linear regression of array displacements obtained from the inversion, their error covariance has a significant effect on the resulting velocities. When the covariance is taken into account, nearly the same velocities are obtained no matter what campaign is taken as reference in the inversion.