周晓慧^1,,
杨艺林¹,
姜卫平^1,2,,,
周星宇²
1. 武汉大学测绘学院, 武汉 400079
2. 武汉大学卫星导航定位技术研究中心, 武汉 400079

基金项目: 国家重点研发计划项目（2018YFC1503600），国家自然科学基金面上项目（41874007），国家****科学基金项目（41525014）联合资助


详细信息

作者简介: 周晓慧, 女, 湖北襄阳人, 副教授, 主要从事GNSS时间序列分析研究.E-mail:xhzhou@sgg.whu.edu.cn

通讯作者: 姜卫平, 男, 湖南溆浦人, 教授, 博士生导师, 主要研究方向为卫星大地测量学的理论及其应用.E-mail:wpjiang@whu.edu.cn

中图分类号: P228

收稿日期:2019-03-06
修回日期:2019-12-20
上线日期:2020-01-05

Preliminary spatial-temporal pattern of vertical deformation revealed by GNSS imaging

ZHOU XiaoHui^1,,
YANG YiLin¹,
JIANG WeiPing^1,2,,,
ZHOU XingYu²
1. School of Geodesy and Geomatics, Wuhan University, Wuhan 400079, China
2. GNSS Research Center, Wuhan University, Wuhan 400079, China


More Information

Corresponding author: JIANG WeiPing,E-mail:wpjiang@whu.edu.cn

MSC: P228

--> Received Date: 06 March 2019
Revised Date: 20 December 2019
Available Online: 05 January 2020


摘要
摘要:地壳垂向形变在连续空间和时间域内呈现显著特征，探求其时空变化特征有助于理解地球物理过程，为研究地球内部相互作用机制提供支持.本文使用美国西部地区PBO与中国大陆CMONOC两个GNSS网测站的坐标时间序列，通过基于中位数并顾及年际差异的非参数方法（MIDAS方法）估计测站的速度与不确定性；建立空间结构函数（SSF）并确定区域内各测站间的相对关系；以此为基础，构建顾及空间结构的滤波器（MSF）以剔除粗差，增强区域共性；最后，基于MSF与图像处理技术对速度场进行空间加密，生成了研究区域空间内连续的地壳垂向形变图，即区域GNSS影像.随后，针对两个研究区域，分别利用MSF验证实验与棋盘格检测验证了GNSS成像方法的合理性及生成GNSS影像的可靠性；并通过对比使用顾及空间结构滤波前后的各测站速度与不确定性生成的GNSS影像，验证了顾及空间结构的滤波方法在GNSS影像生成中的必要性，并分析了其中存在过度平滑与规则圆弧状突变边缘的问题，讨论了可能的解决方案.最终，将两区域GNSS影像结果与已有的大地测量学及地球动力学结果进行了对比，发现美国西部地区的GNSS影像正确反映出了海岸山脉以峰值速度为2 mm·a^-1，内华达山脉以峰值速度为3 mm·a^-1，以及赫布根湖地区以峰值速度为1.5 mm·a^-1隆升；洛杉矶地区（峰值速度为-2.5 mm·a^-1），维多利亚河及其河谷地区（速度为-1.5 mm·a^-1），以及蛇河平原东部、蒙大拿州西南部（速度为-1 mm·a^-1左右）的沉降运动；中国大陆的GNSS影像同样反映出喜马拉雅山脉与青藏高原南部（速度呈现为1.0 mm·a^-1），华北地块与天山地块（速度为1.5 mm·a^-1与0.3~0.6 mm·a^-1）等典型区域的隆升；长江下游地区以苏锡常地区（速度为-2.1 mm·a^-1）为中心，向外速度逐渐减小的沉降运动，以及华南地区（速度呈现为-0.6~-1.5 mm·a^-1）、东北地区（速度呈现为-0.6~-1.5 mm·a^-1）、塔里木盆地（速度呈现为-1.2 mm·a^-1）等区域的沉降运动.因此，本文认为GNSS影像具有合理性与正确性，有助于地壳垂向形变的整体时空分布特征研究.
关键词: GNSS影像/
垂向形变/
时空特征

Abstract:Crustal vertical deformation exhibits significant spatial and temporal variation features. Spatial-temporal pattern is helpful to understand geophysical processes and provide support to the researches on the earth's internal interactions. This paper utilized the GNSS coordinate time series of western US region (Plate Boundary Observation) and Chinese mainland (Crustal Movement Observation Network of China), estimated the velocities and uncertainties of these GNSS stations with a robust nonparametric estimation method, namely, MIDAS; constructed the relationships between stations with Spatial Structure Function (SSF); further, constructed Median Spatial Filter (MSF) and applied to eliminate velocity outliers and enhance regional common characteristics; densified the velocity field on the basis of image processing technology and generated spatially continuous images of crustal vertical deformation (GNSS images) in study areas. In addition, this paper conducted experiment to verify MSF and simulation experiments in these areas with checkerboard test, which verified the reasonability of GNSS imaging and the reliability of the GNSS images generated with the stations in these areas. Moreover, with the comparison on the images generated with the velocities before and after MSF, the necessity of MSF in GNSS imaging was verified and the reason of over-smoothing and the appearance of arcuate abrupt boundaries were analyzed. Also, possible solutions were discussed. Ultimately, the comparison of GNSS images of these areas with existing geodesy and geodynamic results indicated that the GNSS image of western US correctly reflects the uplift of Coast Ranges (Peak velocity value is 2 mm·a^-1), Sierra Nevada (Peak velocity value is 3 mm·a^-1) and Hebgan lake area (1.5 mm·a^-1), and the subsidence of Los Angeles area (Peak velocity value is -2.5 mm·a^-1), Victoria River and its river valley (-1.5 mm·a^-1), eastern Snake River Plain and southwest of Montana (Around -1 mm·a^-1). GNSS image of Chinese mainland also correctly reflects the uplift of Himalaya and southern Tibetan Plateau (1.0 mm·a^-1), north China block (1.5 mm·a^-1) and Tianshan block (0.3 mm·a^-1 to 0.6 mm·a^-1), and the subsidence of lower Yanzte Region (Suzhou-Wuxi-Changzhou area (-2.1 mm·a^-1) as center), south China (-0.6 mm·a^-1 to -1.5 mm·a^-1), northeast of China (-0.6 mm·a^-1 to -1.5 mm·a^-1) and Tarim Basin (-1.2 mm·a^-1). Thus, this paper verifies the reasonability and validity of GNSS imaging, and its helpfulness on the study of temporal and spatial distribution of crustal vertical deformation.
Key words:GNSS image/
Vertical deformation/
Spatial-temporal pattern



PDF全文下载地址:

http://www.geophy.cn/data/article/export-pdf?id=dqwlxb_15295