Page 83 - 卫星导航2021年第1-2合期

Cheng et al. Satell Navig (2021) 2:2 Page 11 of 12

plate motion in updating the CGCS2000 and its accuracy in those complicated geological areas. Te longer the
evaluation. Tese approaches and test results are summa- time from the CGCS2000 reference epoch, the larger
rized below: the diferences between the transformed coordinates
with the linear velocity model and the CGCS2000
1. For selecting global control stations, a more rigid cri- reference coordinates will be.
terion was proposed and implemented using a sta-
tistic method called the supervised clustering based Te key role of the CGCS2000 maintenance lies in
on the plate-fxed coordinate system to identify and obtaining high-accuracy positions at any epoch. Tis
select the stations whose movements agreed well research indicated that various strategies should be
with the movement of the corresponding plate. Tis developed to improve the accuracy of the CGCS2000
criterion and the supervised clustering were applied frame construction and maintenance. However, for a
to the selection of the stations whose velocity and long-term maintenance, the accuracy of the CGCS2000
azimuth residuals were both under their two sigma frame based on a linear velocity model may be at a level
standards. From the initial 126 global candidate sta- of 2–3 cm, or even worse, especially in those unstable
tions, 92 IGS stations were selected as the global con- areas. If the accuracy cannot meet the requirements for
trol stations. some applications, e.g. accurate local monitoring net-
2. For a better alignment of the Chinese regional net- works and regional reference frames for large-scale map-
work with the ITRF2005, ten out of 92 IGS stations ping, the continuous observations for several years at the
in China and its surrounding area were used as the Chinese national GNSS stations are needed to improve
regional control stations and their coordinates were the accuracy of the velocity feld, or the CGCS2000 frame
obtained from a global combined adjustment under is to be updated.
the control of the 92 global control stations. Te
results indicated the accuracy of the alignment of the Acknowledgements
We are grateful to professor Suqin Wu at China University of Mining and Tech-
10 regional control stations with the ITRF2005 was nology for useful discussion and suggestions. This work was also supported by
improved from a cm-level to a mm-level, compared the team members, Xin Chen, Le Gao and Jun Wan who are master students.
with the coordinates of these stations obtained from Authors’ contributions
the velocities provided in the ITRF documentation. PC proposed the research and supervised the experiments. YC performed
3. To evaluate how well a regional network is aligned the research, analyzed the data and wrote the paper. XW processed the GNSS
with an ITRF, the magnitudes of the coordinate cor- data and estimated the velocities of stations. YX developed the software of
correcting GNSS position to CGCS2000. All authors read and approved the
rections at the 10 regional control stations were used fnal manuscript.
as the indicator. Te smaller the correction value, the
better the agreement of the two frames is. For testing, Funding
This study is supported by the National Key Research and Development Pro-
one-month observations from 148 SDCORS were gram of China (2016YFB0501405) and Natural Resources Innovation Platform
processed and analyzed. Te results show a notice- Construction and Capacity Improvement (A19090).
able systematic deviation between the P DRF and P ERF Availability of data and materials
regional reference datum, and the average diferences All the data used in this project, including the observations of a period of
at the 148 stations in the Y, X and Z components are 10 years at GPS stations CMONOC, GPS time series from Scripps Orbit and
about 1.2 cm, 1.0 cm and − 3 mm respectively. Te Permanent Array Center website and local CORS observations in Shandong
province.
alignment of the P ERF with the ITRF is better than
that of the P DRF. Competing interests
4. Te velocities at 1025 Chinese national GNSS refer- The authors declare that they have no competing interests.
ence stations were obtained by a least squares esti- Author details
mation with the mean precisions in the N, E, U com- 1 Chinese Academy of Surveying and Mapping, 28 Lianhuachi West Road,
2
ponents being ± 0.124, ± 0.127 and ± 0.563 mm/a, Beijing 100036, China. Aerospace Information Research Institute, Chinese
Academy of Sciences, No. 9 Dengzhuang South Road, Haidian District, Bei-
respectively. Tey were the sample data, to construct jing 100094, China.
a 1° × 1° gridded velocity feld in mainland China
using the inverse distance weighting interpolation Received: 27 July 2020 Accepted: 1 December 2020
method. Te overall precisions of the constructed
velocity feld in the E, N components were 0.78 mm/a
and 0.95 mm/a, respectively. References
5. Currently, the accuracy of the CGCS2000 frame Altamimi, Z., Métivier, L., & Collilieux, X. (2012). ITRF2008 plate motion
maintenance based on a linear velocity model is at a model. Journal of Geophysical Research, 117, B07402. https ://doi.
2–3 cm level, and poorer in western China, especially org/10.1029/2011J B0089 30.