Skip navigation


You are in:  Home » Publications » Reports » Report 23

Crustal Motion in East Antarctica Derived from GPS Observations

M. Jia, J. Dawson, G. Luton, G. Johnston, R. Govind, J. Manning
Geoscience Earth Monitoring Group, Geoscience Australia, Canberra, Australia

Abstract

Eight years of continuous permanent GPS data and three years of GPS campaign data are used to provide current estimates of crustal motion in Antarctica within the International Terrestrial Reference Frame 2000 (ITRF2000) (Altamimi et al. 2002). Crustal motions derived for this paper are compared with published results from several groups. The crustal motion estimates are consistent with that provided by other groups in the horizontal components but not in the vertical component.

1. Introduction

Current-day velocities of crustal deformation in Antarctica are important indicators for many geodetic and geophysical studies, including plate motion, intra-plate tectonics, Antarctic postglacial rebound and absolute sea level change. GPS geodesy has the potential to measure velocities of the crust directly over periods of maybe a few years, especially in horizontal components, as demonstrated by Dietrich et al. (2001) and Sella et al. (2001).

In this paper, almost eight years of continuous permanent GPS data and three years of GPS campaign data, for a total of 50 sites in the Antarctic and Australian regions are analysed with three strategies (A, B and C). Solution A is a combination of the Geoscience Australia’s IGS RNAAC (Regional Network Associate Analysis Centre) solutions as submitted to the IGS from 1996 to present. Solution B is a combination of the re-processed daily regional solutions only using data observed during the Geoscience Australia Antarctic campaigns of 2001, 2002 and 2003. While solution C is a combination of re-processed daily regional solutions from continuous permanent GPS data from 1995 to 2001 in Antarctica and Australia region.

The crustal velocities in Antarctica relative to ITRF2000 are derived. These results are compared with that provided by several other groups and some conclusions are drawn from this analysis and comparison.

2. Data

This research uses GPS data collected from both continuous GPS networks operated by Geoscience Australia (formerly, AUSLIG) and also other organizations, and additionally includes Antarctic summer campaign data as well.

Continuous GPS sites around this region have gradually increased since 1989. Up to now more than 40 such GPS sites are available around this region. However due to inconsistent hardware and software description and availability of reliable precise IGS orbital products, only the data after 1995 are used in this paper. Data from another 33 Antarctica sites, which were collected during three summer campaigns are also used. Seven Antarctica sites, which have at least a three-year time span of GPS data, are shown in Figure 1. The occupation and duration history of all sites are listed in Table 1.

Table 1 Site occupations used in this analysis (* denotes continuous occupation, numerical values for sites A351 denote occupation days for that year’s campaign)

Site 1995 1996 1997 1998 1999 2000 2001 2002 2003 Span
(years)
Antarctic plate
A351             17 45 46 3
CAS1 * * * * * * * 45 50 9
DAV1 * * * * * * * 45 50 3
DUM1       * * * *     4
KERG * * * * * * * 45 50 9
MAW1 * * * * * * * 45 50 9
MCM1 *                  
MCM4 * * * * * * * 45 50 9
Australia plate
ALIC * * * * * * *     7
AUCK * * * * * * *     7
BUR1         * * *     3
CEDU *   * * * * *     7
COCO * * * * * * *     7
DARW * * * * * * *     7
DST1   * * * * * *     6
GRIM * * * * *         5
HIL1     * * * * *     5
HOB2 * * * * * * *     7
JAB1     * * * * *     5
KARR * * * * * * *     7
KOUC         * * *     3
NOUM       * * * *     4
PERT * * * * * * *     7
STR1       * * * *     4
SUVA         * * *     3
TID1   * * * * * *     6
TOW2 * * * * * * *     7
WEL1 * * *             3
YAR1 * * * * * * *     7
Eurasia plate
BAKO       * * * *     4
BINT         * * *     3
GETI       * * * *     4
NTUS     * * * * *     5
Pacific plate
CHAT * * * * * * *     7
FALE       * * * *     4
KWJ1   * * * * * *     6
MAC1 * * * * * * *     7

Fig. 1 Antarctica GPS sites, which have at least three-year time span of GPS data

3. Data Processing

Three data processing strategies (A, B and C) are reviewed in this paper. The Bernese GPS Software Version 4.2 (Hugentobler, et al., 2001) is used in the daily data processing for all three strategies.

3.1 Weekly Combined Regional IGS RNAAC Solutions (Strategy A)

In the case of processing Strategy A the GPS data from the 16 Australia regional sites (including the Australian Antarctic sites) are processed and combined into weekly SINEX files and stored as products of the IGS RNAAC. The data spans the period 1 January 1996 to 31 July 2003. IGS final orbital and the Earth rotation parameters are used. This data has not been re-processed and the computed solution strategy is not consistent over the period of the solution but generally reflects the IGS standard at the time of computation (around 30 days after observation). For historical reasons ionosphere free floating solutions are used as final solutions.

3.2 Reprocessed Daily Solutions (Strategy B and C)

For processing strategy B and C the GPS data are processed on a daily basis using the Bernese Processing Engine (BPE) of the Bernese GPS Software Version 4.2. Dual-frequency carrier-phase and code data are used. RINEX file sizes less than 70% of normal size are excluded. Code measurements are only used for receiver clock synchronisation. The elevation cut-off angle is 10° with elevation-dependant data weighting. The data sampling rate is 30 seconds for strategy B and 180 seconds for strategy C. Standards and procedures for the data processing are briefly summarised as follows:

4. Data Analysis

Crustal velocity estimates are based on a weighted least squares line fit of the weekly position estimates for strategy A, and to the daily position estimates for the strategies B and C. Twelve IGS core stations around this region comprise the reference network. The reference stations, which are constrained to their ITRF2000 values with weighted constraints on Net-Translation/Net-Rotation/Net-Scale change and their rates, are listed in bold letters in Table 1. Incorrect antenna heights are corrected using Bernese GPS Software Version 4.2. Outliers, defined as both points that lie off the best fit line by more than 3 times the standard deviation and points whose residuals are larger than 3cm for vertical component, and 2cm for horizontal components, are not used in the final combined solutions. In this paper, velocity error estimates account for only white noise and parameters of annual and semi-annual signals are not estimated due to the use of limited campaign data. Therefore, the estimated standard deviations for velocities may be not very reliable at this stage.

Fig.2 Coordinate time series for site MAW1 (strategy A)

Fig.3 Coordinate time series for site A351 (strategy B)

Fig.4 Coordinate time series for site MAW1 (strategy C)

Typical time series plots are shown in Figure 2 for strategy A, in Figure 3 for strategy B and in Figure 4 for strategy C. The estimated velocities and their standard deviations for the three solution strategies are listed in Table 2. .

Table 2 Estimated velocities and their standard deviations for the three solution strategies

Site
Solution
 Velocity
(mm/yr) 		Velocity standard deviation
(mm/yr)
East North Vertical East North Vertical
CAS1
A
 2.6 -9.7 2.2 0.6 0.4 1.5
B
 2.1 -9.5 4.0 0.2 0.2 0.2
C
 3.1 -9.4 5.0 0.2 0.2 0.3
DAV1
A
 -0.8 -5.4 3.5 0.4 0.4 1.1
B
 -1.3 -5.6 3.0 0.3 0.2 0.7
C
 -0.4 -5.2 1.6 0.2 0.2 0.3
KERG
A
            
B
 6.0 -2.7 5.0 0.2 0.2 0.1
C
 8.0 -4.1 4.2 0.2 0.2 0.3
MAW1
A
 -2.2 -3.1 4.3 0.5 0.4 1.1
B
 -1.9 -2.5 3.5 0.3 0.2 0.5
C
 -1.1 -2.9 2.1 0.2 0.2 0.3
MCM4
A
            
B
 6.3 -10.0 -1.1 0.5 0.4 2.0
C
 8.3 -11.5 10.0 0.2 0.2 0.4
A351
A
            
B
 -2.9 -4.7 -0.5 0.3 0.3 1.0
C
 -2.3 -5.1 -0.9 0.8 0.6 3.2

5. Comparison Of Results

The crustal motion velocities derived from GPS are compared with that from other groups. The results of comparisons are listed in Table 3. The NUVEL1A-NNR values are from DeMets et al, 1990 and DeMets et al., 1994. The ITRF2000 values are from Altamimi, 2002. The JPL values are from JPL, 2003. The IGS (weekly) MIT values are from Herring, 2003. The SOPAC values are from Bock, 2003. Table 3 shows that the velocities in horizontal directions from all groups are compatible. The RMS velocity differences are generally less than 1mm/yr and have a maximum RMS of 1.2 mm/yr. On the other hand, the velocities in vertical direction show greater variability than that in the horizontal directions. When the likely outliers (indicated in bold) are included the maximum RMS is 8.1 mm/yr and the all RMS values are larger than 3 mm/yr. When the likely outliers are excluded then all the RMS values are less than 2 mm/yr. Further analysis of the relative motions between plates and intra-plate deformation analysis are beyond of the scope of this paper and will be discussed in the future. .

Table 3 Velocity comparisons (- denotes velocity unavailability. Bold values show likely outliers and values for mean and RMS differences in the vertical are calculated excluding these outliers).

Site Solution Velocity (mm/yr)
East North Vertical
CAS1 NUVEL1A-NNR
2.0
-8.7
-
ITRF2000
2.6
-9.6
3.7
JPL
2.7
-11.2
4.6
SOPAC
1.9
-9.9
3.5
IGS (WEEKLY) MIT
2.0
-9.6
-6.8
IGS (WEEKLY) Official
2.6
-10.2
2.8
A
2.6
-9.7
2.0
B
2.1
-9.5
4.0
C
3.1
-9.4
5.0
MEAN
2.5
-9.9
2.4 3.7
RMS
0.4
0.6
3.5 1.0
DAV1 NUVEL1A-NNR
-2.2
-2.9
-
ITRF2000
-1.5
-4.8
4.3
JPL
-1.7
-5.6
1.7
SOPAC
-2.0
-4.9
3.1
IGS (WEEKLY) MIT
-1.7
-5.0
-8.5
IGS (WEEKLY) Official
-1.6
-5.9
2.4
A
-0.9
-5.5
3.6
B
-1.3
-5.6
3.0
C
-0.4
-5.2
1.6
MEAN
-1.4
-5.3
1.4 2.8
RMS
0.5
0.4
4.1 1.0
KERG NUVEL1A-NNR
6.4
-1.3
-
ITRF2000
6.0
-3.1
5.0
JPL
4.5
-4.3
0.9
SOPAC
5.8
-3.7
4.9
IGS (WEEKLY) MIT
5.9
-2.5
-7.6
IGS (WEEKLY) Official
5.6
-3.6
4.1
A
-
-
-
B
6.0
-2.7
5.0
C
8.0
-4.1
4.2
MEAN
5.9
-3.4
2.3 4.0
RMS
1.0
0.7
4.6 1.6
MAW1 NUVEL1A-NNR
-2.1
0.3
-
ITRF2000
-2.3
-2.6
2.8
JPL
-2.8
-4.7
0.4
SOPAC
-2.0
-3.0
1.1
IGS (WEEKLY) MIT
-2.3
-2.8
-15.0
IGS (WEEKLY) Official
-2.9
-3.5
0.8
Dietrich et al. (2001)
-3.0
-5.0
-
A
-2.3
-3.2
4.5
B
-1.9
-2.5
3.5
C
-1.1
-2.9
2.1
MEAN
-2.2
-3.4
-1.1 2.1
MCM4 NUVEL1A-NNR
7.5
-11.7
-
  ITRF2000
9.7
-11.7
0.8
  JPL
10.0
-11.9
-1.1
  SOPAC
8.7
-11.6
2.5
  IGS (WEEKLY) MIT
9.2
-10.9
-16.8
  IGS (WEEKLY) Official
9.1
-12.0
1.7
  A
-
-
-
  B
6.3
-10.0
-1.1
  C
8.3
-11.5
10.0
  MEAN
8.7
-11.3
-0.6 0.6
  RMS
1.2
0.7
8.1 1.6
A351 NUVEL1A-NNR
-4.3
-2.3
-
  ITRF2000
-
-
-
  JPL
-
-
-
  SOPAC
-
-
-
  IGS (WEEKLY) MIT
-
-
-
  IGS (WEEKLY) Official
-
-
-
  A
-
-
-
  B
-2.9
-4.7
-0.5
  C
-2.3
-5.1
-0.9
  MEAN
-2.6
-4.9
-0.7
  RMS
-
-
-

Summary

Three solution strategies have been used to derive current crustal motion velocities in Antarctica. All results in horizontal directions from the three solution strategies are compatible with that from others. Significant inconsistency in vertical component between different groups exists. Likely outliers in the vertical component are visible for all sites between the IGS (WEEKLY) MIT estimates and all other solutions. Another likely outlier appears between the MCM4 solution C and other solutions, the explanation of which is not clear at this stage. Further analysis of MCM4 in time will provide more conclusive estimates of velocity and perhaps time series discontinuity. Much longer time data spans, more accurate loading corrections and velocity estimation models, which take into account annual and semi-annual signals, may be needed to derive reliable results in the vertical component.

Acknowledgments

The authors are grateful to the IGS community for the IGS products and data used in this analysis, and to Paul Digney for his contribution to the Antarctic field activities. This research was supported by the Australia Antarctic Division (ASAC proposal 1159).

References

Altamimi Z., Sillard P., and Boucher C., 2002, ITRF2000: A New Release of the International Terrestrial Reference Frame for Earth Science Applications, J. Geophys. Res. 107, ETH 2-1 to ETG 2-19.

Bock Y., 2003, Global Time Series, web site: ftp://garner.ucsd.edu/pub/timeseries/

DeMets C., Gordon D., Argus D.F. and Stein S., 1990, Current plate motions, Geophys. J. Int.101, 425-278.

DeMets C., Gordon D., Argus D.F. and Stein S., 1994, Effect of recent revisions to the geomagnetic reversal time scale on estimates of current plate motions, Geophys. Res. Lett., 21, 2191-2194.

Dietrich-R; Dach-R; Engelhardt-G; Ihde-J; Korth-W; Kutterer-H-J; Lindner-K; Mayer-M; Menge-F; Miller-H; Mueller-C; Niemeier-W; Perlt-J; Pohl-M; Salbach-H; Schenke-H-W; Schoene-T; Seeber-G; Veit-A; Voelksen-C, 2001, ITRF coordinates and plate velocities from repeated GPS campaigns in Antarctica; an analysis based on different individual solutions. Journal of Geodesy. 74; 11-12, Pages 756-766. 2001.

Herring T., 2003. Global Time Series, web site: http://www-gpsg.mit.edu/~tah/MIT_IGS _AAC/index2.html

Hugentobler U., Schaer S. and Fridez P., 2001, Bernese GPS Software Version 4.2. Astronomical Institute, University of Berne. McCarthy, 1996, IERS Technical note No. 21: IERS Conventions (1996)

Sella-Giovanni-F; Mao-Ailin; Dixon-Timothy; Stein-Seth, 2001, REVEL; a new global plate velocity model and changes in plate velocities over the last 25 ma. Abstracts with Programs - Geological Society of America. 33; 6, Pages 397. 2001.

JPL (Jet Propulson Laboratory), 2003, GPS Time Series, web site: http://sideshow.jpl. nasa.gov/mbh/series.html