Report of GPS-99 Session 5
Combination of space geodetic techniques for
global dynamics and reference frames
This session was composed of 12 oral talks and 4 poster papers which covered
topics covering a broad spectrum of issues associated with the combination
of space geodetic data. In this summary we highlight the issues and results
presented in this session.
The general areas covered by the session included:
(a) Collocation of space geodetic systems and conventional measurements, and
ties between them; (b) Direct comparison and combination of observations
between different systems; and (c) comparison of time series of position
estimates between difference systems.
Several talks and posters were presented on the collocation of space
geodetic systems. Yoshino Taizoh spoke on the Keystone Project and the
issues of comparison, calibration and combination. The latter being the
ultimate goal once the systematic differences between and within systems are
understood. The difficulties of comparing different space techniques (i.e.,
accurately determining the measurement reference points for each of the
systems) were discussed here and in papers by Yasuhiro Koyama et al. and
Kazuo Shibuya et al. Biases of up to 30 mm still appear different systems as
Kashima Space Research Center although the velocity estimates between
different systems are in reasonably good agreement.
On a global scale, Zuheir Altamimi discussed scale differences and rates of
change between the geodetic systems contributing to the ITRF realization.
There are scale differences of up to 2.3 parts-per-billion (ppb) between the
system and this study addressed some of the possible origins included
atmospheric delay modeling in VLBI (up to 1 ppb) and changes in the
gravitational constant of the Earth, GM, for the satellite based systems.
For SLR, 45% of the change in GM appears to map into scale; for DORIS, about
25% maps in. Tim Springer discussed a more direct comparison of laser
ranging systems to GPS through SLR tracking of Glonass and GPS satellites.
The 50-mm radial-bias between microwave and SLR satellite position
determinations still persists and its origin is not clear. John Dawson also
spoke on the comparison of SLR and microwave orbit determination of Glonass
and while their initially appeared to be no bias in these results,
discussions of the paper suggested that the bias was likely to have be
absorbed in scale parameters estimated in the analysis. Also on this global
scale Thomas Herring spoke about non-secular position changes observed in
GPS analysis and the comparison of these changes with results from
collocated VLBI stations. In some cases, these non-secular variations are
correlated between VLBI and GPS and with atmospheric pressure loading.
However, it is clear that these variations are close to the noise levels of
both systems. He also noted that such comparisons would be more complete
with SLR results but for reasons that are not clear, the SLR community does
not seem to be able to produce loosely-constrained, reasonably high
frequency (1-7 day averages) station position determinations. Remi Ferland
presented results on the combination of IGS SINEX files the improvement in
the precision of these combinations over the last few years.
Other combination results presented included a paper by Govind et al. on the
Asia Pacific regional geodetic project, and a by paper on Jianquing Cai on
comparison of the German geodetic reference system with GPS. One paper was
presented by Per Helge Andersen on the combination of geodetic data at the
observational level rather than a parametric level where most combinations
are made. This is an interesting approach and offers some distinct
advantages if common parameters (such as atmospheric delays and clock
variations) can be made compatible between the difference measurement types.
The results presented were still a little to preliminary to see if there
are significant gains with this type of approach.
In summary, many different aspects of combinations were presented in this
session. There still appears to be systematic differences between systems
as shown by the 50 mm orbits differences and up to 30 mm station position
differences with the latter probably arising from defining the correct
system measurement point. To further complicate comparisons of systems
there appears to non-secular variations in station positions of a few
millimeters horizontally and up to 10-mm vertical.
Thomas A Herring