[IGSMAIL-5518]: GPS week 1400 model changes made at CODE
Stefan Schaer
stefan.schaer at aiub.unibe.ch
Fri Dec 22 06:52:02 PST 2006
******************************************************************************
IGS Electronic Mail 22 Dec 06:52:16 PST 2006 Message Number 5518
******************************************************************************
Author: Stefan Schaer, Urs Hugentobler
Dear IGS colleagues,
A remarkable number of model changes/improvements was activated in
CODE's IGS analysis starting with GPS week 1400. Note that changes that
must be considered as mandatory for IGS analysis are indicated in our
listing with "+":
* Absolute GNSS PCV model (APCV):
+ Switch to the absolute GNSS PCV model IGS05_1400 (latest version:
IGS05_1402). For details see: IGS Mails 5272, 5318, 5438.
+ Consideration of receiver antenna radome codes (actually done since
week 1367) became relevant with the use of the absolute PCV model.
- For ionosphere analysis, this model change has an effect on L1-L2
receiver PCV, but not on satellite PCV correction (L1-L2=0).
* Terrestrial reference frame (TRF):
+ Use of IGS05 TRF realization. Note: IGS05 and ITRF2005 velocity
vectors are identical. IGS05 is conform to the absolute PCV model,
ITRF2005 to relative PCV. For details see: IGS Mails 5447, 5455.
- List of IGS05 fiducial stations was completed with (12) antenna
sharing stations:
-----------------------------------
IGS05 Added Category
-----------------------------------
DARW 50134M001 DARR IGLOS
DAV1 66010M001 DAVR IGLOS
GODE 40451M123 GODZ IGLOS
GOLD 40405S031 GOL2 IGS
IRKT 12313M001 IRKM IGS
MADR 13407S012 MAD2 IGS
ONSA 10402M004 OS0G IGLOS
BAHR 24901M002 S051 NGA
THU3 43001M002 THU2 IGLOS
TIDB 50103M108 TID1/TID2 IGS
YAR1 50107M004 YAR2 IGS
-----------------------------------
Note: An interesting detail is the fact that YAR1 (inactive), not
YAR2 (active) is declared as IGS05 reference station.
* Changes concerning CODE weekly analysis reports:
- Verification of geodetic datum definition: Residuals wrt IGS05 are
provided for the (extended) list of IGS05 fiducial stations in our
weekly analysis summary files. See:
http://igscb.jpl.nasa.gov/mail/igsreport/2006/msg01133.html
http://igscb.jpl.nasa.gov/mail/igsreport/2006/msg01150.html
http://igscb.jpl.nasa.gov/mail/igsreport/2006/msg01168.html
http://igscb.jpl.nasa.gov/mail/igsreport/2006/msg01184.html
http://igscb.jpl.nasa.gov/mail/igsreport/2006/msg01210.html
http://igscb.jpl.nasa.gov/mail/igsreport/2006/msg01234.html
"R" indicates fiducial stations rejected from datum definition,
like, e.g., MKEA (due to recent earthquake).
- The 7 parameters of the corresponding Helmert transformation are
listed. These transformation parameters should be consistent to
those provided, e.g. for week 1400, in:
http://igscb.jpl.nasa.gov/mail/igsreport/2006/msg01137.html
http://igscb.jpl.nasa.gov/mail/igsreport/2006/msg01139.html
- In addition, retrievals of weekly geocenter coordinates are given
(in the TRF used).
* SINEX:
+ GNSS antenna offset parameters (constrained to IGS05 PCV model
values) are included in SINEX for both the GPS and the GLONASS
satellite constellation.
- We do no longer provide CoM weekly station coordinates (translation
defined by geocenter, geocenter coordinates assumed to 0/0/0) but
started to submit SINEX containing CoN coordinates (conform to the
used TRF not only in terms of rotation but also in terms of
translation).
- Geocenter coordinates (constrained to 0/0/0) are included.
* Troposphere modeling:
- Mapping function: changed from Niell to GMF (Global MF).
- A priori model for hydrostatic component: changed from a
Saastamoinen-based to GPT (Global Pressure Temperature) model.
Note: GMF/GPT models were developed by Johannes Boehm, TU Vienna.
- Gradient model: refined.
Note: Elevation mask angle of 3 degrees remains unchanged.
* Solar radiation pressure (RPR) a priori model:
- Start to use CODE RPR model coefficients for GLONASS.
- Use updated model coefficients for GPS. This update is relevant in
particular for all Block II-R satellites.
- Special RPR handling of R18 (behaving differently compared to other
GLONASS satellites) turned out to be erroneous and was switched back
to the zero model (starting with week 1406). This RPR modeling
problem will be further investigated.
* Ocean tide loading (OTL):
+ CMC (center of mass) correction applied in GNSS POD process,
based on the FES2004 OTL model (already used since week 1367).
+ hardisp.f (developed by Duncan Agnew, UCSD) implemented to consider
displacements from a total of 141 constituent tides (see also
IGS Mail 5300).
* Application of Shapiro effect for GNSS:
+ Shapiro (relativistic signal propagtion delay) effect is taken into
account to be in accordance with the absolute GNSS PCV model.
* Mean pole:
- Mean pole computed according to IERS2003 standards.
* Phase windup:
- Not only the geometrical part of the polarization effect is
considered but the effect due to the nominal attitude motion of the
GNSS satellites is taken into account. Note that this model change
does just affect our clock results.
* Format changes:
+ SINEX: Inclusion of satellite antenna offset parameters (see above).
+ SP3: Added 4th header comment line according to IGS standard:
============================================================
/* PCV:IGS05_1400 OL/AL:FES2004 NONE YN ORB:CoN CLK:CoN
============================================================
Legend:
PCV model used
Ocean/atmospheric tidal loading (OL/AL) model used
CMC applied (for OL and for AL, respectively)
Reference frame definition concerning orbits included
Reference frame definition concerning clocks included
+ CLK: Clock RINEX with additional header records to specify
PCV model used and
DCB (differential code bias) model used (see also IGS Mail 5490):
================================================================================
G GPSEST V5.1 IGS05_1400 SYS / PCVS
APPLIED
G GPSEST V5.1 P1C1.DCB @ ftp.unibe.ch/aiub/CODE/ SYS / DCBS
APPLIED
================================================================================
Note: We did not change the clock RINEX format version number (2.00)
because not all mandatory records of format 3.00 can be included.
Best regards,
The CODE AC Team
More information about the IGSMail
mailing list