[IGSMAIL-2899]: Original / New realization of ITRF97

Remi Ferland ferland at geod.nrcan.gc.ca
Thu Jun 22 13:50:29 PDT 2000


******************************************************************************
IGS Electronic Mail      23 Jun 11:22:02 PDT 2000      Message Number 2899
******************************************************************************

Author: Remi Ferland

Dear Colleagues,

	As you are well aware, starting with GPS week 1065
we have switched to a new realization of ITRF97 (IGS00P04_RS51.SNX).
The new realization is a subset of the cumulative solution IGS00P04.snx,
containing weekly solutions from the last 4 years (1996-1999).
It can be found in:
	ftp	igscb.jpl.nasa.gov
	cd	/pub/station/coord
	get	IGS00P04_RS51.SNX.Z

It replaces the original realization of ITRF97 (ITRF97_IGS_RS51.SNX)
extracted by Zuheir from the ITRF97 solution, and available at the same
location.

	The objective of the change was to ensure a better internal
consistency of the IGS products. The new realization has been aligned
to the original realization to minimize potential discontinuities. 

	The magnitude of discontinuity was checked by transforming the
weekly combinations (1060-1064) using the new IGS realization
"IGS00P04_RS51.snx". This week solution (1065) was transformed using
"ITRF97_IGS_RS51.snx". All transformations were unweighted. The
statistics of the 6 weekly transformation are :

Parameter	Average	  Std.		Normalized
---------	-------	  ----		----------
R X (mas)	 0.0152	(+-0.0081)	 1.8
R Y (mas)	 0.0063	(+-0.0041)	 1.5
R Z (mas)	-0.0022	(+-0.0073)	-0.3
T X (m)		-0.0007	(+-0.0003)	-2.1
T Y (m)		 0.0005	(+-0.0002)	 2.0
T Z (m)		 0.0002	(+-0.0002)	 0.9
SCL (ppb)	-0.0807	(+-0.0294)	-2.7

Note:	Signs of GPS 1065 transformations have been reversed to ensure
	consistency.

	Based on the small sample containing the 6 most recent weeks,
there is no significant difference (at 3 sigmas) between the two
realizations of ITRF, when applied to the weekly combination.

	The transformation parameters uncertainty derived from the 
propagation of the covariance information from the weekly solutions
and the ITRF realization is about 0.04mas in rotation,
1.0mm and 1.5mm in XY and Z translation and 0.2ppb in scale. This 
also indicates that the average transformation parameters are
not significant!

	The discontinuity on the cumulative solution (IGS00P23.snx)
expressed in terms of the transformation parameters is:

R X (mas)	-0.00010
R Y (mas)	-0.00004
R Z (mas)	-0.00026
T X (m)		 0.00000
T Y (m)		 0.00001
T Z (m)		 0.00000
SCL (ppb)	-0.00055

d R X (mas/y)	-0.00015
d R Y (mas/y)	-0.00003
d R Z (mas/y)	-0.00033
d T X (m/y)	 0.00000
d T Y (m/y)	 0.00001
d T Z (m/y)	 0.00000
d SCL (ppb/y)	-0.00107

	The discontinuity on the cumulative solution for GPS weeks
1060-1064 is (as expected) almost identical to the transformation given
above. The magnitude of transformations are very small … they are well
within the noise level!

Please feel free to do your own comparison,

Best Regards, 
Remi,
--------------4DE546B51CD206EAA1CE4177
Content-Type: text/x-vcard; charset=us-ascii;
 name="ferland.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Remi Ferland
Content-Disposition: attachment;
 filename="ferland.vcf"

begin:vcard 
n:Ferland;Remi
tel;fax:(613) 995-3215
tel;work:(613) 995-4002
x-mozilla-html:FALSE
org:Geodetic Surveys;Natural Resources Canada
adr:;;456 - 615 Booth Street;Ottawa;Ontario;K1A 0E9;Canada
version:2.1
email;internet:ferland at geod.nrcan.gc.ca
title:Research Officer
fn:Remi Ferland
end:vcard

--------------4DE546B51CD206EAA1CE4177--




More information about the IGSMail mailing list