Date: Fri, 8 Nov 2002 15:14:19 +0100 (Paris, Madrid)
From: Martine Feissel-Vernier
To: Henno.Boomkamp@esa.int
Cc: Jean-Paul Berthias , jries@csr.utexas.edu,
Markus.Rothacher@bv.tu-muenchen.de, John.Dow@esa.int,
Jean-Jacques Valette ,
Laurent Soudarin ,
Pascal Willis , feissel@ensg.ign.fr
Subject: Re: proposed JASON campaign
Dear colleague
I have a few short comments on your plan.
- Proposed statistics (2B, p.2). Statistics in the three directions would
be more informative than 3D statistics (if I did not misunderstand your
wording).
- 2C, p. 2: you could use version numbers as part of the filenames.
- It seems to me that having a number of solutions obtained by three
techniques allows you to go a little further in the comparison
statistics. You could estimate the variances of the individual solutions
(all techniques), and the covariances between solutions from the same
technique, by the "three-corner hat technique", i.e. a least square
inversion of equations of the type
"variance(A-B)=variance(A)+variance(B)+covariance(A,B)"
where variance (A-B) is derived from the orbit differences analysis, and
covariance(A,B) is assumed to be zero if solutions A and B are
obtained by different techniques. The latter hypothesis may be not
rigorously verified (common modelling error effetcs may exist), but
that's life...
What do you expect to be the timetable of the project?
I am glad to see the progress in the JASON orbit comparison campaign.
We view the consideration of DORIS in the game as extremely important.
As you know, we would like to develop orbit comparisons in the framework
of the future IDS. We are sure to learn much from your analysis campaign.
With my best wishes.
Martine Feissel-Vernier
Analysis coordinator,
DORIS Pilot Experiment
--
Martine Feissel-Vernier Internet: feissel@ensg.ign.fr
Observatoire de Paris and
Institut Geographique National Telephone: 33 (0) 1 64 15 32 83
ENSG/LAREG Fax: 33 (0) 1 64 15 32 53
8 Av. Blaise Pascal, Champs sur Marne
77455 Marne la Vallee Cedex 2, France
--