From: Jim Ray "(NGS" 301-713-2850 "x112)" <jimr@ngs.noaa.gov>
Message-Id: <200603241954.OAA17929@ness.ngs.noaa.gov>
Subject: Re: estimating EOP rates for DORIS
To: ids.analysis.forum@cls.fr
Date: Fri, 24 Mar 2006 14:54:23 -0500 (EST)
In-Reply-To: <6.1.1.1.2.20060323113020.021ee7a8@pop.cls.fr> from "laurent.soudarin@cls.fr>" at Mar "23," 2006 "11:30:23" am
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

I would like to express full agreement with Zuheir's views concerning
polar motion rate estimation in DORIS analyses.

There are two basic questions: 1) Is DORIS sensitive to variations in PM
over periods <24 hr?  2) Are the parameters required nonetheless for some
other reason (i.e., multi-technique combinations)?  The answer to both is:
No.

The point made by Daniel -- "The discussions we had last year within the
GPS community showed that pole rates were not useful to improve EOPs." --
is irrelevant and highly misleading.  Parameters should be added into the
geodetic solution when the observational data are sensitive to the physical
effect involved.  That is very definitely the case for GPS and marginally
for VLBI.  Whether the PM offset estimates are "improved" or not is
irrelevant (an in any case a tiny effect).  Failing to adjust the EOP-rates
would introduce a corresponding error in the geodetic results.  The same is
not true for DORIS and therefore the parameters are neither necessary nor
desirable.

PM-rates are also not needed in the multi-technique combination because
DORIS provides daily noon PM estimates, consistent with all the other
techniques except VLBI.

I do not follow the details of how the various apriori PM series are
constructed.  Of course, they should be as close to reality as possible in
order to allow offset-only adjustments to be most meaningful.  In no case
should PM-rates be "fixed" to any apriori value explicitly, although by
not adjusting PM-rates some apriori values will naturally be imposed.
Likewise, there should be no constraints applied to PM-rate parameters to
enforce continuity at day-boundaries (or for any other reasons), as this
will be inherited by any subsequent multi-technique combination.

I view satellite LOD parameters quite differently.  Their adjustment is
probably justified for DORIS, as with GPS, to soak up unmodeled orbital
errors at least as much as true length of day variations (which, by the
way, are much larger than PM variations).

Good luck,
--Jim