Date: Fri, 15 Nov 2002 10:14:34 -0800
From: Pascal Willis <Pascal.R.Willis@jpl.nasa.gov>
To: jean-paul.berthias@cnes.fr
Cc: "John C. Ries" <ries@csr.utexas.edu>,
    Bruce J Haines <bjh@cobra.jpl.nasa.gov>,
    Yoaz Bar-Sever <Yoaz.Bar-Sever@jpl.nasa.gov>,
    nzelensk@geodesy.gsfc.nasa.gov, sluthcke@xyz.gsfc.nasa.gov,
    krchoi@csr.utexas.edu, Martine Feissel <feissel@ensg.ign.fr>
Subject: Re: DORIS phase center determination/ all satellites

Hello Jean-Paul,

thanks for your message.

usually my posteriori standard deviation for the R_OFF parameter are 
around 0.9 cm for spot2, spot4 and 0.5 for spot5; 1.3 for envisat; 
1.4 for topex and 0.9 for JASON. Of course, these depends mostly on 
the weight I use for the data, as well as the a priori constraints I 
use.

The numbers I gave in my table correspond to the dispersion around 
the estimated mean (given in the table). I only use values for days 
with small a posteriori standard deviation to estimate the mean.

The factor of 2 is typical of my runs with the Gipsy/Oasis. When I 
get an posteriori standard deviations, it usually needs to be 
multiplied by 2 to obtain a more realistic value, taking into account 
possible systematic bias.

I will compare my orbits looking at the eclipses aspect that you mentioned.

Have a nice week-end
Pascal