[Issues] Problems with astrometric

[Philippe Heraudeau]

Edwin A. Valentijn wrote:
> It has been a very wise and well thought over decision to decouple the 
> astrometric solution form the re-gridding process.
> It allows to re-process astrometric solutions, and also to arrive at 
> global solutions. The plots of residuals of astrometric solutions, using 
> the input data and not the re-gridded data are formally fully correct 
> and do the right thing and are not misleading- they do what they are 
> supposed to show: the residuals of  the astrometric solution.
> If nevertheless larger errors appear, after re-gridding, than something 
> else is wrong. The standard way to inspect this is the PSF end-to-end check
Hi Edwin,

I hope I am wrong and that there are no problems !

But the problem seems already to appear using source lists extracted 
from the reduced (with the right astrometry)
and therefore does not seem to come from the regridding process.

Maybe Erik could comment on the fact that "the inspect plot  is made 
from source lists extracted from the
reducedframes with 'bad' astrometry to which the astrometric solution is 
mathematically applied before making these
plots but no new image is created as this point."?
This is nice and _has_ to be there but this is not the plot we want to 
show in a paper for example, the one we want must come from
the regriddedframes where astrometry is really applied.

It is difficult to have a look at this with the psf alone at this small 
differences. Also, the psf does not come into play for the inspect, just 
the position of sources matters.

I can only strongly recommend to check this further and make a inspect 
task which will show the residuals
from source lists extracted from reduced (with proper astrometry and 
regridded frames to check that everything is all right
1) with the astrometry and
2) with other steps in the system (e.g. regridding...).
It would be a very nice Quality Control.
The inspect tool exists already in AW, my suggestion is to have a nice 
interface to pass the proper source lists to the inspect method.

Philippe

-- 
==========================================================================
Dr. Philippe Heraudeau
Argelander-Institut für Astronomie Email: heraudeau at astro.uni-bonn.de
Auf dem Hügel 71                   Tel.: +49 (0)228 733673
D-53121 Bonn                       Fax.: +49 (0)228 731775
Germany
=========================================================================