[vsnet-alert 20441] Transient astrometry, was Re: ASASSN-16ob

[Brian Skiff]

On Fri, 2016-12-02 at 06:09 +0200, Berto Monard wrote:
> Hello,
> 
> ASASSN-16ob seems to be a 'new' object with a much larger outburst amplitude.
> 
> Observing images at CBA Kleinkaroo show ASASSN-16ob at a position just off the
> assumed progenitor star.
> Based on UCAC2 reference stars, I measure 06 47 18.83 -64 37 08.8 for the true
> position. Although the differentials might indicate otherwise, the new object
> seems somewhat NE of the USNO 18.4B start.

     Berto's useful note reminds me to suggest observers try
to obtain better coordinates for various transient objects.
While having a position to ~1" or so is sufficient for finding
during an outburst, it is also useful to have much better
coordinates (say, <0".2 real external accuracy) both to identify
the object in archival images/data, and to recover it unambiguously
in the future.  Not so long ago we did not have the means
to go very faint, but that has changed in recent years.

     My suggestion would be to do the astrometry on
several images using only high-accuracy star catalogues as
the reference net.  This can include UCAC2, UCAC4, URAT1,
2MASS, and now GAIA.  For obscured regions in the southern
galactic plane, the VISTA catalogues seem also to be very good.
(Soon enough GAIA will supplant everything else.)  These are
all accessible through the CDS-Strasbourg VizieR catalogue-query
utility, and by other means.

     Perhaps obviously, using a telescope with longer
focal length (say >2 meters f.l.) will in general give better
results than a small telescope or telephoto lens.  (Large
aperture helps, too, but focal-length is more important.)
I would also recommend placing the target in those several
images in different places on the CCD chip, such as half-way
to each corner of the device, as well as in the center.
This will result in using somewhat different sets of reference
stars, and also give you a more realistic estimate of the
uncertainties on the position.  Getting small _internal_
uncertainties in an astrometric solution from a single image
can be misleading compared to real external accuracy.

     Those of us involved in catalogue work, such as with
SIMBAD/VizieR, VSX/GCVS, NED, large databases, etc, will
appreciate your efforts in this regard.


\Brian