[Comm2011] test plate system using 9 pointings

[Jeffrey Bout]

Oh, by the way, in the plot you see only 8 pointings. The one that is
missing, does not know how to behave in a processing queue...

2011/6/13 Jeffrey Bout <jeffreybout at gmail.com>:
> Hi all,
>
> I've been busy with checking the coadds overlap of the 9 pointings for
> testing the plate_kids_decimal plate system.
> In this figure
>
> http://www.astro.rug.nl/~bout/nine_pointings.png
>
> you can see the boundingboxes of sourcelists derived from the coadds.
> the overlap between the coadds in RA is around 2.35 arcmin and in DEC
> around 7.45 arcmin. The figure already shows that there is little
> variation in these numbers.
> Whether these overlaps are enough depends on the dither strategy. Each
> coadd has 5 dithers with 1.7 arcmin variation in RA and 5.7 arcmin
> variation in DEC. These number are smaller than the overlaps between
> overlaps, so when coadds are combined to some supercoadds, no areas
> with less coverage will appear in the coadd-overlaps.
> So I conclude that at low declination the overlap between coadds
> following the plate_kids_decimal is enough.
>
> PS: Here a detailplot of RA overlap between coadds with sources (+ and
> x of 2 coadds).
> http://www.astro.rug.nl/~bout/nine_pointings_detail.png
>
> Jeffrey
>