[daip] SDGRD questions

[Eric Greisen]

Bob Garwood wrote:
> Hi Eric,
> 
> Can sdgrd make images in galactic coordinates?  If the pseudo-UV input 
> file used to get the data into AIPS could indicate coordinate values in 
> galactic coordinates (or other systems, but galactic is the most common 
> after J2000) would that be imported OK and would sdgrd then make an 
> image in galactic coordinates?  I know specifying the center of the map 
> in the sdgrd inputs seems to require a coordinate of RA/Dec and I've 
> always just assumed that that meant it could only make images in RA/Dec, 
> but I'm not sure.
> 
> IF it can only image in RA/Dec would this workaround work:
> 
> lie in the input UV-FITS files. This is what Glen's conversion routine 
> does now.  The coordinates are read from the
> SDFITS file as is, so if they're galactic they stay galactic, and the 
> output UV-FITS file always labels them as RA/Dec.
> 
> make the image in AIPS, write it out to a FITS cube.
> 
> Change the appropriate CTYPEn values in the resulting image from 
> 'RA---GLS','DEC--GLS' to 'GLON-GLS','GLAT-GLS'.
> 
> i.e. I don't think there's anything about the labels RA/DEC that the 
> gridder uses, just the projection type.  So it's fine to not worry about 
> that label until after the image has been made.
> 
> And if there's some better way to do this I'd greatly appreciate hearing 
> about it.
> 
> Thanks.
> 
> Bob

I see no obvious reason why it could not take a data set in galactic 
coordinates, correcly labeled,  and image in GLON and GLAT directly 
without all the fooling around.  I would be interested to see you try it 
and if it fails to send me the info so that I can fix it.  I can't find 
any reference to GLON or RA-- et al in the code at all.  It is inthe OOP 
package part of AIPS and so there may be hidden mysteries but I suspect not.

I vae been on vacation, so pardon the delay in answering.

Eric Greisen