[daip] VLA archive uvfits file antenna positions.

[Dave Mehringer]

Thanks Eric
Are the details of the FIXANT OPTYPE=‘CENT’ transformation from the old VLA data documented somewhere? If possible, we’d like to be able to replicate that transformation in CASA, so that users of both packages can expect the same results independent of the package used. 

Also, since you mention an “old” format, I assume there is a new format. Can you tell me the specifics of that and when the switch was made, or point me to documentation?
Thanks again. 

> On Mar 25, 2020, at 5:04 PM, Eric Greisen <egreisen at nrao.edu> wrote:
> 
> 
> On 3/25/2020 2:09 PM, Dave Mehringer wrote:
>> Hi Eric
>> I was wondering if you could help shed some light on antenna positions in VLA archive uvfits files. Your uvfits spec description regarding the STABXYZ column of the AN table states:
>> 
>> “Station coordinates. The STABXYZ array shall give the coordinate vector (element 1 is the x coordinate, element 2 is the y coordinate, and element 3 is the z coordinate) of the antenna relative to the array center
>> defined in the header, provided that the antenna is not an orbiting antenna. The coordinate system used for the antenna coordinates is indicated by the FRAME keyword in the header. In the absence of FRAME information, the coordinate system longitudes are assumed to be based in Greenwich, England if the array center is all zero and to be based at the array center if it is not zero. Note that this means that the x and y coordinates of the station are rotated wrt the coordinate system used to express the array center and so cannot be added to them in a simple fashion.”
>> 
>> The archive file I’ve been given has no FRAME keyword, and the antenna x-y positions are plotted in the attached figure. Ultimately, I need to transform the positions in the uvfits AN table to ITRF coordinates, since that’s generally what CASA needs. 
>> 
>> I was hoping that you could tell me how to do that. There is some code in our uvfits that suggests that the uvfits AN table positions should be rotated relative to the array center by an angle equal to the longitude of the array center (and indeed this results in the positions in the attached plot being transformed into approximately at least the expected shape of the array with the positive Y axis corresponding to north. In the code, those positions are then added to the ITRF array center coordinates. 
>> 
>> Can you tell me if the coordinate system used in VLA archive uvfits files is such that the above described transformation will indeed result in the ITRF antenna positions? And if not, can you provide details of how this can be achieved? 
>> Thanks 
>> 
>> 
>> 
> The format you are seeing is the antenna format for all old VLA data.  There is a task in AIPS that will convert this correct (but unfortunate) format to an Earth-centric one.  The task is FIXAN with OPTYPE='CENT'.  Note that the WRNG OPTYPE was for a format CASA wrote that pretended to convert JVLA antenna coordinates (Earth centered) to the coordinates they assumed AIPS required.  I got them to stop that.
> 
> Eric
> 
> 
> 
> 
> 
>  	
> This email has been checked for viruses by Avast antivirus software. 
> www.avast.com
> 
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listmgr.nrao.edu/pipermail/daip/attachments/20200325/4d543f63/attachment-0001.html>