[fitswcs] [forwarded] WCS for co-moving maps

[William Thompson]

On 09/16/15 08:02, Ian Avruch wrote:
> Dear William,
>
> On 09/11/15 21:05, William Thompson wrote:
>>> WCSNAME = 'Offset from SSO'
>>> NAXIS   =                    3
>>>
>>> CTYPE3  = 'FREQ    '
>>> CUNIT3  = 'GHz     '           / WCS: Unit axis 3
>>> SPECSYS = 'SOURCE  '
>>> SSYSOBS = 'TOPOCENT'
>>>
>>> NAXIS3  =                 2001
>>> CRPIX3  =               1000.0 / [] WCS: Reference layer index, unit=Scalar
>>> CRVAL3  =            1800.0000 / [] WCS: Wavelength, time, ... of reference
>>> layer; unit=length,time,...
>>> CDELT3  =               5.0E-4 / [] WCS: Scale in 3rd dimension, unit=Length,
>>> Duration, ...
>>>
>>> NAXIS1  =                   11
>>> NAXIS2  =                   11
>>> CTYPE1  = 'OFLN-GLS'
>>> CTYPE2  = 'OFLT-GLS'
>>
>> Shouldn't this be -SFL instead of -GLS?
>
> Interesting, why do you suggest "-SFL" ?
>
> We follow the AIPS GLS definition (AIPS Memo 46 by E.Greisen), old
> as it is. The reference coordinates (alpha_0, dec_0) are the
> instantaneous position of the moving target, so by definition equal (0,0).

This is discussed in section 6.1.4 of Calabretta and Greisen (2002), where they 
describe why the AIPS -GLS projection was renamed to -SFL in the WCS formalism.

Bill


-- 
William Thompson
NASA Goddard Space Flight Center
Code 671
Greenbelt, MD  20771
USA

301-286-2040
William.T.Thompson at nasa.gov