Re: wcscoordsys to eq-j2000 does not work

[Mohammad Akhlaghi]

Hi Alex,

Thanks a lot for reporting this problem, it has been fixed (by updating 
the 'EQUINOX' keyword, see below) in the following commit:

https://git.savannah.gnu.org/cgit/gnuastro.git/commit/?id=470d4434ab

You can get the latest Gnuastro version with this bug fix from 
http://akhlaghi.org/gnuastro-latest.tar.lz

The keywords of the output files had indeed changed (here I am running 
'astfits' on the two files that you had uploaded). To visually see this 
change better, we'll first write the header keywords in plain-text files:

$ astfits NGC6946_scaled.fits -h1 > NGC6946_scaled.txt

$ astfits NGC6946_scaled-eq-j2000.fits -h1 > NGC6946_scaled-eq-j2000.txt

Then we can look at the differing lines between the two files. You see 
that the 'CRVAL1' (RA of reference point), 'CRVAL2' (Dec of reference 
point), 'LONPOLE' (longitude of the pole) and 'LATPOLE' (latitute of 
pole) have indeed changed significantly (on the scale of half a degree 
in some cases!). You also see that 'RADESYS' (RA/Dec System) has also 
been changed from 'FK4' to 'FK5'.



$ diff NGC6946_scaled.txt NGC6946_scaled-eq-j2000.txt
22,25c22,25
< CRVAL1  =               308.45 / [deg] Coordinate value at reference 
point
< CRVAL2  =          59.98333333 / [deg] Coordinate value at reference 
point
< LONPOLE =                180.0 / [deg] Native longitude of celestial 
pole
< LATPOLE =          59.98333333 / [deg] Native latitude of celestial 
pole
---
> CRVAL1  =     308.702781271596 / [deg] Coordinate value at reference 
point
> CRVAL2  =     60.1568349215312 / [deg] Coordinate value at reference 
point
> LONPOLE =        179.564003507 / [deg] Native longitude of celestial 
pole
> LATPOLE =        60.1568349215 / [deg] Native latitude of celestial 
pole
29c29
< RADESYS = 'FK4'                / Equatorial coordinate system 

---
> RADESYS = 'FK5'                / Equatorial coordinate system



The confusion was caused by the 'EQUINOX' keyword not being changed (the 
output still had a vale of '1950.0'). So with the commit above, Gnuastro 
now also updates that (as it should have).

Once you install the latest version you will notice that the only 
difference is in the value of 'EQUINOX' (all the other numbers will be 
the same as the output you sent). This is because the main keyword in 
this aspect is the 'RADESYS' one (which was correctly set before also: 
so in case of conflict 'RADESYS' has priority over 'EQUINOX').

Cheers,
Mohammad