WCS support with AIPS.
Octavi Fors Aldrich wrote on Jan 05, 1999
Hi all! We have been adding WCS support to a set of images with immatchx.ccsetwcs task in IRAF 2.11. The resulting images seem to work with SAOimage and Skycat, but not with AIPS. Here I attach part of the FITS header created by IRAF: RADECSYS= 'FK5 ' EQUINOX = 2000. WCSDIM = 2 CTYPE1 = 'RA---TAN' CTYPE2 = 'DEC--TAN' CRVAL1 = 15.8028232142857 CRVAL2 = 62.3328343915344 CRPIX1 = 597.68161016791 CRPIX2 = 443.244410338209 CD1_1 = -1.1259141406074E-4 CD1_2 = 3.73442780962290E-7 CD2_1 = 4.23897271224069E-7 CD2_2 = 1.12683930664224E-4 LTM1_1 = 1. LTM2_2 = 1. WAT0_001= 'system=image' WAT1_001= 'wtype=tan axtype=ra' WAT2_001= 'wtype=tan axtype=dec' We suspect that AIPS is not able to understand CD matrix notation, and is waiting for the older CDELT,CROTA format. Does IRAF have some task to transform CD to CDELT,CROTA notation? Thanks in advance, Octavi. ================================================================= Octavi Fors Aldrich Astronomy Department Physics Faculty Avgda. Diagonal 647 08028 Barcelona SPAIN octavi@fajnm1.am.ub.es =================================================================
Lindsey Davis wrote on Jan 05, 1999
Octavi, Other users have asked how to convert from CDELT/CROTA2 notation to CD notation and vice versa, so I have taken the liberty of copying a recent question and response on this subject from our sitemail archives. As you can see below the latter conversion is not always possible. Regards, Lindsey Davis PS. Make sure you check that I have the sign conventions correct in the CD to CDELT/CROTA2 conversion, especially in the case of axis flips ... The CDELT/CROTA2 to CD conversion follows the AIPS convention. > >I have several images which use the (what I believe to be normal) WCS keywords >for the coordinate matrix: > > CD1_1, CD1_2, CD2_1, CD2_2 > >I would now like to plot them using a program called WIP, but it appears to >require the keywords: > > CDELT1, CDELT2, CROTA, ... > >Can you remind me how to convert the CD matrix values to these keywords? I >seem to recall that CDELTi = CDi_i, but I'm not sure about the rotation bit >of it, i.e., I don't remember the coordinate transformation expression for >the CDELT form. > The transformation from CDELT/CROTA2 to CD is the following. CD1_1 = CDELT1 * cos (CROTA2) CD1_2 = -CDELT2 * sin (CROTA2) CD2_1 = CDELT1 * sin (CROTA2) CD2_2 = CDELT2 * cos (CROTA2) Unfortunately there is no general transformation in the other direction as the the CDELT/CROTA2 notation does not support skew and no generally accepted way of representing skew was ever adopted in that notation. On the assumption that that there is in fact no skew the reverse transformation is the following abs (CDELT1) = sqrt (CD1_1 ** 2 + CD2_1 ** 2) abs (CDELT2) = sqrt (CD1_2 ** 2 + CD2_2 ** 2) sign (CDELT1 * CDELT2) = sign (CD1_1 * CD2_2 - CD1_2 * CD2_1) As a matter of convention for astronomical images CDELT1 should be assigned to be negative if the above expression, actually the determinant of the CD matrix, is negative. The rotation angle is given as CROTA2 = arctan (-CD1_2 / CD2_2) or CROTA2 = arctan ( CD2_1 / CD1_1) If there is no skew the two expressions for rotation should be the same. If they aren't and the skew is small you can take an average.
Last post on Jan 05, 1999