iraf.net

When running the extinction task I get the following error:Warning: MWCS: dimension mismatch (mw_gltermr)I don't know what this means. Please help! Thanks.

The message usually means the image was dimensionally reduced, i.e. a 2-d image copied from a 3-D cube. What happens is the WCSDIM and/or WAXMAP01 keywords are wrong, the workaround is to delete these keywords.Cheers,
-Mike

Thanks, Mike. I did find that explanation elsewhere in the help files. As far as I know the data has not been dimensionally reduced (it's not my data). It's a simple spectrum. I did remove the WCSDIM keyword from the header and I still get the same error message. Any other suggestions?

Please post a long header of the image so I can see the WCS keywords.-Mike

a0194.ap.imh[2708][real]: V4743 Sgr
No bad pixels, min=unknown, max=unknown
Line storage mode, physdim [2816], length of user area 1904 s.u.
Created Fri 15:23:24 27-Jul-2007, Last modified Fri 15:23:24 27-Jul-2007
Pixel file "newton!/iraf/imdirs/karen/a0194.ap.pix" [ok]
New copy of a0194.r
EXTEND = F / File may contain extensions
ORIGIN = 'NOAO-IRAF FITS Image Kernel July 2003' / FITS file originator
DATE = '2007-07-27T19:08:58' / Date FITS file was generated
IRAF-TLM= '15:08:58 (27/07/2007)' / Time of last modification
OBJECT = 'V4743 Sgr' / Name of the object observed
OBSERVAT= 'MMT ' / observatory
OBSERVER= 'Wagner,Schwarz,Starrfield,Szkody' /
COMMENTS= 'V838 Her/CN program' /
EXPTIME = 5. / actual integration time
DARKTIME= 6. / total elapsed time
IMAGETYP= 'object ' / object, dark, bias, etc.
DATE-OBS= '2003-09-13 ' / date (dd/mm/yy) of obs.
UT = '04:03:40 ' / universal time
TELESCOP= 'test ' / telescope name
DETECTOR= 'ccd35 ' / detector
PREFLASH= 0 / preflash time, seconds
GAIN = 1.2 / gain, electrons per adu
DWELL = 33 / sample integration time
RDNOISE = 2.6 / read noise, electrons
CAMTEMP = -131 / camera temperature
DEWTEMP = -195 / dewar temperature
BIASSEC = '[2689:2708,1:256] ' / overscan portion of frame
TRIMSEC = '[1:2688,1:256] ' / region to be extracted
DATASEC = '[1:2688,1:256]' / image portion of frame
CCDSEC = '[1:2688,1:256]' / orientation to full frame
ORIGSEC = '[1:2688,1:256] ' / original size full frame
CCDSUM = '1 2 ' / on chip summation
INSTRUME= 'test ' / instrument
APERTURE= '1 x 180 slit ' / aperture
DISPERSE= '500 gpm ' / disperser
TILTPOS = '4700 ' / tilt position
INSFOCUS= '2.5 ' / instrument focus
WCSDIM = 1 /2
LTM1_1 = 1.
WAT0_001= 'system=equispec'
WAT1_001= 'wtype=linear label=Pixel'
ZEROCOR = 'Jul 27 15:08 Zero level correction image is Zero'
FLATCOR = 'Jul 27 15:08 Flat field image is Flat.r with scale=9950.837'
CCDPROC = 'Jul 27 15:08 CCD processing done'
BANDID1 = 'spectrum - background none, weights none, clean no'
APNUM1 = '1 1 154.64 164.64'
CTYPE1 = 'PIXEL '
CRVAL1 = 1.
CRPIX1 = 1.
CDELT1 = 1.
CD1_1 = 1.
REFSPEC1= 'comp '
DISPAXIS= 1

I think you may be see an error message that is masking a different error: e.g. the task needs to know things like the airmass and so looks for RA/DEC/ZD etc keywords that you don't have (see help page for the list). Otherwise, I'd delete the WCSDIM and LTM1_1-Mike

Hi Mike,I tried deleting the keywords you suggested but still get the same errors. I moved on to another set of data that has more complete header info (including RA/DEC/ZD/AIRMASS, etc) and I get the exact same error when I try to run the extinction package. Do you have any other ideas of what might be happening? I appreciate the help.Karen

Frank is the expert but is away until the end of the week. Does another task like SPLOT or SLIST open the image correctly?-Mike

I can do splot just fine. It only occurs when I attempt to run the extinction package. Thanks!

Any chance Frank is back? I still have no idea how to deal with this problem. Thanks!

Hi Karen,I apologize for the delay in responding. There are a couple of problems. The first is that extinction is the wrong task to use. You will notice the task is only in the longslit package and the summary help (help longslit) says the task is obsolete. This old task only works on 2D spectra. The correct task to use for both 1D and 2D spectra is calibrate. This can do one or both of extinction and standard star flux calibration.The other problem, since I tried to run calibrate on you sample header to verify my response, is that extinction correction (by whatever task) requires you dispersion correct the data first. In other words, the extinction correction is a function of wavelength and so you need to have a wavelength solution attached to the data.Yours,
Frank Valdes

Thanks, Frank! I actually was trying to run the task on dispersion corrected data (despite the sample header I posted here). I didn't know about the calibrate task. You can tell I haven't done reductions in awhile! I'll try it out. Thanks again!