iraf.net

Hello, When combining 3 different spectra of the the same object (scombine or odcombine), the (summed/averaged -- it doesn't matter) output "combined" spectrum looks basically exactly the same as just ONE of the input spectra (i.e. same noise and signal levels). Strangely, the cosmic ray rejection algorithm works perfectly, but it appears that the combining isn't doing so well. I read a thread about WCS display units being different from actual units of the spectrum causing problems with spectral addition (in splot), and wonder if that wouldn't cause what I'm seeing. It's hard for me to tell, and the fix provided on that forum seemed to be exclusively for splot. Here's what one of the headers of my one-d spectra looks like: SIMPLE = T / Fits standard
BITPIX = -32 / Bits per pixel
NAXIS = 1 / Number of axes
NAXIS1 = 4802 / Axis length
EXTEND = F / File may contain extensions
ORIGIN = 'NOAO-IRAF FITS Image Kernel July 2003' / FITS file originator
OBJECT = 'h2d ' / Name of the object observed
DATE = '2008-02-06T23:53:05' / Date FITS file was generated
IRAF-TLM= '18:53:05 (06/02/2008)' / Time of last modification
CTYPE1 = 'LINEAR '
WAT0_001= 'system=equispec'
WAT1_001= 'wtype=linear label=Wavelength units=Angstroms'
CRVAL1 = 3000.
CDELT1 = 0.75
CD1_1 = 0.75
CRPIX1 = 1.
DC-FLAG = 0
SLITLEN = 60 / slit length in pixels
SLITTYPE= 2 / Type of aperture
CNTRLINE= 32 / Spectrm central row
SLITNUM = 14 / Slit number
AIRMASS = 1.48500 /start airmass
EPOCH = 2008.03620000 /
EXPTIME = 3000.00000000 /exposure time (sec)
LC-TIME = '03:13:36' /local time
UT-TIME = '06:13:36' /
UT-END = '07:03:36' /
DATE-OBS= '2008-01-14' /
RA = ' 09:25:31.7' /
DEC = '-76:38:12.2' /
RA-D = 141.382250000 /ra-degrees
DEC-D = -76.6367222000 /dec-degrees
SITENAME= 'LCO ' /
EQUINOX = 2000.00000000 /
ST = 32596.8 /
IMCMB001= 'h2d_3073_2spec.fits'
IMCMB002= 'h2d_3074_2spec.fits'
IMCMB003= 'h2d_3075_2spec.fits'
NCOMBINE= 3
BANDID1 = 'spectrum: background fit, weights none, clean no'
APNUM1 = '1 1 27.28 39.28'
WCSDIM = 1
LTM1_1 = 1. The 2-d combined image (from which this one was extracted) had different header components -- most notably: CTYPE1 = 'LINEAR '
CTYPE2 = 'LINEAR '
WAT0_001= 'system=world'
WAT1_001= 'wtype=linear label=Wavelength units=Angstroms'
WAT2_001= 'wtype=linear' Thanks for the consideration. I hope I've been able to explain the problem well through my confusion and frustration, but if not, I'll happily try to be clearer. -Jess

Hi Jess,I'm sorry for the frustration. If you could send a listing of the parameters and the command line you use I could try and see what is happening. I can use your header to make three artificial spectra with the same dispersion info. If you think it might be significant you could send header listings of the other two spectra.Yours,
Frank Valdes