Once you have one or more cleaned images, you can use the results in SPICE. Use the “clean” versions of each of the three files in the “Input” tab of SPICE, the same way you would use their non-cleaned counterparts. As noted previously, these will be produced automatically by IRSCLEAN, provided the originals were available in the same directory as the image file to be cleaned.
NOTE: To extract a spectrum properly from a cleaned image, you must use the cleaned versions of the uncertainty and bmask files, in addition to the clean image itself. If you use the original uncertainty and/or bmask files, some cleaned pixels may not be used, and the output uncertainty spectrum may incorporate unrealistic errors for cleaned pixels.
We show below a comparison between the spectrum extracted from the original Long High image SPITZER_S3_13349376_0012_0000_8_bcd.fits using uncertainty SPITZER_S3_13349376_0012_0000_8_func.fits and mask SPITZER_S3_13349376_0012_0000_8_bmask.fits. and the spectrum extracted from the cleaned image SPITZER_S3_13349376_0012_0000_8_bcd_clean.fits using uncertainty SPITZER_S3_13349376_0012_0000_8_func_clean.fits and mask SPITZER_S3_13349376_0012_0000_8_bmask_clean.fits. (The image was cleaned using the rogue mask depicted in earlier examples above.) The blue curve is the original spectrum and the red curve is the spectrum of the cleaned data. Most of the spikes in the red spectrum have been removed by cleaning.
Table 8.1 Extracted spectra for sample bcd data. Shown are the original data (blue) and the cleaned data (red).
That's it! We hope IRSCLEAN helps improve the quality of your data. Feel free to report problems to us as you find them.
