errors iin calibration
giannicatanzaro wrote on Jul 17, 2012
hi all
I have a question regarding lambda calibration.
I realized I did a mistake in the calibration of some spectra, is there
a method to delete the calibration and start from scratch?
If I apply refspec and dispcor, iraf apply a new correction over the previous one.
thanks for your help
gianni
I have a question regarding lambda calibration.
I realized I did a mistake in the calibration of some spectra, is there
a method to delete the calibration and start from scratch?
If I apply refspec and dispcor, iraf apply a new correction over the previous one.
thanks for your help
gianni
Francisco Valdes wrote on Jul 17, 2012
Sorry for the late reply.
I gather you mean that you can't go back to the original uncalibrated data. If you applied the dispersion calibration to "linearize" the spectra, i.e. resampled the raw spectrum, then it is certainly harder. If you use the same database file, that is the file produced by IDENTIFY/REIDENTIFY, on a linearized spectrum it will indeed assume the mapping from pixels to wavelength is the original pixels and the result will not be what you want. I'm not sure what to suggest to recover. If you have the spectrum and the arc both calibrated by the same "incorrect" solution. Then you can use IDENTIFY to make a new solution relative to the old solution and apply it as a delta calibration.
There are a few other things you could do but I would need more details. Again the basic answer is to go back to the data before dispersion calibration but if you can't you will have to be creative with coming up with a delta correction.
Yours,
Frank Valdes
I gather you mean that you can't go back to the original uncalibrated data. If you applied the dispersion calibration to "linearize" the spectra, i.e. resampled the raw spectrum, then it is certainly harder. If you use the same database file, that is the file produced by IDENTIFY/REIDENTIFY, on a linearized spectrum it will indeed assume the mapping from pixels to wavelength is the original pixels and the result will not be what you want. I'm not sure what to suggest to recover. If you have the spectrum and the arc both calibrated by the same "incorrect" solution. Then you can use IDENTIFY to make a new solution relative to the old solution and apply it as a delta calibration.
There are a few other things you could do but I would need more details. Again the basic answer is to go back to the data before dispersion calibration but if you can't you will have to be creative with coming up with a delta correction.
Yours,
Frank Valdes
Last post on Jul 17, 2012