iraf.net

Dear IRAF users, I would like to ask your opinion about this "problem" I found with the optimal extraction of spectra contaminated by second order that may be of general interest.
I want to extract all the flux, also the contaminating one, because I'll apply a procedure to remove it.
I noticed that when the two spectra (first and second order) are slightly separated (because of the cross dispersing effect of the atmosphere at large airmass - but not so large if the seeing is really good) the optimal extraction and the non optimal extraction shows great differences where the second order spectrum has deep features. It seems that the second order spectrum, fainter than the first order spectrum, is less weighted than the first and hence some features (real stellar absorption lines) of the second order, that are well visible in the non-optimal extraction, are lost in the weighted extraction. Non optimal spectra have a larger noise, but preserve all features. I wonder if the correct spectra to be used are the non optimal ones, even if with higher noise and spikes. I think so. Is there a way to preserve the correct flux and all features but also the advantages of the optimal extraction? Cheers
Giuseppe Altavilla
INAF-OABO