supernovae icon indicating copy to clipboard operation
supernovae copied to clipboard

Published 20 hours ago • verified publisher icon astrocatalogs

[SN2005ap] <Spectra over-corrected for redshift>

Open mnicholl opened this issue 8 years ago • 5 comments

Several spectra here look like they have been de-redshifted twice

mnicholl avatar Oct 25 '16 21:10 mnicholl

Pinging @jparrent.

guillochon avatar Oct 25 '16 21:10 guillochon

aha, thanks Matt. Looks pretty bad, I'll fix this and return.

jparrent avatar Oct 26 '16 01:10 jparrent

Alright, this one turned out to be an adventure since there are 3 spectra from SuSpect and 3 spectra from Superfit, with overlap, so I'm going to have to explain the rabbit hole with some drawings.

First, the earliest spectrum from SuSpect should now be reset to a proper observer frame. I have made a PR but I will monitor the changes to fix anything that may still be off. I have also removed the Superfit spectra because (i) the first two were from SuSpect (and have since been slightly trimmed; i.e. duplicates) and (ii) I don't know where the oldest spectrum (``+36'') came from prior to Superfit to know where it actually sits in wavelength; it turns out that it's not always the case that all of the Superfit files are in the rest frame, as they tend to be. So it can come back another time.

Now look this over. screen shot 2016-10-26 at 1 14 18 am

Rabbit hole. Disclaimer: I could be wrong in my assessment here, but ^that seems pretty clear.

The following image is just to show that the files for these two 'identical' spectra are different; i.e. both files represent the 'observer frame' spectrum from WISeREP and the spectrum that has been properly corrected to an observer frame. screen shot 2016-10-26 at 12 21 53 am

BTW, recall WISeREP scooping up spectra from the OSC, including Superfit spectra. Well, those have now been removed from the OSC. screen shot 2016-10-25 at 11 04 24 pm

I will check the OSC viewer after the PR update, but @guillochon is it possible there is something buggy going on wrt how the two SuSpect spectra (above in blue) are being corrected in the OSC viewer? The files are the 2nd and 3rd files below: screen shot 2016-10-26 at 1 03 57 am By eye, they're "nearly there" but not near enough is all.

jparrent avatar Oct 26 '16 05:10 jparrent

@jparrent The currently-displayed spectra on the OSC are clearly a mess, I'm hoping after the daily update (which will complete by this afternoon) the webpage will be a lot clearer.

I think a single text box that allows one to plop down vertical lines in the viewer would be helpful to include on the spectrum plot, what do you think? Can try to hack those in by Thursday.

guillochon avatar Oct 26 '16 09:10 guillochon

@guillochon Yeah that would be helpful, but no hurry.

jparrent avatar Oct 26 '16 13:10 jparrent