search

desihub / desispec

DESI spectral pipeline
BSD 3-Clause "New" or "Revised" License
36 stars 24 forks source link

Occasional residuals around strong sky lines #947

Open kdawson1000 opened 4 years ago

kdawson1000 commented 4 years ago

In general, the sky subtraction is doing quite well. In visual inspections of the LRG spectra, I've noticed only a few cases in tile 68002 where the sky residuals are leading to confusion in the redshift classification. These can affect either [OII] or [OIII] and do produce templates with negative, or otherwise unphysical emission lines. The examples are:

35185953924451944 35185953941229153 35185965894995390 35185965894995908

julienguy commented 4 years ago

Great. Can you please provide the file name of the coadd ?

astro-blue commented 4 years ago

I am adding a few cases from tile 67230 for ELGs. The [OII] lines are next to sky lines which make it hard to tell whether or not they are true signals. 35185760109859639 35185766116101682 35185760097274623 35185735921308626

julienguy commented 4 years ago

For the first target, sky residuals are bad in NIR. This will hopefully improve with better arc lamp calibrations. 

plot_spectra -i $DESI_SPECTRO_CALIB/daily/tiles/68002/20200315/coadd-0-68002-20200315.fits -t 35185953924451944 --rebin 4 --zbest ~/redux/daily/tiles/68002/20200315/zbest-0-68002-20200315.fits

Screenshot from 2020-10-30 08-01-39