desihub / desispec

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

Issues detected on some quasar spectra in iron #2147

Open corentinravoux opened 8 months ago

corentinravoux commented 8 months ago

I investigated outliers in the iron QSO catalogs. My study allows me to extract the spectra that present issues in the Lyman-alpha forest region. For DESI members, you can find more details on the method that I used here: https://desi.lbl.gov/trac/attachment/wiki/LyafMeetings/min23-05-12/

I found four quasar spectra with instrumental problems, and @julienguy advised me to post an issue here. The four quasars have the TARGETID 39628512784550454, 39636538107103557, 39633338260000189, 39627631729051509

You can find here an example of problematic spectra. As you can see, there is a significant peak in the flux, and the noise at the same position does not capture this issue. For the four quasars, I have zoomed on the defect region here. Finally, I have extracted from the spectra fibermap all the information related to those three quasars and put it in this file. The spectra show the same issues both inside the healpix and tiles cumulative folder. I have not tried to plot the individual exposure.

Our first discussions with @julienguy pointed towards a missed cosmics or a defect linked to specific fibers. If you need additional plots or information on this topic, I can provide them.