rongpu
commented
2 weeks ago Here are the redrock redshifts vs time (LASTNIGHT) for the 3 fibers in iron:
The same 3 fibers in jura, and it appears that there are some periods of time (e.g., since December 2023) that are not affected by the issue:
And for reference 3 different fibers in jura:
Re-posting what I posted on slack: Fibers 551-553 have a large number of spurious ELG redshifts in the jura reduction while the same tiles/spectra were fine in iron, which means there is something is wrong in the jura pipeline for these 3 fibers. Other tracers are also affected (with lower redshift success rates) albeit less dramatic.
The 3 fibers are visible ("cosmic-ray-like") in the ELG redshift distributions in jura (the Y1 subset): https://data.desi.lbl.gov/desi/users/rongpu/redshift_qa/y3_vs_y1_comparison/redshift_vs_fiber_elg_norm-jura_y1_subset.png and for comparison the same plot for the iron reduction: https://data.desi.lbl.gov/desi/users/rongpu/redshift_qa/y3_vs_y1_comparison/redshift_vs_fiber_elg_norm-iron.png
Below are example spectra for the 3 fibers:
Fiber 551 iron:
The same spectra in jura:
Fiber 552 iron:
The same spectra in jura:
Fiber 553 iron:
The same spectra in jura:
