See red and blue VS green. Bumps are at around [4, 5, 6, 7, 8]*12.125 Hz or so.
When it was really bad (red), it was easily identifiable in detchar summary page, not so when it was mildly bad (blue).
I ran coherence tool on these lines. I first tried to find a comb over there but failed. Then I tried to find them as single lines. I looked for 6 single lines: 12.125Hz, 48.5Hz, 60.625Hz, 72.75Hz, 84.875Hz, 97Hz. Two of them are not found in the coherence tool. Here are the results:
12.5Hz: https://ldas-jobs.ligo-wa.caltech.edu/~duo.tao/O2_line_12.125/index.html (Weak results and the structure does not look like a single line)
48.5Hz: https://ldas-jobs.ligo-wa.caltech.edu/~duo.tao/O2_line_48.5/index.html (Found in four of the weeks, in some EX and EY magnetometers. There are some channels that do not look like a single line. But there seems to be something going on so I showed them.)
60.625Hz: https://ldas-jobs.ligo-wa.caltech.edu/~duo.tao/O2_line_60.625/index.html (Not found)
72.75Hz: https://ldas-jobs.ligo-wa.caltech.edu/~duo.tao/O2_line_72.75/index.html (Not found)
84.875Hz: https://ldas-jobs.ligo-wa.caltech.edu/~duo.tao/O2_line_84.875/index.html (Weak. Only found in one channel and the structure does not look a single line)
97Hz: https://ldas-jobs.ligo-wa.caltech.edu/~duo.tao/O2_line_97/index.html (Found in many EX and EY magnetometers)
Robert's measurements seem to suggest that this is the resonance of one of the baffles.
https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=35166