I found evidence of possible scattered light while looking at some data from a lock yesterday. Attached is a whitened spectrogram of 30 minutes of data starting at GPS 1425273749. It looks like the peaks are around 28, 38, and 48 Hz, but they are broad and it's hard to tell the exact frequency and spacing. Sheila thinks this may have appeared after Tuesday maintenance. Tagging detchar request so some tests can be run to help us track down the source!
Ryan Short and I have been looking through the summary pages to see what we could learn about this.
Our range has been shaggy and low since Tuesday, this does seem to line up well with Tuesday maintence. Comparing the glitch rate before and after Tuesday isn't as easy, Iara Ota pointed me to the DMT omega glitch pages to make the comparison for Wed when omicron wasn't working. DMT omega pages don't show the problem very clearly, but the omicron based ones do show more glitches SNR 8 and higher since Tuesday maintence, we can compare Monday to Thursday.
Hveto does flag something interesting, which is that the ETMX optical lever vetos a lot of these glitches, both the pitch and yaw channels are picked by hveto, and they don't seem related to glitches in other channels. The oplev wasn't appearing in hveto before Tuesday.
In recent weeks (every day after Feb 26), there have been large jumps in the amplitude of ground motion between 10-30 Hz at ETMX during the night. A good example of this behavior is on March 1 (see the relevant summary page plot from this page). This jump in ground motion occurs around 3 UTC and then returns to the lower level after 16 UTC. The exact times of the jumps change from night to night, but the change in seismic state is quite abrupt, and seems to line up roughly with the time periods when this scattering appears.
Ryan found this alog from Oli: 83093 about this noise. Looking back through the summary pages, it does seem that this started turning off and on Feb 20th, before the 20th this blrms was constantly at the level of 200 nm/s.
Comparing the EX ground BLRMS to the optical lever spectra, whenever this ground noise is on you can see it in the optical lever pitch and yaw, indicating that the optical lever is sensing ground motion. Feb 22nd is a nice example of the optical lever getting quieter when the ground does. However, at this time we don't see the glitches in DARM yet, and hveto doesn't pick up the optical lever channel until yesterday. I'm having a hard time telling when this ground motion started to line up with glitches in DARM, it does for the last two days.