Josh, Elenna, Robert

The especially clean lock stretches of the past few days have contained a number of periods when noise bumps at multiples of 4.05 Hz appear.

• By lining up the bumps with equally spaced lines in Matlab, the series appears to be f(n) = n*4.05+/-0.25Hz. (fig 1)
• The repeating broad bumps suggest scattering or some other non-linear thing like compression or saturation. So far I haven't seen a strong resonance at 4.05Hz in Darm.
• These bumps typically "switch on" for burst of 30-90s in duration. This can happen a few to a dozen times in an hour, polluting one to ten minuts of data. (figs 2,3,5,6)
• Some GPS times of bumps chosen from the spectrograms for further analysis are (figs 4,6):

start: 2023-04-12 07:08:00 (1365318498) len: 120s. 
start: 2023-04-12 07:24:30 (1365319488) len: 120s. 
STRONG start: 2023-04-12 07:39:00 (1365320358) len: 120s. 
start: 2023-04-12 07:41:00 (1365320478) len: 120s. 
weak start: 2023-04-14 07:52:00 (1365493938) len: 120s. 

start: 2023-04-14 07:12:00 (1365491538) len: 120s. 
start: 2023-04-14 07:23:00 (1365492198) len: 120s. 
start: 2023-04-14 07:31:00 (1365492678) len: 120s. 
start: 2023-04-14 07:49:00 (1365493758) len: 120s. 
STRONG start: 2023-04-14 07:54:00 (1365494058) len: 120s. 

The attachment has the same images but in one PDF.