Not quite daylight saving …. but maybe a timezone issue?
Looks very much like a "slow earthquake" or a tremor (i.e. harmonic tremor).
Geophysicsfag here, and haven't looked at the data at all – but a signal of 20 mins with such harmonic pattern, i.e. one very dominant (and low) frequency over such timespan usually indicates that.
One would have to check signal strength depending on station location ….
Shouldn't be too hard to figure out for us seismologists.