At Sheila request, I look into some of the long locks to see if the coupling of SRCL to DARM is stationary or not. In brief, as far as I can tell, it looks stationary.

I took 5 hours of data from 1118016016, and compute the coherence between DARM and SRCL with 0.1 Hz resolution (first plot). There is some coherence, especially in the region between 20 and 70 Hz. 

Then I computed the coherence and transfer function for short 10 seconds-long periods, with a resolution of 1 Hz. I computed the average and standard deviation of the coherence and the absolute  value of the TF. The second plot shows the coherence and the transfer function: the shaded regions correspond to the standard deviation. It seems to me that there aren't large flutuations as was the case in the past.