On several occasions I have set up seismometers on a building slab and on the ground near the slab and found that the slab follows the ground below a few Hz. However, I don’t think that I have previously put these results in the log. 

Recently Jan Harms and I were making Newtonian-noise related measurements and set up seismometers at the locations indicated on the attached plan of the corner station. The top plots in the figures X,Y and Z show calibrated velocity spectra for the X, Y and Z axes (directions indicated on the plan) for seismometers at locations indicated by arrows on the plan in colors matching the spectrum trace colors. The lower plots in each figure show coherence between the seismometer on the slab edge and the one 3.5 meters away, off of the slab in the gravel (RED). For comparison, the figures also show (BLACK) coherence with the seismometer that is 3.5 m in the opposite direction on the slab. Because we used only 2 seismometers, the black traces were taken at a different time. Coherence between the slab-edge seismometer and the off-slab seismometer drops faster with frequency than for the slab-edge and the other seismometer on the slab. 

The frequency at which coherence drops off between the slab-edge and the other on-slab seismometer at a distance of 3.5m away, is typical for 3.5m spacing, even at the center of the slab, and does not result from one of the seismometers being at the edge of the slab. The drops in coherence in the 1 Hz region, where the velocity is lowest, are also present in spectra from the huddled seismometers, indicating that a significant fraction of the signals in these regions are due to electronic noise. The black traces were made at lower resolution than the other traces and so do not extend down to the lower frequencies.

The on-slab and off-slab spectra are very similar and the coherence is high below about 3 Hz in X, 5 hz in Y, and 8 Hz in Z. For fundamental Rayleigh waves, the wavelengths at these cutoff frequencies are thought to be of order 100 meters, based on propagation velocity measurements.

Robert S., Jan H.