Great.

Also...

A way to conclude ground motion is present is to have an observer actually see the lever arm/beam moving up/or down during the event.

Another way to be confident is to ensure that the seismograph electronics are all shielded and grounded (in a copper / faraday cage). This blocks EMF in the same passband from getting into the signal prefilter/amp. ( A geophone which supposedly captures EMF {but not ground motion?} co-located could be used to clean up the signal from the seismo by subtracting out the EMF signal fromt he geophone (using a Kalman filter design).

Another thought is to examine the effect as a function of distance from the polar regions. If it decays with distance then it's probably EMF noise (ghost ground movements).

There's other info too but I'll wait....