The first bit of related-but-not-directly information was that the same author wrote on the effects of the 1964 Alaska earthquake (magnitude 9.2, the second strongest ever recorded) that were noted outside of Alaska--effects from all around the world, in fact. That report is online, and wow! was that some earthquake. Water levels in some Georgia wells jumped 10 feet or more in response to that quake, and wells on the Georgia coast were shaken enough to dislodge accumulated iron sulfide deposits on the well casings, and pumped blackish water for weeks. Seismic seiches (waves generated by earthquakes, like tsunamis, but on streams and lakes) were also observed, lots of them in Georgia and Florida. There's also the neat tidbit of those days of less-than-instantaneous information transmission that the Alaska earthquake was spotted by the author on the recorder chart for that Dawson County well, and he called it in to the Atlanta papers with his estimate that the magnitude was "greater than 8.3".
And the other bit of interesting information was that the reason we had a recorder on that Dawson County well in 1956-1958 was because the well was on the site of the Georgia Nuclear Reactor site, also called the Georgia Nuclear Laboratory. This was a Lockheed plant site where they were trying to develop a nuclear-powered airplane, before someone thought better of the idea of having nuclear reactors in vehicles that might crash and scatter parts over wide areas. It also seems that the management of nuclear waste was not well thought out beforehand, which led to USGS being involved in a ground-water/surface-water study to see what was happening to the waste that was put in pits on the site. (My quick scan of the resulting report saw a conclusion that the travel time from the site to the Etowah River was a year or more, and that this was considered sufficient decay time. I wonder what the isotope was they were using, or what level of radioactivity was considered to be 'of concern'?)
And all this leads to the thought that our current technology for ground-water monitoring (and surface-water, mostly) misses all this cool stuff. Electronic data loggers make a reading every 15 minutes, or 30 minutes, or hour--surface water may be more frequent, but ground water might be even less often. We get near-real-time hydrologic data relayed to the Web by satellites, but the incidental seismic data is very unlikely to be captured, and wouldn't give enough information if it was.