For Steve. Microseisms are the continuous background vibrations of the Earth observed between earthquakes and created by wave action in the deep ocean and reflection from the shoreline(in reality it is a complex interaction) To all intents and purposed it is disregarded as noise.
Storms over the oceans also represent major sources of microseismic activity. Research has suggested storms may be tracked by responses from geophone arrays, besides more conventional techniques. This would indicate that storm events can be discriminated from "noise" by their unique seismic signature. It would seem a not unreasonable assumption that in a stressed area a little nudge may be all that is needed to trigger an earthquake. Probably in most cases a very minor one, but nevertheless an earthquake. I do not dispute the fact that the major quakes occur at plate boundaries and major interplate quakes are rare beasts. But the fact that faults exist everywhere and that each represents at least one seismic event(or multiple events)would suggest to me that various mechanisms are at work. It is probably a fair argument that convection in the mantle is a major cause of stress in the crust all over any particular plate but the stress release mechanism can be any of a number of factors, some of which have already been mentioned.Also relaxation of stress does not have to be one big response, it could be a sequential series of small stress releases as with the idea of water injection on major faults to lubricate and hence cause failure at a much lower intensity and therefore less damaging. There is a body of research that does offer some support for these ideas but more research is needed to gain general acceptance. It is really not so much an idea of what causes the earthquake but more a consideration of potential triggers or repetitive events creating a trigger by exceeding thresholds. This is not really something that can be hammered out in a couple of paragraphs especially when everyone has a different starting point in their understanding. And at the end of the day everyone knows that a true geologist has spent at least three years working at every permutation of "perhaps" or "maybe". A straight yes or no with no qualification would be a disgrace to the profession.