The shallow events are thought to result from the release of crustal stresses. Only about a dozen of these events were recorded and no obvious patterns were seen. Moonquake seismograms are quite different from terresterial ones. The movements are much higher in freq., and reveberation continues for a very long time .... for the shallow and deep events up to 50 minutes and for impacts... well in excess of an hour for the larger ones. It was hard to pick phases and there were no clear surface waves, the very dry lunar surface is thought to be responsible.
The number of events recorded by each station per year varied between 600 - 2000 depending on the site. A major portion of these were very small and total energy released over a year was barely sufficient to produce a M2 quake. The text goes on to say that these numbers did not include the very small thermally induced events that accompanied the abrupt temperature changes at lunar sunrise and sunset. It is thought that these events resulted from small scale cracking and slumping of the surface. The crust was found to have an avg. 'P' - velocity of ~6.7km/s and the mantle has a fairly uniform 'P' vel. of ~ 8 km/s