Aftershocks from August 23 earthquake in central Virginia from August 23 to September 19. Interactive map. (USGS)

I posed this question to Morgan Page, an aftershock expert at USGS. Here was her response (note: “M” stands for magnitude):

It will be a few years before activity returns to the low levels observed before the earthquake, and that is provided a large aftershock does not occur. (For very large earthquakes that have many more aftershocks, aftershock sequences have been known to continue for a century!)

A rule of thumb is: if you have X number of aftershocks the first day, then you can expect about X/2 aftershocks the second day, and X/3 aftershocks the third day, and so on. Currently the rate of aftershocks is less than 1 per day (there were 5 aftershocks above M2 in January, so that’s about 1 every 6 days). That rate will continue to drop as time passes.

So at what point is a tremor considered a new or independent earthquake, rather than an aftershock? Page responded as follows (note : “M” stands for magnitude) :

...with any particular earthquake we cannot tell. But given that the number of independent “background” (that is, non-aftershock) earthquakes expected is about 1 M2+ earthquake per year in this area [central Virginia], that’s about how many per year are “independent”. The remainder are aftershocks. We can’t tell which earthquakes are the independent ones, but it’s a pretty small proportion at this point!

David Applegate, Associate Director for Natural Hazards at USGS, added the following:

We keep referring to aftershocks as long as there are more quakes happening in the vicinity than the background rate before the mainshock occurred. The rate of aftershocks decay with time (a statistical pattern known as Omori’s law), but the magnitude does not decay, so even though there will be fewer with time, it doesn’t mean that they will all be much smaller.

Particularly in stable continental crust, this elevated rate of earthquakes can go on for months, years, even decades. There are seismologists who argue that much of the elevated seismicity in the Central U.S. are still aftershocks of the series of magnitude-7+ earthquakes that struck two centuries ago

Related: USGS definition of aftershock