NEIS is now placing at least one of the two unusual events off the coast of
No. California at a more mundane epicenter, it would not surprise me if the
other will soon be relocated as well. The new epicenters appear to be around
40.3N 127W in an area where such events are common.
In general practice, unusual epicenters given by local networks, especially
for events outside the network coverage, should be viewed as suspect. They
are usually wide off the mark. Only with global coverage can these events be
placed within reasonable distance of the "true" epicenter. It is always wise
to look at the quality rating of an epicenter - this will aid the user of earthquake
lists to be a bit wiser in their use.
My post on "poor locations" below shows how this applies to these two
poorly located (by NCSN) events. This is a good chance for all of us to learn
a bit about wise use of earthquake catalogs.