We just had a workshop on this topic last week. I couldn't find the original source of the information in the story, but there are some problems with the stated conclusions in the news article.

They say:

The last known great earthquake in the Pacific Northwest -- the Cascadia Earthquake -- struck in January, 1700. While such phenomena are infrequent, with a cycle of every few hundred years, the region is due for a big one.

The Geological Survey of Canada added for an average week, the chances of a mega thrust earthquake occurring is 0.0005 per cent (or one in 200,000); during a week of Episodic Tremor and Slip this increases to 0.026 per cent (or one in 4,000) -- a 50-fold increase.

My comments:

We are in the time window in which a megaquake is possible, but it is not "due". The recurrence rate for M9s is every 500 years and it has been 310 years.

1/200,000 is one every 400 years - should be 1 in 500 years, from the paper GSC just wrote, not so bad. Maybe more often, if there are some 8-8.5 quakes, as Chris surmises. Geologic guesswork has an M8-8.5 perhaps every 500-1500 years in northern Cascadia, in addition to the full length M9s every 500 years. There are probably more 8-8.5s in southern Cascadia, as Chris Goldfinger at OSU has been saying. All this means the 400 yr number is ok.

However, this calculation also seems to assume that Cascadia megathrust events are mainly triggered by Vancouver ETS, not ETS further south in Washington, Oregon, or northern Cal., which is very hard to defend. Also, the effectiveness of ETS in triggering megaquakes is completely unconstrained by data - ETS may not be ANY more dangerous than other times, we just don't know now. So to argue for a 50-fold increase in the chance for an M9 is somewhere between highly speculative and flat-out wrong.

This
http://www.canadianunderwriter.ca/issues/story.aspx?aid=1000394447
is not a step in the right direction, either.