The Really Big One

Yes. It's not quite so distracting in the print edition, where they tend to be off to the side and don't look like part of the article.

I've lived in Seattle for 20 years and the only earthquake I've ever actually felt here was the Nisqually quake in 2001. Contrast that with some parts of Japan where you feel quakes on a monthly basis or more.

I used to live in Portland, and thought about this impending catastrophe a lot. Once -- under the influence of intoxicants, most likely -- I came up with a way of dramatically reducing its impact. It's a batshit crazy solution which will never actually happen -- but even when I'm sober, I still think it would work.

Here's what I'm thinking: the major problem with a disaster like this is the uncertainty surrounding its timing. "Maybe it'll be some other generation's problem. Maybe we don't actually need to let it affect our thinking about where we live and what we build." This kind of thinking is expedient and economical right up to the point where it gets absolutely everybody killed.

There's a second, somewhat less fundamental problem: the causality count can depend dramatically on exactly when the earthquake happens. It'll be worse if it happens at high tide rather than low tide (the tsunami will reach farther). Worse if it happens at night rather during the day (darkness is harder for emergency services to cope with). Worse if it happens in the winter rather than the summer (more deaths from exposure), etc.

These problems aside, a 9.0 with a tsunami is eminently survivable. You just need to be in the right place at the right time, surrounded by the right types of buildings / geology / topography, when it happens.

Anyhow, how do we solve these problems? Stopping the earthquake is out of the question ... but making it happen isn't. So here's what I propose: we announce an exact date and time for the earthquake.

Give everybody, say, 30 years to prepare for it. That's a decent-enough timescale for the sort of population displacements which will need to happen. So we'll have the earthquake in 2045. Let's have it in the mid-summer, at 10AM during a low tide, to minimise all the timing-related casualties. Then, at the appointed time and date -- if the earthquake hasn't happened yet -- we set off howevermany scores of nukes it's been calculated to take, drilled into the subduction zone, to lubricate the fault enough for a full-fault rupture. Shouldn't take too much of a push to get things going.

Over the next 30 years, the absolutely certain foreknowledge of this event would alter the planning, policy, and insurance environments dramatically. Tsunami walls etc. would be built in areas which are planned to be defended. In areas which cannot be defended and must be abandoned, the property prices, development activity, and eventually population levels would gradually crash. Within defended / geological stable zones, buildings which you want to be standing in 31 years would still need to be retrofitted to survive a 9.0 quake; otherwise, they'd be torn down before that point. Finally, on the appointed day of the quake, everybody could gather in the most geologically-stable high-grounds they can find, and just sit back and watch the fun.

Of course there'd be a danger that the quake might arrive at a bad time before then -- but even if it does, the incremental changes in land-use and the built environment, which such a policy would provoke, would result in far fewer casualties than our current trajectory.

Alas, it'll never happen, if only because a policy which results in the intentional destruction of millions of peoples' homes is a bit difficult to sell in a democracy, even if it would save tens to hundreds of thousands of lives.

That does seem likely, though as far as I know, AWS's biggest zone is located in Northern Virginia. And, I'm sure the AWS people have planned as much as they can for this eventuality. (Um, I hope.)