“You could put [a port or city] on the east side of the bay if you want to move away from the San Andreas fault, but that puts you closer to the Hayward fault.”

That would not make much of a difference. The Loma Prieta earthquake of 1989 occurred roughly 100 km south of San Francisco and Oakland. If either the San Andreas or the Hayward fault ruptures, the whole area will shake. Maps of predicted shaking intensity don’t suggest that anything within a hundred miles is particularly safe. And I don’t want to commute that far, whether by car or by tcp/ip.

http://www.nnvl.noaa.gov/MediaDetail.php?MediaID=704&MediaTypeID=1

The above details every tornado in the US between January 1, 1950 and December 31, 2010 – nearly 60,000 touchdowns. Basically, most of the eastern two-thirds of the United States gets hit on a regular basis. So that’s one hazard that’s difficult to avoid, unless you move west, where instead of tornados the more common risks are earthquakes, mudslides, and wildfires.

San Francisco, for instance, is one of two or three places on the west coast of North America (Vancouver is the other definite; Portland is debatable) to have both a good natural harbor and reasonably good access to a portion of the interior. So it makes some sense to put a city there (and at Vancouver and Portland, which have risk from being in the Cascadia subduction zone), as long as you make sure that the buildings can withstand a strong earthquake. You could put it on the east side of the bay if you want to move away from the San Andreas fault, but that puts you closer to the Hayward fault.

Similarly with New Orleans. You can argue that the ocean to river port ought to be further upstream (like, say, Baton Rouge), but the need to ship grain from the Midwestern US to the rest of the world implies that you have to have such a port in that general area. In doing business there, you accept the risk that an occasional hurricane will flood you out.

In those two cases, you can argue that people who are not willing to take those risks should not live there, and in other cases like river flood plains, you can argue that there is no economic benefit that justifies anybody taking that risk. With tornadoes, pretty much everybody has to assume some risk (tornadoes have been reported in all 50 states, and the top ten deadliest tornado list includes one that hit Worcester, MA). All you can do about tornadoes is (1) try to be somewhere else when it goes through and (2) have a backup plan in case (1) fails.

Timberwolf: That is the largest scale of the different levels I discussed, and i had planned to mention it but ended up wanting to keep the post shorter. It s also a hard topic.

In Massachusetts after the great storm of 1970-whatever washed away 200 houses, they made a new rule: This (pointing at a line on the map) is the sea shore. You can’t live there. If you live there now, fine, but the moment the storm damaged your home past a certain point, it comes down and we buy the land from you (or whatever).

Buyouts in floodplains are getting more common.

But as you point out, there are limits, and tornadoes are one of those things I don’t think we can avoid this way.

The people of at least one town all pulled up stakes and moved after the Mississippi flood of 1993. But Limon, Colorado, which was devastated by a tornado in 1990 ago, got rebuilt in the same place.

And the SF Bay Area? We’re expecting a devastating earthquake in the next thirty to sixty years. There are ways to strengthen a house and to prepare for survival until FEMA gets its act together. The SFFD teaches classes in earthquake survival and how to assist afterward. Move? Maybe. After the Big One hits, people will say to me, “You should have moved.” Yep â?¦ and so should everyone in New Orleans, and they should not move back.

We live on an active planet. The best we can do is learn how it works and prepare for likely events.