Matt Springer has been giving presentations around San Francisco about home earthquake preparedness since 2008 (for more information about the presentation, go to his earthquake preparedness website). This Quaketips blog is devoted to posts ranging from technical "how-to" articles to more philosophical "should-you" topics. New articles will be posted at most about once a month, so people who subscribe won't be subjected to lots of e-mail.

Saturday, July 9, 2011

Should I stay or should I go?...the inevitable dilemma about staying in the building during an earthquake

This month, we are taking a break from the actual preparedness issues and will instead delve further into this annoyingly counter-intuitive recommendation to not run out of buildings during earthquakes.

If you have attended or otherwise watched my talks about earthquake preparedness, you know that I make a point about not running out of a building during an earthquake.  I know, it just seems like the thing to do, when everything is shaking around you, and you picture yourself trapped under pancaked and collapsed buildings.  But the point I make in my talks, and want to emphasize with a bit more discussion here if you read further, is that at least in modern industrialized nations, statistically, you are taking a bigger risk of being injured as you run from your nice strong building into the DANGER ZONE that exists right next to buildings. 

While the structure is designed to remain intact, lots of less-secure parts of the building are attached on the outside with much less thought (or attached with nothing BUT thought…), waiting to come crashing down as it shakes (bricks, window glass, ornamental masonry, signs, etc.).   From what I hear, most injuries from earthquakes in places like California are sustained by people getting hit by the OUTSIDES of buildings, not the insides. 

(Here’s an important disclaimer: I have not sat there and watched people get hit indoors or outside during large earthquakes.  What I know comes from the official recommendations of groups like the American Red Cross and the US Geological Survey, which I am passing along.)

Many people assume that inside a building = not safe, and outside of the building = safe, based in part on photos they see after major earthquakes.  The problem is that when large earthquakes occur, do you think the photographers are interested in photographing the store sign over the front door that crashed down to the ground, or the metal light fixture that fell and dented a sidewalk, or a piece of window glass that fell from the neighboring 15-story building like a falling dagger?  Of course not, even if things like that were 99.9% of the damage.  They want to photograph the extremes, so after the Christchurch, New Zealand quake in early 2011 that killed a couple of hundred people, I saw a bunch of photos of collapsed buildings on the web and was just starting to think that all buildings in Christchurch looked alike when I realized that I was seeing the same three collapsed buildings over and over.  (I’ve since read that there were 4 high-profile building collapses; of course that means that the vast majority of buildings DIDN’T collapse.)  

The Pyne Gould building, for example, showed up a lot; amazing collapse; check out these photos (from AFP/Getty Images through, and National Geographic).  Of course, I’m guilty of this also in my talks; when I show notable damage from the Northridge quake for example, I show a collapsed department store, a collapsed apartment building, and a partially collapsed parking structure; and I guarantee you that I did not randomly select a cross section of the typical damage!  I went for the photos that make the biggest impression, and then of course I have to mention to people that most of the damage didn’t look like that.

If I knew for a fact that the building I was in was going to completely collapse, I might be tempted to try to get out; but of course we can’t know this.  Our buildings’ structural supports are much stronger than their “skins”; the outer walls and windows and all the trimmings.  Even if the building does not collapse, those outer walls still can collapse and they tend to collapse outward.  During a large quake, I’d be happier in an open field than inside a building, but where I want to be the least is next to buildings for those reasons.  In fact, I’ve heard that there is no safe place to be outside in a city downtown area, where any skyscraper window glass that dislodges can sail diagonally like an airborne machete…and I suspect that if I were on the streets of San Francisco’s Financial District, I might just be running into a building while everyone else is running out of it!

Here’s the unfortunate fact: we just can’t know ahead of time.  Let’s say a large quake strikes and one out of a thousand buildings completely collapses, killing or injuring or trapping a total of 50 people.  And let’s say that the other 999 out of 1000 buildings don’t collapse, and people who remain inside are fine under desks, etc., while 200 people who run out of the buildings get crushed by falling bricks and rubble.   Was staying inside the “right” thing to do?  Well, not if you are one of those 50 people who were in the few collapsed buildings; but without knowing ahead of time if yours is one that will collapse, you would have had the best odds of surviving if you had not run outside.  Of course, I completely made up those numbers to serve my purpose, but I think it is a reasonable scenario for a Northridge or Christchurch style quake.  It is the kind of situation where it’s simply not valid to show a photo or two and say “see, this proves what people should have done.”

Speaking of which, I will now show a photo or three and say “see, this proves that being next to a building CAN be a pretty bad idea.”  These photos from Christchurch came from and from CNN and are all over the web with numerous attributions.  (I am focusing on Christchurch rather than Japan because the Japanese tragedy was compounded by the tsunami and it seems most of the damage photos are tsunami-related; it’s difficult to know when photos of rubble are entirely due to the quake itself.)  I also think these photos are a good illustration of why it is not automatically a good idea to be NEXT to an object during a quake (think back to my article posted a couple of months ago about the Triangle of Life spam e-mail).

And in case you are thinking that all this rubble fell into the streets AFTER people finished running out of buildings and on to safety, this quote is telling: "There are bodies littering the streets, they are trapped in cars and crushed under rubble," Police Supt Russell Gibson said.” (BBC News, Feb 23, 2011)

At any rate, I wish there was an iron-clad rule: “do this and you will be fine.”  Unfortunately, all we can do is play the odds.  Back in Christchurch, in the collapsed Pyne Gould or Canterbury TV buildings, well, I would NOT want to be inside those buildings.  The much publicized ordeal of the woman who was trapped in the building and called her family to tell them she was probably not going to make it before being rescued a whole day later, and her co-workers who indeed did not make it, make us feel like clearly we should get out in case the building collapses; but people who ran out of non-collapsing buildings only to get buried underneath the rubble of the outer walls collapsing into the streets would have been better off staying inside.  How many of us know someone who says that if they HAD been wearing their seatbelt, they would not have been thrown free of the car wreck before it exploded, etc., but we still (hopefully) wear our seatbelts because it gives us the best odds of being protected from harm. 

To me the safest strategy to consider ahead of time if inside a California building in a large quake is to plan on finding a relatively safe spot under a table and to hope that my building won’t be one of the rare buildings to completely collapse, unless I have reason to believe in that specific instant that the building really is going to collapse.  I’ll follow the plan that might hurt me in 1% of the possible outcomes and will save me in 99% of the possible outcomes (again, numbers are just for illustration). 

And the next time I get in my car and I’m about to put on my seatbelt, I might have a brief moment of uncertainty, but then I’m pretty sure I’ll buckle up.

>>back to blog


  1. Interesting read. I agree with what you say in regards to buildings in California. However, the situation is very different in Portland, OR. There's been little to no seismic retrofitting and only buildings built 1995 and on will stand a chance. On top of that, the quake will likely be bigger than what California deals with. Doesn't look good for us up here.

    I'm currently in one of the brick buildings that will certainly come tumbling down in the great cascadia quake and I regularly run the "stay or go" scenario through my mind. What if it hits at night? What if I'm blackout drunk and I don't wake up? I'm seriously reconsidering my habits because of this thing. ;)

    In my case, I think I have less than a minute to escape the perimeter of my two-story structure before bricks start pelting everything below. Best chance: I think I'll have to grab my go-bag and put a helmet on and hope for the best, but chances are I'm a dead man.

    If anyone wants to know more about the quake and how to prepare, check out The Big One:


    1. Thanks for the comment. It's always a tough call, as I discussed in this article and also you might want to check out the further musings in my 4/25/15 article. I especially stressed in that later article that there's no perfect advice here; we are just trying to minimize the odds of a problem. Importantly, and I like to repeat this whenever I can, if everyone follows the advice that will save the most people, some of them won't be saved and they might have been saved if they hadn't followed that advice. However, if everyone does the opposite, then more people won't be saved. The tough decision is to decide what advice it is that really saves the most people, and under what circumstances to disregard that advice.

      If you feel that your building won't withstand a large quake, then you may have a more compelling reason to behave differently, although keep in mind that if half of the buildings collapse, many of those that don't will probably have outer perimeter failures like those described here and it is hard to predict which will actually fall.

      On the other hand, if you feel that your building won't withstand a large quake, that sounds like a good reason to retrofit or move to a different building! (But I know doing that as easy as writing that...)

  2. Thank you for this article and your blog!


COMMENT POLICY: Comments on blog posts can be very useful, raising issues and adding helpful information. However, some people attempt to post generic comments with embedded links to irrelevant websites. Due to this comment spam, all submitted comments will be verified by me first so there will probably be a delay before legitimate comments get posted. If your comment is taking a while to show up, it probably just means that I have not checked my e-mail yet. NOTE THAT COMMENTS THAT ARE ACTUALLY ADVERTISEMENTS WILL NOT BE POSTED.