More Power, Less Acceleration

By Ken Collins 16/03/2011

Time Ball Station Aerial_600x400
Lyttleton's Time Ball Station from the Stuff web site

Just a quick update today, following on from the theme of the last post, and the horror of the devastation Japan is now experiencing.

With the NZ government announcing a Royal Commission of Enquiry into the building collapses in Christchurch, it has been interesting to observe people’s perceptions, from politicians all the way down (or should that be up?).

The disconnect is partly in trying to understand why there was so much damage in Christchurch for a relatively small 6.3 magnitude quake, as opposed to the massive 9.0 quake seen in Japan.

So the follow on from my last post on ’Buildings are not Designed to be Race Cars’ where I talked about Peak Ground Acceleration (PGA), I thought it would be interesting to make some comparisons.

The Geonet map from the February quake is here and shows central Christchurch had a PGA of between 0.6g and 0.8g with up to 1.88g in the eastern suburbs and an incredible 2.2g at the epicentre.

Japan Intensity Shaking Map from USGS web site
Japan Intensity Shaking Map from USGS web site

The PGA maps from the U.S. Geological Survey show that in Sendai (about 130km from the epicentre), the PGA was 0.21g, with surrounding areas experiencing between 0.35g and 0.65g. In Tokyo the PGA was 0.17g. Have a look at the maps here and you can mouse over recording points to see the PGA expressed as a percentage of g.
eg 100% = 1g, 20% = 0.2g.

Even at these levels of acceleration it would appear that some buildings in Japan suffered structural damage as a result of the earthquake rather than the Tsunami.

So here’s the kicker. At the earlier reported magnitude of 8.9 in Japan (it has now been updated to M9.0) the energy released was 8000 times greater than in Christchurch, but the Japanese mainland experienced a significantly lower PGA than Christchurch did.

The Christchurch experience was reported as a short sharp jolt that was extremely violent. The Japan experience has been reported as a very long sway that just continued to build and build in intensity.

Location, proximity, soil types, rupture dynamics and many other factors mean that how each eathquake is expressed (and felt) at the surface is different.

And now back to the design of buildings. How do you best account of these huge differences so that you can structurally design buildings with some certainty? Can a building be made to reliably resist a PGA of say 1.5g (when a Formula 1 car accelerates at 1.4g)? Can older buildings be retrofitted to even remotely approach this? Assuming not, what level is an acceptable level to get older buildings up to? Will we ever experience that sort of PGA again in NZ? So is it worth designing buildings to resist that?

Engineers can now point to real life examples to answer many of those questions and scientists have more data to analyse than they have ever had before .

There are many, many newer buildings in Christchurch that did survive remarkably intact, despite the PGA they experienced (apparently) exceeding their design load state. For instance, a new-ish building around the corner from the CTV building hasn’t even a broken pane of glass.

Again, I must stress that I am not a structural engineer, and these are my thoughts as an Architect. But these are all questions that we have been discussing in our office, with no clear answer. The Department of Building and Housing and the Royal Commission will certainly have their work cut out trying to make sense of it all. And lets hope that sense does prevail.

0 Responses to “More Power, Less Acceleration”

  • Unfortunately, this is categorically untrue.

    Above the epicentre in Christchurch, the peak ground acceleration (PGA) was a whopping 2.2g, which nobody disputes is huge. 1.88g was recorded in the city and the reason for the scale of the destruction there (the Darfield 7.1 was only 1.26g).

    But here you have referred to the Sendai quake as generating a PGA of just 0.35 – 0.65g which is incorrect. The highest PGA recorded on the Japanese mainland was in fact 2.7g Iin Miyagi Prefecture) which is monstrous, and unprecedented. Other sensors in the region recorded well over 2 as well. Even the page that you link to demonstrates this quite emphatically.

    This error travelled through New Zealand media like wildfire with cascading iterations which incorrectly suggest that the Christchurch quake was more destructive than the Sendai quake in terms of shaking, and quoting this PGA. Perhaps it was perpetuated because it was a comfortable scientific myth that gives us all a reason why there was so much destruction wrought on Christchurch.

    • Thank you for the update James. When that blog was written, I carefully went through the USGS information available at that time as well as other sources, and the information I presented was correct. However the USGS site has been significantly updated since then, with what appears to be thousands of sensor reports, as opposed to the perhaps hundred or so sites that were being shown at that time. Therefore I was going on the best information available at the time of writing. I was also refering to Sendai as the closest major city as a comparison to Christchurch and not all of the outlying areas and towns. Even now, mousing over a reading in Sendai shows 0.6g and a reading directly adjacent to this showing 2.0g.

      However I would also observe that the ground shaking in Christchurch, for the magnitude, was also unprecedented. What you have shown is that the ground shaking for a magnitude 6.3 quake as in many places similar to a magnitude 9 quake that released 8000 times the energy. The major difference is that the Japanese quake is spread over a huge area, while the Christchurch quake is relatively contained.

      With the media at the time seriously questioning the strength of our buildings and the standards they are constructed to, in order to resist a relatively small 6.3 magnitude quake, the ground acceleration provides some context. Especially as the structural standards refer to a measure of ground acceleration or movement that a building must resist, and not the magnitude of the quake. It also demonstrates that the magnitude is not a good gauge of how destructive a quake can be to the buildings sitting on the ground.

      Certainly there was no attempt to imply or create the illusion that the Christchurch quake was anywhere near the destructive power of the massive quake in Japan. That is quite clear from the TV pictures. However there was a disconnect in NZ media reporting between magnitude and PGA and what that means for buildings. Perhaps a better title is now More Power, Similar Acceleration.

  • James,

    I have to concur with Ken. I looked at the USGS figures (also on wikipedia) around the same time (perhaps a little before Ken did?) and saw 0.35g, etc., values. I presume these have been updated since.

    (In fact I have a comment pointing to this at the end of this thread – I’ll add a further comment to update this later tonight.)

    The wikipedia entry for PGA has been updated, too. It earlier showed much lower PGA figures for the 2011 Sendai earthquake as my comment shows. For what it’s worth, this table now reports a whopping 4.36 vector sum PGA for the 2008 Iwate-Miyagi Nairiku earthquake (magnitude 6.9/7.2).

    While I’m writing, one of my (minor) gripes is with the media relying overly on magnitude rather than intensity measures (i.e. impact at a particular area on the ground) to convey the import of an earthquake. I pointed to a geologist writing about this sometime ago (March 1st, 2010 on my blog; I’d include a link but a second link will likely see this comment treated as spam!)