[CEUS-earthquake-hazards] Question-

Thu Feb 28 16:34:58 MST 2008

To Laurence Wright,

  You ask an excellent question that deserves an answer (I hope you'll get 
many answers).  Certainly, facilities of critical importance, such as a 
regional blood bank, need to be designed so that they will still be 
operational after the next moderate or large earthquake in the vicinity, 
if there is a significant probability of that earthquake occurring during 
the lifetime of the building. 

  One of the major reasons for using terminology such as a "2% chance of 
exceedance in 50 years" is to express the probability that a building will 
experience ground motions larger than a certain value over its approximate 
lifetime.    If there were no factor of safety engineered into a building, 
designing to ground motions with a 2% chance of being exceeded in 50 years 
would imply that there is a 2% chance of building damage or failure over 
50 years.  Here I am assuming that having ground motions larger than the 
design value will cause damage or failure of the building and I am 
ignoring the variability in construction. This 2% probability takes into 
account the probability of having a moderate or large earthquake over 50 
years and the probability that such an earthquake would produce a certain 
level of ground shaking or larger if it occurs. 

  It would be up to you and the Red Cross to decide if  2% (in this 
example) is an acceptable risk over the lifetime of the building, keeping 
in mind the factor of safety engineered into the structure.    Structural 
engineers would provide you with information on this factor of safety. 
Structural engineers could also tell you the expected consequences to the 
building  for different levels of shaking.
I am not making a statement on the appropriateness of  using 2% chance of 
exceedance in 50 years as a design probability level for your building.

  A useful check on this decision of what probability level to use in 
design is to compare the design motions to the ground motions expected for 
the next 1811-12 type New Madrid earthquake, which the USGS estimates to 
have a 7-10% chance of occurring in the next 50 years (see previous 
emails).  Do you want to design for the median ground motions expected for 
such an earthquake?  Of course, these ground motions would be exceeded 
during the next 1811-12 type earthquake at 50% of the locations, on 
average.   This is caused by random variability; we don't know if one of 
these locations with higher ground shaking would be where your blood bank 
is built.  This 50% chance of having ground motions larger than the design 
value during the next earthquake might not be acceptable to you.   It 
turns out that ground motions in Memphis with a 5% chance of being 
exceeded in 50 years approximately equal  the median ground motions 
expected for the next 1811-12 type earthquake.  You could design to higher 
ground motions that have lower probabilities of being exceeded in 50 years 
that correspond to, say,  the median plus one standard deviation ground 
motions that will be  exceeded at only 16% of the locations during the 
next 1811-12 type earthquake.     You could also do this comparison using 
the ground motions expected for a nearby magnitude 6 earthquake. 

 Joe Tomasello in a recent email suggested that you use a design with 10% 
probability of exceedance in 50 years for your facility.  As I pointed out 
in a previous email, this design value would be far less than the median 
ground motions expected for the next 1811-12 type earthquake.   It might 
also have the consequence that your regional blood bank in Memphis  would 
have a higher probability of failure from earthquake shaking over its 
lifetime than, say, a WalMart in Nashville that was built with design 
ground motions of 2/3 times the value with 2% probability of exceedance in 
50 years according to the International Building Code.

   As a seismologist, I am not qualified to give you a specific 
probability level for design.  I can provide our best estimates of the 
ground motions with various probability levels of occurring during the 
lifetime of your building.  I can also tell you the range of ground 
motions expected during the next moderate or large earthquake and the 
range of estimates of the probability of that earthquake occurring over 
the lifetime of your building. These are scientific issues that are 
addressed through analysis of data on earthquake recurrence and ground 
motions.

  I hope this reply is helpful.   I would be happy to talk to you in more 
detail about the scientific aspects of this issue. 

-Art

Art Frankel
Coordinator for Earthquake Effects Research
U.S. Geological Survey
MS 966, Box 25046
DFC
Denver, CO 80225
phone: 303-273-8556
fax: 303-273-8600
email: afrankel at usgs.gov

<WrightLa at usa.redcross.org> 
Sent by: ceus-earthquake-hazards-bounces at geohazards.usgs.gov
02/28/2008 07:40 AM

To
<WrightLa at usa.redcross.org>, <ceus-earthquake-hazards at geohazards.usgs.gov>
cc

Subject
Re: [CEUS-earthquake-hazards] Question-

I find it intriguing that nobody on the list has actually answered the
questions I asked, which were-

"Assume that you are designing a blood processing center for the Memphis
Tennessee area which will serve approximately 10% of the US population.
You are being asked to provide design standard requirements for this
facility on which 10% of the United States population will depend for
blood products.
The base facility is being designed for a 45 year useful life.

What seismic standard will you use?

Why?"

So why is it that everyone is willing to speak in generalities, but is
apparently not willing to state whether they'd use 2%/50 as the MCE, or
something else? I am not asking for someone to do any design, or go
through the various evolutions of site evaluation, I just want to know
what standard
you, if you were the designer of record, would use as the basis of
design.  And why.  I'm not going to  criticize anyone, and this is a
hypothetical situation, but I am really interested in knowing what
standard the experts here would use, and why they'd use that standard.
I think most of us would find that enlightening.

Laurence M. Wright
Senior Project Engineer
Manufacturing Engineering
202-303-4457 (o)
202-549-5843 (c)

_______________________________________________
CEUS-Earthquake-Hazards mailing list
CEUS-Earthquake-Hazards at geohazards.usgs.gov
https://geohazards.usgs.gov/mailman/listinfo/ceus-earthquake-hazards

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://geohazards.usgs.gov/pipermail/ceus-earthquake-hazards/attachments/20080228/5e7bc49c/attachment-0001.html 