<HTML>
<HEAD>
<TITLE>RE: Article on Central U.S. Earthquake hazard by expert panel</TITLE>
</HEAD>
<BODY>
<FONT FACE="Calibri, Verdana, Helvetica, Arial"><SPAN STYLE='font-size:11pt'>------ Forwarded Message<BR>
<B>From: </B>Art Frankel <<a href="afrankel@usgs.gov">afrankel@usgs.gov</a>><BR>
<B>Date: </B>Thu, 28 Apr 2011 16:02:24 -0600<BR>
<B>To: </B>Oliver Boyd DOI/GD/Golden <<a href="olboyd@usgs.gov">olboyd@usgs.gov</a>><BR>
<B>Subject: </B>please post<BR>
<BR>
I think it doesn't make sense to say the occurrence of the Haiti, Wenchuan, and Tohoku earthquakes invalidates seismic hazard mapping and probabilistic seismic hazard assessment (PSHA). It's all a matter of how good the inputs to your PSHA are and whether you consider earthquakes with longer recurrence times. The inputs to the U.S. national seismic hazard maps include paleoseismic and geologic slip rate information that provide estimates of earthquake recurrence rates from thousands of years of data. <BR>
<BR>
First of all, the 1999 GSHAP hazard map for Haiti and the Caribbean did not even include the hazard from specific faults, such as the Enriquillo fault. Their map only used the recent seismicity. Thus, their hazard map was not based on the methodology used in the U.S. national seismic hazard maps. Our new hazard maps for Haiti do include the major faults in Haiti and show quite high hazard in Port-au-Prince, for example (see USGS Open File Report 2010-1067). The hazard from these faults was estimated using slip rates constrained by GPS data from Eric Calais and his colleagues. Future paleoseismic investigations will improve estimates of these slip rates and recurrence times.<BR>
<BR>
The Tohoku and Wenchuan earthquakes apparently have long recurrence times of perhaps a thousand years or more. It is necessary to consider paleoseismic data and geologic slip rates in order to accurately assess the hazard from these long-recurrence time events. For the U.S. national seismic hazard maps, we used paleoseismic and paleotsunami data when evaluating the hazard from the Cascadia subduction zone. We also considered paleoseismic data to evaluate the hazard for the New Madrid region and other areas. By using geologic slip rates to estimate recurrence times for hundreds of Quaternary faults, we evaluated the hazard for long recurrence rate faults that don't exhibit substantial recent seismicity. <BR>
<BR>
So the recent occurrence of these long-recurrence time earthquakes actually supports the methodology we apply to make the U.S. national seismic hazard maps. It also supports the idea of using low probability levels of, say, 2% probability of exceedance in 50 years for building codes. Designing for these low probability ground motions provides protection for long recurrence time, high consequence earthquakes such as the Tohoku and Wenchuan earthquakes. <BR>
<BR>
<BR>
<BR>
<BR>
<BR>
Art Frankel<BR>
U.S. Geological Survey<BR>
University of Washington<BR>
Dept. of Earth and Space Sciences<BR>
Box 351310<BR>
Seattle, WA 98195-1310<BR>
phone: 206-553-0626<BR>
email: <a href="afrankel@usgs.gov">afrankel@usgs.gov</a><BR>
<BR>
------ End of Forwarded Message<BR>
</SPAN></FONT>
</BODY>
</HTML>