[CEUS-earthquake-hazards] comment on Richard Kerr's article

[John Vidale]

I'll restate the comment I added on the Science page after Richard Kerr's article:

It's not clear to me why the spectacular earthquakes in Haiti, China, New Zealand or Japan would bode ill for the USGS way of estimating hazard, nor motivate zeroing the New Madrid Seismic Zone hazard.  
 
Using the current USGS methodology, the rates of activity geologically visible on the Haiti and Chinese faults would lead to a decent estimates of the hazard, and would have even before the earthquakes had occurred. While the magnitude of the Japanese earthquake was a surprise in its contrast to the local hazard estimate, the USGS invokes more comprehensive logic trees in California and elsewhere that could well have captured some component of the compound-segment hazard in Japan.  
 
It's true by definition that very rare earthquakes, such as those in China, New Zealand, and Japan that had recurrence intervals of upwards of a thousand years, are unlikely. Those low-frequency faults shouldn't appear as hotspots on hazard maps, even though they make up a substantial fraction of ongoing activity. I agree probably hazard maps should be smoother, maybe that's Seth's only point.  
 
However, the history of repeated large earthquakes in New Madrid earthquakes is not well enough understood to lower our guard just yet. If anything, the lesson of surprising earthquakes is that we should not write off threats that we do not yet understand.

John
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://geohazards.usgs.gov/pipermail/ceus-earthquake-hazards/attachments/20110429/69199f16/attachment-0001.html>