Heartland wife went to a New Madrid workshop in Memphis the last couple of days (but did not stop at Graceland). She brought home some USGS brochures. One shows a hazard map with the highest hazard where the 1811-1812 quakes occurred. They give the 50 year probablilty of a repeat of the 1811-1812 quakes (M7.5 to 8) as 7% to 10%. This is similar to the 30 year probabilty for Carrizo plain on the San Andreas fault. H.W. says there was a paper in Geology last year where sandblows could be caused by floods without any earthquakes. I think that the best evidence for repeated New Madrid quakes is the dating of buried sand blows. But, the area may be so prone to liquifaction that blows could be caused by smaller local quakes or large more distant quakes (Sue Hough idea?), and some by floods. I have not read much on the area but have heard Hough talk, and read the articles in EOS on the debate a few months ago (?). I think there is getting to be a consensus that GPS does not show any strain within the resolution of the data (could be a mm or so a year, or could be nothing). So, one would think that the only way to get a major (M7+) quake on the same parts of the same faults is if faults store differential stress. As discussed in the debate with John in a thread below, from a structural geologic point of view I have a problem with this: I think that strain (motion) has to occur to build up the stress. I don't think the Wrightwood results require something different because I visited the site there on a field trip and the slip per event is based on a model and is very shaky.
For some place like Sumatra, sure, some part of the subduction zone that only slipped 1 m rather than 20 m might not take too long to build up enough stress to fail again....but I would expect even that would take decades...even at 50 mm/yr.
I suppose one interesting concept is the out of sequence thrusts: the big, steeper thrusts above the main plate boundary: Can they fire off a tsunamigenic (sp?) M7.5 or 8 soon after the fault beneath slips: or does the subduction quake relax the stress on the fault above.....
Chris