Someone said that to get a M8.6 on the San Andreas you need a 800 km-long rupture with 20 or 25 m of slip (presumably average slip). I don't think there is any evidence for that in the paleoseismic record. In fact, what were once thought to be 10 m slip events every 300 years in the Carrizo plain are now thought to be multile quakes much more often.

But, the paleo-seismology data may be permisssive of "wall to wall" events having happened (?). Wall to wall would be from Salton Sea to Parkfield. But such a quake would not be 8.6: would be more like 8.0 or 8.1. There is a lot of interest in whether a San Andreas quake could rupture through the San Gorgonio knot.

There was a talk on the 8.6 and 8.2 in the Indian Ocean and the 8.6 was a crazy quake: jumping between perpendicular faults. They worked out the rupture history pretty well from seismology.

I'm worried about career again: A proposal for Ross Sea stratigraphy 34 million to 14 million years ago, with me as lead PI, was rejected and with very weak reviews and panel comments. The comments seem so strange that I'm about to check whether they even were using the final version of the proposal. My/our mistake (or one of them) was to not suggest capable reviewers. We may have ended up with theoretical geophysicists or something. But the NSF panel did not like it either.

Chris