Earthquakes were not ignored in the preliminary HSR studies. There *was* an alternative that went pretty much direct paralleling I-5. In order to be on the surface in the Gorman area (crossing the San Andreas, which with the potential for 20+ foot offsets you don't want to do in a tunnel) the tunnels (especially on the north side) needed long grades of 4% or more - while not impossible with electrified trains, difficult, and best suited to a maglev setup which was out of the question for other reasons. Also, it required crossing the CA aqueduct without messing up its tunnels, a very long, high fill on the northern approach, thoroughly unknowable geology due to the depth (though bad rock would be a pretty good guess), etc. You can imagine the cost. And just imagine the runaway with blown brakes popping out of the tunnel at 500 mph - better hope the next 250 miles of track are very straight and smooth.

The whole Tehachapi Mtn range is active - even where there aren't big faults it's always deforming and folding. Some areas are folding fast enough to see or at least measure over the course of 10 years or so. I'd rather be on top of that for maintenance and repair purposes than deep inside it.

The Tehachapi-Mojave-Palmdale route is a bit longer (my guess is 20-30 minutes longer - the prelim EIR thought it would be less), but has the advantage of naturally crossing all the major faults on the surface and providing potential service to the Lancaster-Palmdale area (a political plus even if relatively few people might use it given pricing issues vs. Metrolink). With shallow tunnels at most, the geology is more discoverable ahead of time. Grades overall will be less, which reduces peak energy needs - not a bad thing.

The HSR people discovered the same thing the RR people in the 19th century did after balancing everything - Tehachapi is the better way to go if you're using steel wheels and rail. HSR just had to pay a lot more for consultants to organize everything and write it up.