San Francisco already had a dozen or so steel skyscrapers by the time of the 1906 earthquake. They all survived the earthquake well, though some succumbed to fire.
The earthquake damage was inconsiderable. Every
building on both side of Market Street stood against
the earthquake. The modern steel-frame buildings
were unhurt, and that style of structure stands vindicated.
The city has to rise from the ashes of
conflagration, and not from the ruins of an earthquake.
(California State Board of Trade, 1906).
I work in one of those pre-1906 buildings--a 10-story steel skyscraper that was one of the very first on the West coast. Whenever there's a small earthquake the whole building jumps, like a truck hitting a bump in the road. While disconcerting, from everything I've read[2] these buildings are quite safe. They were built to be earthquake resistant (though obviously the science and engineering was nascent back them), but more importantly they were built with riveted steel framing. Riveted connections have proven to be extremely resilient and in combination with other (perhaps accidental) behaviors of the materials and framing techniques they perform remarkably well.
Seismic safety isn't necessarily difficult and certainly isn't anything new. But because labor is expensive (it's why nobody rivets anymore and bolted connections aren't more common) seismic engineering in our era is dominated by high-tech solutions and designs that work with cheaper building methods and cheaper materials. Old tech isn't per se unsafe, and some old tech is better than modern tech, it's just not cost effective.
[2] From literature about the 1906 earthquake and other major earthquakes, various seismic studies (see paper, above), and random stuff like articles discussing the strength of riveted hulls in ship building. Caveat: I'm not an architectural engineer or anything of the sort.
A 9.0 quite simply won't happen in CA, because our fault lines aren't capable of generating that magnitude.
The Tohoku earthquake in Japan for example wasn't predicted.
"And yet the disaster of five years ago—a magnitude-9.0 megathrust earthquake off Tohoku, followed by a major tsunami and nuclear accident—came as a surprise. Until March 11, 2011, the consensus among seismologists was that a particular stretch of fault would observe certain rules, rupturing at consistent intervals in events of similar size." [1]
Another Tohoku earthquake quote:
"For decades, seismologists had believed that Japan could not experience an earthquake stronger than magnitude 8.4. In 2005, however, at a conference in Hokudan, a Japanese geologist named Yasutaka Ikeda had argued that the nation should expect a magnitude 9.0 in the near future—with catastrophic consequences, because Japan’s famous earthquake-and-tsunami preparedness, including the height of its sea walls, was based on incorrect science. The presentation was met with polite applause and thereafter largely ignored. Now, Goldfinger realized as the shaking hit the four-minute mark, the planet was proving the Japanese Cassandra right." [2]
This is an active research area and seismologists have a lot of interesting questions to figure out. For example, the Napa fault was discovered... after the Napa earthquake. [3]
And USGS scientists recently found that "the Hayward fault also extends south to connect with the Calaveras fault" and separately "the Hayward fault bends about 10 degrees to the right to connect smoothly with the Rodgers Creek fault underneath San Pablo Bay". [4]
There isn't really anything you can do with this info, a 9.0 is very very unlikely in our lifetimes, but I think it's interesting to know what the state of the art is.
[1] https://www.newyorker.com/magazine/2015/07/20/the-really-big...
[2] https://www.newyorker.com/tech/elements/shaken-beliefs-seism...
[3] http://www.latimes.com/local/california/la-me-quake-napa-fau...
[4] https://www.theverge.com/2016/10/19/13335816/earthquake-faul...
https://www.newyorker.com/magazine/2015/07/20/the-really-big...
It's not so bad! And it could be really bad!
The not so bad case. The 6.9 Loma Prieta 1989 quake killed 63 people. 42 of those deaths were on the highway collapse. And since then we reinforced the highways and any building touched by construction in the last 30 years was seismically upgraded. Lots of property damage, but we've already paid for that with insurance and taxes. Btw SF's many wood frame buildings that lean against each other do quite well.
From:https://pubs.usgs.gov/fs/2015/3009/pdf/fs2015-3009.pdf
The bad news is the 4% chance event. An 8 or (omg!) 9.0 earthquake. There are few buildings that would survive, some hospitals and data centers that are built to very high standards. (Like the datacenter at 365 Main for example, cool video here [1] about how that works. [2]).
Everything else is likely to be in really bad shape, with thousands of deaths and whole blocks leveled. Current standards don't go that high because it would be much too expensive. Imagine spending $5 million on every house and office building in San Francisco.
The other it-might-be-really-bad direction is that NY Times story below [3] points out that we don't really know how our new skyscrapers will behave even in 7.0 earthquakes, it seems quite possible that a few could fall over and wreck 2-3 blocks of other buildings.
Keep in mind that about 2,500 people die in car accidents every year in California. And that the 4% 8.0+ event only happens every 825 years.
So... not so bad! Could be really bad!
[1] https://www.youtube.com/watch?v=8OVDKZB9r1Q
[2] http://www.datacenterknowledge.com/archives/2009/04/14/quake...
[3] https://www.nytimes.com/interactive/2018/04/17/us/san-franci...