Modeling Earthquakes

Quote of the Day

Earthquakes are difficult to forecast but we’ve made great strides with this new model. The update includes more faults, better-characterized land surfaces, and computational advancements in modeling that provide the most detailed view ever of the earthquake risks we face.


Mark Petersen
USGS geophysicist
January 16, 2024
New map shows where damaging earthquakes are most likely to occur in US 

When the validity or failure of your model won’t be known for another century or two it is reasonably easy to be confident enough to sell your services and act convincing.

I don’t know if Petersen and company are knowingly unethical, and perhaps they really do have an earthquake model that is worth more than the bits of memory used to store it, but I have my doubts.

I trust this model even less than the climate models. Many of the forces are deep in the earth acting on materials with unmeasurable characteristics over time periods of 1000s of years.

It does match the “gut feel” of what might expect. And that is probably all that is necessary to get paid and become an “expert in the field”.


9 thoughts on “Modeling Earthquakes

  1. It looks like a reasonably accurate map of historically known earthquake events in the USA. If it happened at this place, this often, then it’ll probably happen there again…. that sounds good to me. When? Ah, now that is the question….. I’m sure that there bottle of Snake Oil and Skunk Piss they’re selling will predict it with 100% accuracy.

    • And the Gris-gris and various strange lizard parts.

      About all we can say is that Parkfield California has a lot of them, so nobody will take the bet that an earthquake won’t happen there.

      The other problem is that every earthquake, from the New Madrid 1811 earthquake Fault in Eastern Missouri to the 1971 Sylmar earthquake and the Northridge/Santa Monica quake in 1994 happened on deep faults that had given no sign they even existed.

      When my mother worked in downtown Santa Monica in 1945, she experienced an earthquake while walking on Fifth Street. She said the sidewalk waved up and down like swells on the ocean, and the shop windows flexed in and out, but not enough to break.

      Similar to the idea of extinct volcanoes, I think the idea of inactive faults will have to change. and somehow the stress on “Inactive faults” will have to be measured to make whatever model is finally developed actually more than halfway useful. .

  2. Mr. Huffman:

    This appears to be simply an updated seismic hazard map, which has been available for decades. Here’s one example (there are more at the USGS web site)

    These are used largely by structural engineers to provide design criteria for vertical and horizontal accelerations used for designing buildings and structures. They are usually fairly conservative values and provide input to local building codes so that objective criteria can be used for structural building codes.

    They are not intended to predict any particular earthquake at some particular time, instead they offer a consistent set of values for input to structural models for use in design.

    Back when I was an engineer designing building-size structures I used to depend on the models done by our structural engineers of my designs, using these acceleration values as inputs, to determine if the structure I needed to support my components was sufficiently robust to handle the worst-probable earthquake for a given location.

    Just as a side note, that area of extreme hazard in MIssouri is the location of the New Madrid fault, where in the early 1800’s an earthquake made the Mississippi River run backwards for two days and rang church bells in Chicago. A far worse quake than has ever hit California, and most people don’t even realize that there’s a mid-continental fault right there.

  3. When I lived in the Bay Area I spent a lot of time worrying about EQ’s. We moved to Idaho during the Pandemic and experienced our first EQ within six weeks of our arrival. Nothing since.
    There is a community/town in Central California called Parkfield and is known for rather specific intervals between significant EQ’s of between 12 and 32 years. The last one was in 2004. Parkfield is rather unique in this phenomenon and I don’t know of any other place that has a similar experience.

  4. Well, you got to do something with all those newly minted college grads. Give’em a shiny new computer and a tell them to predict earthquakes. And it kept’em busy for awhile.
    Unfortunately, we got some very predictable results. A big waste of money on nothing more than a make-work shit detail.
    And just like the FBI and school shooters. When something happens, it will have been on their list.
    “Ya, we’ve been watching that fault for some time now.”

  5. I’m not much impressed by those people in the USGS. Apparently, they have never bothered to do any research by asking people who live in earthquake country what actually happens during one.
    “When my mother worked in downtown Santa Monica in 1945, she experienced an earthquake while walking on Fifth Street. She said the sidewalk waved up and down like swells on the ocean…”
    I saw the same thing during a ‘quake in Sunnyvale ~1979. This action is where the ground acts like a fluid, and why things fell down during the ’87 ‘quake in the SF Bay area.

    They claimed to have not known it could do that.

    I saw it again during that ‘quake.
    It is really memorable to see waves in a blacktop parking lot!

    Prevalent in “fill”, ground that was dumped to make it level, or to fill in a water area to extend the land surface, which is MOST of the land surrounding the Bay. If you are near the Bay, and it is level, it is almost guaranteed to be Fill. The Bay surface is a LOT smaller than it was when the Gold Rush happened.

    That ‘quake in New Madrid wasn’t just a one day deal. Aftershocks lasted months, IIRC. If that blows again, we are in deep kimchee. The river didn’t just run backwards, it relocated as far as six miles at one location. The first steamboat on that river was on it’s first trip south when it occurred. They reported that the river ended at a hole in some farmer’s field that took them hours to back away from. Wild ride!

    • Wow. I did not know all that about the New Madrid earthquake. In fact, I only know what I reported because that’s what Allan Eckert wrote in his book about the conflict with the Americans in the Ohio River valley where he described how Tecumseh was a prophet who tried to unite the tribes against the Americans.

      The Kimchee will be deeper than they think there. California, Oregon and Washington have building codes that have been improved over the years to adapt to earthquakes. Nothing of the sort has happened essentially east of the Rocky Mountains, so when the next one hits, NO ONE will be ready for it.

      Thanks for enhancing and expanding my knowledge of this event.

  6. Smashed crockery in Boston, rang church bells on the East Coast.
    Most all the Corp of Engineers work on all those rivers in the Twentieth Century were not designed to handle earthquakes. Kiss it all goodbye. Wet places got dry and dry places got wet. 50 thousand square miles of land got rearranged. The next one will kill millions instead of thousands.
    Might not be as bad as the Three Gorges Dam letting go, as far as death numbers. IIRC, the estimate for that is 1/3 of the Chinese dying. That would be well into 9 figures. I’m thinking that building it might be the single most stupid thing humanity has ever done. The Chinese have no concept of quality control. A foreign concept to them. Lots of cracks are appearing in the dam structure, may just be a matter of time.

Comments are closed.