4 ms·

Is there any work on inducing earthquakes to relieve stress before it builds to catastrophe levels?

iamcurious 11y ago
I like how you think. If we are going to have an earthquake, at the very least we should have an scheduled organized mild one.
- nappy-doo 11y ago
  On the Richter scale, the difference a full step (7.0 to 8.0) is about a factor of 32. So if we're trying to mitigate an 8.0 earthquake every 100 years, you're talking about a magnitude 7.0 every 3 years to release the same amount of energy. Most engineering structures on the west coastare built to withstand a 6-7. When a magnitude 8 comes (and it will come, and in some sense we're coming out of the stress shadow from the 1906 quake just now) the devastation will be horrendous. *Editting to add: The structures on the west coast are engineered to withstand a SINGLE 6-7 magnitude event. Not one every 3 years.
  - freehunter 11y ago
    And even if they were built to withstand a mag 7 earthquake every three years, the things inside of them likely won't be. Imagine the engineering that would have to go into designing every single item in your house to withstand a deadly earthquake on a regular basis. No more popping out to Ikea to grab the same furniture everyone else uses. Your TV stand is on springs. Your TV is on springs connected to the TV stand on springs. Your cupboards have notches where your dishes can lock into them so they don't slide. The cupboard doors have latches that let them open when pulled slowly, but resist opening when sudden force is applied. Your refrigerator bolts to the floor and is filled will elastic pockets that hold every item individually. Its either that, or spend a significant amount of time every few years packing all of your items into boxes to ensure they won't be damaged when it's earthquake time.
    - chipsy 11y ago
      My mom's strategy for the biggest items: lash them to the walls so that they don't topple or walk out of position. This doesn't solve the problems of falling plates, but it does mitigate death-by-cabinet. That said 80% of the benefits are in location. If you're on flat bedrock, located away from the coast, you just don't feel the most intense shaking, and you won't face a tsunami.
    - kaybe 11y ago
      Well, it would be an amazing business opportunity for yacht outfitters.
- sevenless 11y ago
  We already know through fracking that injecting fluid into subduction zones can trigger quakes. I wonder about injecting a very strongly non-Newtonian fluid (a super-high tensile version of oobleck, maybe with nanotubes) into subduction zones. At high shear the fluid would be very viscous and might allow gradual release of energy.
rudolf0 11y ago
I feel like one could accidentally trigger the full quake by doing this, though I'm no seismologist.
- mokus 11y ago
  Even if that was a possible outcome, it seems like the advantage of having it at a time of your choosing would be significant.
  - iSnow 11y ago
    The legal liability would be significant too, I guess.
cossatot 11y ago
It's something that most seismologists have thought of before. Unfortunately, there is no way of knowing how much stress has already been accumulated. We also don't know nearly enough about earthquake triggering to be able to do it in a controlled fashion once, let alone repeatedly. You'd have to do this thousands of times (log scaling, remember!), in a new location each time, to produce enough small earthquakes to relieve the stress accumulated on a locked subduction zone fault: Large earthquakes produce relatively similar magnitudes of stress drop to smaller ones; this stress drop is just spread over a much larger area. So you'd have to basically tile the fault with drill holes and then do whatever to produce earthquakes. And hope it doesn't cascade. Thousands of times.
- pixl97 11y ago
  Nothing like a cascade failure to make an already stressful project into a terrifying experience.
jofer 11y ago
It's a matter of magnitude. A magnitude 2 earthquake releases ~30x the energy as a magnitude 1 earthquake. A magnitude 3 releases ~900x the energy of a magnitude 1. You can see where this is going. (Note, the "shaking intensity" of the earthquake doesn't increase as quite as rapidly as the amount of energy released, but the relationship depends on a lot more than the earthquake's size.) Let's say we were to trigger many magnitude 5's. We'd need to trigger 810,000 of them to equal the energy released in a Mw 9.0 earthquake. Interestingly, though, there are other types of earthquakes that _do_ release accumulated elastic strain over large areas. Cascadia is actually where they were first noticed. Every ~11 months, the cascadia subduction zone has a magnitude 6 or 7 earthquake. You'll never notice in Seattle, though, because it occurs over about a month. These "slow earthquakes" _do_ release significant amounts of accumulated strain. Sometimes they trigger "normal" earthquakes and vice versa. We don't really understand the details. The most interesting part (and the part most relevant for seattle) is that the same fault can have both these "slow earthquakes" and a full on >Mw 9 "normal" earthquake. This is one of the "hottest" topics in earthquake seismology and convergent margins research at the moment.
- tempestn 11y ago
  So, my understanding from what you wrote is that it may indeed be possible to slowly release the accumulated pressure, but we don't really understand how that occurs naturally yet, let alone how to trigger it artificially.
  - dangerlibrary 11y ago
    Given the regularity of previous large quakes, and the fact that the described magnitude 6 quake takes place over a month, it would take longer to relieve the pressure via month-long magnitude 6 quakes than we have until the next large quake. Best case scenario - you could put it off for a few dozen years.
    - tempestn 11y ago
      Even if you could relieve some of the pressure, you could theoretically make the big one less intense. It sounds like we're not even remotely close to being able to do it in practice though. (Although if we were, there's no reason why the described magnitude 6 over a month would be a limit.)
Someone 11y ago
I wouldn't call it work yet. "Long-term research" already is somewhat of a compliment. One problem is that experimentation is risky. If you do an experiment, and a tsunami follows, you are likely to be sued, even if it happens months later. Conversely, there's no way we know of to prove that a man-triggered magnitude quake prevented a more destructive one a few years later. So, you couldn't even prove that that quake with a few billion dollars of damage and a few hundred deaths relatively speaking is a good thing. Dreamers even envision extracting energy from tectonic plates. Small-scale, that is trivial (you can easily stretch a rubber band a few centimeter a year across a fault line). Large-scale, it could be a huge energy source.
protomyth 11y ago
There was a bit of discussion on that in this thread: https://news.ycombinator.com/item?id=9733034 https://news.ycombinator.com/item?id=9733034
- protomyth 11y ago
  Ok, now I have to ask, why the down votes on a link to a previous article directly related to the parent's question?

I like how you think. If we are going to have an earthquake, at the very least we should have an scheduled organized mild one.

nappy-doo 11y ago
On the Richter scale, the difference a full step (7.0 to 8.0) is about a factor of 32. So if we're trying to mitigate an 8.0 earthquake every 100 years, you're talking about a magnitude 7.0 every 3 years to release the same amount of energy. Most engineering structures on the west coastare built to withstand a 6-7. When a magnitude 8 comes (and it will come, and in some sense we're coming out of the stress shadow from the 1906 quake just now) the devastation will be horrendous. *Editting to add: The structures on the west coast are engineered to withstand a SINGLE 6-7 magnitude event. Not one every 3 years.
- freehunter 11y ago
  And even if they were built to withstand a mag 7 earthquake every three years, the things inside of them likely won't be. Imagine the engineering that would have to go into designing every single item in your house to withstand a deadly earthquake on a regular basis. No more popping out to Ikea to grab the same furniture everyone else uses. Your TV stand is on springs. Your TV is on springs connected to the TV stand on springs. Your cupboards have notches where your dishes can lock into them so they don't slide. The cupboard doors have latches that let them open when pulled slowly, but resist opening when sudden force is applied. Your refrigerator bolts to the floor and is filled will elastic pockets that hold every item individually. Its either that, or spend a significant amount of time every few years packing all of your items into boxes to ensure they won't be damaged when it's earthquake time.
  - chipsy 11y ago
    My mom's strategy for the biggest items: lash them to the walls so that they don't topple or walk out of position. This doesn't solve the problems of falling plates, but it does mitigate death-by-cabinet. That said 80% of the benefits are in location. If you're on flat bedrock, located away from the coast, you just don't feel the most intense shaking, and you won't face a tsunami.
  - kaybe 11y ago
    Well, it would be an amazing business opportunity for yacht outfitters.
sevenless 11y ago
We already know through fracking that injecting fluid into subduction zones can trigger quakes. I wonder about injecting a very strongly non-Newtonian fluid (a super-high tensile version of oobleck, maybe with nanotubes) into subduction zones. At high shear the fluid would be very viscous and might allow gradual release of energy.

rudolf0 11y ago

I feel like one could accidentally trigger the full quake by doing this, though I'm no seismologist.

mokus 11y ago
Even if that was a possible outcome, it seems like the advantage of having it at a time of your choosing would be significant.
- iSnow 11y ago
  The legal liability would be significant too, I guess.

cossatot 11y ago

It's something that most seismologists have thought of before. Unfortunately, there is no way of knowing how much stress has already been accumulated. We also don't know nearly enough about earthquake triggering to be able to do it in a controlled fashion once, let alone repeatedly. You'd have to do this thousands of times (log scaling, remember!), in a new location each time, to produce enough small earthquakes to relieve the stress accumulated on a locked subduction zone fault: Large earthquakes produce relatively similar magnitudes of stress drop to smaller ones; this stress drop is just spread over a much larger area. So you'd have to basically tile the fault with drill holes and then do whatever to produce earthquakes. And hope it doesn't cascade. Thousands of times.

pixl97 11y ago
Nothing like a cascade failure to make an already stressful project into a terrifying experience.

jofer 11y ago

It's a matter of magnitude. A magnitude 2 earthquake releases ~30x the energy as a magnitude 1 earthquake. A magnitude 3 releases ~900x the energy of a magnitude 1. You can see where this is going. (Note, the "shaking intensity" of the earthquake doesn't increase as quite as rapidly as the amount of energy released, but the relationship depends on a lot more than the earthquake's size.) Let's say we were to trigger many magnitude 5's. We'd need to trigger 810,000 of them to equal the energy released in a Mw 9.0 earthquake. Interestingly, though, there are other types of earthquakes that _do_ release accumulated elastic strain over large areas. Cascadia is actually where they were first noticed. Every ~11 months, the cascadia subduction zone has a magnitude 6 or 7 earthquake. You'll never notice in Seattle, though, because it occurs over about a month. These "slow earthquakes" _do_ release significant amounts of accumulated strain. Sometimes they trigger "normal" earthquakes and vice versa. We don't really understand the details. The most interesting part (and the part most relevant for seattle) is that the same fault can have both these "slow earthquakes" and a full on >Mw 9 "normal" earthquake. This is one of the "hottest" topics in earthquake seismology and convergent margins research at the moment.

tempestn 11y ago
So, my understanding from what you wrote is that it may indeed be possible to slowly release the accumulated pressure, but we don't really understand how that occurs naturally yet, let alone how to trigger it artificially.
- dangerlibrary 11y ago
  Given the regularity of previous large quakes, and the fact that the described magnitude 6 quake takes place over a month, it would take longer to relieve the pressure via month-long magnitude 6 quakes than we have until the next large quake. Best case scenario - you could put it off for a few dozen years.
  - tempestn 11y ago
    Even if you could relieve some of the pressure, you could theoretically make the big one less intense. It sounds like we're not even remotely close to being able to do it in practice though. (Although if we were, there's no reason why the described magnitude 6 over a month would be a limit.)

Someone 11y ago

I wouldn't call it work yet. "Long-term research" already is somewhat of a compliment. One problem is that experimentation is risky. If you do an experiment, and a tsunami follows, you are likely to be sued, even if it happens months later. Conversely, there's no way we know of to prove that a man-triggered magnitude quake prevented a more destructive one a few years later. So, you couldn't even prove that that quake with a few billion dollars of damage and a few hundred deaths relatively speaking is a good thing. Dreamers even envision extracting energy from tectonic plates. Small-scale, that is trivial (you can easily stretch a rubber band a few centimeter a year across a fault line). Large-scale, it could be a huge energy source.

protomyth 11y ago

There was a bit of discussion on that in this thread: https://news.ycombinator.com/item?id=9733034 https://news.ycombinator.com/item?id=9733034

protomyth 11y ago
Ok, now I have to ask, why the down votes on a link to a previous article directly related to the parent's question?