Hacker News new | past | comments | ask | show | jobs | submit login
South Korea accepts geothermal plant probably caused destructive quake (nature.com)
321 points by bookofjoe on March 22, 2019 | hide | past | favorite | 195 comments



---- From the article:

A South Korean government panel has concluded that a magnitude-5.4 earthquake that struck the city of Pohang on 15 November 2017 was probably caused by an experimental geothermal power plant....

Unlike conventional geothermal plants, which extract energy directly from hot underground water or rock, the Pohang power plant injected fluid at high pressure into the ground to fracture the rock and release heat — a technology known as an enhanced geothermal system....

---

This is both horrifying and incredible. They managed to fracture enough rock to generate energy contained in a mag-5.4 earthquake.

That's around 200 KT of TNT [0]. That's a lot of useful energy if contained. I wonder what the efficiency of the plant would be then.

South Korea already has 23 nuclear plants though [1], I wonder why they decided to go this route, perhaps easier setup and lower associated costs ?

[0] https://science.howstuffworks.com/environmental/energy/energ...

[1] http://www.world-nuclear.org/information-library/country-pro...


> South Korea already has 23 nuclear plants though [1], I wonder why they decided to go this route, perhaps easier setup and lower associated costs ?

Despite South Korea being among the only countries that can currently do successful large nuclear builds, South Korea's government is fairly anti-nuclear, reflecting fear of the public post-Fukushima [1]. This is really sad because the skilled workers and construction management expertise required to accomplish this are very rare, and this team could be instrumental in very rapidly decarbonizing the world if deployed strategically. S. Korea also has some of the best shipyards. Turning them into assembly lines for GW-scale nuclear plants (floating or embanked) is one of the more interesting ways to rapidly and cheaply build out terawatts of clean, safe energy [2].

[1] https://www.theguardian.com/world/2017/jun/19/new-south-kore...

[2] https://news.mit.edu/2015/new-look-floating-nuclear-power-06...


A floating nuclear plant sounds like of one the most dangerous things a society can do. One of the problematic aspects of atomic fission energy plants are their failure behaviour. One of the most important aspects there is the containment of the pollution (in Fukushima they literally freeze the ground water under the plant so it is contained from below, too). If you are free floating in water, possibly even an ocean, you are about as uncontained as you can possibly get.


That's a common reaction, but it doesn't stand to much scrutiny. Before radiation gets to the public, about 4 or 5 different barriers have to fail. The first is the fuel pin, then the cladding metal, then the coolant itself (which can often absorb problematic fission products), then the reactor vessel, then the containment, and then dispersal. You're focusing on containment/dispersal.

But how do the first ones fail? The answer is that lack of decay heat removal allows the earlier barriers to heat up, melt, and fail. Well, if you have an intimate connection to an infinite heat sink (the sea), you don't ever lose decay heat cooling. You can't! So your fuel and clad stay intact in almost all scenarios.

Earthquakes? No problem, the sea buffers you. Tsunamis? No problem, stay in moderately deep water and the wavelengths are so long that you'll barely notice them. Heavy weather? The world's largest ship (Prelude) is designed to stay operating (it's a LNG facility) during Cat 5 cyclones. Military attack? Sink and cool passively until a designed recovery operation can occur Ship collision? Stay out of shipping lanes; worse case, sink and don't leak.

Also, keep people out of your exclusion zone by being a few km offshore.

Honestly it's a pretty slick low-carbon rapid deployment scenario that improves construction cost and safety. Operation will likely be more expensive, but maintenance maybe not (since you can go home to the shipyard and be relieved by a spare).


Most energy profit comes from externalizing the costs. Floating reactors would do that fantastically!

Love the "sink and don't leak" requirement.


Better stated: sink and don't leak because you are intimately linked to a near-infinite heat sink, and heating up/melting are a prerequisite to leaking.


that doesn't discuss corrosion though, the ocean is full of salt, how many half-lives until corrosion prevents containment?


The ocean would take a while to corrode through a couple dozen cm of steel, especially in cold water. But you're right that eventual leakage is a concern. A viable design of this kind of system would have to come with a sink-safely-and-cool design fully engineered as well as a designed recovery process. In other words, it should be expected that a recovery and disposal operation will be required (even though it's unlikely to be needed). This system should be designed so the salvage/recovery operation is easy.

Nuclear accidents generally worry about something called Large Early Release Frequency. Some of the most bioactive/dangerous fission products decay away in a few days. This kind of scenario completely eliminates those FPs from concern, though we do still have to worry about the longer-lived ones.


is that assuming steel at the same temperature as the surrounding salt still water, or assuming steel that is hotter than the constantly convecting stream of fresh salty water?


The way the heat transfer would work in this scenario would have small temperature gradients on the outermost layer of heat transfer because steel and water are good heat transfer mechanisms.

This isn't hypothetical. This list may interest you: https://en.wikipedia.org/wiki/List_of_sunken_nuclear_submari...


I wasn't asking about a list of sunken submarines?


They are scuttled floating nuclear reactors at the bottom of the sea. These are real-life examples very similar to the scenario you are inquiring about. They have not corroded away and released wholesale nuclear waste after many decades. If that's not relevant to your line of inquiry then I must be totally misunderstanding you.


quoting myself, emphasis added:

>... the ocean is full of salt, how many half-lives until corrosion prevents containment?

quoting you:

>They have not corroded away and released wholesale nuclear waste after many decades. If that's not relevant to your line of inquiry then I must be totally misunderstanding you.

You don't misunderstand me, you purpousely misinterpret my questions so you can give easy answers...


Ah, I see what's going on here. Please review the HN guidelines.

I-131 has an 8-day half-life and is the primary threat to populations in large early releases. The direct answer to your question for I-131 is at least 2,000 half-lives. Sr-90 and Cs-137 have 30-year half lives, so for them it's at least 2. As you surely know, the longer half-life nuclides release energy more slowly and are therefore less dangerous to biological systems. At the extreme, U-238 has a few billion year half-life and can be handled safely without shielding.

In the scenario I'm painting, the reactor would be recovered from the sea within ~5 years so none of this matters. The corrosion will not fail the system within those 5 years. I do not propose to just leave any failed reactor down there indefinitely.


You must be referencing the following rule:

>Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize. Assume good faith.

Which is exactly what I was accusing you of before you reflected the accusation. Please note there are 2 components in this rule:

1. Please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize.

2. Assume good faith.

I will discuss part 1 in the context of our discussion, but first point out that 2: does not mandate to keep and maintain the a priori assumption of good faith, it only mandates to assume good faith.

Now for part 1, lets personally dissociate and review the discussion as being held by Alice and Bob:

After Bob states,

>Better stated: sink and don't leak because you are intimately linked to a near-infinite heat sink, and heating up/melting are a prerequisite to leaking.

Alice asks a concise question:

>that doesn't discuss corrosion though, the ocean is full of salt, how many half-lives until corrosion prevents containment?

and later Alice adds the question:

>is that assuming steel at the same temperature as the surrounding salt still water, or assuming steel that is hotter than the constantly convecting stream of fresh salty water?

All the while Alice is a priori assuming good faith on behalf of Bob.

Now Bob can give multiple interpretations to Alice's question, and he is required to please respond to the strongest plausible interpretation of what someone says, not a weaker one that's easier to criticize.

Bob can use interpretation 1 interpreting Alice as Alice1 implying all of the following:

* 1A) Alice is worried about shortlived isotopes

* 1B) moreover she seems to believe steel corrodes in a matter of days in the salty sea, Alice probably never heard of the Titanic recovery, Alice believes that ships can't be reused because after every trip they are decommisioned and a new ship is built for every trip.

* 1C) Also Alice seems to be unaware that Iodine is the most easily mitigated isotope since we can bulk manufacture Iodine tablets containing non-radioactive isotopes.

* 1D) Alice seems to be uninformed about all the above topics despite referencing concepts like nuclear half lives, the corrosion of metal in salty water, convection of hot water in cold water, and the concentration and saturation of metal ions in aquaous solutions...

This interpretation of Alice is easy to criticize, for obvious reasons

or Bob can use interpretation 2 interpreting Alice as Alice2:

* 2A) Alice is worried about longlived isotopes

* 2B) Alice is worried about the influence of energy release in the long tail of nuclear decay: consider a simple system of N identical unstable isotopes decaying to a stable isotope (thats ignoring the worse long decay chains), after one half life, half the number of remaining radioactive particles has halved, but half of the energy that will eventually be released as heat (not temperature!) is still contained in that long tail. Alice wonders if that energy can speed up the corrosion process on long time scales. When salty water dissolves metal, theres a thin layer of water that is saturated by dissolved metal which acts in a self-limiting way. But if the heat causes convection, that thin layer of saturated water will be constantly replenished with fresh unsaturated salty water. Similarily evaporation is much enhanced if convection or wind carries away the saturated air, which is why we like to hang our clothes to dry outside...

If Bob chooses interpretation 1 (which is easier to criticize) over interpretation 2, then it is Bob who is acting in violation of part 1 of the rule from the guidelines...

If Bob then at some point replies "They have not corroded away and released wholesale nuclear waste after many decades." Then Alice can only conclude that Bob has chosen the weaker interpretation Alice1 over Alice2. At that point she simply corrects her a priori assumption that Bob is acting in good faith, and she explicitly points it out.

Then Bob escalates by reflecting the identical accusation in a vague reference to the guidelines, simply because Alice is open about her founded conclusion on Bob's behaviour, while Bob never explicitly states he chooses interpretation Alice1 over Alice2 even though it is evident to any reader... Alice did assume good faith on behalf of Bob, but Bob's replies imply he chose the weaker interpretation Alice1. That is unless Bob genuinely believes people like Alice think ships are one-time-use items, that Iodine tablets do not exist, ...

I hope someone (dang?) who can prove their association with the platform can clear this up, perhaps in your favour perhaps in mine (don't care really, I would just like clarity / precedent, so that we maintain equality before the guidelines)

*

Also you keep changing attention to a lesser problem of containment, the short-lived nuclides, for example you state:

>Nuclear accidents generally worry about something called Large Early Release Frequency.

Why are you personifying the accident events? Surely you mean nuclear experts instead of accidents? Let me explain why they focus on the short-lived nuclides: because they can be affordibly mitigated with measures like Iodine tablets. abstaining from eating produce from the affected area for a few days, etc...

The longer lived ones are not necessarily safer, they are simply not affordibly mitigatable over longer timespans! (In case of consumption, the shortlived ones have a higher activity of course, but the longer-lived ones with a lower activity would be consumed for long timespans, such that DNA damage can integrate over time)

*

Regardless of these issues, would you consider it prudent for mankind to explicitly define an absolute reference background energy-spectrum of radio-activity? i.e. for each gamma energy bin some typical but from then on fixed reference background activity? Because the only references to background I find are currently comparing with whatever local background is found away from a target of investigation, which is good enough on short timescales, but how will future generations be able to compare their background with ours? It seems we keep assuming that the natural background can not be influenced by human activity, which seems dangerously close to the original fallacy that human activity can not influence atmospheric CO2 concentration...


"So your fuel and clad[ding] stay intact in almost all scenarios."

With that much "no problem", it seems like something may have been overlooked in your summary above.


Are you referring to "almost all"? Experienced engineers never speak in absolutes.


It's the breeziness and imprecision of "almost all" and "no problem" that gives a bad impression about how far these words can be trusted. What does such a claim actually mean?


Ah, I see. Well I'm definitely trying to whip up excitement to the public here. If I tell you that the LERF goes from 1e-3/year to 1e-8/year, a lot of people wouldn't understand the implication, which is, effectively, "no problem". Nuclear engineers tend to speak in acronyms and numbers, I try to distill it down. There is weight behind these kinds of estimates though.


On the other hand, one would not want to oscillate between breezy "no problem" claims, and engineering jargon. Because jargon is likewise not convincing. It might be helpful to just link to the reliability assessment. The document linked from the MIT website is just a conference publication (I looked it up) which is pretty vague and is really just a concept overview.

Note: I acknowledge that you have much more expertise here than I do, but it's not translating well. OTOH, I'm receptive to actual analysis - I have a PhD and work in a model-intensive engineering field.


Fair. Finding the balance for a wide audience between believable and exciting is tough. I'm usually good at it but in this case, it's my favorite concept in my field of expertise so I get a little too enthusiastic.

Here is a master's dissertation from MIT on the topic that goes into lots of good analysis: https://dspace.mit.edu/handle/1721.1/103707


Huh. In this entire conversation you've come across as a person way-overstating their case, and as a result being totally unbelievable and unconvincing. And in a thread which is off-topic for the posting, which is a poor choice of a place to engage at all.


I said one thing, people continued the conversation, I continued it with them. There's a collapse thread button for a reason. Yet here we are talking.

What part is least believable for you? Shipyard construction being cheap? Floating nukes being safe? Nukes being safe in the first place? Nukes being low-carbon? Many of these thing sound surprising because they go against pop culture but they're interesting in that the scientific consensus is fairly opposite of pop culture on this topic.


That risk assessment assumes only the technical aspects. But there are still institutional, management and other 'people' issues that could render any engineering guarantee fruitless. Otherwise, the delivery of nuclear plants would have no way to off schedule.

I have no objection to the development of nuclear technology. But as I understand, what concerns people is not entirely the 'likelihood' of the disasters but rather the 'severity' of them. After all, People make mistakes and organizations corrupt. So, I think any tech progresses on the scale-down of the worst case where all safety is off would be far more helpful in convincing the public.


In order for the sea to act as a heat sink, many of the initial barriers will have failed. Only then can sea water touch overheating fuel. How does being in the sea stop the fuel rods from melting (or whatever they do when they're going bad)?


Natural circulation heat exchangers. The AP1000 has this huge tank of water on the roof that can cool things for 72 hours. Then it runs out of water. At sea with some displacement you use this system but never run out of water.


The U.S. Navy seems to operate a bunch of floating (and some submersible) nuclear power plants in the ocean without too much trouble.


Here’s a list of sunken U.S. and Soviet nuclear submarines. And by “nuclear”, it means propulsion/reactor, not just nuclear warheads.

https://en.m.wikipedia.org/wiki/List_of_sunken_nuclear_subma...


And as far as I can tell, none if those nuclear reactors have been leaking radioactive material. The fact that close to a dozen nuclear power plants have been sent to the bottom of the ocean and have maintained their integrity is testament to just how safe nuclear power is when implemented correctly.


But how long will the containment stay effective, and will it be effective longer than the nuclear activity and risk it contains?


These subs are continuously monitored. Articles on the internet say Kursk had a minor leak that was quickly repaired, without coming close to any health concerns.

Remember nuclear pressure vessels are subject to some of the most harsh conditions when in operation. They're built to withstand contact with extremely hot water under pressure. Will salt water eventually corrode through it? Maybe. But remember energy through fossil fuels and organic matter kill 3 and 4 million people per year respectively. Those concerned with nuclear safety often fall into the fallacy of letting perfect be the enemy of good. The best solution is the least-bad solution.


Keep in mind we've also blown up 500+ nuclear bombs including almost 10 underwater. Not a great idea in hindsight, but overall we didn't experience any major problems as a result.

Seems like even in the worst and extremely unlikely scenario of full detonation, we'd still be fine.


> overall we didn't experience any major problems as a result.

Some Marshall Islanders would like a word[0]

[0] https://www.theguardian.com/world/2014/mar/02/bikini-atoll-n...


The difference is that those tests were done on or near land, the fallout settled either on land or in shallow water, close to marine life.

The explosions which happened in deep water (Wigwam in 1955 for instance) had practically zero lasting effects beyond radioactive steam entering the atmosphere.


The scale is a little different though. Say a 300MW submarine reactor runs for 10 years, the energy produced is about 20 megatons. The energy in a bomb might be 400 kt, half of it from fission, so the reactor will contain a hundred times more spent fuel.


We've blown up nukes all the way to 50Mt. But granted, the underwater tests were in the kt ranges.


"""

Of the nine sinkings, two were caused by fires, two by explosions of their weapons systems, two by flooding, one by bad weather, and one by scuttling due to a damaged nuclear reactor.

"""


Well we all know the one scuttled due to damaged reactor was actually a cover story for defecting with first strike technology to the Americans...


No one got the reference. Unfortunate.

(for those looking, it's to the movie The Hunt for Red October)


...and to live in Montana, owning a pickup truck..


with two wives!


and yet despite that, we have yet to see significant effects on any of our daily lives from any of it.


Well that's a glib way of evaluating anything. I'd be concerned about the impact on sea life here.

School shootings in the USA and ISIS beheadings don't affect my daily life, either.


Effects would be on sea life far removed from our daily lives. Seafood can be imported from other locations.


That we know of


The sites are well studied.


There are no diesel electric ballistic missile submarines, with only two probable exceptions being North Korea and Israel


Correct me if I'm wrong, but I'm pretty sure the scale is much different.


Not too different. Subs can be a few hundred megawatts, vs. a few thousand in power plants.


Only an order of magnitude or two.


There are no single reactors putting out tens of GW, so just one (at most).


You can fit about 2 GW-scale AP1000 reactors on a big ship.


But what about scale? Sub reactors are tiny! Never more than a few hundred MW, with relatively little fuel.

Typical powerplant reactors operate in the TWs.


Your scale is off a bit. The Fukishima reactors were 484 MW to 1.1GW. Diablo Canyon, a fairly "new" (and now unused) Nuclear reactor setup in California were both units under 1.2GW. The reactors on the "Ford" class aircraft carriers are reported to produce about 700MW each.


Arg, I can't believe I brain-fumbled that. I meant to type GW. Wish I could still edit.

Yes, it's 1-2 orders of magnitude.


*Typical power plants reactors operate in the GW, not TW. So there's about an order of magnitude difference between subs and power plants, not 4 orders of magnitude.


Without too much trouble that we know of. And a small number of small reactors.

I think it is extremely irresonsible to build military equipment with nuclear reactors. They are destroyed in conflict, polluting everything.


Ah, the classic "this old power plant design had this failure mode, therefore all nuclear does this" argument. Do you know when Fukushima was built and what year the designs were drafted for it?


I don't know enough here to comment on who's right, but when the magnitude of catastrophic failure is large enough, model error is appreciably important. So in those cases past experience is a very important input into the decision, even if not perfectly correlated.

If anything, it proves that identifying all failure modes is challenging, and theoretically safe is not the same as practically safe.


"So in those cases past experience is a very important input into the decision, even if not perfectly correlated."

Of course it is, which is why modern reactor designs have incorporated safety features based on these accidents and older designs have been retrofitted (with some exceptions that are legitimately concerning). It's entirely true that it is impossible to predict every mode of catastrophic failure, but that does not mean it's impossible to create designs that are resilient to unplanned disasters. No type of power plant can be perfectly safe, but for assessing practical safety records, in terms of deaths / TWh generated, even estimating conservatively nuclear power is safer than any other source of power (including wind and solar). Some references for this:

https://www.statista.com/statistics/494425/death-rate-worldw...

https://ourworldindata.org/what-is-the-safest-form-of-energy

https://www.nextbigfuture.com/2008/03/deaths-per-twh-for-all...


The magnitude of the error has been overstated though. There were issues, but a coal powerplant is worse for public health than the known disasters.


Hydroelectic as well technically. A set of hydroelectric dam failures in China in 1975 killed more people directly from the event (171,000) than is estimated for the direct and future deaths caused from Chernobyl and Fukushima combined.


The Boing accidents showed that technological progress doesn't help in minimizing errors, as there's always a higher incentive of minimizing costs in the implementation, even if the plan is safe. A very important thing we can do is to limit the possible worst case scenario.


You do that by having strict standards. Boeing was allowed to skirt the standards because the FAA is in shambles.


Do you have any evidence that the same thing can't happen to a nuclear reactor? (Actually there are quite a few for the opposite as far as I know)


It probably can, but there's a fundamental difference between a nuclear reactor and an airplane.

There's a handful of nuclear reactors compared to planes, and new nuclear reactor designs, where there's an opportunity to cut corners, are not being introduced at the same rate as new plane designs.

A reactor is a large, heavy, stationary thing. Economic concerns exist, but they're not going to make engineering decisions based on weight like you would in a plane where every kilogram of material costs a fortune in fuel over the lifetime of the plane. An extra chunk of concrete in a nuclear plant costs nothing, operationally speaking.

We're just lucky that the planes aren't nuclear despite many wildly ill-advised attempts to make this a reality.


If you look at Fukushima disaster, it was not a design problem either, but maintainence problem. It was probably running more without modifications than it was designed for.

There were updates suggested to modernize the facility, but for cost cutting purposes they were ignored:

https://m.phys.org/news/2011-03-iaea-japan-nuclear-quake-wik...

It always comes down to cost cutting when the accidents are rare enough.


It was a design problem and a maintenance problem. A lot of the design decisions made in that era later proved to be Very Bad Ideas, like how there was no proper hydrogen containment above the reactor vessel.

Newer designs have suffered more major faults and managed to contain virtually all of the radiation. American designs, in particular, place great emphasis on having an extremely resilient containment structure above the reactor. A lot of things can go horribly wrong but so long as the extremely radioactive gas is contained it can later be cleaned up. These radioactive elements are extremely toxic, but also very short lived. You just need to buy time.

The Fukushima design may as well have had a tin roof, it exploded almost immediately and exposed the reactor to the elements. If that's not a design flaw, I don't know what is.

That and a number of the systems necessary to keep the reactor under control depended on poorly positioned generators that weren't flood-proofed. This seems like a major oversight on a building located in a tsunami and typhoon zone.


And the DOE & NRC aren't?


Especially, let's not forget, when the person running the DOE doesn't think it should exist.


I agree, Fukishima was designed about 15 years after nuclear power was first invented, it isn't that far off from complaining about how unsafe cars are in crashes using Model-Ts as an example. Fukushima should have been decommissioned before it was even 'completed' but because of the political barriers in building a new nuclear plant(s) they just stuck with it since it had already got over the regulation hurdles.

I think our problems with nuclear power are 95% political and maybe 5% or less technological. Canada for example has nuclear plants that use a neutron moderating coolant. If they lose coolant, the reaction stops. Unlike earlier designs where a loss of coolant lead to overheating and potential explosions. Not to mention new nuclear plants need 6-7 completely independent shutdown/safety features which is unheard of for earlier plants that have at best 1 or 2 emergency shutdown procedures that interfere with each other.


When humans are involved there are always failure modes.


Water is a great radiation barrier. The "nuclear" failure option for a floating plant, sinking it, is reasonable. There are already a few reactors on the ocean floor (subs). That is far safer than having them in the atmosphere.


Floating nuclear power plants are already quite common and have a spotless track record in the past century. That's wht most of America's aircraft carriers and modern submarines are.


This neglects the failure modes of nuclear power. The biggest danger, by far, is overheating, fire, and melting.

Paradoxically, a nuclear plant could be made extremely safe if it just dumped all the fuel into the ocean at the first sign of trouble.

With ulimited cooling the fuel can't melt so it will be safely contained within the rods/pellets.

Yes the radiation near to the fuel would be insane, but water is so dense it would be safe to swim around maybe 50 feet away.

There's never been a bad accident in spent fuel pools even though they contain orders of magnitude more fuel than operating reactors. This is just because it's really hard to melt something sitting in thousands of tons of water


South Korean government is not against anti-nuclear. It's CURRENT administration of South Korean govt is. Once the president and his cabinet members leave in due time, anything can change.

Pros and cons of a democratic government I guess.


"The government" and "the current administration" are the same thing. The government can change, but right now it is anti-nuclear.

Anyway, you have it backwards. The S Korean government is not against anti-nuclear, it is anti-nuclear. But I suspect that was a typo.


> This is both horrifying and incredible. They managed to fracture enough rock to generate energy contained in a mag-5.4 earthquake.

Or they fractured enough rock to cause what would have been an earthquake of magnitude >= 5.4 at a future date to happen at that time, which I think is more likely.

As I understand it, earthquake's are opposing forces of tectonic plates that finally overcome their coefficient of friction. Affecting that coefficient of friction slightly could cause it to happen sooner or later.


Alternative alternative explanation: they caused what would have been 4096 magnitude 3 earth quakes (0.2 magnitude difference corresponds to 2x energy per wikipedia) to merge together into a single magnitude 5.4 earthquake. Since they reduced the coefficient of friction enough to allow all that energy out at once instead of in small steps.


Reducing static friction results in more frequent releases of the same energy.


Assuming the energy input is constant that would be impossible. Why is the energy input non constant?

Is static/dynamic friction even a semi-accurate model for plate tectonics?


I had a physics professor whose day job was geophysics. We were discussing earthquakes, and he said he really wanted to investigate the possibility of eliminating the threat from the San Andreas fault by intentionally triggering it at some regular interval, before the opposing forces built up to dangerous levels. He was onto something!


Yes, precisely. The problem is what to do with all the currently built up energy. We know that a big one is coming, we don't know when. Before we can implement the program for slow steady release of energy, we will have to suffer through one big one.

Fukushima is in a better place today in that it's past the big one, and could try to release energy slowly.

Regardless, it'll never happen in the USA, given the anti-science political climate. Remember when they were about to launch the CERN LHC and everyone freaked out about "they're going to create black holes and kill us all!?" Now imagine California soccer moms and their Facebook groups spreading misinformation about "they're triggering earthquakes and will kill us all (and also vaccines cause autism!)" …


Heh, yeah. Same as the backlash to proposals for geoengineering our planet’s temperature and atmospheric composition. The way I see it, if we expect to inhabit this place for the next several hundred thousand years, we’ll either 100% have to have that ability, or instead be satisfied riding our Hobbyhorse of Environmental Virtuousness right into extinction or evacuation as whatever meager efforts we make are swamped by natural variations. Might as well start doing the science now.

Back to earthquakes though. Somewhere in HN there’s a comment on a story about the Cascadia subduction zone suggesting that we publicly select a date some many years in the future on which we’ll get everyone to a safe place, then pop that fault with a bunch of buried nukes. Would make some great TV!


That was considered in the 1970s.[1] Generally considered to be a really bad idea.

[1] https://physicstoday.scitation.org/do/10.1063/PT.5.2036/full...


Unfortunately, your article doesn't explain why the idea isn't taken more seriously.


> They managed to fracture enough rock to generate energy contained in a mag-5.4 earthquake.

No, they managed to fracture enough rock to destablize the system and allow the fault to slip and release the energy it had built up over decades.

In effect, this pulled the trigger. The energy was in the action, which was cocked by the earth.


While true, it's mostly pedantic. OP is talking about a potential earthquake harvesting power plant. So generating in this sense means releasing the trigger, in the same way that burning coal is generating energy. It's just releasing the stored chemical energy of the coal, but we still say that coal powerplants generate energy.


I was thinking about the same question and came at it from the other side, which is "How big a plant and how far away do you have to situate it to destroy a city?" Which is to say, what would someone need to do to weaponize geothermal fracking? Urban warfare is siege warfare and being able to knock down the castle without having to expose yourself to enemy fire would seem to be an advantage. The fault lines would need to co-operate too of course.


Distance and size of plant would also be heavily dependent on the fault lines. Could be a very useful military tool for attacking particularly vulnerable places (if you can fortify a nearby position against enemy attack for long enough to build such a thing). Seems overly-situational without some extremely rapid assembly.

Good for softish sci-fi, though!


The energy stored in the fault was already there pent up. They just released it. Which if you actually look at it from a damage perspective. Might be sensible. Not allow those mega-nukes like the san andreas fault to build up tons of energy. Instead frack vibrate the fault loose in lots of micro quakes.


Well, the energy is already there, the fracking will have lubricated the rock enough that it could release the energy.


>>This is both horrifying and incredible. They managed to fracture enough rock to generate energy contained in a mag-5.4 earthquake.

It was probably hanging in a balance. Then boom!


Did they actually manage to pump that much energy in it, or was it that they disturbed the fault enough to cause it give?


They did not generate that energy, better to say, they liberated it.


>South Korea already has 23 nuclear plants though [1], I wonder why they decided to go this route, perhaps easier setup and lower associated costs ?

Atoms scare uniformed voters.


I guess you meant "uninformed" not "uniformed"?


Yes. I must missed a key, and it looks like it's been too long to edit the mistake.


Just to clarify. The source of the energy of this earth quake was not man-made. It was pre-existing gravitational potential energy.

Similarly, a low-energy scream in the montain can trigger a high-energy avalanche.


> Similarly, a low-energy scream in the montain can trigger a high-energy avalanche.

While this is a popular myth, it's actually entirely false. You can shout and scream all you want - there's not enough energy transmitted to the snowpack to actually trigger an avalanche (by at least two orders of magnitude). This has been debunked by multiple studies, one example here: http://www.gblanc.fr/IMG/pdf/reuter2009.pdf


It is one of those areas where definitions matter more than reality. Many avalanches happen without external input. At a fraction of a second before one will happen, any extra energy will "cause" it to happen slightly earlier than otherwise. So one could biuld a situation where a shout, or a mouse, could indeed cause an avalanche.

It is like heating a pot of water to 100c. It is about to boil. Shout at it and it will start to boil a picosecond faster than if you didnt shout at it. Be really careful with the temperature and maybe your shouting makes it boil when otherwise it never would.


Not all potential avalanches happen. As a positive feedback loop the difference between an avalanche that does or does not happen can be arbitrarily small.

Back country snowboarding is a great example where small avalanches are much more common even if most of them don’t become significant. But, because the many small avalanches happen close together they are much more likely to grow. This is also why sound is more effective than you would expect, getting a huge number of tiny avalanches to happen at the same time makes it significantly more likely to form a large one.


Sound alone, even a sonic boom, cannot trigger an avalanche. You need explosives in contact with the snow.

Sources: personal experience in AIARE training [0] and Utah Avalanche Center via Myth Busters [1]

[0] https://avtraining.org/ [1] http://www.discovery.com/tv-shows/mythbusters/mythbusters-da...


>You need explosives in contact with the snow.

Then how do they happen naturally?


Sorry, I didn't mean to suggest that only explosives can trigger an avalanche, obviously that is not true. Lots of natural and human activities can trigger an avalanche, shouting and even very loud soundwaves are simply not one of them. More here if you're interested: https://utahavalanchecenter.org/avalanche-problem-toolbox


Isn't an explosion just a very loud soundwave? This doesn't pass the physics sniff test, but maybe I am just being pedantic.


At some point something is so loud it stops being "sound" and starts being "shock" and imparts so much energy it can break things or set things on fire.


If you want to draw a distinction, ~194 dB is probably the place to do it (in Earth atmosphere). That's where the sound pressure oscillates between 0 and 2 atm and any louder would necessarily distort.


Regarding the distortion, the error due to wrongly assumed superposition becomes noticeable before that. At this point, the non-linearity of soundwaves becomes glaringly obvious.


One way is wind loading... snow falling in a storm gets blown around and piles up on a slope that's already loaded. Another type - a wet slab avalanche - triggers when the sun melts snow and causes water to flow between the various layers in the snowpack, causing two layers to slide.

The goal of the bomb is to thump the snow hard enough that if any weak layers of snow exist below a hard slab of solid snow, the slab will crack, slamming air through the weak layer, breaking the support of the rest of the slab and causing a slide.

Naturally occurring avalanches usually need some obvious change of force on the snowpack, or some reason why a weak layer becomes weaker. I'd guess that a sonic boom, or definitely a scream, couldn't do enough in almost any situation to tip the balance.


Yes absolutely, we're just exploiting potential energy here, and that was my point, perhaps I wasn't clear enough.

That's why the efficiency of the power plant (if this energy would have been captured) would have been a sight to see on paper. The energy they put into the fracking would have been tiny compared to this.

edit: spelling


Are you suggesting capturing the energy of the earthquake?


That's literally what he said, yes.

He did imply it would likely not be possible. He said that if it could be captured, it would be an absolutely mind-boggling ROI in terms of energy in vs energy out.


But how renewable/reliable would it be? A few years of quaking, and we'd hit peak quake energy. All down hill from there.


That seems like it would be a feature, lots of free energy and no more earthquakes killing people, destroying infrastructure (which must cause massive amounts of pollution), etc.


Or all the continental plates have shattered into a billion pieces and molten rock is bubbling up through the innumerable fractures.


Moot point-- it's not possible currently.

However, think of the sheer amount of potential energy cause by plate tectonics... we're talking about continents... moving.


I mean, if you want to be technical about it, it would definitely not be downhill from there because all the hills would be gone.


So a good PR could frame this as having pre-empted the quake, and launched an era of managed, controlled quakes.

Also, just like after that scream, there is no avalanche risk, since all the snow slid down the valley, there should now be no quake risk for quite a while at that site.


Exactly. It would make very little sense if a power plant meant to extract energy from the ground somehow inserted a lot more instead.


Would it be worth it for something like this to be used in places that often have catastrophic earthquakes, to instead release scheduled mini-quakes every so often to relieve the pressure?


It would be electromagnetic potential caused by strain on the rock, right?


If the energy was building up anyways, then wouldn't it have come out in an earthquake eventually? In fact, the later earthquake would have been worse. Maybe people will start intentionally fracking near earthquake zones on a regular basis to prevent the energy buildup from becoming dangerous.


Yes. In fact, years ago it was suggested that we do that for the San Andreas Fault to prevent a catastrophic "big one" release.

I read occasionally that fracking causes earthquakes, too. While there is controversy around fracking, I see the systematic de-stressing of the earth's surface as a benefit.


According to the USGS, 1-2% of induced earthquakes are caused by fracking. [0]

Apparently the rest of them are mostly caused by a similar process of injecting wastewater deep under ground. Of course, fracking is a major producer of wastewater, so it is therefore an indirect cause of many more than 1-2% of induced quakes.

[0]https://earthquake.usgs.gov/research/induced/myths.php


"I see the systematic de-stressing of the earth's surface as a benefit. "

Those of us in geological fields totally disagree with you. If you start a small earthquake in San Andreas, say up in the northwestern section, you can trigger a full-length chain reaction that culminates in unleashing the big one which will rip all the way through Mexico. What is worse is that the earthquake will get stronger as it propagates along the fault line. What you call the big one in the northwest section where it all began is thousands of times weaker than what gets felt further down the fault.


What makes you think a man-made stress release will be any different from a natural one?


Timing, for one thing. "The big one" might not hit the San Andreas Fault for 10,000 years for all we know.

It might not ever hit. Crustal rocks can permanently store tremendous amounts of energy as stress; if the stress is changing, the fault might only move enough to release some small delta of the total stored stress. But: grease the system and you might release a lot more, and all at once.

It's very hard to predict what will happen, either naturally or human-triggered. IMO this is a good argument to not rush it. Every day that passes gives us more time to understand and prepare.


So you would rather just wait for it to randomly unleash as the big one?


It does seem better than purposely releasing the big one, or doing so accidentally, which would be a risk until we can build much more sophisticated models that would guide these release efforts.


Absolutely. Would you rather it be triggered by some person or corporation, and then have the problems of trying to hold them responsible for the deaths and damage that will happen?


Maybe. It's hard to know if a later earthquake would have happened or been worse; it is normal for crustal rock to store some level of stress that is never released. You would need to know if the stress was changing, and how fast. If it wasn't changing pretty rapidly, it's possible it would have just sat unreleased for a very long time from the human perspective. Major earthquakes are rare; stress is not.

There is also the question of what happens now that it has been released. Rocks are all connected, so a release of stress in one place might result in movement somewhere else eventually.

BTW it is difficult to measure stress in crustal rocks remotely, without releasing some of it. So all of this is hard to know, and therefore hard to account as risk in new hydraulic injection projects.


I suppose the amount of stress/strain held within these rocks will be hard to ascertain. If these earthquakes occur periodically, sure that gives us a better estimate, but we still don't know the unintended consequences of triggering a quake. Not to mention the property damage.


In theory, 10 magnitude 5.4 quakes will cause a lot less damage than a single 6.4 magnitude quake. But of course if that energy were to be dissipated in micro-quakes instead the 5.4 quake is much worse.


It isn't certain that an earthquake would happen either way. Take the example of a spaghetti noodle. If you bend it and it breaks, it releases a lot of energy due to the bending stress. If you introduce a defect in the middle of it, it breaks before building up that much stress. However, we don't know if whatever is bending the noddle would eventually do so until it breaks. Maybe it wouldn't.


Plus, knowing exactly when the quake will happen can reduce a lot of damage. You can stop surgeries, trains, and stuff during that time if you have a warning.


"Hot Dry Rock" geothermal was pioneered by people at Los Alamos:

http://coloradogeologicalsurvey.org/energy-resources/geother...

It was a strange story. It started out with ideas for super-high-temperature mobile reactors that could melt a shaft ten miles down to tap geothermal energy. In the early 1970s the enthusiasm for that kind of thing declined, so they got an idea to tap geothermal energy using "fracking" technology which existed back then but wasn't anywhere near as capable as it is today.

One thing I remember about their work is that they didn't observe any fracking-related seismic activity at all: when you frack sedimentary rock you get the rapid formation of large cracks with obvious seismic activity. At least in that experiment they found that fracking would open up micro cracks in bedrock slowly and that there was no obvious activity from it.


I wonder how close this process is to hydraulic fracturing to extract fossil fuels. Is the volume of liquid pumped into the ground similar? I imagine geothermal is pumping more and deeper to get to where the earth is warmer, but I am not sure.

In my area, Pittsburgh Tristate area (WV,OH,PA) fracking is big business. Seems we are fortunate to not have a major fault line running through the area.

edit - clarifying that I am talking about fracking for fossil fuels


Perhaps this could be thoroughly tested by performing high-pressure hydraulic fracking in very close proximity to gas pipelines.


Something like this happened in Basel, Switzerland in 2007 [1]. I remember that during their work on the enhanced geothermal system small quakes were triggered in the region. They had to cancel the project thereafter.

[1] https://en.wikipedia.org/wiki/Induced_seismicity_in_Basel


Interesting. I always assumed that the energies involved in major earthquakes were far too large for us to affect them in any way.

If we can cause them, I wonder, will we be able to prevent the natural ones as our knowledge increases?


Earth's crust "wants" to move. One thing is triggering an earthquake by fracking, another one is stopping the crust with all the mantle below pushing heat into it. That heat has to go somewhere. I guess the best we'll reasonably do is forecasting quakes.


Hypothetically, if you could cause earthquakes in a controlled way, it may be possible to perform periodic small stress relief quakes rather than allowing one big one to occur naturally. We can't stop it moving, but maybe we can make it move on our terms.


Exactly, this seems like an interesting avenue to explore.


I think to remember that wet faults (as in "there is water mixed with the rocks") slide more frequently that dry faults and yield energy in less powerful bursts. If so, lubrificating faults could be a solution. Unfortunately I found a paper from 2017 [1] that suggests that it's not so simple. There are many more micro conditions that can trigger an earthquake.

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5580450/


You only need the energy to pull the trigger of a gun to release the energy of a bullet.


I wonder if someday we'd be able to trigger small earthquakes to release pressure gradually and prevent larger and more destructive earthquakes.


Or if there's a military that will use this to trigger earthquakes.


It would be the equivalent of controlled burns to prevent larger forest fires. That would be very interesting to see applied to earthquakes.


Seems unlikely, because the Richter scale is logarithmic. For example, to bleed off the energy released by this magnitude 5.4 earthquake, you'd have to trigger 125000 magnitude 2.0 quakes. Fun little calculator on the USGS site: https://earthquake.usgs.gov/learn/topics/calculator.php


Sure, but isn't this literally just a notation convention?

There's big quakes and there's little quakes. But energy is still linear. The fact that a quake releasing 1/3 of the energy of a 6.0 is not denoted as 2.0 quake, does not mean that it isn't releasing 1/3 of the energy.

I'm not a quakologist. Am I wrong?


This may actually be useful in Chile because Chileans don't lose any sleep over a 5.4. Even a 7.0 is meh. You could even have timed earthquakes or just diffuse your earthquakes at noon on Sunday every week.


California had best take heed: The Geysers and Salton Trough have taken way more heat out of the Earth, and I would argue delayed, shifted, and intensified the big one the Hayward Fault is overdue for.

There's been a great deal of ridicule placed on opponents of geothermal energy, saying that geothermal could never contribute to an earthquake. I'm hopeful the ridicule will start to go away now.


At least according to the second paragraph of TFA, the problem wasn't the amount of energy they took out of the ground. It was all the fracking.


Which shouldn't be confused with the type of fracking we normally do.


How is it different?


Presumably the answer is “because magic”


Normal fracking doesent create earthquakes.


I've not seen a credible source say that.

The closest I've seen is sources saying that fracking itself only explains a minority of the increase in earthquakes in places like Oklahoma, and that the majority are more properly attributed to other things, particularly the wastewater injection wells that fracking operations use to dispose of all the wastewater that fracking produces.

Which is such an amazingly damn hair-splitty distinction to most of us, but I can see where a spin doctor could have lots of fun with it.


Although it’s controversial, I don’t think there have been tsunamis of ridicule from every direction hitting people saying fracking increases the odds of an earthquake, which is what this is about.


It's unlikely that all of that energy was injected into the earth by humans. The high-pressure fluid injection generated cracks and lubricated the surrounding rock, facilitating a release of existing stresses. The area around Pohang is riddled with faults. There's a lot of energy down there just waiting to be released.

South Korea, like most other developed nations, is trying to increase the use of renewable energy. There are many hot springs in Korea, especially around active faults. Geothermal is an obvious candidate in those areas. Of course they're also doing wind and solar where feasible.


I'm seeing a lot of folks making a connection between hydraulic fracturing in the US and this earthquake. In fact, the earthquakes in Oklahoma were caused by waste water wells: https://en.m.wikipedia.org/wiki/Injection_well

Fracking is happening all over the US (Appalachians, the Dakotas, Texas), but for some reason Oklahoma only gets the quakes. I think it's because only Oklahoma has these bad wastewater wells. Apologies, can't find the sources I found a few years ago that convinced me of this.


Statistics for increase in rate of earthquakes in US: https://en.m.wikipedia.org/wiki/File:Cumulative_induced_seis...

The keyword is induced seismicity. I hadn't heard of that before.


Yes, waste water is the larger cause of these induced quakes. However, fracking produces a significant amount of wast water, so they indirectly contribute to induced quakes in that way as well.


Was there a similar causality between Puna Geothermal and the vigorous lava vent popping up next to it that was entertaining us last year, extending Big Island of Hawai'i by a large chunk, covering whole Kapoho Bay?


Side question: why did a magnitude 5.4 quake cause so much grief?

A 5.4 is a quake where Californians finally start paying attention, but generally doesn't cause a lot of grief unless something very unlucky happens.


Nuclear obviously has its risks. Exploring alternatives is a viable strategy. Also do not be mistaken to think that nuclear power is unlimited. It requires the mining of a resource that is very very limited.


Next step: harness energy released by earthquakes.


Sounds like something out of a conspiracy theory.


What? There is a huge amount of data demonstrating that the rate of earthquakes increases significantly anywhere that any fracking happens, and the rate of increase is proportional to the amount of fracking that occurs in the region.

What is the conspiracy theory here?


Could we use this method to prevent large quakes by splitting them into a series of smaller ones?


Earthquake magnitudes are in logarithmic scale so magnitude 1.0 difference is 32 times more energy. So to break up a 7.0 earthquake into a series of 5.0 earthquakes, it'll take 1000 small earthquakes. With our current technology it's not very feasible.


Im well aware.


So some of those conspiracy theories about earthquake weapon tests could be credible?


And here I thought earthquakes were the one kind of natural disaster we still couldn't bring upon ourselves


Would this be the first nation to denounce fracking?


no, there is quite a long list of countries that banned it. also a few counties and states in US have also banned it


If I'm recalling the timeline correctly, France placed a moratorium [1] on all fracking activity many years ago. I believe they were the first to try to ban fracking in 2011. That was fought over in court, with the government winning. Then they became the first nation to ban all fossil fuel extraction in 2017.[2] France has two large shale areas that would have been good for fracking, the Paris basin and the South-East basin.[3] Energy companies were eager to try to develop those.

[1] https://www.nytimes.com/2013/10/12/business/international/fr...

[2] https://www.theguardian.com/environment/2017/dec/20/france-b...

[2] https://i.imgur.com/QO68cyO.jpg


I wonder if any countries are trying to weaponize this?


I suspect we'd notice if North Korea started drilling wells in California. It's not exactly a suitcase nuke level of concealability.



South Korea accidentally discovered a bunch of tunnels that the North had been prepping. They'd gone unnoticed until performing some sort of water investigation iirc. You used to be able to go into them (near the DMZ) on tours. The walls are black and the guide I had noted that upon discovery, the fleeing workers painted the wall with coal to "make it seem like it was a mining operation all along".

So, the "drink your milkshake" level access isn't crazy.


Using tunnels offensively and using earthquakes offensively are enormously different things, though.

A tunnel is useful; you can get past the landmines and other defenses. Causing an earthquake a few miles across the DMZ is far less useful.


No no, the comment was merely to show that North Korea did in fact go unnoticed in their building of vast tunnels (wide enough for tanks).


No body/tank is getting through the ironically named Demilitarized Zone dividing South and North Korea, without getting lit up with fast moving pieces of metal hitting with high kinetic energy.

Hence the tunnels.

North Korean government's long standing, highest priority goal is conquest of South Korea through any means, including military invasion.


I'm not disputing any of that.

I'm disputing "earthquakes as offensive weapon" being useful. Not tunnels.


But you could be just across the border and digging in.


So the epicenter ends up in the attacking country’s border (or at least within a few miles)? Doesn’t sound like a particularly useful weapon when you’ve got to attack your own country with it too.


Lots of borders that are asymmetrically populated. (Feels weird to be thinking this way)

The entirety of Himalayas is earthquake Zone-4|5 and with many such borders.


that was a plot in one of the paperbacks in 198x - back then there was a spike in detection of USSR submarines violating Sweden waters, and in the story the submarines were performing under water drilling and injecting some stuff to be able to trigger earthquakes on demand when the time comes.


This is a purely political decision.

The current government, which ousted the former goverment with absurd scandals, is under a huge bribery and corruption scandal which includes kpop stars, police and prosecutors alike.

The current government always tried to shift the media’s attention when they were in trouble with former government’s incidents and this is one of them.

The government is actually trying to indict former government’s staff and let them take all the media’s blame, while they clean up their own bribery scandal.


Here are the abstracts of the two Science papers. This isn't my area, and although it would be hard to prove causation, they do have good evidence that causation is plausible. That's probably as strong a claim as anyone would be likely to be able to make, given the constraints on what is known.

The moment magnitude (Mw) 5.5 earthquake that struck South Korea in November 2017 was one of the largest and most damaging events in that country over the past century. Its proximity to an enhanced geothermal system site, where high-pressure hydraulic injection had been performed during the previous 2 years, raises the possibility that this earthquake was anthropogenic. We have combined seismological and geodetic analyses to characterize the mainshock and its largest aftershocks, constrain the geometry of this seismic sequence, and shed light on its causal factors. According to our analysis, it seems plausible that the occurrence of this earthquake was influenced by the aforementioned industrial activities. Finally, we found that the earthquake transferred static stress to larger nearby faults, potentially increasing the seismic hazard in the area.

The moment magnitude (Mw) 5.4 Pohang earthquake, the most damaging event in South Korea since instrumental seismic observation began in 1905, occurred beneath the Pohang geothermal power plant in 2017. Geological and geophysical data suggest that the Pohang earthquake was induced by fluid from an enhanced geothermal system (EGS) site, which was injected directly into a near-critically stressed subsurface fault zone. The magnitude of the mainshock makes it the largest known induced earthquake at an EGS site.


"The results support the findings of a pair of studies published in Science1,2 last year, which suggested the plant as a likely cause of the quake."

"Earthquakes have been linked to geothermal power plant in other parts of the world. But the Pohang quake is by far the strongest ever tied to this kind of plant — 1,000 times mightier than a magnitude-3.4 quake triggered by a plant in Basel, Switzerland, in 2006."

If you take the word `purely` out of your statement, it is a political decision. Seems to be based on science though. Everything else becomes irrelevant. This also reminded me about the `fracking` debated in US. If I am understanding this link [0] correctly, it's not the fracturing of the rocks but the high pressure liquids that induces earthquake? I am still unclear on the detailed mechanism.

[0] https://earthquake.usgs.gov/research/induced/myths.php


> The current government (...) is under a huge bribery and corruption scandal which includes kpop stars, police and prosecutors alike.

What are you talking about. The bribery/corruption scandal of the "Burning Sun" nightclub is linked to Gangnam police office, which is not exactly known for being Moon's supporters. We're talking about old boys with old corruption that dates back years. Today there was another revelation (although still unconfirmed, it seems) that the guy who started it all by beating up a nightclub patron was none other than the nephew of Choi Soon-sil, the crazy shaman friend of former president Park whose friendship basically brought down the previous government.

The other two sex scandals are even worse: they involve a high-ranking government officer of PGH government, and someone in the owner family of the ultra-conservative newspaper Chosun Ilbo, respectively.

So there's no reason why Moon would want to manufacture an earthquake controversy to turn people's eyes from these sex scandals. If anything, it should be the other way.

Get your conspiracy theories straight, geez.


I think this accident was expected. All construction works in South Korea are corrupted now. Why nobody knows?? It is because these kinds of works are under controlled by some companies. Not only this part but also nuclear power plant part is the same as this catastrophe. Just check out the Nuclear power plants in UAE. Those NEVER going to generate electricity because those are based on this culture...

https://medium.com/@cyb999999999/nuclear-energy-clean-but-no...




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: