QuakeFinder: Is It Possible To Reliably Predict Earthquakes?

massivepanic writes "Unlike most natural disasters, earthquakes strike without warning. Between large quakes and resulting tsunamis, millions of lives have been lost because science has been unable to provide accurate, useful earthquake predictions. Stellar Solutions' QuakeFinder Division hopes it can develop the tools to dramatically reduce this loss of life. Its network of over 100 sophisticated sensor stations has detected patterns of electromagnetic pulses several days before a number of different earthquakes, moving the possibility of earthquake prediction closer to reality."

Comments

answer:

[Anonymous Coward]

no

Re:answer:

[Anonymous Coward]

Maybe?

http://american-marconi.org/

According to this site they have already done it.

One recommendation

[Anonymous Coward]

It does not matter how reliable this technology is, I just hope they don't deploy it in Italy.

Is It Possible To Reliably Predict Earthquakes?

[ackthpt]

Yes.

With any kind of accuracy? Mmmm. No.

We are getting pretty good at predicting where they can happen, though.

I use QuakeSpy

[goldspider]

Oh, so this article submission is NOT a repeat from 1997?

Yay, Quakefinder

[flogger]

Yay! Quake finder is making a comeback. This just makes my day, I've never been a fan of gamespy and find a good old fashioned game of legacy quake I on quakeworld or quakefinder just hasn't been the same. I mean it has been years since...{{rereads the summary}}

Oh. Well. I'll just go in a corner and look for my quake I discs...

First you need to answer a question

[Anonymous Coward]

Are you at the controls of haarp technology which can CAUSE an earthquake?

Other wise within mainstream (broken) science I would say
Volcanoes yes, Earthquakes no.

Red Wood Ants

[BeCre8iv]

Red Wood Ants exhibit specific behaviour before earthquakes

http://io9.com/can-ants-predict-earthquakes-472681784
for example

Apply webcam and simple pattern recognition algo and presto. ...Now off to read TFA

Re:Red Wood Ants

[antdude]

http://www.accuweather.com/en/weather-news/can-ants-predict-when-an-earth/10211815 as well.

Re:Red Wood Ants

[Xest]

Yes, the only flaw is that said ants only live in a tiny fraction of the world where quakes are a threat.

Re:Red Wood Ants

[Hognoxious]

They're small, poor, and can't vote. Pick the little buggers up, put them in a tank, and put the tank in your quake zone.

Re:Red Wood Ants

[Xest]

Are we still talking about the ants here or is this some sort of scheme to use migrant workers to clean up after a quake?

preemptiv

[BeCre8iv]

From TFA

"First, it’s important to understand the difference between true earthquake prediction and the simpler idea of earthquake alerts."

GPS

[Anonymous Coward]

what happened to all that stuff on GPS reception going down the drain in areas prior to an earthquake?

I thought that looked rather promising

Why yes, you can predict earthquakes.

[Nyder]

https://www.google.com/#safe=off&output=search&sclient=psy-ab&q=animals%20flee%20before%20earthquake&oq=&gs_l=&pbx=1&bav=on.2,or.r_qf.&bvm=bv.45175338,d.cGE&fp=144025172efe81&biw=1155&bih=844&pf=p&pdl=300

You will find some animals flee before earthquakes, some act peculiar, as in, will be afraid or seemingly barking for hours before they hit.

I'm sure you wanted a more scientific answer, but sorry, while the scientist do great stuff, predicting shit that happens isn't one of them.

Re:Why yes, you can predict earthquakes.

[ShanghaiBill]

You will find some animals flee before earthquakes, some act peculiar, as in, will be afraid or seemingly barking for hours before they hit.

No you won't. You will find the opposite: Several have experiments have found no correlation between animal behavior and upcoming earthquakes. No experiments have found a correlation. There is a good summary here: Earthquake prediction - Animal behavior.

This just an old wives tale based on a few anecdotes. There is no evidence that animals can predict earthquakes. How could they? Unless you believe that animals have some ESP that science isn't aware of, they would have to be relying on some sort of infra-sound or ultra-sound, or maybe a magnetic disturbance. If that was true, we would be able to detect the signal directly, and cut the animals out of the loop. But no one has found any reliable signal.

Re:Why yes, you can predict earthquakes.

[excelsior_gr]

Actually, "predicting shit that happens" is the only thing that scientists do. The problem is, the prediction must be based on models that are in turn based on measurable and reproducible observations. This means that your anecdotal "animals flee before earthquake" is out, unless you can prove that a certain species will exhibit a certain measurable behavior (i.e. "flee", but how far? Do they flee in the direction of the epicenter or away from it?) with a given certainty at a certain time-point before the quake strikes (by the way, you will also have to define when the quake "strikes": do you take only the time-point of the largest peak or do the smaller ones recorded before that also count? If so, how small?). And you'll have to show that there is not only just a correlation, but also that the animal exhibits the behavior because of the earthquake itself. Otherwise, the animal is just a proxy of another parameter that can be directly taken into account (thus sparing having to turn the geology department into a zoo). If, for example, the animal flees because the ground trembles, well, we can already measure this variable with good accuracy and take it into account in the modeling. Plus, other things cause the ground to tremble (again, I'm taking this as an example), like, i.e. passing trucks. So it is not "animals flee before earthquake" but rather "animals flee when ground trembles". Of course, the sensory apparatuses of a certain animal may be finer than our respective instruments (so then you know what to do in order to improve your prediction), or, maybe an animal takes something into account that we aren't, like smell. Then you'll have to figure out what component in the air jumps/sinks to a concentration level different than normal ("normal" being a state that you'll also have to define, of course) before an earthquake.

It is not the cryptic terminology that makes an answer more scientific, it is how you got it. We certainly have a lot to learn from animals, but we have to be careful.

Re:Why yes, you can predict earthquakes.

[Alomex]

And you'll have to show that there is not only just a correlation, but also that the animal exhibits the behavior because of the earthquake itself.

You are wrong. If all you are trying to do is predict earthquakes correlation is enough. Just like the "canary-in-the-coal-mine" was an effective way to measure CO2 and/or methane levels even if we didn't quite understand the mechanisms by which it works from the get go.

Re:Why yes, you can predict earthquakes.

[excelsior_gr]

Well, you got me. I'm a theoretician. Of course you start with the correlation, but I like getting to the bottom of things, if I can.

Re:Why yes, you can predict earthquakes.

[Lord Crc]

This just an old wives tale based on a few anecdotes. There is no evidence that animals can predict earthquakes. How could they?

The only conjecture I heard which didn't sound totally "out there" was that the animals sensed the changes in the electric field caused by the piezoelectric effect due to stress changes in the rocks before an earthquake.

Something along those lines at least.

Re:Why yes, you can predict earthquakes.

[Anonymous Coward]

If the correlation is strong enough, then that is all you need. If the correlation is not strong enough and there are too many false positives, better understanding the actual connection between the two would help you improve the usefulness, especially if there is a fear false positives could undermine such a system.

Re:Why yes, you can predict earthquakes.

[ShanghaiBill]

This just an old wives tale based on a few anecdotes. There is no evidence that animals can predict earthquakes. How could they?

The only conjecture I heard which didn't sound totally "out there" was that the animals sensed the changes in the electric field caused by the piezoelectric effect due to stress changes in the rocks before an earthquake.

Something along those lines at least.

If this conjecture was true:
1. We would be able to detect the electric field changes with instruments.
2. Animals would respond to artificial experimental changes to the electric field. Fish do respond, but I haven't heard of them predicting earthquakes.
3. Someone would have used animal behavior to actually predict an earthquake rather than just making after-the-fact statements.
So while it is possible that an electric field is involved, Occam's Razor says that it is all a bunch of nonsense.
 

Re:Why yes, you can predict earthquakes.

[Anonymous Coward]

No you won't. You will find the opposite: Several have experiments have found no correlation

If that was true, we would be able to detect the signal directly

You're trying to argue that because humans haven't found a signal yet, there cannot be a signal. That's a well known fallacy. If it were true there could be no new discovery of anything, because we would have already discovered it before.

Re:Why yes, you can predict earthquakes.

[MickLinux]

1. We can. google Earthquake lights. Light is a valid member of the electromagnetic spectrum. or be lazy, and try this link:

http://www.youtube.com/watch?v=y196J86YrRU

2. Animals do respond to various electromagnetic changes. That doesn't mean that animals that are good earthquake predictors are necessarily responding to those changes-- there are many others that they could be responding to others. There are noises, gas releases, earthquake clouds, and other such signs of an imminent (read ongoing, but not noticeable to us) quake that they might respond to.

See the other posts on the redwood ants.

3. Some have. I think there was a guy named jobs who did this. The chinese did this once. Yes, they are discounted and explained away, but science has to start somewhere.

The best explanation that I have seen for why so many politically powerful experts dismiss earthquake prediction out of hand, is because instead of persuing a causal analysis, back in the 70's, the field turned to statistical analysis, which by nature cannot predict quakes.

The best way to predict quakes is to start with ones we know are predictable--those generated by nuclar tests, for example.

Then try to decrease the field of unknown, and increase the field of known.

Re:Why yes, you can predict earthquakes.

[ShanghaiBill]

1. We can. google Earthquake lights.

No we can't. Earthquake lights have only been observed while the earthquake is occuring. Claiming they can predict earthquakes is like saying that violent shaking of the ground can predict earthquakes.

Re:Why yes, you can predict earthquakes.

[MickLinux]

The holy grail of earthquake prediction is not predicting what will happen next year: it is predicting when one is already coming with enough warning to save lives.

In another post on this story, I noted that prediction is about forecasting what is actually an already-occurring event, that we just haven't noticed yet.

In line with that, the proposed California early-warning system is a very feasible plan, just way too expensive for my taste.

Re:Why yes, you can predict earthquakes.

[Hognoxious]

it is predicting when one is already coming with enough warning to save lives.

Do you know what the prefix pre means? It means before, and dict means say. It's therefore logically impossible to predict (say before) something which is already happening.

That would be condicting, which doesn't even exist.

Re:Why yes, you can predict earthquakes.

[Hognoxious]

What's this alleged signal made of? Magitrons? Has someone dissected an ant and found a mysterious organ that doesn't apparently do anything?

Re:Why yes, you can predict earthquakes.

[MickLinux]

Let me ask you: if you have an otto cycle engine rigged up with all kinds of sensors, and you start it up and run it for 40 minute under constant load, can you predict the characteristics of the next full cycle?

You can, though it hasn't happened yet, because the process isealready underway. In the same way, the earthquakes that accompany a volcanic eruption are absolutely predictable, even though --and because--the process is already underway. For example, if you see that the depth of certain shocks is decreasing at a fairly constant rate, you can predict that the next shock will be at a depth that is a function of the time elapsed.

In the same way, the proposed California Early Warning system is a prediction system that depends on processes already underway. Yes, the earthquake is already going. But it's not yet going strong. The early warning system announces that according to the shock pattern underway, there is a high likelyhood that this will be a big/small one.

Since you asked the question,

[Celeritas 5k]

Apparently not.

Making a mighty big assumptiom

[frovingslosh]

The OP failed to offer any support for the position that those "millions of lives" should be saved.

Best not to try

[Anonymous Coward]

You may get sent to jail if you don't predict right.

Yes...

[tangent3]

It's possible. That's why six scientists were jailed for manslaughter after failing to predict an earthquake.
http://edition.cnn.com/2012/10/23/world/europe/italy-quake-scientists-guilty

Re:Yes...

[Robotbeat]

That Italian court is such utter /bullshit/. What is the motivation to enter seismology studying earthquake prediction? Great pay? Nope. Recognition? Not really (no Nobel Prizes), at least not the kind you'd want. And with austerity, you're not going to get stable employment, either. When scientists are crucified for not being exactly right, all there is is downside, the risk of being accused of manslaughter if you say what you think. Congratulations, Italy, you might as well be in the Middle Ages.

I'd understand these charges if seismologists were as well-paid as surgeons are. But they flat /aren't/. If this is how members of society treats scientists, then members of society /deserve/ to lose their lives in earthquakes. But not their children. The blood of those who die is on the hands of those who blame the scientists and litigate. Anti-science is why we don't have better prediction.

-Chris

Re:Yes...

[idontgno]

And we have the answer: Darwinian Reduction.

Every scientist is obligated (at gunpoint if necessary... lives are at stake here!) to predict the next earthquake. Time, place, and intensity.

All the ones who predict incorrectly are imprisoned.

Repeat with all the remaining ones.

I figure after two or three rounds you'll have some damn good earthquake predictors out and about. Keep them close and remind them periodically about their former colleagues entertaining "Bubbah" back in the klink, by way of encouragement.

Voila! Reliable earthquake prediction.

Re:Yes...

[lesincompetent]

ENOUGH WITH THIS BULLSHIT! Even though *i dissent with the sentence* i must clarify that they have been charged for clearly stating that the situation was OK, that people should stay at home and that people should just ignore those pesky FORESHOCKS that lasted for MONTHS with INCREASING magnitude. I was fucking THERE at the time (well... not exactly there as i knew that we were in for quite a ride). Enough with this disinformation.

Re:Yes...

[painandgreed]

It's possible. That's why six scientists were jailed for manslaughter after failing to predict an earthquake. http://edition.cnn.com/2012/10/23/world/europe/italy-quake-scientists-guilty

They did not go to jail because they failed to predict an earthquake. They went to jail because they stated there would not be an earthquake in order to discredit somebody who claimed he had a system for predicting earthquakes. If they had stuck to procedure and official press releases, they'd have been fine and some quack would have a following and some new time, but they wanted to discredit the other guy so much they held and unofficial news conference. While starting off ok, they quickly devolved into making what sounded like confident claims there would be no earthquake under the questioning of the locals.

Re:Yes...

[MickLinux]

My memory was that their offense was --for economic reasons--silencing a scientist who was publishing that suddenly increased rates of radon release in the region indicated the probability of a large quake in the near future.

Oh, and the scientists who were charged were those who had the specific job of earthquake preparedness.

That said, I suspect a major part of their fault was not in caring too much about tourist dollars, but rather in their knee-jerk reaction that earthquakes can't be predicted, therefore anyone who predicts them is a charletan.

Which is what they were taught in standard courses, which is why they had the certifications, which is why they had the job.

But certifications isn't science. Certifications is politics. In fact, most of what you think of as science isn't science, but politics. Science is researching, making predictions, doing experiments, analyzing data, and analyzing results..

Re:Yes...

[Anonymous Coward]

Sorry for my English,

No, it is not so, it is not the work of scientists in Italy to inform people, Civil Protection and local politicians should do it, but they did not.

If you check you can see that the headchief of the National Earthquake Center GR was present at the meeting as an observer and not as a participant in the report was made four days after that meeting by others.

it is as if you are a bus driver, a thief uses the bus to go to a store for a robbery and the judge condemns you.

None of the politicians who had to inform the people was investigated.

Unwise ideas...

[hahn]

The Italian government is convinced you can.

I wouldn't ever claim to be able to accurately predict earthquakes even if I knew I could. To do so is to also claim responsibility.

Re:Unwise ideas...

[scatteredthoughts]

Fortunately, only a regional court which is convinced of that http://www.bbc.co.uk/news/world-europe-20025626

Re:Unwise ideas...

[Rhacman]

"I am here to officially report that my findings regarding any potentially imminent seismic activity are inconclusive. On an unrelated note I will be on foreign travel for the next two weeks with all of my possessions that do not function well at the bottom of a gaping chasm."

Re:Unwise ideas...

[Anonymous Coward]

Nobody in Italy is convinced.

The smart asses were sentenced because they said that a big earthquake was impossible (not unlikely, but impossible) and they didn't give the population infos about what to do in case of a big earthquake (and that was part of their jobs).
Why they didn't? Because they were nominated to that role by politicians and they were loyal to them; in that situation politicians said that they had to reassure the population and they did what the government told them, not what an honest man of science would have done.

Re:Unwise ideas...

[Anonymous Coward]

The irony is big on this one.
The sentence to which you are referring is the result of the opposite of what you are saying.

The Italian government wanted to oppose the media hysteria around an idoitic guy who was saying that he was able to predict an earthquake; so they held a conference to make the people know that you can't predict earthquakes.
The problem is that speakers "overdid" their job and made false and dangerous statements to get the point across.

The conviction is the result of the usual blame shifting from politics, with an iconic post-disaster scape-goating and witch-hunt.

Release Date

[Anonymous Coward]

Even more important would be this tech predicting the release date of Quake V.

It's easy!

[FuzzNugget]

Just use Twitter!

Re:It's easy!

[Anonymous Coward]

I'll rely on the USGS NetQuake network, one of which is installed at my house in the San Francisco Bay area. It's a very robust blue box, contains battery backup and runs Linux! The box is quite sensitive, constantly recording data and sends 'interesting signals' back to the USGS. It frequently detects far away minor quakes I do not feel.

Of course we can, in theory.

[Immerman]

Didn't we just get an articel yesterday about some species of ant whose behaviour reliably changes a day in advance of a magnitude 2 or greater quake? Obviously *something* is happeneing that they can detect, and if nothing else we might be able to transplant the ants to risk-prone areas and monitor their behavior. Better still if we could figure out what they're detecting and create our own sensor networks we might be able to improve detection rates and/or times even further.

Quake

[Bengie]

QuakeSpy was great at finding Quake servers.

History of failure

[ahziem]

According to Nate Silver in The Signal and the Noise, earthquake prediction has a long history of failure. Unlike weather, there is not much understanding or useful data. Many have claimed otherwise, but when applied to the future, their models failed.

No, not possible.

[houghi]

Way to many parameters.

People still need to do tests to see when a glue breaks and it breaks at different moments. That is just a simple cable that was made by a factory under reasonably controlled circumstances with a specifically know amount of glue with many known parameters.

And you want to know when it happens with many unknown parameters and a vastly huge area? Nope. Not possible.

Didier Sornette

[Anonymous Coward]

Didier Sornette may have provided some steps in the right direction. He began with analyzing the fuel-tanks of rockets for micro fractures, using frequency analysis -- and it worked well enough that such methods are apparently still used today. He's also predicted several major stock-market events well in advance. He worked on earthquake prediction too, applying similar principles, but that, I guess, is pretty damned tricky. However, I suspect whoever does eventually discover a suitable method of prediction, will be sure to mention Sornette.

Anyway, that's what I gathered from The Physics of Wall Street: A Brief History of Predicting the Unpredictable -- a pretty fine book, IMO.

- P

Re:Didier Sornette

[Anonymous Coward]

I forgot to mention that certain patterns which lead to success with the (rocket) fuel tanks exhibited fractal tendencies, so you might want to throw Mandelbrot in the equation too. Sornette was certainly well-aware of Mr. Benoit -- and "fractal" certainly seems like the right word, for the subject at least.

- P

Japan

[Anonymous Coward]

There's a site in japan (and in Japanese - can't remember it now) that my ex-partner used to frequent and they were having very promising results with local changes in the earths magnetic field being predictive of earthquake events. The changes were miniscule, of the order of nanotesla but consistently detectable and from what the ex could translate for me helped predict a number of earthquakes in the region being studied.

Unfortunately the site is also heavily tainted by lots of UFO and HAARP type conspiracy theories, which I found to be a great shame since I am a physicist and the magnetic field technique seems quite promising but will probably go nowhere as it is tainted by association.

Re: Is It Possible To Reliably Predict Earthquakes

[fermion]

On the US pacific coast they evidently have a system that can reliably provide warning for earthquakes, which is about the best we can hope for. Electrical signals travel faster than seismic waves, so in many cases there will be time to automatically prepare connected infrastructure. We are talking about the 30 seconds it takes to turn off devices that might cause damage, open emergency doors, sound a siren to give a few seconds notice to pull the car over. This is what other countries do, but I guess the US is not that technologically advanced.

In reality I suspect predicted earthquakes will be like predicted hurricanes landings. The false positives will kill it. Everyone will get prepared the first few times, but after a few false positives people will just assume the next one is as well. But we can't risk a false negative, so what is there to do? And people do get killed in hurricanes because they do have the sense to stay off the street.

For most natural disasters the best thing to do is simply to engineer the safety into the products. For hurricane it is buildings that will not flood and will not blow away. For earthquakes it is building that will fail gracefully.

Yes!

[Anonymous Coward]

Ask an I-Tie!

Did anyone else read "quakefinder" and think somet

[davydagger]

Did anyone else read "quakefinder" and think something else.

like discovering network multiplayer games of one of the more popular trilogies of first person shooters?

What about ...

[FlutterVertigo(gmail]

piezoelectricity from the friction & pressure?

# Brave Sir Robin ran towards...

[Hognoxious]

i.e. "flee", but how far? Do they flee in the direction of the epicenter or away from it?

My understanding is that flee sort of implies away, but I could be wrong. Perhaps that's what happened to the dinosaurs.