Slashdot Mirror
Earthquake Prediction Months In Advance
eegad writes "A UCLA seismologist named Vladimir Keilis-Borok claims earthquakes can be predicted months in advance. In the article at the University of California Newswire, he claims that the "team including experts of pattern recognition, geodynamics, seismology, chaos theory, statistical physics and public safety ... has developed algorithms to detect precursory earthquake patterns." It also says "the team's current predictions have not missed any earthquake, and have had its two most recent ones come to pass." They predict "an earthquake of at least magnitude 6.4 by Sept. 5, 2004, in a region that includes the southeastern portion of the Mojave Desert, and an area south of it." We'll see if they're right."
They got rid of the old guy with his knee that "acts up" right before an earthquake?
well, it's nothing one behind the ear wouldn't cure
While this is a great advance, the real deal will be when we get to the point we can predict precisely enough to WARN the people living in these areas.
As in, hey two weeks from friday, leave the area for a day or two.
Sent from your iPad.
that if they published this information a bit earlier, or used the tech worldwide a bit earlier, lives could have been saved in Iran
In June of 2003, this team predicted an earthquake of magnitude 6.4 or higher would strike within nine months in a 310-mile region of Central California whose southern part includes San Simeon, where a magnitude 6.5 earthquake struck on Dec. 22.
In July of 2003, the team predicted an earthquake in Japan of magnitude 7 or higher by Dec. 28, 2003, in a region that includes Hokkaido. A magnitude 8.1 earthquake struck Hokkaido on Sept. 25, 2003.
In 6-9 months there will be an earthquake within 310 miles of San Francisco of at least 4.0.
This is fun!
"I only speak the truth"
Karma: null(Mostly affected by an unassigned variable)
Coming soon to a TV near you: The earthquake channel! Get your 10 day seismic activity forecast!
clifgriffin > blog
For years scientists have known about the signs that the faults give off before an earthquake occurs, but most scientist are skeptical that they'll ever be able to accurately predict them because there are so many environmental factors to consider. Read more on PBS's microsite called Savage Earth, The Restless Planet: Earthquakes. It talks about prediction and whatnot.
There's been other studies like this.
For example, 30-odd years ago, some school did research looking in newspapers of the last 30 days before an earthquake for missing dog reports. Their results showed a large increase right around the time an earthquake happened in the area of the quake.
Blogzine
He made an earthquake prediction in Japan based on radio waves, and he actually came pretty close. Close enough that his ideas are worth more investigation.
If this turns out to be true, it would be a disaster for the economy in an area. Would you hang around or invest in a place where there's a big quake known to be coming in the next few months? It'd be like being told you've got a 100% chance of contracting cancer in the next few months. Although it helps you prepare, life can't be normal after that.
Seriously, I imagine if this sort of thing holds up, authorities will. Although this warning is so vague, it's only enough to get people to load up on emergency supplies, and possibly local governments to review disaster policies. Not that that accomplishment should be minimized, but something more certain a day in advance would be great.
-Looking for a job as a materials chemist or multivariat
This Russian group first got attention in the US seismology community when it "predicted" the Loma Prieta (Silicon Valley) quake of 1989. The technique performs spatial-temporal statistical analysis of weaker earthquakes that proceed large quakes. The first President Bush even asked the US Geological Survey to look into this.
The method may work, but it has not yet passed the scientifically required of repoducibility by scientists outside the Russian research group. Several leading US seismologists have tried reproducing this analysis method without success. Either the method is devilishly difficult to reporduce, important details have [perhaps intentionally] not been published, or it really doesn't work. Furthemore, you dont see the US results in press, because people generally dont publish negative results. Hopefully the reproducibility issues will be resolved and there will be a successful prediction method.
(Read my lips: cold fusion)
People expect that earthquake prediction would be accurate to within a few hours, so that evacuations can be accomplished, while avoiding unneccessary evacuations. The trouble is, evacuations are expensive, have their own hazards, and it's going to be incredibly hard to choose the lesser evil of bad evacuation timing, versus the present practice of not evacuating and being unprepared for the quake.
What would really help is a preparation protocol that can be syncronized more accurately with risk. If an earthquake could be predicted with a graduated probability, then gradually more disruptive preparation steps could be taken as the risk rises.
There is no doubt in my mind that this is a breakthrough in earthquake science, and that the researchers who developed this so called "tail wags the dog" method should be congratulated for their achievement.
One thing bothers me, however. Okay, so we know that there's going to be an earthquake somewhere in the world. The question is, what can we do?
In an affluent country/county, with educated individuals and a well organized emergency response force, there are several things to be done. First, evacuation procedures are begun. Secondly, the rescue and medical teams can be put on standby. Many similar actions can be taken.
However, the vast majority of the world that experiences earthquakes with some consistency can't do quite as much with such foreknowledge. First, most of their buildings are not specially enhanced to survive earthquakes (witness Iran, an extreme case of unpreparedness I admit but it serves my pont). Secondly, the population is highly dense and these people don't necessarily comprehend the danger, making evacuation procedures much less effective. Thirdly, the emergency police/medical presence in such areas is pitiful. Finally, the state itself does not have the necessary resources to carry out effective measures - they have to wait until foreign aid pours in. Now, the question is, will the U.S. grant emergency aid to, say, Iraq, because someone predicted that an earthquake would occur? Not likely. And if they don't get the money, these emergency operations don't get underway in any meaningful manner.
It seems to me that the focus has been diverted from building the infrastructure necessary to cope with earthquakes (in terms of buildings as well as emergency care) to instead predicting them in advance. As I said, if predicting them won't do too much good, why are we concentrating more in that area than in the one that actually WILL make a difference.
Hell, its probably the same deal as with research in diseases. The people with the money to conduct research don't have the same priorities/problems as those for whom research could benefit most.
Maybe I'm just pessimistic.
It is apparently now possible to locate the epicenters of tiny earthquakes ("microquakes") that occur very often, and they found that these often occur in the same spot, which would tell us that that location is a place where no bigger Earthquake could happen, as the tension is released often.
Even if we assume that we can conclude the other way round (saying, if the microquakes cease for a while, the probability of a bigger quake right in that spot would rise - which is probably true sometimes), still there would be no information about when the bigger quake would occur or how much bigger it was.
Sure, one could estimate the energy buildup (maybe, in some way), but the time when the bigger quake happens is still unknown. Also, the absence of microquakes is just telling that no more of these are happening - noone can know if this is because tension is building up or if for some reason this place is now lubricated better and tends not to lock anymore.
What one would need is a reliable way to measure the tension underground, and still it wouldn't be possible to know when a big quake happens. It would give a result like "Uh this tension is really high. Better we leave right now and dont come back until the big quake happened."
So far, the only sensible protection against Earthquakes is either buildings that withstand earthquakes (or dont kill people when they collapse... well the first approach sure is favoured ;) or not building at all where quakes happen.
Here's my prediction: "somewhere on earth, before the end of time, the earth will have at least a 0.1-magnitude earthquake!"
The point is, that only claimed that that had no false negatives. But they didn't discuss another critical aspect: how many false positives they had, and how tight their specificity is.
Without those details, you miss a lot.
This all seems like a hoax to me, BUT... lets pretend for a moment that it is absolutely true.
If earthquake prediction became the norm, imagine the damage to local economies here in the US!
Imagine this scenario...
"Earthquake, 2 months from now, Seattle area".
Ok, what do you, a business owner, do? Pack up and get out. Hell, you've got 2 months to do it.
Ok, what do you, a would-be tourist on vacation, do? Pick an alternate destination.
Ok, what do you, a local citizen, do? Panic. Perhaps pack the family and leave. Perhaps stay and stockpile supplies if your employer hasn't left yet.
I think it's very obvious that natural disaster prediction would be devastating for local and regional economies. In the big picture, as local economies start their own self-destruction, it'll have a bigger effect on the nation as a whole.
Skiers and Riders -- http://www.snowjournal.com
Scientists have known about these advanced prediction techniques for decades...
Unfortunatly, the original research was destroyed in an earthquake in 1987.
How do we know this guy isn't some sort of evil arch villian with an earthquake machine? How?? Huh? How??
The key is to start out general and work down to more specific predictions as the methodology is improved and you can build an adequate database and figure out what parts of your model are hogwash and which parts contribute. If they can get one or more parts of the prediction accurate at the 9 month mark, then there is a chance that they can become even more accurate over lesser ranges as time progresses. Also, the ability to predict a major earthquake out at the 9 month mark would be quite welcome for municipalities who are planning emergency preparedness. Imagine being able to budget so that your emergency personnel have the materials on hand that they need. Imagine being able to say "Ok, no-one go on vacation during the August-September timeframe as we are probably going to need all the bodies we can lay our hands on. Just because in the past it has not been possible to predict this sort of thing accurately does not mean it will not be possible in the future and therefore is not worth spending money on. Robert H
Giving a value on the Richter scale is not really meaningfull. You can have a 7 earthquake doing almost no damage if it happens far below earth surface and big damage with a 4 one near the surface in a low developped country.
It all depends on where the earthquake takes place.
You should use an estimate on the Mercalli scale. I find it more relevant.
Richter scale is all about energy released, Mercalli scale is all about damage/lost of lives which really is what matters.
Iraq: war to save the U
One of my fondest memories from high school was Iben Browning's earthquake prediction. He claimed a massive earth quake was going to shake the New Madrid fault around 1990. See http://geology.about.com/library/weekly/aa030903a. htm. Several months before the predicted date we had a 4.x quake during school. Everyone thought this was clear evidence Browning's prediction would come true. The school board cancelled class for 2 days surrounding the predicted date. No earthquake ever occurred. He helped us out and made the merchants in our area who jacked up their prices rich.
Who remembers the San Jose earthquake back in the early 90s? who here knew that there was a conference on Seismology there that same day.
probably the same number of seismologist that knew an earth quake was coming.
BTW, the conference was cut short.
The most important prediction method is to antipicate the maximum horizontal force resulting from an earth quake. A force execeeding 10% the amout of earth's surface gravity, called a "g", at one Hertz can collapse a poorly designed building or overpass. 200% g is observed in the largest quakes. A guide to destruction in terms of "g" is here .
The United States Geological Survey has spent a lot of effort predicting maximum forces. this is based on the location of previous large earthquakes and local soil conditions among other factors. This has resulting in relatively low death rates of quakes of similar size. For example last month's central California quake and Iranian quakes were about the same size with death tolls of 3 and 30,000. Ditto 1994 Northridge and 1995 Kobe Japan with tolls of 55 and 6,000.
Wow, I knew grep is powerful, but not that powerful;-)
I wonder what this might do to the insurance business. Lets say perhaps they predict a 7 or greater in LA in the next 4 months... Now a smart person living in that area would go beef up their earthquake or homeowners insurance (or buy some if they don't have it already).
But a smarter insurance company might decide not to sell any more quake insurance until after the deadline if you live in that area.
So now we know they are coming but can't do much to protect ourselves other than getting out of the area.
Apple free since 1990!
Reading the /. headline, you'd think that "scientists have learned how to predict earthquakes", but the glaring hole I'm seeing in the article is the absence of the a success rate. Sure, it "predicted" a couple of quakes, but how many false positives did it produce? How accurate were the predictions? Was it "a 95% chance of an earthquake between 4.5 and 4.6 magnitude within 100km of x? Was it "an unknown percent chance of an earthquake between 4.0 and 9.0(a really huge difference) "somewhere in California"?
This article is extremely vague about the accuracy or precision of the method, and limited to small test areas.
Don't get me wrong, I'd like us to be able to predict devastating earthquakes to help minimize casualties, but this is way too early to call it news.
This is a manual virus. Copy it to your sig and help me spread!
What's really informative about all of these models is that they pretend to model chaotic events. The lessons taught by Dr Lorenz fall on greedy ears.
They can go around predicting earthquakes, but miss just one and their creditbility, and funding, dry up. And miss one they will. These boys need to move their focus to modeling ground water movements. There's government money to be made doing that, or you can supress property rights or free enterprise, and no one will get a chance to criticize your work because the government and the biggest special interest groups are behind it. So, how do you avoid the strange attractor and arrive at previously determined conclusions? Simple. You use the big, second order differential equations as eye candy to blind the ignorant, then you substitute linear equations, disguised with a lot of greek letters, super and subscripts, amid a flood of jargon. Then you run your model backwards! Yup! You start with your desired conclusion and run your model backward to a set a 'inputs', adjusting co-efficients along the way to help out. It doesn't take long to find those 'inputs' in the huge pile of 'data' you've collected. That makes it easy to avoid the insensitivity, nonuniqueness and instability that is common in non-linear systems. Non-linear? That's what the atmosphere, ground water and earth movements are. That they could be accurately and fairly modeled by what are essentially y=mx+b (linear) equations is foolish, if not dishonest.
http://www.pha.jhu.edu/~ldb/seminar/butterfly.h
Of course, that doesn't stop some people from claiming that all they need to do to circumvent Chaos is discover more 'accurate' models. These folks also while away the hours inventing perpetual motion machines or over-unity power sources. Why not? They spent the better part of 50 years writing papers based on the Piltdown Man. http://www.clarku.edu/~piltdown/pp_map.html
And what did they do after the hoax was discovered? They claimed they knew it was a hoax all along! In the meantime, over 500 'learned' papers were written using the Piltdown Man as proof of all sorts of Evolutionary theories. Who knows how many Doctorates were handed out on the basis of that scam. But, who cares? Lots of grants were given, salaries funded and careers made using those phony bones. The scams are the same, the bones have changed.
Running with Linux for over 20 years!
You know, if you *really* want to test your own systems, go grab a copy of OpenSTA.
Reasonably flexible and GPL'd.
Wolde you bothe eate your cake, and have your cake?
As of this instant, your server is downright swift. [wanders around] Oooh, this works nice, and no evil javascript required.
My university had the complete USGS survey book, the big thick one with maps of everything anyone ever tracked, from climate to weeds. (Wonderful resource, that book.) I remember the earthquake details as compiled up thru 1958, and that if you want a quake-free location, the closest to that is North Dakota (only 3.n magnitude or less on the map). And it's amazing how many major metro areas are planted directly atop historical large-quake clusters.
~REZ~ #43301. Who'd fake being me anyway?
that whatever methods used are not specific to trends of the seismic regions they studied (i.e. California).
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
The only long term solution is to depopulate the vulnerable areas, locating all industry and housing in stable regions. The earthquake zones are OK for agriculture, provided the small number of buildings needed are properly designed.
Does anyone ever wonder why, in spite of many historical disasters, the US population in the principal earthquake regions continues to grow? Who profits from this? And who will benefit by allowing a situation lie this to continue, despite the certainty of there being 7-digit (yes, I mean 7-digit, a million or more) casualties in a major city within the next 50 years?
Look elsewhere, to Naples, and you will see the potential, indeed the certainty, of a million or more casualties if Mt. Vesuvius enters a new eruptive cycle, the first eruption of any cycle being likely to be massive, and pyroclastic. The time to evacuate the population is several days, from nothing to a much bigger than Mt. St. Helens eruption, only a few hours. Yet people live there... (In fairness to the Italian government, they are trying very hard, but have so far failed to slow down the population growth).
I ask myself why, when there are better places to live. Governments, including the US, should be planning a phased withdrawal from earthquake zones, instead they encourage the setup of vital industries in these regions. The Japanese are no better, remember the semiconductor shortage because one factory in the whole world that made a certain epoxy resin was situated in Kobe, and was out of action for many months? It all makes no sense.
Nor will the actions of the local government when a quake is predicted on the west coast of the US. Sadly, we will not have to wait long to find out how irresponsibly it will be handled.
Having said all this, my mother has recently moved house and lives on a fault line, the mighty Ochil Fault in Central Scotland. The vertical displacement of this fault is at least 5km (yes, really, over 3 miles, vertically, maybe a lot more, because even the coal mining industry has never bored deep enough to find out) at its maximum, and at least 3km wher it passes within 100 metres of the house. But, and it is a very big but, the last quake was about 2 on the Richter scale, way below the threshold where damage occurs, a few years ago. It is a very old fault, dating from the Carboniferous era or maybe before, and nowhere near a modern plate boundary. However, if that region did revert to its original level of activity, life would not be possible within 200 miles, or maye even nowhere in the UK, as there would also be massive pyroclastic volcanism. But, reactivation of such regions is unlikely, and certainly does not happen within a human time scale. However, it does make one think....
The problem is that the otherwise-more-desirable areas economically (good harbours, good rail access, natural resources, etc.) also tend to lie along areas high in natural disasters. As you say, it hasn't stopped anyone from living there, even tho we KNOW New York City is long-overdue for an 8+ quake, as are Seattle and the Bay area, and there's also some question about St.Louis (where historically, we already once had what some regard as the most economically-destructive quake in history). And hurricane season doesn't stop anyone from building in Florida or the Carolinas or Halifax. Being tornado alley's ground zero (likewise a predictable annual event) hasn't depopulated Oklahoma City, either. If you have any brains, you build to account for whatever your region throws at you. Proper foundations in quake zones, brush clearance and ceramic tile roofs in wildfire zones, shutters in hurricane zones, etc.
:)
But as another poster points out, to some degree the rest of us pay for all this, primarily with increased insurance rates. Shoddy construction in Florida results in houses being blown away -- funny how a much bigger hurricane can hit Halifax and do far less damage! (And ordinary winter winds on the Great Plains are routinely hurricane force, yet houses don't fall over there.) Turns out building contractors in Florida often (illegally) use staples instead of nails, and staples pull out when stressed, so in a high wind these houses literally fall apart. But meanwhile, insurance skyrockets for everyone, deserving or not.
Here in California, you can no longer always get private homeowner's insurance if you're in a severe wildfire or earthquake zone, but you can get federally-funded (ie. taxpayer-paid) insurance.
No good answers, just a pile of observations.
And remember.. the four California seasons: Fire, Flood, Riot, and Earthquake!
~REZ~ #43301. Who'd fake being me anyway?
Robert,
That's like saying, for example, that just because working perpetual motion machines haven't been made in the past doesn't mean they won't be made in the future.
Such a statement does not take into account the physical reality of the Three Laws of Thermodynamics. 1) You cannot get more energy out of a process than you put into it. 2) Not only can you not get more out, you can't even get out what you put in. 3) To get out all of what you put in, your process must vent waste energy to ZERO degrees Kelvin, which is impossible to reach... hence, you can't get out of the game.
As far as weather, water and earth... energy inputs to those systems cannot be mapped to specific output results.... they are not deterministic! Small changes in inputs can result in wildly different outputs (insensitive to initial conditions), or a given input doesn't always give the same output (nonuniqueness) or the system goes into wild oscillations (instability). Man HAS NO CONTROL over how much energy is put into these systems, even if he could measure them, and their models cannot reliability make any predictions as to the result of those inputs. It doesn't matter if they are considered linear or nonlinear systems. The best one can do is graph the strange attractor that resides behind a particular system. For a given input, the longer the process is allowed to continue, the more unpredictable the results will be. The best weather "models" can only go about 10-14 days into the future, and the results are given only in percentages in an area. They do that by running the same data in several different models and averaging the results. And, although they may "predict" a 30% chance for rain in your area, you have no assurance that it will rain at all on your house, your block or your city. Perhaps not even on your county or that area of your state.
It is intersting to note that the stock market is a chaotic system too. That's why you don't see any models predicting the price of Gold or any other stock on January 12, 2005 at 1:43 PM to within ten cents per ounce... or even a dollar per ounce. He who can do that rules the market. If these people truely had the ability to create models which accurately predict the dynamics of chaotic systems they'd test them first in the stock market. That they don't says volumes.
Running with Linux for over 20 years!
... your post seems to have little actual merit.
That's like saying, for example, that just because working perpetual motion machines haven't been made in the past doesn't mean they won't be made in the future.
No it's not. Predicting earthquakes is not known to be impossible, whereas perpetual motion machines are.
As far as weather, water and earth... energy inputs to those systems cannot be mapped to specific output results.... they are not deterministic! Small changes in inputs can result in wildly different outputs (insensitive to initial conditions), or a given input doesn't always give the same output (nonuniqueness) or the system goes into wild oscillations (instability).
This isn't true... the system is deterministic, disregarding quantum effects. It is chaotic, which means, as you said, that small changes can result in wildly different outputs. You also don't cite any evidence that earthquakes are highly chaotic. Given that many of the effects leading up to an earthquake take place on long timescales, chaos isn't so much of an issue as in weather prediction. Also, the amount of energy involved could make it like predicting a hurricane. You know there will be hurricanes in the Carribbean when the energy is there, and once one starts forming, you can tell approximately where it will hit. And unlike weather, the stress patterns that control earthquakes don't move much at all.
What's more of a problem is getting the data. It's hard to tell the stress on the rocks several miles beneath the surface, so a detailed model that allows computation of these stresses from other data is key.
If these people truely had the ability to create models which accurately predict the dynamics of chaotic systems they'd test them first in the stock market. That they don't says volumes.
Err? What are you talking about? You can't take a model for one chaotic system and port it to another, entirely dissimilar one (except for a few error-bounding theorems and the like). The forces in the market are entirely different from those under the earth, and modeling one does not mean you can model another. Furthermore, the market is a minority game, so any improvement in modeling has a tendency to cancel itself out as more people begin using it.
I hereby place the above post in the public domain.