Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
Chaos: The Tarot of Temblors

By imrdkl in Science
Fri May 07, 2004 at 10:53:47 PM EST
Tags: Technology (all tags)
Technology

Earlier this year, a Russian geophysicist named Vladimir Keilis-Borok, who currently works at UCLA, shook up the scientific world by predicting that a major earthquake will hit southern California before Sept. 5, 2004. Ordinarily, such claims are dismissed out of hand by the scientific community, but Keilis-Borok's team successfully predicted two other major earthquakes in 2003. Then in March, the California Earthquake Prediction Evaluation Council (CEPEC) issued a report which states that, while not yet fully substantiated, the methodology behind the prediction is legitimate, and the results are provocative. CEPEC recommended that all existing earthquake preparedness plans be reviewed.

Since the announcement of his "major breakthrough" in earthquake prediction, Keilis-Borok has received a lot of attention in the press, with all of the sensationalism and doomsaying that you might expect, but not much detail. His technique for making these predictions is usually described with a paragraph or two, as being based on identifying patterns of small tremors, with an occasional short reference to something called "chaos theory". Now, one popular analogy for chaos theory in action involves a butterfly flapping its wings in South America resulting in a hurricane on the east coast of the US. One might then wonder if perhaps a big rock which crumbles 15 miles underground in Mexico might lead to a bunch of new oceanfront property in Arizona...

Well, read on before you contact your realtor.


Charles Richter, the famed American seismologist and inventor of the earthquake magnitude measurement which bears his name, once said that earthquake prediction, "provides a happy hunting ground for cranks and publicity seeking fakers". Richter, who spent most of his life looking for ways to predict earthquakes, to little or no avail, lamented the "rush to any suggestion of earthquake prediction like hogs toward a full trough," by journalists and the general public. Richter, however, didn't have the advantage that today's scientists have - because Richter's work came before scientists began to recognise chaos in natural processes.

It turns out that many of the patterns which are to be found in analyzing seismic activity leading up to an earthquake are self-similar. Self-similarity is an important notion which comes out of the study of chaos and nonlinear equations, and it's really much easier to visualize, than to define. Very informally though, when speaking in terms of simple surfaces or curves, self-similarity exists when any portion of the surface or curve (picture) can be magnified (or scaled/zoomed), resulting in a new curve or surface which is identical to the original. When this independence of scale is observed, the surface or curve is called a fractal. Some of the more interesting examples of so-called "deterministic" fractals include the Mandelbrot set, and the Sierpinski Gasket, which some readers may already be familiar with.

In the study of seismic premonitory patterns which predict earthquakes, Kellis-Borok claims that self-similarity can also be observed - the groups of small and medium-sized tremors which precede a large earthquake tend to "look alike", for any given magnitude of the earthquake to be predicted. Formally, according to Kellis-Borok, "earthquakes follow a general hierarchical process that proceeds via a sequence of inverse cascades to produce self-similar scaling."1 This type of self-similarity, which usually results from a natural process, gives us so-called "random" fractals which are said to be "statistically self-similar". Instead of being exactly mappable onto itself for some scaling, a random fractal looks similar to itself at all degrees of magnification. Using this result, along with several well-known patterns, Keilis-Borok claims that it's possible to predict earthquakes more reliably, and on a more timely basis, than ever before.

The study of seismic event patterns is nothing new, though. There are hundreds of published scientific works which investigate every conceivable aspect of major earthquakes and the events which lead up to them. Many of these works propose so-called "functionals", which are defined on a sequence of shocks within a given area and magnitude range. These functionals eventually capture a specific pattern of seismic activity over a given area, a pattern which the author then may claim is premonitory to a larger event.

There are two primary problems with all of the functionals (and their associated patterns) which have been proposed to date, either they aren't timely enough - they precede large earthquakes by a period of years instead of months or weeks; or they aren't reliable enough - they precede large earthquakes by relatively short time periods, but they also give "false alarms". Therefore, to come up with a short-term predictor of large earthquakes, Keilis-Borok had to step back and see the bigger picture, so to speak, much as we do when we "zoom out" on a small bit of the Mandelbrot set. To accomplish this, he used a technique called renormalization, a rather complex data-reduction process, but which essentially means that he applied a scaling factor to existing functionals via manipulating their adjustable parameters so that their tendency to correctly predict large earthquakes was maximized.

In plain English, Keilis-Borok let the tail wag the dog - and in more ways than just renormalization of the data - but that's not necessarily a bad thing, as you'll see next.

Keilis-Borok's team utilizes three distinct, and previously known patterns when issuing a prediction. Two of these patterns, named ROC and Accord, were apparently discovered and proposed by Keilis-Borok himself during the last two decades. The ROC pattern analyzes the (relatively) simultaneous occurrence of medium sized shocks at relatively "large" distances from each other which form "chains", while the Accord pattern looks for uniformly distributed shocks within a "grid" of land tracts imposed upon some large land area. The third pattern, simply called U, analyzes the relative increase in frequency of these same shocks anywhere within the area under study. Each of these patterns, when taken individually, demonstrates one or more of the problems discussed previously when used to predict earthquakes. Accord and ROC can predict earthquakes months in advance, but they yield numerous false alarms. Pattern U is a more reliable predictor, but increased frequency of shocks within a given large area can begin years in advance of a major earthquake, and all three are subject to bias introduced by residual shocks (aftershocks) from previous earthquakes within the same boundary. When utilized together however, Keilis-Borok's team claims that these three patterns can provide a much more reliable short-term prediction.

To establish this claim, Keilis-Borok post-analyzed a large area of southern California, in a scientific paper published late in 20022. Five large (> 6.4 magnitude) earthquakes which had occured four to five years apart during the last 20 years in the given area were considered. In each case, the individual patterns, ROC, Accord, and U, were somewhat reliable as predictors, but also subject to false alarms, or long waiting periods. Taken together though, they reliably predicted all of the earthquakes in the study to within a period of months. This was a significant result - enough to get them published, in any case - which has not been accomplished with any other functional/pattern. The study concluded by emphasizing that the new technique reliably predicted earthquakes independent of scale, magnitude, and time, implying that the technique is truly able to capture the chaotic, but self-similar nature of the earthquake process.

The combined technique which was developed out of that study is now referred to by Keilis-Borok as "tail wagging the dog"3, and since then it's been used to predict two actual major earthquakes - a magnitude 8.1 quake near Hokkaido, Japan, in September of 2003 and a magnitude 6.5 shock that struck Paso Robles, California in December 20034. In both cases, earthquakes of similar magnitudes had not occurred in the respective areas for years preceding the predicted event.

The technique developed in the original study had to be adapted for real-life predictions of course, to see the "tail" first, and then analyze the bigger picture - thus "wagging the dog". To that end, the Kellis-Borok team first looks for the ROC pattern, a "long chain" of earthquakes occurring within a relatively short time frame, and relatively widely dispersed - specifically, there must be at least 6 quakes of magnitude 2.9 or greater along a chain of at least 110 miles (175km) . The ROC pattern is the natural place to begin looking, as it does not require a bounding area be introduced beforehand. When such a chain is found, the bounding area is then set, and the other two patterns, Accord and U, are searched for. If they're found within the area defined by the original chain, then a warning is issued for the subsequent 9 months.

That's pretty much how it works, this "holy grail" of seismology - now the question is, how reliable is it? By most accounts, Keilis-Borok is two for two, but one seismologist, Susan Hough from USGS in Pasadena, who is mentioned in only one of the various articles written about Keilis-Borok to date, claims that there have actually been three predictions issued by the Keilis-Borok team, and that one of them was incorrect. This claim seems to be countered by the January press release, which states that, "The team's current predictions have not missed any earthquake, and have had its two most recent ones come to pass.", and this reporter was unable to verify her claim. In any case, it's safe to say that if the Keilis-Borok team is correct in their current prediction, they'll likely garner more than a few new followers and believers in their "tail wags the dog" technique.

For now, although the scientific community is "split" about the findings, a "wait and see" approach is prevalent. For one thing, the area targeted by the current prediction is very large - indeed, the probability for an earthquake of magnitude 6.4 or greater occurring randomly in the area is at least 10%, according to the CEPEC report mentioned earlier. And while CEPEC is cautiously optimistic about Keilis-Borok's work, they also feel that the "results do not at this time warrant any special public policy actions in California". Keilis-Borok remains undaunted by the skepticism however, saying, "Application of nonlinear dynamics and chaos theory is often counter-intuitive, so acceptance by some research teams will take time. Other teams, however, accepted it easily."

Indeed. The more we learn about chaos and nonlinear dynamics, the more we understand that they are inherent in most, if not all, natural processes - and any attempt at predicting the outcome of such processes reliably must take chaos into account. At the same time we should pay heed to Richter's admonishment, and not "rush like hogs to a full trough" towards Keilis-Borok's prediction - and furthermore, how many of us really want the prediction to be true? It seems just slightly morbid to hope for an earthquake...

In any case we must ask, if Keilis-Borok's prediction actually is borne out, how useful is it, really, to be able to predict earthquakes months, instead of years in advance? Clearly, a months-long warning is too short to change building codes, and much too long to order evacuations. Nevertheless, if Keilis-Borok is right, it will be a significant victory for science and understanding the world we live in, and will certainly help to persuade preventive efforts, and planning in the future, if nothing else. That fact alone could save millions of lives.

References:

  1. Earthquake Prediction: Basics, Achievements, Perspectives - Keilis-Borok
  2. Premonitory patterns of seismicity months before a large earthquake: Five case histories in Southern California - Keilis-Borok, et al. - 2002
  3. Earthquakes can be predicted months in advance, report UCLA scientists who predicted San Simeon earthquake - Press Release, UCLA, Jan. 6, 2004
  4. Documented prediction of San Simeon Earthquake 6 months in advance: Premonitory change of seismicity, tectonic setting, physical mechanism - SEISMOLOGICAL SOCIETY OF AMERICA Abstract, 2004
  5. The Fractal Microscope - a java-based set of tools for playing with fractals.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Related Links
o Russian geophysicist
o predicting
o report
o "chaos theory"
o popular analogy
o earthquake magnitude measurement
o said
o Mandelbrot set
o Sierpinski Gasket
o large area
o scientific paper
o claims
o press release
o Earthquake Prediction: Basics, Achievements, Perspectives
o Premonitor y patterns of seismicity months before a large earthquake: Five case histories in Southern California
o Earthquakes can be predicted months in advance, report UCLA scientists who predicted San Simeon earthquake
o Documented prediction of San Simeon Earthquake 6 months in advance: Premonitory change of seismicity, tectonic setting, physical mechanism
o The Fractal Microscope
o Also by imrdkl


Display: Sort:
Chaos: The Tarot of Temblors | 99 comments (79 topical, 20 editorial, 1 hidden)
-1, crap. (1.50 / 26) (#3)
by rmg on Fri May 07, 2004 at 05:33:17 PM EST

why is it that every time something involves "chaos theory" we get the same old story about fractals and all the other "neato" bullshit we've all heard before? mandelbrot set, renormalization, blah, blah, blah.

why can you not present what these things actually are? why do you mention these things at all if you have absolutely no plans to talk about them in any meaningful way? you mention these "functionals" that you say are at the core of the research, but there is no explanation of what they are, except by the standards set in popular science.

if you don't understand the research, fine. just don't write about it. if you do, you owe it to the readers to write a useful, interesting exposition of it, not a piece of claptrap like this that gives us no more insight into earthquake prediction than gleick's piece of shit of a book gives the reader into chaos theory.

in short, you could have conveyed everything here by writing "someone has done some research leading to a few accurate predictions of earthquakes using chaos theory."

please, in the future, do not insult us with bullshit analogies about fractals and dogs. if you can't explain it in explicit, useful terms, don't.

----

dave dean

No insult intended (3.00 / 5) (#6)
by imrdkl on Fri May 07, 2004 at 05:58:28 PM EST

The last time that I tried to present even the most elementary of mathematical basis for one of my articles shall likely remain so. Even the rudimentary coverage given to the functionals in this article will cause some eyes to glaze over. Neverhtheless, I am open to discussing in more depth the notions of self-similarity and even renormalization (at an undergraduate level, granted) with you in more detail, if there's something which you wish to enlighten us with. The functionals described in this paper, however, are quite rudimentary - making use of simple binomial probability and exponentials to meet threshold values. The prediction works because the data is self-similar, and because earthquakes are a chaotic process, not because I gave some fancy-pants proof.

So why dont you take your mamby-pamby ass back to class, son?

[ Parent ]

something *i* wish to enlighten *you* about? (2.14 / 7) (#12)
by rmg on Fri May 07, 2004 at 06:22:55 PM EST

you, as the author of this piece, are charged with enlightening us.

if indeed it is "rudimentary," i am quite sure you could go into greater depth without prompting. and on the matter of self-similarity, no, i certainly don't have anything to say about it. that is the very substance of my complaint. after reading your article, i couldn't tell you one god damned thing about the topic. you don't tell us why the data is self-similar beyond some half assed crap that basically amounts to a tautology.

it is quite possible, as you suggest, that there's a great deal you know about this topic that i, with my meager background could not even begin to understand. please, impress me with the depths of your knowledge!

----

dave dean
[ Parent ]

jeez, no creativity (none / 0) (#31)
by Wah on Sat May 08, 2004 at 02:14:47 AM EST

after reading your article, i couldn't tell you one god damned thing about the topic. you don't tell us why the data is self-similar beyond some half assed crap that basically amounts to a tautology.

Umm, small rocks break before big rocks.  If you are watching closely enough, this 'pattern' expresses itself in the same way as other patterns smart people like the one you are posing as have been aware of for a couple of generations.  And 'watching closely' involves computing a great number of iterations of the rudimentary concept (small rocks break before big rocks).

That doesn't seem so complicated to me, and seems to be the gist of the article.

Did cts steal your shift key?
--
'The Matrix' is a better interpretation of quantum mechanics than Copenhagen.
[ Parent ]

sure. (none / 2) (#33)
by rmg on Sat May 08, 2004 at 02:28:05 AM EST

i agree that this treatment if very thorough and comprehensive and even ponderous if you are looking for Science as in Popular Science.

but this is kuro5hin, "a site for people who like to think." i would hope for something more like Science as in the National Science Foundation. that is, a _useful, precise, interesting account of the research.

now certainly, these kinds of things don't occur to a computer janitor such as yourself. you simply think "iterations! fractals! rocks! it all makes so much sense!" -- in short, you do not have the level of education (or i would claim common sense) to realize what you don't know, and in fact, as evidenced by the moderation on my original post, neither does kuro5hin at large.

and that, friends, is what is at the bottom of the decline of kuro5hin. mediocrity has become the norm, and in fact the enforced rule. halfwits like wah here take it upon themselves to heap abuse on the few worthwhile members this community has left simply to get a laugh! YHBT he cries!

----

dave dean
[ Parent ]

yes (none / 2) (#34)
by Wah on Sat May 08, 2004 at 02:58:19 AM EST

it's quite true. All of the stories you have submitted and gotten posted have been completely and utterly thorough.

So the 'worthwhile' members of k5 bitch about stories not being written to challenge their vast intelligence, and yet never write stories themselves.

Your perspective is intriguing.

No wait, it's bullshit.

you do not have the level of education (or i would claim common sense) to realize what you don't know,

There's a great many things I do not know.  Up til today, I didn't know that 'chaos theory' (which real mathemagicains such as yourself eschew) could be used to "accurately" predict earthquakes.  

What are all those quotes on your user page for again, biatch?
--
'The Matrix' is a better interpretation of quantum mechanics than Copenhagen.
[ Parent ]

let's not go down this road. (none / 2) (#37)
by rmg on Sat May 08, 2004 at 03:08:47 AM EST

it's been established in previous discussions that my claims are not exaggerated as far as education is concerned. many here can attest to it, having personally benefited from my knowledge and expertise. i doubt many can say the same of you.

on the other hand, i will readily admit (and indeed proclaim!) my infamy and my aversion to slinking around and using alternate accounts to post stories. i've also mentioned on numerous occasions my intense aversion to the sort of discussions that go on around here when the readers are faced with anything scientific or mathematical. these factors greatly limit my desire to contribute anything but the best in cutting edge, avant garde trolling to this particular community.

----

dave dean
[ Parent ]

but of course (none / 1) (#39)
by Wah on Sat May 08, 2004 at 03:14:56 AM EST

education by the truckloads.

and a thimble full of wisdom.

these factors greatly limit my desire to contribute anything but the best in cutting edge, avant garde trolling to this particular community.

figures, fucktard.
--
'The Matrix' is a better interpretation of quantum mechanics than Copenhagen.
[ Parent ]

wah, the biggest troll on kuro5hin (nt) (none / 0) (#85)
by circletimessquare on Tue May 11, 2004 at 09:23:53 AM EST



The tigers of wrath are wiser than the horses of instruction.

[ Parent ]
not chaos (none / 3) (#47)
by martingale on Sat May 08, 2004 at 04:50:28 AM EST

If prediction does indeed work, then it's precisely because chaos is *not* involved.

A chaotic process generally moves about its state space in such a way that within a short time, the distance from where it started is potentially so large that it cannot be usefully bounded.

If the state of the system giving rise to earthquakes were chaotic in this sense, then knowing a sequence of intermediate states would be essentially completely useless for prediction. But your article (I haven't looked at the references closely) suggests the opposite.

Most continuous random processes of interest have erratic paths. In fact, often these paths are not differentiable. However, that doesn't make the paths chaotic. For example, a brownian motion, variations of which are commonly used to model stock market statistics, has a derivative almost nowhere. It is literally impossible to draw its path. However, the distance traveled by the path in a short time is very short, and eminently predictable.

It appears to me that in this case, some time series are modeled as self-similar processes (the Brownian motion being self-similar as an example). It's hard to be sure because mathematical terminology used by scientists is often both imprecise and unfortunate.

Nevertheless, the idea of using functionals is just a fancy terminology for the idea of computing functions of the observed path. Renormalization, which is a dirty word invented by physicists, is most likely related to process convergence as certain parameters are tweaked. This follows the venerable tradition of so-called diffusion approximations. In terms of a picture, it usually amounts to squeezing and expanding the time series graph coordinates until the graph looks like a known process.

Nearly all probability theory can be called measure theory in hairy infinite dimensional function spaces, but frankly that's just wankery, like invoking chaotic behaviour on everything. If the time series being studied is well modeled by a self similar process, that's already a valuable discovery.

[ Parent ]

Strangely enough... (3.00 / 5) (#51)
by imrdkl on Sat May 08, 2004 at 08:45:49 AM EST

Newton's attractor is bounded, even though the path may be infinitely long. The mandelbrot set is certainly bounded. Generally, it's unclear how you can dismiss the theory due to the tendency of some generators to go off in the weeds. Anyone can recognize that the boundary for any given seismic event is defined and limited by the length of the tectonic plates which are beneath it.

Chaos isn't randomness. It seems random, but the observation of self-similarity debunks the notion that true randomness is at work, at least for the statistician. Your dismissal of probability theory belies your prejudice in this regard.

Similarly, physics was fundamentally affected by the notion of renormalization. It might be a dirty word for the purist, but the fact remains that the process has advanced understanding of processes which were totally incomprehensible before its introduction. I like this explanation:

Any given theory of physics has some energy scale where the theory breaks down. Renormalization not only allows you to perform calculations below that scale, but through the renormalization group equations, tells us where that scale is! This allows people to predict where to find new physics.


[ Parent ]
Correction (none / 1) (#53)
by imrdkl on Sat May 08, 2004 at 10:12:24 AM EST

That's the lorenz attractor, a bit after newton's time. And yes, it will diverge in the long run, but geologically speaking, there's no harm in assuming short term stability (eons). Some think that tectonic plates migrated from a single land mass, which may evidence divergence as you say.

[ Parent ]
it's a fatal attraction (none / 3) (#55)
by martingale on Sat May 08, 2004 at 10:23:05 AM EST

Newton's attractor is bounded, even though the path may be infinitely long. The mandelbrot set is certainly bounded. Generally, it's unclear how you can dismiss the theory due to the tendency of some generators to go off in the weeds.
Boundedness isn't the issue. You can have a chaotic system inside a small box, if you like. The issue is with pinpointing the path within such a box. When your system is chaotic, then knowing where it was a moment ago is useless ... it could be literally anywhere, within the box of course.

Contrast this with a non-chaotic system. For such a system, knowing where it was a moment ago is useful - if it was in the lower left corner of the box, it's probably still there if the time interval is short. If it's attracted towards some stable equilibrium, again you know where to look and you don't have chaos.

The point about Lyapunov functions is that they allow bounds on the behaviour of the dynamics to be computed. But if you have a chaotic system, then no "useful" bounds exist. In the case of the box, you can always say that the dynamics remain within the box - but that's far from useful, even though it's a bound.

When you take a random process such as BM, you don't completely lose the ability to predict the path. Sure, you can never be certain what the path looks like exactly, but you can make good guesses with high probability. That's not the same as a chaotic system.

Chaos isn't randomness. It seems random, but the observation of self-similarity debunks the notion that true randomness is at work, at least for the statistician. Your dismissal of probability theory belies your prejudice in this regard.
Hmm, I didn't intend to dismiss probability, rather I dismissed the direct connection to chaos. But it seems you're happy to accept the distinction yourself, so I guess I overdid it ;-) I however tend to dismiss the overemphasis on self-similarity, which I think of as a fad.

Self similar processes in probability have a well defined meaning, but self similarity is not sufficient to capture all sorts of long term effects. In fact, many self similar processes have short memory. That is one of the reasons applying them to the stock market is an imperfect fit.

Similarly, physics was fundamentally affected by the notion of renormalization.
I doubt that's the kind of renormalization they're talking about. Here's a direct quote from the second reference:
We renormalized in time and space three previously known premonitory seismicity patterns, which reflect the increase of seismic^ activity and increase of earthquake correlation range.
That sounds a lot more like simple rescaling than expanding quantities via a renormalization group. To be sure, they don't offer a lot of details, but I think you may be off in assuming they use the same method the physicists use for field theories.

[ Parent ]
Zoom out - see the big picture (none / 1) (#63)
by imrdkl on Sat May 08, 2004 at 04:58:58 PM EST

The only thing that Keilis-Borok is claiming here is that, given that seismic processes are chaotic, and likewise display self-similarity, his (their) technique bears out that assumption - since it scales to all magnitudes of earthquakes. They're not basing their methodology or their claims on chaos theory, they're only saying that their results indicate (verify) chaos in (yet another) natural process. The functionals are based entirely on simple probability estimators from discrete distributions, and have no chaos blended in.

About the only thing that I can think of, that actual chaotic analysis tools would be good for in the study of earthquakes, is to simulate them. And while Keilis-Borok has apparantly done this, it really isn't relevant to this work.

Finally, I suspect that you're right about Keilis-Borok's usage of the renormalization term, though. In fact he indicates that by his description - which sounds much more like rescaling.

[ Parent ]

Wrong sense of boundedness (none / 3) (#59)
by Coryoth on Sat May 08, 2004 at 01:38:25 PM EST

By definition a chaotic system ought to be bounded -  that is the orbits lie in a compact space.  That's what you get with the Mandelbrot set and  Newton's attractor.  ying in a bounded space doesn't really give you anything.  The grandparent poster meant bounded in the sense that the space of paths through the (bounded space) is unbounded.  That sounds weird, but does in fact make perfect sense.  There aren't bounds on possible paths.

Chaotic systems are, indeed, not random.  However, they are such that given a dataset you cannot make any sensible extrapolation due to the sensitive dependence on initial conditions - you will always have a margin of error in measurement, and as any time series in infinitesimally close to a completely different one in a chaotic system, your measurement error means prediction is meaningless.

Of course that doesn't mean you can't still extract meaningful information from the system, which would seem to be what is going on here.   I haven't read enough of the details to be sure.

But I agree with the grandparent poster, in terms of the mathematics renormalization and random fractals aren't on solid ground.  That's not to say this scheme doesn't work, just that it has shaky foundations (but so did a lot of physics once upon a time).

Jedidiah,

[ Parent ]

Thanks for that clarification (none / 1) (#64)
by imrdkl on Sat May 08, 2004 at 05:05:11 PM EST

I got what martingale meant after he restated it, but as I point out in my reply, the foundation of Keilis-Borok's work isn't based in fractals, it's based on simple probability distributions and exponentials.

Keilis-Borok isn't basing his technique on chaos, he's simply saying that his technique validates chaos in the process. Nothing more, and nothing less.

[ Parent ]

Last time (none / 1) (#68)
by Scrymarch on Sun May 09, 2004 at 08:34:35 AM EST

That forest restoration article was great, actually, even if it was a lot of effort to write.

[ Parent ]
Functionals (3.00 / 6) (#16)
by Coryoth on Fri May 07, 2004 at 06:51:58 PM EST

Functionals shouldn't be that hard to provide some sort of explanation for that can be understood by a general audience.

One can think of them as a special kind of mathematical function.  Normally a function takes one or more numbers as arguments and outputs a number or a vector.  A functional takes a function as an argument and outputs a number.

If you want a way to think about it, think of a normal function as an infinite dimensional vector, and that's the input to the functional - so in a sense a functional is function that takes an infinite number of numbers as input.  That's not very rigorous, so don't go writing it in your math finals, but it should give you a decent conceptual idea of the sort of object we're talking about.

Now, how does that relate to this earthquake prediction?  Well, that's less immediately clear to me - I'm a mathematician, and I don't know much about earthquakes.  I suspect it is along the lines of taking a time series of data and viewing that as a function (that is, you interpolate the discrete sampling), and thus when you apply a function to that, you are really using a functional.

Again, that's not very rigorous, and would probably be better explained in terms of function spaces as vector spaces, but I think I'd be writing too long, and I'm guessing as to how they're using the term anyway.

HTH

Jedidiah.

[ Parent ]

i'm well aware of what a functional is. (none / 2) (#17)
by rmg on Fri May 07, 2004 at 06:54:30 PM EST

i was pointing to the lack of explanation for an inexpert audience and especially the particular functionals in question and their relation to the research.

----

dave dean
[ Parent ]

and yet (none / 2) (#30)
by Wah on Sat May 08, 2004 at 02:02:19 AM EST

despite this awareness, you are unable to explain the things yourself.  Instead chosing to call out someone else for not doing what you also did not do.

What part of 'hive mind' do you not understand?

How long and boring did you want this article to be?
--
'The Matrix' is a better interpretation of quantum mechanics than Copenhagen.
[ Parent ]

i have no time for scurrilous trolls (none / 3) (#32)
by rmg on Sat May 08, 2004 at 02:17:03 AM EST

like yourself.

what is at issue is the article. i gladly admit to not knowing much about the PR juggernaut that is "chaos theory" because i am a pure mathematician and in pure mathematical circles, chaos theory is widely disregarded. indeed, my knowledge of it is limited to what i have gleaned from a textbook on dynamics and bullshit treatments of the subject like the one presented in this article.

however, if you have a substantive meta-critique, i'm sure everyone will be very interested to hear how you think i could have improved upon my criticism of imrdkl's article.

----

dave dean
[ Parent ]

prove it (none / 0) (#36)
by Wah on Sat May 08, 2004 at 03:01:42 AM EST

i am a pure mathematician and in pure mathematical circles, chaos theory is widely disregarded.

I'm waiting.  

Is that some sort of moral purity.  What, are you some type of Jehovah's Mathemetician?

i'm sure everyone will be very interested to hear how you think i could have improved upon my criticism of imrdkl's article.

You mean by reading my comment to you?
--
'The Matrix' is a better interpretation of quantum mechanics than Copenhagen.
[ Parent ]

lame. (none / 2) (#38)
by rmg on Sat May 08, 2004 at 03:10:34 AM EST

playing stupid can only take you so far. remember that.

----

dave dean
[ Parent ]

from a troll... (none / 0) (#40)
by Wah on Sat May 08, 2004 at 03:18:07 AM EST

that means so very, very much.

And yes, your criticism was quite lame.

--
'The Matrix' is a better interpretation of quantum mechanics than Copenhagen.
[ Parent ]

i'm only trying to help you. (none / 2) (#41)
by rmg on Sat May 08, 2004 at 03:20:23 AM EST

only the arrogant disregard sound advice.

----

dave dean
[ Parent ]

please stop (none / 0) (#42)
by Wah on Sat May 08, 2004 at 03:23:02 AM EST

the irony is painful.
--
'The Matrix' is a better interpretation of quantum mechanics than Copenhagen.
[ Parent ]
Your only hope is to stop replying. (none / 0) (#49)
by finality on Sat May 08, 2004 at 07:52:47 AM EST

Much like an excited baboon manipulating his penis, rmg will not stop replying like a five year old once he has himself a captive user. Break the cycle: rate him zero and leave.
This account has been anonymised. If you can give a good reason why, email rusty@kuro5hin.org, as he is obviously lacking one.
[ Parent ]
oh really now, that's too much! (none / 1) (#60)
by rmg on Sat May 08, 2004 at 02:13:46 PM EST

if i find an eager fool willing to post page after page of offtopic crap under the guise of trollbusting, can i really, in good conscience, pass it up?

----

dave dean
[ Parent ]

not in your good conscience... (none / 0) (#79)
by Wah on Mon May 10, 2004 at 10:39:29 PM EST

...wait.

You call that 'good'?

if i find an eager fool willing to post page after page of offtopic crap

where'd you put my mirror?
--
'The Matrix' is a better interpretation of quantum mechanics than Copenhagen.
[ Parent ]

well i never! (none / 1) (#80)
by rmg on Tue May 11, 2004 at 01:09:52 AM EST

are you insinuating that i post offtopic crap? i can't believe this!

----

i ♥ legitimate users.

dave dean
[ Parent ]

absurd (none / 0) (#81)
by Wah on Tue May 11, 2004 at 01:24:28 AM EST

like chaos theory even.  Can anything useful possibly come out of it?
--
Help us cross the digital divide, yo.
[ Parent ]
wah, the biggest troll on kuro5hin (nt) (none / 0) (#84)
by circletimessquare on Tue May 11, 2004 at 09:22:41 AM EST



The tigers of wrath are wiser than the horses of instruction.

[ Parent ]
thanks (none / 0) (#86)
by Wah on Tue May 11, 2004 at 10:26:54 AM EST

you may now bask in my glory.
--
Help us cross the digital divide, yo.
[ Parent ]
what a poor sad pathetic creep (nt) (none / 0) (#87)
by circletimessquare on Tue May 11, 2004 at 10:34:25 AM EST



The tigers of wrath are wiser than the horses of instruction.

[ Parent ]
my thoughts exactly (nt) (none / 0) (#88)
by Wah on Tue May 11, 2004 at 10:40:56 AM EST


--
Help us cross the digital divide, yo.
[ Parent ]
you have a reponse for me everytime (none / 0) (#89)
by circletimessquare on Tue May 11, 2004 at 10:50:21 AM EST

but do you have any response to this, trollboy?

http://www.kuro5hin.org/comments/2004/5/8/54352/79317/160#160

you are really sad and pathetic

why don't you stop biting at rmg and i?

why don't you give up your multiple accounts?

why don't you stop 0 rating and modbombing all the time?

why don't you admit what you are...

for whatever i am, whatever kind of troll i am, your behavior makes you something out as a lower life form on kuro5hin than i or rmg ever could be

you're pure scum, really

your behavior is way lower than rmg or i could ever, or would ever, stoop to

rusty shouldn't ban you (yet another difference between my behavior and you: i don't ask for people to get banned), you should be here forever

why?

because you define the lowest form of existence here

you serve as the ruler against which other account's behaviors are to be judged

with you serving as the definitive example of the lowest puerile, sycophant, disgusting behavior

you're pure scum wah

you really really are

i could never stoop as low as you, i could never dream to be as low as you


The tigers of wrath are wiser than the horses of instruction.

[ Parent ]

yes (none / 0) (#90)
by Wah on Tue May 11, 2004 at 10:55:34 AM EST

I've got a nice anecdote for you. And him.

you are really sad and pathetic

Yes, this behaviour is.

why don't you stop biting at rmg and i?

Who's biting who?

why don't you give up your multiple accounts?

Never had 'em, never will.

you're pure scum wah

No, I'm acting like it.  Intentionally, as you well know.  Why does it bother you so?

...............
...............
...............
...............
...............
...............

But you can be quite insightful when you are angry.

you serve as the ruler against which other account's behaviors are to be judged

And as such.  I rest my case.

--
Help us cross the digital divide, yo.
[ Parent ]

what you is (nt) (none / 0) (#91)
by circletimessquare on Tue May 11, 2004 at 10:57:22 AM EST



The tigers of wrath are wiser than the horses of instruction.

[ Parent ]
rmg! rmg! (none / 0) (#83)
by circletimessquare on Tue May 11, 2004 at 08:43:44 AM EST

check out this thread!!!!

http://www.kuro5hin.org/comments/2004/5/8/54352/79317/122#122

The tigers of wrath are wiser than the horses of instruction.

[ Parent ]

bwahahaha! (none / 0) (#93)
by rmg on Tue May 11, 2004 at 01:08:54 PM EST

the funny thing is that he is exactly right, except about the rmg being a mean son of a bitch part. i'm actually a nice, friendly troll!

----

i ♥ legitimate users.

dave dean
[ Parent ]

aren't we all? (nt) (none / 0) (#94)
by Wah on Tue May 18, 2004 at 02:27:13 PM EST

XOXO
--
Help us cross the digital divide, yo.
[ Parent ]
Chaos theory (3.00 / 5) (#44)
by Coryoth on Sat May 08, 2004 at 03:34:35 AM EST

Well, presuming you know what you're talking about we'll start with a definition

A chaotic system is a dynamical system which is topologically transitive, and periodic orbits are dense.  

All "chaos theory" says is measures of quantitive and qualitative system behaviour are still possible.  The actual measures, like Bifurcation diagrams, or recurrence plots can be mostly qualitative, or for things like fractal dimension, purely quantitive.  The problem occurs when one notes that Bifurcation diagrams and fractal dimension really have very little in common.  In that sense "chaos theory" doesn't exist as any one thing, and this is where many of the misconceptions come from.  Its kind of a blanket descriptor thrown over a wide variety of potentially vastly differing approaches toward extracting some meaning from chaotic systems (which are otherwise extremely hard to deal with -  as you can see from the properties of the definition).

The media only makes it worse by pretending that it is all one coherent theory.  Furthermore, the hype has brought in a lot of random people from other fields applying the ideas in a half assed way so that they can claim they are "using chaos theory".  These "uses" normally resemble a sociologists use of statistics.

HTH

Jedidiah.

[ Parent ]

meta-critique (3.00 / 4) (#52)
by codemonkey_uk on Sat May 08, 2004 at 09:30:33 AM EST

Comment Type: "Editorial" would have been an improvement.
---
Thad
"The most savage controversies are those about matters as to which there is no good evidence either way." - Bertrand Russell
[ Parent ]
He gave perfectly sufficient pointers. (3.00 / 14) (#18)
by mcc on Fri May 07, 2004 at 07:02:47 PM EST

While light on technical details the article at least gave a vocabulary and pointers to relevant work in the field fully sufficient for someone to easily perform the research to uncover and understand the technical details themselves.

Mathworld describes a functional as "a real-valued function on a vector space V, usually [a vector space?] of functions". A quick google search for "functionals earthquakes" returns a number of results which seem to indicate this is in fact the correct terminology to be looking at, and that this is a concept which seems to be widely applied in the analysis of statistical models. I must admit ignorance as to exactly how one goes about creating a vector space of functions, or as to what exactly... well, any of the terminology in the links Google turns up on means (I assume a "Lyapunov Functional" to be a functional consisting of Lypunov Functions, but as far as what a "Feynman-Kac Functional" is or what it has to do with markov chains I am somewhat lost). However this at least gives an idea of the context, which is the important thing.

Looking in this article's references, the first link appears to be aimed at the public, or at least whoever is awarding these people grants, and contains little beyond an argument that research about earthquakes is important. Upon looking at the second paper, however, it is immediately clear that this is the "useful, interesting exposition" you were looking for. The full set of mathematics in their model is exposed, as is at least one possible theory as to why that set of mathematics might describe reality. I must admit that in order to understand these mathematics at any level beyond that of "okay, they're doing something to some data" I personally would probably have to simultaneously follow along with a few statistics textbooks just to understand the meanings in context of the terms used. However, I see no reason whatsoever why this paper would not be entirely accessible to anyone who was willing to take the time to look up the contextual significance of the relevant vocabulary as they read.

This article gives little indication of the historical context of these findings; little idea of whether the idea that statistical models based on functionals are useful in modelling the behavior of faults is one that is widely accepted, or just one that this single Keilis-Borok group is pushing; and little indication is given of what chaos theory has to do with this is given except to the extent of using the term as a loose umbrella for the idea of charting patterns found within systems which appear from the outside to be too dense in complexity to analyze from a standpoint of pure mechanics. However, as far as the first two things go this article is outwardly stated as being a presentation of one group's work-- not a summary of an entire field-- and it achieves that stated goal. As far as the chaos theory junk science bit goes, self-similarity is indeed a feature of their mathematical model, and lacking any other popularized vantage points within the relevant materials to provide to the reader, it is not unreasonable to reference chaos theory just because it is relevant yet something a layman is likely to have heard of.

There is nothing wrong with summarization; that is, after all, to some extent the entire point of journalism. Were Mr. Dkl to assume a functional understanding of the underlying mathematics, he would have produced an article few would have been able to understand and of no more practical use to the reader than had he just pasted in the text of his second reference ouright. Attempting to explain the underlying mathematics and present the article in that context would prove little more practical, as one could easily spend the entire length of this article just trying to provide a suitible explanation of one of the simplest terms involved in all of this, for example "vector space". So what we're left with instead is a perfectly passable summary, and those not content with a summary are free to read the further details in the bibliography; that is, after all, what I assume the point of a bibliography to be.

---
Aside from that, the absurd meta-wankery of k5er-quoting sigs probably takes the cake. Especially when the quote itself is about k5. -- tsubame
[ Parent ]

Nod (2.66 / 6) (#20)
by imrdkl on Fri May 07, 2004 at 07:24:24 PM EST

I was just about to say that. :)

Actually, the chaos, or at least the self-similar result of a chaotic underpinning, is intended to be more than a loose umbrella in this article. In particular, Keilis-Borok claims, and I carefully point out that the their technique (with renormalization) applies to all magnitudes of earthquakes. Thus, when the pattern(s) are met, micro-tremors predict small tremors, small tremors predict medium tremors, medium tremors predict large tremors, and so on - the scaling effect works down to any level of magnitude. This is the very basis of the chaos/self-similar claim, and clarifying it more deeply than that would require much more than graduate-level math, since the mathematics of random fractals in nature is not well-defined, and certainly not found in any textbook.

I do wish he had some cool pictures of data, though.

[ Parent ]

textbook (2.75 / 4) (#46)
by martingale on Sat May 08, 2004 at 04:11:38 AM EST

graduate-level math, since the mathematics of random fractals in nature is not well-defined, and certainly not found in any textbook.
Most of those ideas are due to Mandelbrot. He and his school are the ones trying to convince everyone to model every process as random fractals. Whether that has anything to do with the stock market crashes of the last ten years is debatable ;). As textbooks go, you can't do better justice to these ideas than his books. You can find more rigorous sources, but frankly, rigour in this case is only as good as the underlying assumptions.

[ Parent ]
Sorry, no time to read it (1.00 / 17) (#10)
by Badger Patrol Leader on Fri May 07, 2004 at 06:13:28 PM EST

Could somebody who read the article tell me when California is gonna sink? Thx!

--
Reminder: Patrol meeting at 7:00 at the Jensen's. Mike will bring snacks, and we will have a presentation on cold-weather camping.

Not soon enough. <nt> (none / 3) (#27)
by godix on Fri May 07, 2004 at 10:37:28 PM EST



Thank god I'm worth more than SilentChris

[ Parent ]
Be proactive (3.00 / 10) (#28)
by fenix down on Fri May 07, 2004 at 10:37:48 PM EST

You can't just wait around for California to sink on it's own, take the bull by the horns, buy a shovel and get to work.

[ Parent ]
a big shovel (none / 0) (#70)
by khallow on Sun May 09, 2004 at 02:11:44 PM EST

I think a series of 100 megaton nukes buried along the San Andreas fault should help with the digging. And be sure to buy up a bunch of future beach front property first.

Stating the obvious since 1969.
[ Parent ]

Like (none / 0) (#78)
by emmons on Mon May 10, 2004 at 10:30:15 PM EST

As much of Nevada as possible. It's pretty arid right now so the land is cheap. Give it a few years of being on the coast and it'll green-up nicely for a healthy profit.

Now if you could only get your hands on a few of those nukes...

---
In the beginning the universe was created. This has made a lot of people angry and been widely regarded as a bad move.
-Douglas Adams

[ Parent ]

Great Stuff (1.25 / 4) (#22)
by OldCoder on Fri May 07, 2004 at 07:57:26 PM EST

But what is the next predicted earthquake?

--
By reading this signature, you have agreed.
Copyright © 2003 OldCoder
at the top (none / 3) (#24)
by llimllib on Fri May 07, 2004 at 08:21:12 PM EST

Earlier this year, a Russian geophysicist named Vladimir Keilis-Borok, who currently works at UCLA, shook up the scientific world by predicting that a major earthquake will hit southern California before Sept. 5, 2004.


Peace.
[ Parent ]
-1, Witchcraft (1.05 / 19) (#29)
by egg troll on Fri May 07, 2004 at 11:50:58 PM EST


He's a bondage fan, a gastronome, a sensualist
Unparalleled for sinister lasciviousness.

W-1tchcraft (none / 2) (#48)
by Karaoke God on Sat May 08, 2004 at 05:20:43 AM EST

Would've been funnier :)

[ Parent ]
i can see it now (none / 3) (#43)
by the77x42 on Sat May 08, 2004 at 03:34:15 AM EST

Arnold running around saving lives during an earthquake yelling, "DAMN YOU VLADIMIR!!!"


"We're not here to educate. We're here to point and laugh." - creature
"You have some pretty stupid ideas." - indubitable ‮

butterfly causality (none / 0) (#50)
by limivore on Sat May 08, 2004 at 08:43:49 AM EST

Causality is the cheap frozen pizza of models. Seriously unfulfilling to the connaisseur.
What do you mean "it works"? Do you mean you like those little fake pepperonies?

not really usful yet (none / 1) (#54)
by modmans2ndcoming on Sat May 08, 2004 at 10:16:02 AM EST

until they can predict with in a week or so, it will not be useful.

are people suppose to leave for months at a time?

solution (none / 0) (#56)
by martingale on Sat May 08, 2004 at 10:26:19 AM EST

Nuke'em from orbit. It's the only way to be sure.

Oh, wait, wrong problem. Disregard kthx.

[ Parent ]

Actually, it works. (none / 3) (#58)
by mcc on Sat May 08, 2004 at 01:21:33 PM EST

Your solution is really applicable for any problem, so long as you are willing to be flexible about your definition of "success".

[ Parent ]
Still helps. (none / 1) (#61)
by WhiteBandit on Sat May 08, 2004 at 03:42:05 PM EST

While it seems preposterous to expect someone to leave for months at a time, it still helps in regards to planning. Families have a chance to stock up on emergency supplies that can possibly hold them over for a few days in case a quake happens.

Frankly though, going 2 for 2 is not enough evidence to determine if this is accurate or not. It will be interesting to see what happens this summer though. Even if they are wrong, a lot will be learned and they can go back and refine their data.

Also somewhat exciting for me at least is the fact that I'll be back down in So Cal this summer, working an internship in seismology at UCLA for the Southern California Earthquake Center. I'm kinda of anxious for something to happen. :)

Last thing to note though, they are predicting a 6.4 to strike somewhere in the Mojave Desert. Depending on where it strikes, it really won't affect all that many people or cause much damage. The Hector Mines quake in 1999 was a 7.0 and while most of Southern California felt it, there was very little damage done due to its location.

[ Parent ]

The 6.4 magnitude (none / 1) (#62)
by imrdkl on Sat May 08, 2004 at 04:35:30 PM EST

Is a minimum. The threshold parameters for U() and ROC() were both exceeded, but it's unclear by how much. (and whether that means anything) The parameter for the U() functional was 6, but the actual number of qualifying tremors that occured to give a significant value, as well as the length/strength of the ROC chain is not specified. I dropped a line over to the good doctor to ask about this, but he's too busy for some wannabee from a website with a hard-to-pronounce name. Perhaps you can drag it out of him. :)

[ Parent ]
Thanks for helpful summary (2.57 / 7) (#57)
by decon recon on Sat May 08, 2004 at 12:29:51 PM EST

Thanks very much for this informative article. I had wondered what was up with this new earthquake prediction method.

Now I get the general idea: large earthquakes might be able to be predicted within months by a method analyzing the self-similarity of tremor patterns, using several types of tremor patterns.

<i>In any case we must ask, if Keilis-Borok's prediction actually is borne out, how useful is it, really, to be able to predict earthquakes months, instead of years in advance?</i>

There would be a very serious economic impact on travel plans and civil and medical planning could be more effective.  

<i>if Keilis-Borok is right, it will ... certainly help to persuade preventive efforts, and planning in the future, if nothing else. That fact alone could save millions of lives.</i>

On preventive efforts: If a prediction is made of a "> 8.0" Richter scale earthquake several months in advance with an accurate methodology, then, yes, many millions of lives could be saved through significant civil, public health, medical and travel preparations.
(On Richter scale: http://www.seismo.unr.edu/ftp/pub/louie/class/100/magnitude.html)

So, the Keilis-Borok work could be a very significant achievement in terms of public safety monitoring.

Btw, I disagree with some of the criticisms of the article: A little math wouldn't detract from the main article, but that is not needed to get the ideas across. In a future article, if the math is really new, perhaps you could try a technical afterword?

Thanks again.

anagram( 'Keilis-Borok' ) == 'risible kook' |nt (2.60 / 5) (#65)
by Wise Cracker on Sat May 08, 2004 at 07:41:26 PM EST


--
Caesars come, and Caesars go, but Newton lives forever
Chaos on the brain (none / 0) (#66)
by dollyknot on Sat May 08, 2004 at 09:34:58 PM EST

Try this for size

They call it an elephant's trunk, whereas it is in fact an elephant's nose - a nose by any other name would smell as sweetly.
Bullshit! (3.00 / 4) (#67)
by sakusha on Sun May 09, 2004 at 03:37:51 AM EST

Keilis-Borok's predictions are complete horseshit. I've read his papers, and they all refer to explanations in previous papers but none of the papers actually explain the basis for his "predictions" in detail. This is the signature of quackery, claiming your predictions are accurate but not providing any data that can be independently verified. This is not the Scientific Method, it's witchcraft.
This whole scenario is like people who claim to have found the underlying pattern behind the NYSE stock index. Sure, with only a few variables, you can create a function that almost exactly tracks the index over past history. But there are an infinite number of those functions that match past history, only one of those functions will track the future performance. As the disclaimers always say, "past performance is no predictor of future performance." The odds are infinity:1 against KB having the correct function that accurately predicts future earthquake performance.
And BTW, yes, I previously worked as a data analyst for professional seismologists. I know a quake crackpot when I see one. KB is a crackpot.

Thanks for the heads up. But... (none / 1) (#92)
by decon recon on Tue May 11, 2004 at 12:52:12 PM EST

It doesn't *logically* follow that because the detail of a method is not presented that the method is incorrect. Keilis-Borok may be protective of details for whatever reasons.

Still, your evaluation makes one wonder.

Btw, I worked as a data analyst for 6 years in the public health field. If I remember correctly, in most public health papers, methods are only outlined generally. In a paper dedicated to methodology, you would find details.

Does Keilis-Borok have paper(s) specifically dedicated to methodology?


[ Parent ]

The Butterfly Effect (none / 2) (#69)
by localroger on Sun May 09, 2004 at 12:02:48 PM EST

Somewhere in here is a fundamental misunderstanding of what the "butterfly effect" means.

It's not an analogy; it's an actual effect that can be observed in any model of a complex system that satisfies the conditions for chaos. Running the model twice with a single, very tiny change to an insignificant initial condition can cause totally different output after the system has iterated enough.

This was first noticed in simple models of the weather, and it's why people who understand what chaos theory is all about know that it will never be possible to predict the weather more than a few days in advance with any accuracy. You cannot make an accurate enough measurement of the real world's initial state to start up a model that won't diverge significantly from reality in this time frame.

Now what Chaos taketh away Chaos also giveth, and it turns out that some of these butterfly-effect prone systems are drawn toward certain highly ordered output series called "strange attractors." If you can identify a system's strange attractor (if it has one at all) then you don't need to model it to make predictions. All you have to do is figure out where you are on the strange attractor's cycle and the rest follows.

Unfortunately, self-similarity and all that don't guarantee that a system will have a strange attractor. The butterfly effect is about unpredictability of such systems. Frankly this sounds like the seismological version of Elliott Waves; yes, the data do look like that but if it had predictive value the technical analysis folks would own the entire stock market by now.

What will people of the future think of us? Will they say, as Roger Williams said of some of the Massachusetts Indians, that we were wolves with the min

Not every butterfly will cause a hurricane (none / 0) (#71)
by darkonc on Sun May 09, 2004 at 02:18:02 PM EST

Only a butterfly that is near a a couple of specific locations(and times) can shift the generation of a hurricane.

Consider: Normally me digging a hole with a pickaxe isn't going to cause a noticable seismic event. If, on the other hand, you put me near one of those huge 'balancing rocks' (Nevada, is it?) I can probably cause a seismicly noticable event (and get myself quite throroughly squashed) with a couple hours of hard work. Similarly with butterflies.
Killing a person is hard. Killing a dream is murder. : : : ($3.75 hosting)
[ Parent ]

Exactly, however... (none / 1) (#73)
by localroger on Sun May 09, 2004 at 03:21:57 PM EST

...the whole point of the butterfly analogy is that there is no way to tell which butterfly is at the sensitively dependent point. The model cannot tell you that unless it is infinitely accurate, which is impossible in practice.

What will people of the future think of us? Will they say, as Roger Williams said of some of the Massachusetts Indians, that we were wolves with the min
[ Parent ]
Apparently conflicting, but Not inconsistent (none / 3) (#72)
by darkonc on Sun May 09, 2004 at 02:36:36 PM EST

Susan Hough from USGS in Pasadena, ... claims that there have actually been three predictions issued
....."The team's current predictions have not missed any earthquake, and have had its two most recent ones come to pass.",

This could easily be taken to mean that there have been 3 published predictions of the team, with the last two having been correct. I think I remember an article that stated that the third only missed it's mark by a couple of months (i.e. wrong, but close).

In any case, 2/3 is not bad for predicting earthquakes 9 months out. It's far from perfection, but it'd be enough to cause me to check my first-aid supplies and flashlight batteries if I lived in the prediction area.

I'd also note that the team predicting an earthquake an a region which normally has a 10% chance of an earthquake in the next 9 months time is still worth noticing. If the prediction fails, it brings the team down to a 50% batting average Vs boosting them to a 75% average.

Unlike your average quacks, they're making specific predictions that go beyond what the current standared can predict -- Predictions which can be properly marked as accurate or not.

If they're really quacks, then the statistics will catch up to them after a time.
Killing a person is hard. Killing a dream is murder. : : : ($3.75 hosting)

No. (none / 0) (#75)
by sakusha on Sun May 09, 2004 at 09:27:48 PM EST

No, that's not at all what the claims mean. What it REALLY means is that of the 2 previous predictions they chose to publicise, both of them were correct. What is NOT said is how many predictions they made, it could have been dozens, but they chose not to publicise those.
These predictions are not hard to make. The upcoming California prediction is something any untrained idiot could make. The prediction claims a magnitude 6 quake will happen in one of the most seismically active areas in the world, there are mag 6 quakes there almost every year. This isn't sticking your neck out very far.

[ Parent ]
I recall an old stock market scam (2.80 / 5) (#76)
by Shren on Mon May 10, 2004 at 03:30:42 PM EST

You have a big handful of email addresses. To half of them you send predictions that stock FOO is going to go up - to the other half, you send predictions that stock FOO is going to go down. Wait.

Discard the half of the email addresses that you made the incorrect prediction to, and repeat the process with the half that you made the correct prediction to. Tell half of them that stock BAR is going to go up, and tell the other half that stock BAR is going to go down.

After several iterations, instead of having a large number of email addresses, you have a small number of email addresses. But everyone on that small list of email addresses has seen you predict the stock market, so they're probably ready to buy your thousand dollar guide to the stock market... and all you started with was a large list of email addresses.

Whenever I see someone claim to be predicting the future, I remember this story.

You start off with a thousand seismologists. Half guess that there's going to be an earthquake in Brazil and half guess that there isn't...

the same Keilis-Borok (none / 3) (#82)
by onemorechip on Tue May 11, 2004 at 02:31:46 AM EST

This is the same Keilis-Borok who developed a model for predicting the outcome of presidential elections back in the early 80s. Here are a couple of links on this: 1 2
--------------------------------------------------

I did my essay on mushrooms. It's about cats.

based on his analysis (none / 0) (#95)
by simul on Fri May 21, 2004 at 09:19:23 PM EST

based on his analysis there's little chance of Bush winning. he should probably add another variable: "did the incumbent spend billions on new electronic voting machines?"

the answer "yes" would tip the odds in the incumbent's favor.

Read this book - first 24 pages are free to browse - it rocks
[ Parent ]

This just fresh from /. (none / 0) (#96)
by xutopia on Wed Jun 16, 2004 at 02:14:23 PM EST

http://science.slashdot.org/article.pl?sid=04/06/16/1438224&mode=nested& tid=103&tid=134&tid=99

I offer this link... (none / 0) (#97)
by the on Wed Jun 16, 2004 at 08:07:49 PM EST

...just to increase the overall level of incestuousness around here. Link.

--
The Definite Article
[ Parent ]
September 13, 2004 (none / 0) (#98)
by m42gal on Mon Sep 13, 2004 at 06:51:25 PM EST

And no major earthquake hit SoCal prior to 5 September, 2004....maybe the indicators were off on the prediction?

St.

Is it common to use Spanish words? (none / 0) (#99)
by dsanfte on Wed Oct 13, 2004 at 11:24:39 AM EST

"Temblor" is spanish. Why use it instead of "Tremor", or even just "earthquake"? What specific meaning does it hold?

I live in Canada and I've never heard this word used before. I had to look it up. Everyone around here uses Tremor or Earthquake...

Chaos: The Tarot of Temblors | 99 comments (79 topical, 20 editorial, 1 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!