We Have A Problem: Earthquake Prediction

graph

Nepal | 25 April 2015 | 11:56 NST

It was a typical day for the 27 million residents of Nepal – a small south Asian country nestled between China and India. Men and women went about their usual routine as they would any other day. Children ran about happily on school playgrounds while their parents earned a living in one of the country’s many industries. None of them could foresee the incredible destruction that would soon strike with no warning. The 7.8 magnitude earthquake shook the country at its core. 9,000 people died that day. How many didn’t have to?

History is riddled with earthquakes and their staggering death tolls. Because many are killed by collapsing infrastructure, even a 60 second warning could save many thousands of lives. Why can’t we do this? Or a better question – why aren’t we doing this? Meet [Micheal Doody], a Reproductive Endocrinologist with a doctorate in steel rodphysical biochemistry. While he doesn’t exactly have the background needed to pioneer a novel approach to predict earthquakes, he’s off to a good start.

He uses piezoelectric pressure sensors at the heart of the device, but they’re far from the most interesting parts. Three steel balls, each weighing four pounds, are suspended from a central vertical post. Magnets are used to balance the balls 120 degrees apart from each other. They exert a lateral force on the piezo sensors, allowing for any movement of the vertical post to be detected. An Arduino and some amplifiers are used to look at the piezo sensors.

The system is not meant to measure actual vibration data. Instead it looks at the noise floor and uses statistical analysis to see any changes in the background noise. Network several of these sensors along a fault line, and you have yourself a low cost system that could see an earthquake coming, potentially saving thousands of lives.

[Michael] has a TON of data on his project page. Though he’s obviously very skilled, he is not an EE or software guy. He could use some help with the signal analysis and other parts. If you would like to lend a hand and help make this world a better place, please get in touch with him.

He makes a great point during his narration in this video: earthquakes disproportionately affect the poor because they live and work in lower-cost structures unlikely to be outfitted to withstand earthquakes. Shoring up infrastructure is a huge and costly undertaking. Discovering early warning systems like the one [Michael] is testing here will have an immediate and wide-ranging impact at a minimum cost.

71 thoughts on “We Have A Problem: Earthquake Prediction

    1. Because there is no pattern to analyze. It’s not that no one has thought of this approach before, it’s that this approach doesn’t work. Just ask the Italian seismologists who have been imprisoned on manslaughter for not predicting a devastating earthquake in 2009 near L’Aguila based on many smaller scale earthquakes.

        1. The last article I can find says the guy who made the press releases is still in prison on over a dozen counts of manslaughter. The trial was a misuse of power and public funds either way.

    2. Seismometers are designed to measure traveling waves and detect quakes. If the goal is to characterize seismic noise and detect deviations from its normal pattern, a different design is needed and a different way of interpreting its data is needed. If you please, earthquakes measure and extract information from seismic ” sounds”, while this device measures and gets information from the noise that is otherwise a nuisance to other seismometers.

  1. Here’s a recent paper that was randomly denied for publication that points out a correlation between the oscillating Solar Polar Fields (magnetic fields coming from the north and south poles of the sun) with magnitude 8+ earthquakes. They selected 41% of days in the past 38 years near the SPFs’ peaks, troughs, and polarity switches, and found 78% of M8+ quakes fell within the windows. I am not completely convinced that the windowing techniques aren’t just cherry picking days near known quakes in the training data, but time will tell whether future quakes fit the model as well as past quakes.

    Paper and R code: http://spaceweathernews.com/spf/

    1. The problem with these sorts of papers is that they fall prey to the multiple correlation fallacy.

      One in 20 studies of anything will show a correlation P < 0.05, so one way to get published is to take many *types* of measurements and look for correlations. With N*(N-1)/2 possible pairs to compare, you're likely to see at least one correlation of significance.

      One way around this is to divide your observations into two groups and look for correlations in one half, then see if the correlation still holds in the other half. That narrows the probabilities of random correlation to 1-in-400.

      Of course, if the number of correlation pairs is much larger than 400, you're still going to get publishable results.

      It's possible to get statistical correlations from all sorts of things this way. I've even had a go at this myself: does the gravitational influence of the moon tend to cause earthquakes? It's a fun pastime.

      Note that global warming correlates with the decline in piracy over the past few hundred years.

      1. That doesn’t answer his question. Unless you expect him to infer that since the pole is vertical that you would be able to best detect changes orthogonal to the orientation of the device.

    1. The vertical component is easily calculated. When the device moves up, all of the sensors are compressed equally. Likewise when the device moves down, they all relax equally. Any horizontal movement will average out to zero with this geometry, so the average of all three sensors is only equal to the vertical component. This is subtracted from each sensor’s output to give that sensor’s horizontal component. Trigonometry allows for the calculation of 3d force vectors from these 4 values.

  2. A novel implementation but modern seismometers are extremely accurate and are arranged in this 120 degree configuration. You can get a few seconds of warning by tracking p waves depending on distance from the focus. This already done by the USGS. However it is still of limited effectiveness because the time it takes for the information to diffuse to people.

    1. Exactly. The problem is no longer a *detection* or even *prediction* problem. I live in Japan. There is a network here that can anticipate quakes as much as 30 seconds in advance, with a norm of around 15.
      Given the predicted magnitude, trains will do emergency stops within that 15 to 30 seconds, such that at worse, the train is already down to a slow speed by the time the quake hits.
      The REAL problem is warning. How do you get that message out to millions of people in 1 second? The cell phone networks here have bulk message system that can send out a warning to everyone in an area all at once. The message triggers a background task audible and vibration warning unlike any you would normally get from your phone.
      They dont bother to trigger it for small quakes, and since it is area based, can trigger close to the epicenter and not at all in areas far away.
      The thing is, this network only functions in a technically developed area. You need the sensors, the network, the carrier backbone and agreements, and finally everyone needs a phone with active service.

      At best, a single standing system like this could sound a loudspeaker alarm DURING an earthquake, to the neighborhood, as long as it had power. Does not much good.
      P wave detection requires a network of sensors and a robust communication package. While radio will solve the second issue, the sensors themselves along transmitters, solar power systems etc will require significant effort to reduce costs. Plus that data has to go somewhere, checked, assessments made, and warnings issued. All in the period of 1 or 2 seconds, every second, every day. A long way to go before we can distribute to poor countries.

  3. Lots of people with PhD’s in seismology have and continue to work on the problem. There are very basic and difficult reasons that there has been so little practical progress on warning. Sensor designs are NOT the problem.

    Relative to a 3 axis MEMS sensor this design is quite expensive. It’s a nice decoration, but no more. A P wave sensor can provide a warning for people at some distance from the epicenter. The S and surface waves which cause the most damage arrive later But near the epicenter the time difference is too small to react.

    For poor people even a $1-2 device would have a much poorer cost-benefit ratio than more food.

    1. Right. Modern detection networks use (relatively) cheap sensors. The expense is in the NETWORK, and the functionality to distribute messages quickly.
      In poor countries, mitigation is far more effective than prediction. Most deaths in earthquakes are caused by poor and ill-informed construction techniques. Spending a bit more money up front on the construction of a building will save lives later.
      Post Haiti, the push has been to ensure construction workers are trained in modern building techniques and designers have the skills to design in disaster mitigation.

    2. It’s true that MEMS sensors are sensitive enough to detect local earthquakes, but the are not sensitive enough at seismic frequencies (0.01 – 10 Hz) to detect anything at all in the seismic noise floor. Traditional mechanical sensors are not frequency insensitive at all, and this design is designed to be ultra sensitive with virtually no frequency dependence.

  4. The current consensus in the scientific community is that we cannot predict when an earthquake will happen, but we can predict where it will occur. For most places there are maps depicting the seismic hazard. Earthquake fatalities therefore have less to do with prediction, and more with poor urban planning and construction. Earthquakes cause most victims in countries too poor or too corrupt to implement proper building standards.

    The use of seismic noise, or rather very weak signals has been gaining some traction in the scientific community. It usually goes by the names microseismicity and nanoseismicity. For example, there is a lot of research going on into using seismic noise to predict landslides.

    This is an impressive build for a three axis seismometer. It would be interesting to know its sensitivity. An affordable seismometer could have many applications saving lives. Bringing down the cost of seismometers would really widen their application and our ability to monitor the subsurface.

    1. I’m guessing sensitivity isn’t what it could be. He’s mounted it on rubber feet and has it placed on top of furniture, rather than bolted directly to the house foundation or a concrete pad in the earth. Best case scenario is [Michael]’s digitized a Lehman style seismometer, but even that isn’t quite right since his is oriented vertical rather than a known, non-vertical angle.

      Lehman/garden gate style seismometers can measure ~M4 if they’re within a ~100 mi/ 160km http://www.wylie.org.uk/technology/seismographs/AW1/results.htm

      1. The rubber feet are for the purpose of excluding high frequency movements. The vibrations of the house are exactly what I want to investigate during this phase of the development of the project. A Lehman style seismometer is NOT what we need at all, because it responds to its resonant frequency primarily – we are interested in noise, which is by definition a composite of all frequencies in the pass and of the device

      2. By the way – the data I included with the project log shows the P and S wave components of a Magnitude 4.4 earthquake 350 miles away, with a signal about 20% full scale and 30 times above the noise floor of the instrument as it was then configured. I have changed the op amp chip and the sensitivity of the instrument looks like it is now about 3X greater than it was then. It might be even better when I tweak the amplifier a little bit. Again, however, this seismometer is not about detecting earthquakes. There’s nothing new in that!

        1. Mike – Now I understand. I was like everyone else here thinking just the opposite. With the recent EQ’s in Tennessee you should have some real data to analyze now?

          Region: TENNESSEE
          Geographic coordinates: 35.659N, 89.678W
          Magnitude: 3.4
          Depth: 12 km
          Universal Time (UTC): 25 Aug 2015 13:26:15
          Time near the Epicenter: 25 Aug 2015 08:26:15
          Local standard time in your area: 25 Aug 2015 08:26:15

          Location with respect to nearby cities:
          10 km (6 mi) NNW of Covington, Tennessee
          36 km (22 mi) SE of Blytheville, Arkansas
          38 km (23 mi) W of Brownsville, Tennessee
          40 km (24 mi) N of Arlington, Tennessee
          258 km (159 mi) ENE of Little Rock, Arkansas

    2. Being in the field, this is exactly correct. (more or less a textbook answer)

      Now, the main focus of most modern seismic solutions has been to allow for advanced warning through earlier detection of motion. Meaning more sensitive instruments, Mapping of hazard zones, Identifying triggering factors, … etc. Today’s microseismic equipment can give a few seconds warning to a major event, BUT most minor events are still beyond the scope of current measurement.

      I can say for a fact that the community loves innovation. So, IF you have a new sensor you could become famous.

      An nice link to view live seismic data from SMU’s TXAR array: http://www.geology.smu.edu/~dpa-www/txar/index.html

  5. This project could definitely benefit from a proper AD convertor front end (16 bits or better) and more horsepower in the form of a Raspi or Beagle bone. Then, not only could you make the device cloud connected and remotely monitor-able with a ton more storage space, most of the data binning and pre-processing would be tons easier than on an ATMEGA.

    Very neat idea, and with some more work, has the potential to be something very useful that could have some really cool benefits.

    Man made seismic noise from surroundings is one of the biggest problems encountered when trying to work with data from seismometers, and really limits the possible locations for installation and monitoring that are feasible. The approach taken here attempts to mitigate that problem with signal processing, allowing wider application of this type of monitoring equipment.

    The more data we can gather about Earthquakes, the better we will be able to predict them.

    1. Laser, detector, and a few meters of optic cable wrapped around 4 posts mounted to a base. Very low cost and simple design with high sensitivity and environmental immunity. It does lack directionality though.

      1. Is this based on a TDR (time domain reflectometer)? If so you will also need a high-speed oscilloscope to measure the time a laser pulse travels through the long optic cable. That sets the standard during quiet periods. Any future deviation indicates the structure has moved slightly and causes a reflection off of the sides which should show up as a glitch on the oscope. This method is used for high tech security fence monitoring. It can detect a mouse farting on the cable thousands of feet down range. Anyone climbing on the fence makes a big indication.

    1. Elephants are even better due to their very low-frequency hearing and vocalization. Cats can hear a wider range of frequencies then do dogs. If kitty is getting spooky and irritable it may mean a low-frequency trembler is present. Pretty much a precursor to a quake, She may hiss, scratch, and hide under bed for no reason. They do this for thunder too.

  6. I would like to share some my experiences about earthquake prediction.(a) Seismo-electromagnetic effect: about ten hours before earthquake the reception of television signals on potential epicentral area experiences audio, visual and specral disturbances. The number of disturbances goes on increasing at time approaches towards earthquake occurrence. (b) Reception of radio signals shifts ahead. If a signal is rated to be received (and transmitted) at 1000 kHz, then the same would be received in the potential epicentral area about ten hours (or less )at higher frequencies such as 1100, 1200,1300 …. 1500 kHz. (c) The Cell phone (Mobile) telephones start mal-functioning and the become non-functioning about 100 to 120 minutes before earthquake. If all mobile / cell phones become non-functional then only this could be related to earthquake. The above readings are observable before medium to large magnitude (magnitude = or > 6.0) only. These are not applicable to small magnitude earthquakes.

      1. That’s called BFS or benign fasciculation syndrome. Reduce your daily stress. Common ways to reduce stress include: exercising more, sleeping more, working less, meditation, and eliminating all forms of dietary caffeine (e.g. coffee, chocolate, cola, and certain over-the counter medications)

        Best way to survive an earthquake? Stand in your exterior doorway (lentel/portal). Notice they are the only thing left in post-quake photos..

    1. Arun Bapat – If what you say is true then it sounds like somehow an approximate 100 Khz natural emission is generated near the epicenter 10 hours before a >6.0 magnitude quake. That would cause a superheterodyne-like mixing of the radio broadcasts causing reception similar to a I.F. (intermediate frequency) stage at increments of 100 Khz (and each of it’s harmonics in multiples of 100 Khz).

      Why cell phones malfunction is a mystery. You need to describe how the cell phone acts or reacts. Maybe the radio harmonic reaches into the cell phone radio spectrum disrupting reception of the cell towers? Maybe it interferes with the internal oscillators in a cell phone? How did you observe this happening? What is your source of this information?

      Only bats and cetaceans can hear 100 Khz. But mice come very close to hearing it at about 91 Khz.Maybe this is the mystery signal some animals are tapped into? It may not just be a fixed 100 Khz. It may be a whole wide range of ultrasonic signals surrounding a fault line. If people had a ultrasonic receiver they would be introduced to a totally new world of sounds around their homes and environment only bats are privy to.

      http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6396590&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D6396590

  7. Some good old common sense is my fallback, as I’m not a seismologist or an engineer or even an architect or building contractor. Although the heady discussion here is fascinating about quake warning systems, this is what came to my mind — although it is emphatically non=-empirical and I’d be curious to know if there have been efforts made in this area:

    What about upping the basic safety standards for infrastructure world wide? I’m not saying it has to be costly, state-of-the-art quake retrofitting stuff, but at least (for example) instead of rickety shacks passing as acceptable residences, homes could be built sturdy enough so that fewer of them would collapse under duress.
    Again, using common sense, I would bet that many structures in modern, non-impoverished nations that experience quakes probably don’t have the latest/greatest quake-proof stuff. But I’d feel conifident in knowing that they had to follow basic building codes/safety standards for load-bearing walls and roof trusses and the like, and thus they probably would hold up a lot better in any turbulence dealt by Mother Nature. == Respectfully submitted from the peanut gallery of lay persons.

    1. I wish this was possible. You have to understand poor counties though. The infrastructure that is lacking isn’t brick and mortar and concrete with rebar in it. It’s the presence of governments that care about the welfare of the poor. An earthquake will kill the poor people and spare the well- off in the very same city.

  8. If men birthed i believe they’d think like this. Ur headed in the right direction. Possibly just under thinking the situation. Try this: the earth is a biological organism, just like humans. Biological systems r simple. They have senses. Sound is just one. Smell is the most accute. I’m an mmp. When i encounter an abnormality in the body i identify cause by it’s smell first. Animals raise their noses and smell just prior fleeing a dangerous situation. Sound is secondary. Still, it shifts/changes according to event. Human perception of smell n sound have been grossly dampened. There is however a vibrational equivalent which allows interpretation. We don’t use just one sense to diagnos, rather the combination of our sensory perceptions to determine event. So the question is, how can measure the smell of our environments? How can we combine this with measuring vibrational attributes of our environment? When i look at the ww map of seismic event, i see a woman giving birth. Shallow event equals stretchmarks. Deeper event muscle/bone movement. The deepest the infant itself positioning for impending birth. A womans pregnancy is first confirmed by hormones in the blood, then by sound in her belly. Both of these cause a shift in how she smells. In how her usual perfume smells when she uses it. How things taste to her. Etc… I’ve no doubt in my mind, if u were to approach ur developments with a greater understanding of bilogical creation; using what we know about the birthing phenomena, sensory integration and vibrational math, interpretation and prediction of earth movement and event would succeed. We need to smell and taste our oceans, even before we feel n see them. Just a thot.

  9. Here’s an earthquake prediction method for you:If your local Natural Gas drilling company is doing any hydraulic fracking in your neighborhood,,, you are GUARANTEED to start experiencing seismic anomalies where none have occurred before. Tennessee, Vermont, New York, Missouri, South Carolina, etc. are all experiencing S.A.’s as of TODAY! Guess why?

    1. The vast, vast, vast majority of these are actually due to injection wells. The earthquakes that are triggered or generated by fracking are of insignificant magnitude.
      There’s no need for the chicken little routine.

      1. *generally of insignificant magnitude. I’m sure somewhere out there there’s data for a larger, but still undamaging ‘quake. Hydroelectric projects are a bigger risk to induced seismicity than fracking.

      2. Yes I guess that’s what I meant by hydraulic fracking; the injection of hydraulic fluids into a known (or unknown) fault line causing a seismic event. They deny they are triggering them but the proof is in the pudding. USGS is recording 1.0 to 2.0 magnitude quakes in clusters all around fracking operations. Either their financial greed is clouding their judgement or they truly are dumb as hell.

        Not only are they triggering “small” quakes, which is probably what you meant by “insignificant”, but they are also contaminating ground water supplies with diesel fuel and other injection fluids. Another anomaly they wish you don’t discover is the mysterious fish & bird kills. Somehow they are also triggering some sort of “calthrate gun” effect in lakes and ponds which will kill fish and birds over the area. Not sure if it is a methane release or some other gas but either way a massive calthrate gun event can account for many mysterious problems with boats and airplanes going missing at sea.

        1. Injection wells are not synonymous with hydraulic fracturing.
          Injection wells are used to dispose of hazardous material and/or carbon dioxide, though I don’t believe down the same well. This act, along with geothermal wells have been linked to many quakes, including those of significant magnitude(+~M3.5).
          For the M1-2 ‘quakes’ you’re referring to, yes those are insignificant. They may not even be quakes in the traditional sense but energy from the drilling process itself (ie; the perf gun being used). Even if they are, they are no threat to anything. A million M1-2 quakes in one spot does not mean a M9 is just around the corner. These quakes aren’t even felt by anything but the nearest seismographs. It’s not until ~M4 that any structural damage is possible, below that grandma might lose her favorite tea cup but that’s the worst of it.

          As for the rest of your claims, they don’t have anything to do with causing earthquakes. Most of that can be written off to bad execution or poor adherence to best practices, illegal practices in a few cases. But in and of itself, fracking is one of the safer ways we get energy.

          1. Leithoa – granted the tiny quakes are in themselves harmless. However, recently they have been coming in swarms and clusters EVEN WHEN THERE IS NO DRILLING OCCURRING. They do cause panic. Also the more you keep working a fault line the more likely one day it may trigger a major slippage of a plate. Also all those ruptures down there are introducing what to the ground water table? Some scientists do not agree with you that fracking is safe at all. It’s expected that a stock holder would say that as they would have a vested interest in it’s future.

            Now I’m watching this SPF thing which shows promise for EQ Predictions. Very Interesting:
            http://spaceweathernews.com/wp-content/uploads/2015/08/Relationship-between-M8-earthquake-occurrences-and-the-SPF.pdf

          2. >They [micro-quakes] do cause panic.
            They shouldn’t. It’s like saying because it’s windy today there’s a hurricane/cyclone or tornado coming. Yes wind is associated with hurricanes and micro-quakes can associate with destructive quakes but there is no predictive value. You can post the video version of Plato’s link to SWN, that doesn’t make it any more valid.
            > Also the more you keep working a fault line the more likely one day it may trigger a major slippage of a plate.
            Even at enhanced geothermal and waste injection wells the largest quake triggered was 5.6. But keep in mind these wells are putting things underground not taking them out. We’re not going to trigger the next New Madrid quake by removing* petroleum products.
            >Also all those ruptures down there are introducing what to the ground water table?
            Provided the rig operators are competent and casing the well properly, not skimping on the quality of concrete or pipe: absolutely nothing. The water table in many areas is less than 100 feet below the surface, maybe a few hundred in arid regions. Natural gas plays targeted by fracking wells are +5000 feet. It would take a perfect storm of mistakes and poor subsurface for the two to meet.
            In a properly installed well there is at worst a negligible risk to the water table from the well itself. Most of the horror stories are a result of improperly handled fracking fluid disposal, not the result of installing the well. The water table is in negligible danger from the act of drilling its self.

            > [hydroelectric projects causing earthquakes]’s a new one on me. Care to site examples or evidence of that?
            I made a post, I thought, it may be locked in moderation since it was mostly just links. Suffice to say googling ‘reservoir induced seismicity’ will get you the same links. Dams in China are responsible for many +M6 quakes and even a M7.9 that killed thousands.

            *Some ‘enhanced recovery’ techniques involve injecting fluids down the well to displace hydrocarbons, this may lubricate faults in the same way as other injection wells. But even this is just a matter of the wrong technique for the wrong spot. No fault, no problem.

      3. Really? 4.6 is not so insignificant and its happening more and more in north-eastern BC, as a direct result of fracking, they have to shut down operations and get inspected when they trigger 4+.

    2. All have significant faulting in the areas. No need to point a blame where it is not due.

      The US has significant intra-plate faulting that scientists have known about for longer than there has been Natural Gas and Oil recovery in the areas.

      Take the 1895 New Madrid quake for example. No Nat Gas drilling and located in the central US. Mangitude 6.5 (officially). The unofficial reportings seem to indicate that the movement may have been greater.

      1. I’ve been monitoring USGS Earthquake monitoring on Internet for years. Places in USA where EQ NEVER happens are now occurring with more and more frequency. And these places just happen to be in proximity to Hydraulic Fracking operations. Also the frequency started when these plants came online. The coincidences are very suspicious.Scientists are pretty much in agreement that it is a reality just like Global Warming is real too.

    3. I agree. Beyond the known attrocities of fracking, it’s just common sense: When a hole is created in an environment it is nature to fill it. As we suck the innards out of the earth out, what choice does earth have but to fill them? Just sayin.

      1. That’s not the end of the story: Recently the oil industry found an Irish journalist who used to report on the Northern Irish groups against the UK. Now he’s reporting for the FRACKING industry that there is actually no problems with it, we just imagined it, no water pollution, etc. He even went to Pennsylvania to find some paid shills to lie about their water NATURALLY having methane and the water catches on fire when lit with a match! I mean come on that’s not natural. Fracking uses DIESEL fuel and that also contaminates water tables. I watched this propaganda on YouTube. The only other laughable YouTube I watched was the anti-oil Nuclear folks finding shills to say nuclear is actually safe now! I mean which is it? Are we Americans really this dumb to believe everything “they” tell us? Well I guess the answer is “yes” and they know it.

  10. EARTHQUAKE PREDICTORS: Obviously animals are tapped into something that gives them a lead time on seismic events. How can we humans tap into that same energy? Well essentially an earthquake is when two plates underground shift or move against each other. These are called fault lines. Obviously what USGS and others are recording is the massive movement sound waves that occur AFTER the quake starts and the aftershocks.

    What if the minor tectonic plates give off a very low frequency or a very high frequency both outside human perception. Cats can hear sounds upward into 64 Khz.and as low as 45 Hz. An elephant can hear 16 Hz to 12 Khz. A human can hear 64 Hz to 23 Khz depending on age. (source: http://www.lsu.edu/deafness/HearingRange.html)

    What animal reacts to earthquakes before it happens? What is it’s range of hearing? Can that be used to pinpoint what sound energy is being released prior to big quakes? It seems that rats and mice are pretty good with sounds and do react to earthquakes. Cetaceans and bats have them beat but don’t react so much to quakes due to their normal environment is not conducive to it.

    So maybe scientists can embed sensitive transducers or microphones into an active fault line like say San Andreas. Using a spectrum analyzer, look for any signs of a constant cluster of low-intensity frequencies. Record it indefinitely. When a seismic event occurs correlate it to the anomalous sound energy. If you need a fast-track experiment go to a hydraulic fracking site and do the same with any fault line near it. Also have some guinea pigs (literally) on hand in a cage with web camera to watch for anomalous behavior. Of course you will have to feed them and maintain the cages.

    If we could pinpoint what they are picking up on we might be able to predict quakes ahead of time with some accuracy.

    1. This is from USGS web site:

      Can animals predict earthquakes?
      “The earliest reference we have to unusual animal behavior prior to a significant earthquake is from Greece in 373 BC. Rats, weasels, snakes, and centipedes reportedly left their homes and headed for safety several days before a destructive earthquake. Anecdotal evidence abounds of animals, fish, birds, reptiles, and insects exhibiting strange behavior anywhere from weeks to seconds before an earthquake. However, consistent and reliable behavior prior to seismic events, and a mechanism explaining how it could work, still eludes us. Most, but not all, scientists pursuing this mystery are in China or Japan.

      Uhhhh China or Japan? Why not USA? Because hydrologists like Steve Sobieszczyk at USGS just don’t believe in it. They think it’s just coincidence. Scientists in USA need to throw a little science research at this phenomena and not worry about job security..”Still eludes us”?

      1. Why are you citing a hydrologist when you’re talking about seismology? Why would you expect him to be up to date on current research in a field other than hydrology? Or were you just poisoning the well because he’s a government scientist?

        1. Leithoa – Steven Sobieszczyk is a Hydrologist with the USGS in Portland, Oregon. He has a B.S. degree in Geology from the University of Wisconsin-Oshkosh and a M.S. degree in Geology from Portland State University. Coming out of school, Steven spent time with the USGS modeling landslide and seismic hazards in northern California before he moved to Oregon in 2002. He has been with the USGS studying water quality in Oregon streams ever since. Steven is interested in landslide hazards, sediment transport, and stream ecosystems and has published various reports on these topics. He is also deeply involved in science communication efforts and has produced a variety of videos for USGS Oregon Science Podcast, the USGS CoreCast, as well as the Oregon USGS YouTube Channel. He is a regular contributor to the USGS on Twitter, Facebook, and YouTube.

          He us the USGS resident expert on earthquakes who is willing to do social media and be the face of USGS scientists. He was just the “low hanging fruit”. There are many others at USGS but not willing to do social media and be a target for people like us.That animal earthquake sensing posting I re-posted was from him.

  11. Earthquake prediction is “solved” by looking at very long-wave EMF (google Eric Dollard earthquake), but the best you can do is about 2 days in advance. This is because earthquakes (and many other natural phenomena) are emergent phenomena that are the result of an incredibly large web of interacting objects – this applies to earthquakes at all scales; exactly the same things cause small earthquakes as large ones, it’s just that there are a lot more of the former than the latter (although there is a relationship between the frequency and size of the events). This is very well explained in book “How Nature Works” by Per Bak, which I highly recommend. Of course, we have the ability to temporarily lessen or increase the stress in the earth and hence the number of earthquakes (and potentially volcanoes) in a given area by creating dams, major earthworks, fracking and the like.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.