Predicting earthquakes is not possible. after

Jan 12th 2022

One of the most common questions asked by the USGS is whether earthquakes can be predicted. Their answer is an unconditional “no.” The relevant page on the agency’s website notes that no scientist has ever predicted a major earthquake, and they don’t know how such a prediction could be made.

But that may stop being true soon. Although after decades of failed attempts and baseless claims about earthquake prediction, certain skepticism is warranted—Paul Johnson, a geophysicist at Los Alamos National Laboratory, downplays the predictive power of what he intends to do—but nonetheless it is part of the Investigations aimed at better understanding seismology, he and his team have developed a tool that may make earthquake prediction possible.

As with many scientific investigations these days, their approach relies on artificial intelligence in the form of machine learning. This in turn uses computer programs called neural networks that rely on a simplified model of the way neural systems are thought to learn things. Machine learning has boomed in recent years, making inroads in areas ranging from speech-to-text conversion to cancer detection from computerized tomography (CT) scans. Now, it is applied to seismology.

Slithering Slithering Away

The difficulty with doing this is that neural networks need huge amounts of training data to teach them what to look for – something that earthquakes don’t provide. With rare exceptions, large earthquakes are caused by the movement of geological faults at or near the boundaries between the Earth’s tectonic plates. This tells you where to look for your data. But the earthquake cycle in most faults involves a process called slip slippage, which takes decades. First, there is a little bit of movement on the error as stress builds up, and so there are few data points that need to be fed into the machine learning software. Then there is the sudden and catastrophic slippage to release the accumulated pressure. This certainly creates a lot of data, but nothing particularly useful for prediction purposes.

So Dr. Johnson thinks you need about ten cycles of seismic data to train the system. And as a young science, seismology is a long way off. For example, the San Andreas Fault in California (pictured) causes a major earthquake every 40 years or so. But there is currently only about 20 years (in other words, half a cycle) of data detailed enough to be useful.

However, in 2017, Dr. Johnson’s team applied machine learning to a different type of seismic activity. Slow slip events, sometimes called silent earthquakes, are caused by plate movement. The difference is that while an earthquake usually ends within seconds, a slow slip event can take hours, days, or even months. From a machine learning point of view, this is much better, because such a lengthy process generates a lot of data points on which to train the neural network.

Johnson’s classroom is the Cascadia subduction zone, a tectonic feature that stretches 1,000 kilometers along the coast of North America, from Vancouver Island in Canada to northern California. It is the boundary between the Explorer, Juan de Fuca, and Gorda Plates to the west, and the North American Plate to the east. The constant motion of the last plate over the previous three generates a slow slip event every 14 months or so, and geophysicists have recorded this activity in detail since the 1990s. This means that there are a lot of full cycles of the data – and the machine learning system trained on these by Dr. Johnson was able to “predict” the past slow slide based on the seismic signals that preceded it, and “predict” when it would occur inland. A week or so from the time it actually happens.

The next test of the technology, which has yet to be implemented, will be an actual anticipation of the slow slip event. But even without that happening, Dr. Johnson’s Slow Slip project suggests that machine learning techniques are already working with seismic events, and thus could be extended to earthquakes if there was only a way to compensate for the lack of data. To provide this compensation, he and his colleagues apply a process called transfer learning. This works with a mixture of simulated and real information.

get real

“Lab earthquakes” are miniature earthquakes that are generated on a laboratory table by slowly compressing glass beads in a piston, until something suddenly gives off. This has proven to be a useful alternative to the sticky gliding motion. Dr. Johnson’s team created a numerical simulation (a computer model that captures the basic elements of a physical system) of a lab earthquake and trained their machine learning system on it, to see if it could learn to predict the path of alternate earthquakes.

The result was somewhat successful. But what really makes a difference is augmenting the trained system with additional data from actual experiences – in other words, imparting learning. Combining finely tuned simulated data with a bit of the real thing is significantly more effective in predicting a laboratory earthquake.

The next step toward earthquake prediction would be to apply the same approach to a real geological fault, in this case most likely San Andreas. The machine learning system will be trained on data from a numerical error simulation, plus half the available live data cycle. Dr Johnson’s team will see if this is enough to stop events not included in the training data. He mentions the 6.0-magnitude Parkfield earthquake in 2004—a slide in San Andreas that caused minimal damage, but is well studied—as a potential target.

For now, Dr. Johnson’s aspirations are limited to anticipating the timing of an impending earthquake. The full forecast will also need to include where and how much the error will occur. However, if the timing could indeed be predicted, it would certainly motivate efforts to predict these other parameters as well.

He hopes to have preliminary results in the next three to six months, but cautions that it may take longer than that. If these results are truly promising, there’s no doubt that other teams around the world are trying to do the same, using historical data from other seismic faults in order to validate the technology. This, in turn, should improve the basic model.

If all is in vain, nothing will be lost, because Dr. Johnson’s work will certainly provide a better understanding of the physics of large earthquakes, which is valuable in itself. But, if it wasn’t for nothing and instead created software capable of predicting when major earthquakes would occur, it would be an earth-shaking discovery.

What Are The Main Benefits Of Comparing Car Insurance Quotes Online

LOS ANGELES, CA / ACCESSWIRE / June 24, 2020, / Compare-autoinsurance.Org has launched a new blog post that presents the main benefits of comparing multiple car insurance quotes. For more info and free online quotes, please visit https://compare-autoinsurance.Org/the-advantages-of-comparing-prices-with-car-insurance-quotes-online/ The modern society has numerous technological advantages. One important advantage is the speed at which information is sent and received. With the help of the internet, the shopping habits of many persons have drastically changed. The car insurance industry hasn't remained untouched by these changes. On the internet, drivers can compare insurance prices and find out which sellers have the best offers. View photos The advantages of comparing online car insurance quotes are the following: Online quotes can be obtained from anywhere and at any time. Unlike physical insurance agencies, websites don't have a specific schedule and they are available at any time. Drivers that have busy working schedules, can compare quotes from anywhere and at any time, even at midnight. Multiple choices. Almost all insurance providers, no matter if they are well-known brands or just local insurers, have an online presence. Online quotes will allow policyholders the chance to discover multiple insurance companies and check their prices. Drivers are no longer required to get quotes from just a few known insurance companies. Also, local and regional insurers can provide lower insurance rates for the same services. Accurate insurance estimates. Online quotes can only be accurate if the customers provide accurate and real info about their car models and driving history. Lying about past driving incidents can make the price estimates to be lower, but when dealing with an insurance company lying to them is useless. Usually, insurance companies will do research about a potential customer before granting him coverage. Online quotes can be sorted easily. Although drivers are recommended to not choose a policy just based on its price, drivers can easily sort quotes by insurance price. Using brokerage websites will allow drivers to get quotes from multiple insurers, thus making the comparison faster and easier. For additional info, money-saving tips, and free car insurance quotes, visit https://compare-autoinsurance.Org/ Compare-autoinsurance.Org is an online provider of life, home, health, and auto insurance quotes. This website is unique because it does not simply stick to one kind of insurance provider, but brings the clients the best deals from many different online insurance carriers. In this way, clients have access to offers from multiple carriers all in one place: this website. On this site, customers have access to quotes for insurance plans from various agencies, such as local or nationwide agencies, brand names insurance companies, etc. "Online quotes can easily help drivers obtain better car insurance deals. All they have to do is to complete an online form with accurate and real info, then compare prices", said Russell Rabichev, Marketing Director of Internet Marketing Company. CONTACT: Company Name: Internet Marketing CompanyPerson for contact Name: Gurgu CPhone Number: (818) 359-3898Email: [email protected]: https://compare-autoinsurance.Org/ SOURCE: Compare-autoinsurance.Org View source version on accesswire.Com:https://www.Accesswire.Com/595055/What-Are-The-Main-Benefits-Of-Comparing-Car-Insurance-Quotes-Online View photos