logo2

ugm-logo

Using Machine learning tools to gain new insights from Earthquake data

Scientists at the Columbia University have discovered a totally new way to study earthquakes. They picked out different types of earthquakes from three years using machine learning algorithms. According to them, these machine learning methods pick out very subtle differences in the raw data that we’re just learning to interpret.

Scientists particularly identified earthquake recordings at The Geysers in California, one of the world’s oldest and largest geothermal fields. They assembled a catalog of 46,000 earthquake recordings, each represented as energy waves in a seismogram. They then mapped changes in the waves’ frequency through time, which they plotted as a spectrogram—a kind of musical roadmap of the waves’ changing pitches, were they to be converted to sound.

Seismologists ordinarily dissect seismograms to evaluate a quake’s size and where it started. However, taking a gander at a seismic tremor’s recurrence data rather enabled the scientists to apply machine-learning tools that can pick out patterns in music and human speech with minimal human information. With these instruments, the scientists diminished every seismic tremor to a spectral “fingerprint” reflecting its subtle contrasts from alternate quakes, and after that utilized a clustering algorithm to sort the fingerprints into groups.

Using this machine learning algorithms, they found repeating patterns of earthquakes appear to match the seasonal rise and fall of water-injection flows into the hot rocks below, suggesting a link to the mechanical processes that cause rocks to slip or crack, triggering an earthquake. It also helped them in making a link to the fluctuating amounts of water injected below ground at The Geysers during the energy-extraction process, giving the researchers a possible explanation for why the computer clustered the signals as it did.

Felix Waldhauser, a seismologist at Lamont-Doherty said, “The work now is to examine these clusters with traditional methods and see if we can understand the physics behind them. Usually, you have a hypothesis and test it. Here you’re building a hypothesis from a pattern the machine has found.”

Scientists noted, “These methods could also help reduce the likelihood of triggering larger earthquakes — at The Geysers, and anywhere else fluid is pumped underground, including at fracking-fluid disposal sites. Finally, the tools could help identify the warning signs of a big one on its way — one of the holy grails of seismology.”

The exploration became out of a bizarre aesthetic coordinated effort. As a musician, Holtzman had for quite some time been receptive to the odd hints of quakes. With sound designer Jason Candler, Holtzman had changed over the seismic floods of chronicles of outstanding quakes into sounds, and after that speeding them up to make them understandable to the human ear. Their joint effort, with examine coauthor Douglas Repetto, turned into the basis for Seismodome, a recurring show at the American Museum of Natural History’s Hayden Planetarium that puts people inside the earth to experience the living planet.

As the exhibit evolved, Holtzman began to wonder if the human ear might have an intuitive grasp of earthquake physics. In a series of experiments, he and study coauthor Arthur Paté, then a postdoctoral researcher at Lamont-Doherty, confirmed that humans could distinguish between temblors propagating through the seafloor or more rigid continental crust and originating from a thrust or strike-slip fault.

Encouraged, and looking to expand the research, Holtzman reached out to study co-author John Paisley, an electrical engineering professor at Columbia Engineering and member of Columbia’s Data Science Institute. Holtzman wanted to know if machine-learning tools might detect something new in a gigantic dataset of earthquakes. He decided to start with data from The Geysers because of a longstanding interest in geothermal energy.

Paisley said, “It was a typical clustering problem. But with 46,000 earthquakes it was not a straightforward task.”

Thus, Paisley found a mind-blowing solution of a topic modeling algorithm that picks usual frequencies in the dataset. When applying another algorithm, they identified the most common frequency combinations in each 10-second spectrogram to calculate its unique acoustic fingerprint. Finally, a clustering algorithm, without being told how to organize the data, grouped the 46,000 fingerprints by similarity.

At the point when the specialists coordinated the groups against normal month to month water-infusion volumes crosswise over Geysers, an example hopped out: A high infusion rate in winter, as urban areas send more run-off water to the territory, was related with more quakes and one kind of flag. A low mid-year infusion rate compared to fewer tremors, and a different flag, with transitional flags in spring and fall.

Now, scientists are planning to apply these methods to recordings of other naturally occurring earthquakes.

Preparation key to surviving disaster, emergency planner says

Bryan Lee was a typical preoccupied college student, taking classes at Humboldt State University in Arcata, California, when TV news footage of the 2011 earthquake and tsunami in Japan gave him a jolt of reality.

“It hit me that I was focusing too much on normal college things and not aware of what’s going on in the community,” he said.

He had already seen how periodic landslides could cut his Northern California community off from the outside world for days at a time, leaving gaps on grocery store shelves and driving up the cost of fuel at local gas stations, and he decided he wanted to be better prepared to face the next crisis.

Lee took an emergency medical technician course and some classes in wilderness medicine, and after graduation he got a job in emergency services with the Humboldt County Sheriff’s Office.

About two months ago Lee took a similar position with the Benton County Sheriff’s Office, where he now works under emergency services program manager Kevin Higgins as the county’s emergency services planner.

Part of Lee’s job is to think about the unthinkable — how would the county respond, for instance, to a massive wildfire, a major volcanic eruption in the Central Cascades or the sort of devastating subduction-zone earthquake that scientists say is overdue in the Pacific Northwest?

If the Big One hits, modeling scenarios suggest that highways, bridges and railroads could be badly damaged, disrupting transportation networks and leaving the Corvallis area cut off from crucial supplies of food, fuel and medicine for months on end. Basic services such as electrical power, natural gas, clean drinking water and even sewer service could also be cut off for extended periods.

The county has plans in place to address these problems, but there’s no magic bullet — and the bigger the disaster, the harder it’s going to be to restore services and provide food, water and shelter to people who need them.

Most people rarely think about these kinds of things, so when Lee talks to the public about disaster preparedness, he gives them the bad news first.

“I tell people, ‘I’m from the government, and I’m not going to be able to save you,’” Lee said. “We can’t help every household.”

Then he gives them the good news: Even though it could be days or weeks before law enforcement or disaster relief agencies are able to get supplies to everyone in the county, there are plenty of things you can do to take care of yourself and your family in an emergency.

“I really try not to do gloom and doom,” Lee said. “I really try to preach that these events are totally survivable.”

The key to surviving a disaster, Lee said, is personal preparedness.

That means having at least two to four weeks’ worth of food and water on hand for every member of your household, plus a well-stocked medical kit.

In general, Lee said, it makes sense to stock up on the kinds of nonperishable foods you like to eat, but you’ll also want to think about what can give you the most nutritional bang for the buck.

“In terms of long-term storage of food,” he said, “it’s lentils and rice.”

For water, plan on stashing one gallon per person per day. It’s also a good idea to have a means of purifying water, such as a backpacker’s filtration system or iodine tablets. In a pinch, you can use two drops of unscented household bleach per quart.

But disaster preparedness doesn’t have to stop at the household level. Lee encourages people to get to know their neighbors and talk about ways to help each other in the event of an emergency.

Who has skills that could be put to use? Who is most vulnerable and in need of help? Figuring these things out in advance could make a big difference if and when disaster strikes.

You could also join the local Community Emergency Response Team, a group of trained volunteers who could be called upon to assist professional responders in a disaster. Free CERT training is available through the Sheriff’s Office.

“It’s finally getting traction,” Lee said of the CERT program. “We’ve got about 100 people trained, and more and more people are becoming interested.”

Lee recognizes that many people find the idea of preparing for a major disaster daunting. They feel like they don’t have the time, the money or the space to stockpile supplies for surviving a doomsday scenario.

He advises starting out small — for instance, by putting together an emergency kit for the car. After that, stock up gradually on food, water, medical supplies and other items.

Pretty soon, he predicts, you’ll find you have enough set aside to take care of yourself and your loved ones in an emergency — and you’ll feel better for it.

“It sounds overwhelming when you’re first starting out, but once you start preparing … it’s more of an empowering experience,” Lee said. “(You realize) ‘I could help my family, I could help my neighbors, I could help myself.’”

More Articles ...