Before the advent of artificial intelligence, experts often detected earthquakes by analyzing seismic data by hand. Unsurprisingly, this approach takes a long time, is cost-intensive, and usually identifies fewer earthquakes. Machine learning has now changed virtually everything.<\/p>\n
Using artificial intelligence, researchers have retroactively identified and assigned magnitudes to ten times as many seismic events in historical earthquake data from the Yellowstone caldera from 2008 to 2022 as had been previously registered.<\/p>\n
\u201cIf we had to do it old school with someone manually clicking through all this data looking for earthquakes, you couldn\u2019t do it. It\u2019s not scalable,\u201d Western University engineer Bing Li explained in a university statement<\/a>.<\/p>\n \u201cWe leverage leading-edge deep learning algorithms and a detailed three-dimensional velocity model to construct a 15-year high-resolution earthquake catalog of the Yellowstone caldera region,\u201d Li and colleagues explained in their study, published earlier this month in the journal Science Advances<\/a>. Thanks to this approach, Yellowstone caldera\u2019s earthquake data from 2008 to 2022 now consists of 86,276 earthquakes, all of which strengthen seismologists\u2019 understanding of volcanic and seismic systems.<\/p>\n Calderas, like Yellowstone\u2019s, are essentially ancient volcanoes that erupt and then collapse into their emptied magma chamber, leaving a large depression in the land. Many features related to volcanoes can cause<\/a> earthquakes, and most of the ones at Yellowstone<\/a> are brittle-failure events\u2014when stress in the crust causes rocks to break.<\/p>\n The study found that over half of Yellowstone\u2019s earthquakes are part of earthquake swarms<\/a>\u2014sequences of earthquakes that don\u2019t follow the mainshock-aftershock seismic pattern that most people are familiar with.<\/p>\n \u201cWhile Yellowstone and other volcanoes each have unique features, the hope is that these insights can be applied elsewhere,\u201d said Li, who is also an expert in fluid-induced earthquakes and rock mechanics. \u201cBy understanding patterns of seismicity, like earthquake swarms, we can improve safety measures, better inform the public about potential risks and even guide geothermal energy development away from danger in areas with promising heat flow.\u201d<\/p>\n According to the study, earthquake swarms beneath the Yellowstone caldera took place along more immature fault structures\u2014underdeveloped<\/a> faults that have accumulated fewer \u201cslips\u201d than mature fault structures.<\/p>\n<\/p>\n While researchers expect more aftershocks along immature faults, \u201cto a large extent, there is no systematic understanding of how one earthquake triggers another in a swarm. We can only indirectly measure space and time between events,\u201d said Li. \u201cBut now, we have a far more robust catalogue of seismic activity under the Yellowstone caldera, and we can apply statistical methods that help us quantify and find new swarms that we haven\u2019t seen before, study them, and see what we can learn from them.\u201d<\/p>\n