Tiny seismic rumbles can warn that human activities such as fracking are putting enough stress on underground faults to trigger a larger earthquake.
Analyses of a decade of seismic data from Alberta’s Western Canada Sedimentary Basin shows that 92 percent of earthquakes induced by fracking were preceded by smaller foreshocks, researchers report August 27 in Science. But there are many lingering uncertainties about why and how these foreshocks occur, the team notes. That means that, while still useful as short-term warnings, foreshocks may produce both false positive and false negative earthquake predictions.
A landmark study in 2015 was the first to conclusively demonstrate that fracking triggered dozens of earthquakes in Ohio. Since then, numerous studies have shown that human activities that involve injecting fluids into the ground — for hydraulic fracturing or fracking, enhanced geothermal energy or wastewater disposal — can trigger earthquakes. Most of the rumbles are relatively small. But some human-caused quakes are powerful enough to shake up nearby communities, such as a magnitude 5.6 quake in 2022 initiated by wastewater injection during oil extraction activities near the town of Peace River, Canada.
Regulatory agencies around the world have devised a “traffic-light protocol” to try to manage this growing seismic hazard. A green light means injection operations are full steam ahead. When seismic data tick up and suggest a potentially damaging quake might be about to occur — a yellow light — operations halt. Ideally, there’s enough warning that a shutdown would happen before the light ever turns red.
Small foreshock quakes are the primary way industrial operators at fracking or other sites know the light has flipped to yellow. But the problem is that there’s little understanding of what causes foreshocks to occur, and how often they are truly harbingers of a more dangerous mainshock, say Bei Wang, a geophysicist at Zhejiang University of Technology in Hangzhou, China, and his colleagues.
Wang and his team analyzed seismic data from western Canada from 2014 to 2024, which included about 70,000 earthquakes. Of those, the team identified 77 mainshocks of at least magnitude 3 that were linked to fracking. Then, the team hunted for foreshocks, smaller quakes that might have occurred within 5 kilometers and five days of each mainshock.
Foreshocks occurred before 71 of those fracking quakes, or 92 percent of the time. That means that most of the time there is a yellow warning light that would give operators time to hit the brakes on fluid injection, says Ryan Schultz, a seismologist at ETH Zurich who was not involved in the new study. The bad news, of course, is that that means that “8 percent of the time there’s no warning; you would skip right past the yellow light.”
Even within the 92 percent, there was a lot of inconsistency. In some cases, there was a single foreshock; in others, there might be as many as 700 tiny quakes within the foreshock window for a single mainshock.
Wang and his colleagues propose three general paths by which injecting fluids underground might lead first to foreshocks and then to a mainshock: The fluids might, for example, progressively weaken the main fracture so that it slowly slips; they might just increase strain on the region until it suddenly fails; or they might create a domino effect, producing slip on several fractures in succession.
These are ideas of earthquake-earthquake interaction that have been around since the 1990s, Schultz says. But they remain hypotheses, and different underlying geological systems can have different triggering mechanisms. One thing that will be essential to help improve understanding of these mechanisms tremendously is open access to the seismic and pumping data collected by companies during their activities, he says. That data availability isn’t guaranteed; it varies from country to country, or in the United States, from state to state.
“A lot of what this study does is raise a problem,” noting the flaws in the traffic-light protocol, particularly the possibility of jumping from green straight to red without any warning. “This study is quantifying a bit how often these jumps happen, and what sorts of physical processes are related to creating these jumps.”
Another aspect of improving the protocol is defining what constitutes an unacceptable risk, in terms of damages, Schultz says. “Then you set the red light some ways behind that. You want to hit the brakes before you’re at the wall, not as soon as you’re at the wall. There are a lot of complaints about it, but it’s the best solution we have at the moment.”
