Capturing carbon dioxide from smokestack emissions and pumping it deep underground may not be as useful a tool for dealing with rising greenhouse-gas levels as advocates suggest, according to a new analysis.
The reason: Rising pressure from the enormous amounts of CO2, which would have to be stored for centuries to a few thousand years, could trigger earthquakes. The temblors might do little more than rattle Grandma's china at the surface, but they still could be strong enough to crack rock above the formations used for storage, providing pathways for the buoyant CO2 to leak back into the atmosphere.
Moreover, while some underground formations are well suited for sequestration, they could represent far less storage capacity globally than required if the approach is to be a significant tool for holding down atmospheric concentrations, according to Mark Zoback, a geophysicist at Stanford University in Palo Alto, Calif., and the lead author of the analysis, which appears in this week's Proceedings of the National Academy of Sciences.
Carbon capture and storage "is generally a good idea and can be done safely in many places," Dr. Zoback says. "But we question whether it's a practical thing to do" at the scale of storing 1 billion tons of CO2 a year, which would be needed to help bring CO2 emissions down to 2000 levels by midcentury.
"The volumes that would have to be injected are so enormous ... and in many parts of the world being considered it may well be impossible because of the triggered-earthquake problem," he says.
The topic was part of a broader discussion about earthquakes and energy-related activities at a Senate Committee on Energy and Natural Resources hearing Tuesday. The hearing was tied to the release last Friday of a report on whether hydraulic fracturing – forcing fluids under high pressure into certain shale formations to crack the rock and release the natural gas – could increase the risk of earthquakes.
That report, by the National Research Council, concluded that "fracking" presented little risk of triggering quakes that could be felt at the surface. But it added that injection wells used to dispose of waste from fracking and other forms of oil and gas extraction posed a higher risk of triggering temblors than fracking itself. The study pointed out that little was known about the quake-triggering potential of carbon capture and storage.
Carbon capture and storage has long been considered a potentially potent arrow in the greenhouse-gas-control quiver. In the United States, the climate bill Congress considered but failed to pass in 2009 would have invested some $60 billion by 2025 in research and demonstration projects.
Globally, 29 large-scale sequestration projects at power plants have been undertaken during the past several years, according to a database maintained by the Carbon Capture and Sequestration Technologies program at the Massachusetts Institute of Technology in Cambridge. Of those, 10 are in the US but four have been canceled, largely due to a rocky economy and uncertainty regarding US climate policy.
If the goal is to sequester 1 billion tons a year of CO2 globally by 2050, and utilities were to aim only at the most-suitable formations, some 3,500 such sites would have to be uncovered and at spots convenient enough to be economical, Zoback and Stanford colleague Steven Gorelick found in their analysis. Some 85 sites a year would have to come on line between now and 2050 to meet that goal.
Yet by some estimates, 3.5 billion tons a year would need to be sequestered to reach emissions goals countries have been discussing internationally to curb global warming.
It might be possible to find large-scale formations that could serve an entire region, the researchers say. One such formation lies beneath the southern border of Indiana and Illinois. But the formation borders on a fault zone known for quakes that have reached magnitude 7.
If 100 million tons of CO2 were injected into the formation each year through 2050, the sequestered carbon would place a large amount of pressure on the fault zone. The 100 million ton figure is a small fraction of the amount power plants in the region emit today, the researchers estimate.
The other alternative is to try to sequester the carbon close to the plants. But during the past few decades, increasingly dense arrays of earthquake sensors have shown that earthquakes originate deep in the interiors of continents, far away from the boundaries between Earth's vast plates of crust usually associated with earthquakes and volcanoes. Such sensors also have picked up quakes from oil and gas recovery efforts. Last year, waste-water injection appears to have triggered at least three earthquakes with magnitudes ranging from 4.0 near Youngstown, Ohio; and 4.7 near Guy, Ark.; to 5.3 near Trinidad, N.M.
Many of the faults within a continent's interior are on the verge of snapping, but don't for long periods because it takes so long for strain to build up in a continent's interior. Forcing fluid CO2 into the wrong formations could unlock a fault and allow it to rupture.
The tough faults to find are not the long ones with large, obvious offsets, even though deep underground. They are the smaller faults with shorter displacements, but still capable of creating a quake powerful enough to compromise a repository.
"I agree that these risks are serious," writes Ruben Juanes, a geophysicist at the Massachusetts Institute of Technology who studies the flow of fluids through soil and rock, in an e-mail. Researchers currently have no models to forecast the effect that injecting fluids below ground could have on earthquake activity. And with no carbon-capture-and-storage project yet operating at the large scales needed to offset carbon dioxide emissions, no one has field experience on which to draw to characterize the risk.
On the other hand, he continues, the US hosts some 30,000 waste-water wells, and only eight examples exist where the use of a well triggered a moderate earthquake.
The notion that the technique could founder on the shoals of earthquake hazards is a bit premature, he suggests. Still, he says, the paper "points to the need for more research" on the issue – one that is "an important, but well-known concern."
Zoback's and Dr. Gorelick's calculations are not meant to be taken as a nail in the coffin for sequestration, Zoback emphasizes, but rather a starting point for discussion about getting a better handle on the risks potential repositories face from earthquakes as the formations receive CO2.