Observing Buried Carbon Dioxide

A project proves that millions of tons of the sequestered gas can be safely monitored.

David Talbotarchive page

November 20, 2008

Scientists say that fighting climate change will require pumping billions of tons of carbon dioxide underground. But will it be possible to monitor such large-scale sequestration to make sure it’s not leaking? Evaluations at a remote CO₂-burial site in Saskatchewan suggest that the answer is yes.

**Flow chart:** Seismic-analysis images of a 20-square-kilometer area near Weyburn, Saskatchewan, reveal the spread of injected carbon dioxide 1.5 kilometers beneath the surface. The top image was made in 2001; the bottom image was made in 2007. The yellow areas show where the carbon dioxide has spread in a limestone layer. The images suggest that CO2 can be accurately monitored after it is injected underground.

“We have demonstrated fairly convincingly that you can monitor the CO₂ underneath the surface using seismic technologies,” says Don White, a research scientist at the Geological Survey of Canada, who presented the latest analyses of the site in Weyburn, Saskatchewan, at a conference in Washington, DC, this week. “The results have been positive so far. If we went to regulatory hearings and were asked, ‘How do you know it’s safe?’ we’d say, ‘We’ve demonstrated that it works and that we can monitor it.’”

Weyburn is one of the leading facilities in the world for studying underground CO₂ storage. Located just north of North Dakota, it consists of two old oil fields that use carbon dioxide pumped underground to increase oil production. The site also accepts carbon dioxide piped from the Great Plains Synfuels Plant.

To date, Weyburn has buried 11 million tons of CO₂, most recently at a rate of three million tons per year. To put this in perspective: an intermediate-size coal plant emits about two million tons of carbon dioxide each year–and there are about 600 coal power plants in the United States. So the annual amount that Weyburn accepts is equivalent to roughly one-quarter of one percent of the CO₂ emitted by U.S. coal power plants.

Once underground at Weyburn, the carbon dioxide sits under a bedrock formation 1.5 kilometers below the surface. There it settles in layers of porous limestone, changing the densities of these layers in ways that are visible during seismic tests–planned explosions and sensitive measurements of how vibrations propagate. The resulting images clearly show the expanding CO₂ deposits with an area of roughly 20 square kilometers. “This seems to give a pretty good representation of where CO₂ is moving in the reservoir,” White says. And this will allow engineers to monitor any changes, including any leakage, he says.

Pumping CO₂ underground is conceptually simple–oil companies have been doing it for years to force more oil to the surface. And various geologic formations are known to be capable of accepting carbon dioxide. But adapting these practices for permanent and large-scale greenhouse-gas mitigation will require long-term monitoring. A release of carbon dioxide would defeat the purpose of avoiding greenhouse-gas emissions. And if it occurred in a populated area, it could be deadly to humans and animals.

**Model map:** The Weyburn sequestration site sits within the red dot at the center of the image, in southeastern Saskatchewan, where carbon dioxide is being pumped underground for enhanced oil recovery (labeled “EOR”). Researchers are trying to model how CO2 behaves within about 10 kilometers of that spot, and they’re studying the geology of a far larger area (200 kilometers by 200 kilometers) within a regional geological depression called the Williston Basin. Vertical scale for depth is greatly exaggerated. The buried CO2 lies at a depth of 1.5 kilometers, and the depth of the study area is about 2.5 kilometers.

Larry Myer, a geophysicist at Lawrence Berkeley National Laboratory, who is developing sequestration plans in California, says that the Weyburn experience is showing the way toward broader implementation of accurate monitoring technologies. “In these early stages, one of the most important questions we have to answer is, what techniques work best under what conditions?” Myer says. “Certainly one of the key things they showed was the value of seismic technologies for mapping where the CO₂ is going. It provides tremendous confidence that we can apply this broadly for monitoring.”

White says that the Canadian agency is hoping to install a permanent seismic array and a new sensor-laden monitoring well to further improve the tracking of the CO₂, and ultimately develop a computer model that can be used by future projects around the world. “We want to understand the different [geologic] trapping mechanisms that will be active and the applicability and usefulness of different monitoring techniques, and how they should be applied over time,” he says.

Ultimately, CO₂ sequestration will require more than just knowing the fate of the carbon dioxide: it will require understanding the full range of impacts on everything from groundwater to natural-gas deposits. “No one has injected 10 million tons of CO₂ for 50 years–anywhere,” John Bradshaw, CEO of Greenhouse Gas Storage Solutions, an Australian petroleum consultancy, pointed out at the conference. “How we model that, how we regulate that–on groundwater, oil, gas, CO₂–is something we will have to work together on how to handle.”