Editor's Note: This is the fourth in a series of five features on carbon capture and storage, running daily from April 6 to April 10, 2009.

The Scurry Area Canyon Reef Operators Committee oil field, better known as SACROC, near Snyder, Tex., has slurped 140 million metric tons of liquid carbon dioxide (CO2) since 1972—80 million metric tons of which has stayed trapped in the reservoir. Pumping all that CO2 down has meant pumping more oil out.

For 36 years, oil services companies like Denbury Resources and Kinder Morgan have piped carbon dioxide from naturally occurring reservoirs in Colorado to the declining oil fields of the Permian Basin in West Texas.

The U.S. has at least 100 such projects like SACROC and 3,100 miles (5,000 kilometers) of CO2 pipelines. All told, companies have injected some 10.8 trillion cubic feet of the greenhouse gas since the 1970s, according to petroleum engineer R. Tim Bradley, Kinder Morgan's president of CO2, to raise the yield from oil fields by some 650,000 extra barrels a day—more than 10 percent of daily U.S. total production.

Most important with respect to carbon capture and storage (CCS), the Great Plains Synfuels Plant in North Dakota has pumped as much as two million metric tons of carbon dioxide a year to the Weyburn oil field in Saskatchewan since 2000.

"The Dakota gasification project is creating synthetic gas and taking the CO2 from that process," then pipelining it to the Weyburn oil field, observes carbon storage development coordinator Kurt Waltzer of the Boston–based environmental group the Clean Air Task Force. "In effect, you have demonstrated all the components of doing a CCS project."

In all of these projects, the CO2 basically scours more hydrocarbons out of the oil field. When injected into the oil reservoir, it mixes with the oil and mobilizes more of it—like turpentine cleaning paint—and then allows it to be pumped to the surface.

Using carbon dioxide to churn out more fossil fuels—and permanently storing some of the CO2 in the process—might sound counterproductive to limiting climate change because those fuels, when burned, put more CO2 into the atmosphere. But it does reduce overall emissions by at least 24 percent, calculates petroleum engineer Ronald Evans, Denbury's senior vice president of reservoir engineering: every recovered barrel of oil eventually puts 0.42 metric ton of CO2 into the atmosphere, but 0.52 to 0.64 metric ton are injected underground recovering it. In fact, Kinder Morgan's Bradley estimates that enhanced oil recovery in the U.S. could reduce CO2 emissions by 4 percent, if done correctly.

The great fear commonly associated with carbon sequestration is that trapped CO2 might suddenly escape to the surface with deadly consequences, as happened in 1986 at Lake Nyos in Cameroon. That volcanic lake had naturally accumulated two million metric tons of carbon dioxide in its cold depths; one night it spontaneously vented, displacing the oxygenated air, and suffocated more than 1,000 nearby villagers.

Yet in all three decades of commercial use of CO2 for EOR, there have been no dangerous leaks. CO2 from leaks and ruptured injection wells has always dispersed too quickly to pose a threat.

For example, prospectors in Utah drilling for natural gas in 1936 accidentally created a CO2 geyser. It still erupts a few times a day as pressure builds but is "so unhazardous that it's a tourist attraction, not a risk," says hydrologist Sally Benson, director of the global climate and energy project at Stanford University. In fact, air concentrations of carbon dioxide have to build up to more than 10 percent to be hazardous, which is difficult to achieve, according to modeling from Lawrence Livermore National Laboratory (LLNL).

The reason is that CO2 belching from a volcanic lake creates conditions very different from those of the gas escaping from a wellhead or seeping into a basement, explains Julio Friedmann, leader of the carbon management program at LLNL. At Lake Nyos, an abrupt release of the CO2 allowed dangerous concentrations to pool in low-lying surrounding areas. Pressurized gas escaping from a wellhead or crack simply mixes rapidly with the atmosphere, presenting no danger, much as the use of a fire extinguisher is not hazardous. In situations where atmospheric mixing is minimal, such as for a slow leak into a basement, the problem can be eliminated by simply installing a sensor and a fan, as in apartment buildings today near natural CO2 seepages in Italy and Hungary.

The greatest risk is from the wellheads themselves leaking:  one in 12,000 injection wells leak, according to LLNL. And, not unlike a vase that is glued back together, a wellhead provides the crack where a new break will most likely form, particularly if CO2 is injected too fast and too much pressure builds up deep underground.

Most wellheads, though, seem to hold up. For example, oil wells drilled in 1944 near Cranfield, Miss., are not reacting to extra pressure from injected CO2, according to geologist Susan Hovorka of the University of Texas at Austin, who is running the test. "I'd like to congratulate the roughnecks that drilled those wells," she says, "because they seem to be holding pressure just fine."

At a demonstration project in Japan, even a magnitude 6.8 earthquake didn't shake injected CO2 loose from a deep saline aquifer; the wellheads did not so much as leak. Big earthquakes might cause leakage, but in many cases, they will not, Friedmann says.

The U.S. Environmental Protection Agency (EPA), under the terms of the Safe Drinking Water Act of 1974, is currently crafting new measures to regulate wells for CO2 injection—final rules are set to be adopted by 2011 to protect groundwater sources from CO2 in the subsurface, according to Stephen Heare, director of the EPA's drinking water protection division. "There are 800,000 wells out there injecting almost everything imaginable," Heare says. "We think the technology is there and we can move ahead safely."

Nevertheless, "the first CCS project that is done badly is the last CCS project that will be done," warns Mark Brownstein, New York-based managing director of business partnerships in the climate and air program at the Environmental Defense Fund. "In this respect, it is very similar to nuclear power."