BOSTON -- Concerns have cropped up in recent years that the methane leaks from oil and gas fields could be large enough to pose a significant climate threat.
In various studies, scientists have estimated that anywhere between 1 and 10 percent of natural gas produced leaks through wellheads, transmission lines and compressor stations. Recent measurements from the Uinta Basin of Utah taken by the National Oceanic and Atmospheric Administration set leakage as high as 9 percent.
Now, preliminary results are suggesting that the highest leakage estimates are not broadly representative of the industry and should not be used in policy decisions. The natural gas industry globally leaks "significantly less" than 10 percent of methane throughout the life cycle of production, transportation and distribution, said Stefan Schwietzke, a doctoral student at Carnegie Mellon University who is working on the problem in collaboration with NOAA's Earth System Research Laboratory.
The work was presented at the American Association for the Advancement of Science meeting in Boston earlier this month.
The scientists have not yet computed exactly how much less than 10 percent of methane leaks, but based on preliminary analysis, Schwietzke is most comfortable with estimates of 6 percent or lower globally. That is over the entire lifetime of natural gas, from drilling to transportation to distribution.
Unlike scientists who measure the leakage at ground level armed with flyby planes and mobile equipment laboratories, the new research uses atmospheric inversion models on the global scale.
The method's advantage over the ground-level studies is that it is not constrained in space and time. In contrast, ground-up studies tend to be snapshots in time.
Schwietzke is using methane data from 1980 to the present collected by a global network of more than 100 air monitoring stations operated by NOAA. The stations record the total amount of methane in the atmosphere, which came to about 500 teragrams annually over this time frame.
Scientists can filter out the methane by its various sources, such as cows that emit methane during digestion, or landfills. The scientists pick out the methane that is emitted specifically by coal, oil and natural gas production and use. Using easily observed global concentrations to arrive at the root causes of the emissions is called inversion modeling.
Schwietzke and his collaborators are refining the inversion models to arrive at methane emitted specifically by the natural gas industry. Such a refined model can be used to ground-truth which of the existing emissions of leakage rates, from U.S. EPA's 2.8 percent to NOAA's 9 percent, could be true.
So far, the model has suggested that 9 percent leakage is unlikely. At about 10 percent leakage, about 170 teragrams of methane would have leaked from natural gas production in 2010. But that number is too high to be true, said Schwietzke. EPA estimates that the entire oil industry, including oil and natural gas production, emitted only 64 teragrams of methane in 2010.
In the future, the scientists will use the modeling to suggest their own estimate of how much methane could be leaking from natural gas drilling operations.
Although the study is primarily for conventional gas wells, the leakage rates are likely to be comparable for unconventional gas wells. That's because studies have found that both conventional and unconventional well drilling processes are similarly polluting (EnergyWire, Nov. 29, 2012).