From cow farts to factory emissions, there are a lot of ways to add methane to the atmosphere. Since the Industrial Revolution, the concentration of this potent greenhouse gas has risen rapidly and steadily, climbing from 700 parts per billion (ppb) in 1750 to more than 1800 ppb in 2015. But from 1999 to 2006, that increase temporarily leveled out, mystifying scientists. Now, a new study identifies the likeliest culprit behind the plateau—and singles out what may have kick-started the latest methane jump.
Scientists had a lot of suspects to choose from. Natural sources of methane include wetlands and methane hydrates (methane trapped in ice and buried deep under ocean sediments), whereas human sources range from fossil fuel emissions to the burning of crops and trees to the cow and sheep "emissions" that are a byproduct of large-scale livestock farming. And then there are the sinks—the processes that remove methane. The largest methane sink is the atmosphere itself, where a series of chemical reactions converts the gas into carbon monoxide, carbon dioxide, and water. But which of the processes was to blame for the plateau?
“People were thinking in terms of a temporary suppression of sources,” says Heinrich Schaefer, an atmospheric scientist at New Zealand’s National Institute of Water and Atmospheric Research in Wellington, and the lead author of the new study. “They could point to different things that may have contributed, but none was expected to be permanent.”
To find out what happened, Schaefer and his New Zealand-based team joined forces with researchers from the University of Colorado, Boulder, Institute of Arctic and Alpine Research and Heidelberg University in Germany. To get a global look at methane concentrations before, during, and after the plateau, the team amassed atmospheric methane concentration data from measuring stations from Canada to China to Australia, spanning a period from 1984 through 2015. They also examined previously published methane data from Antarctic ice cores extending back 2000 years to the near present.
From there, they began to construct a model, using the yearly concentration changes to calculate changing emissions. The data also include carbon isotope values for the methane concentrations. Carbon isotopes, atoms of carbon that have different masses, are particularly helpful for identifying methane sources: Different sources have different relative amounts of carbon’s two nonradiogenic isotopes, carbon-13 and carbon-12. Processes like photosynthesis or microbial oxidation serve to “fractionate” the isotopes, increasing the proportion of carbon-12, which then gets translated to the emitted gas. As a result, methane emissions have distinct isotopic values: Methane emitted from any microbially driven source such as wetlands or agriculture have values of about -60‰ (signifying a relatively low ratio of carbon-13 to carbon-12); oil, gas, and coal emissions have an average carbon isotopic value of -37‰; and tree and crop burning averages about -22‰.
Once they had their data, the scientists looked at what might have been behind the plateau. They found a sharp dip in methane concentrations after 1992; that dip corresponded with a decrease in a source with a carbon isotopic value of about -40‰. “That squarely fits the fossil fuel signature,” Schaefer says. The data don’t themselves prove what led to such a dramatic decrease in emissions, but Schaefer’s team had a guess: the collapse of fossil fuel production in the Soviet Union following its 1991 breakup.
So why did methane emissions start to climb again around 2006? Once again, the team ran models to test various inputs and see how they matched global station measurements. This time, the dominant carbon isotopic values in the new inputs were about -60‰, pointing to a microbially driven source rather than fossil fuel inputs. Given the size of the source, the likely culprit was either an increase in wetland emissions or in agricultural production. To figure out which one was ultimately responsible, Schaefer and his team turned to satellite data, which revealed that the largest post-2006 increases in atmospheric methane were occurring in China, India, and Southeast Asia.
That helped narrow down the sources, Schaefer says, because different types of wetlands have different isotopic signatures. While permafrost thawing or boreal wetlands in high latitudes have values of about -60‰, tropical wetlands—such as would be found in those regions—have slightly less negative values, about -52‰. But most tropical wetlands are in the southern hemisphere—not the region identified by the satellite images. That strongly implicated agriculture as the driver for the latest methane increases, the team reports online today in Science. Schaefer says that both rice farming and livestock have likely contributed—although ruminants like cows and sheep overall contribute three times the amount of methane to the atmosphere.
That agriculture, rather than fossil fuels, is driving methane poses a new set of problems for governments trying to fight climate change, Schaefer says. “They have to weigh mitigation of climate change against food security,” he says, which means exploring technical solutions that can optimize food production and minimize greenhouse gas emissions. One example: Recent research has shown that changing the flooding practices in rice terraces can reduce emissions while keeping harvests steady.
This paper “is timely and an important step forward in understanding changes in the global methane budget,” says Isobel Simpson, an atmospheric chemist at the University of California, Irvine, who was not involved in the study. She notes, however, that other recent work on ethane emissions—which can co-occur with methane—suggests a considerable contribution from fossil fuel sources to the recent methane increase. That research suggests fossil fuels are behind at least 28% of that increase—so, she adds, more work is needed to reduce uncertainties and reconcile the ethane-based and isotope-based conclusions.
Schaefer agrees that this is an open question, noting that in the United States there has been an increase in methane leakage from gas facilities, which also leak ethane, he says. The magnitude of those emissions is among “the next questions we’ll have to look at.”