Current official inventories of methane emissions, a potent greenhouse gas released from landfills, livestock ranches and oil and gas facilities, may be underestimated both nationally and in California by a factor of about 1.5, according to new research from the U.S. Department of Energy’s Lawrence Berkeley National Laboratory (Berkeley Lab) and others.

A pair of new studies by Berkeley Lab scientist Marc Fischer and colleagues strongly suggest that methane emissions from oil and gas production may account for a significant portion of the underestimated emissions both in California and nationwide, and may be as much as five times greater than the current inventory estimates of EDGAR (Emissions Database for Global Atmospheric Research), the most comprehensive global methane database.

In “A multitower measurement network estimate of California’s methane emissions” published recently in the Journal of Geophysical Research (JGR), Fischer’s team combines methane measurements from five towers located throughout the Central Valley with model predictions of expected methane signals, a method known as “atmospheric inverse modeling.” They found that California’s total methane emissions are 1.3 to 1.8 times higher than the current official inventory by the California Air Resources Board (CARB).

Separately, Fischer and Berkeley Lab scientist Sebastien Biraud contributed to a study led by Steven Wofsy of Harvard University that also used an inverse modeling approach to find that total methane emissions for the continental United States are 1.5 times the current Environmental Protection Agency (EPA) inventory, and as much as 2.7 times greater for the south-central part of the country, which represents 24 percent of total U.S. emissions. “Anthropogenic emissions of methane in the US” will be published this week in the Proceedings of the National Academy of Science (PNAS).

“This is the first study to quantify methane emissions at regional scales within the continental United States with enough spatial resolution to significantly criticize the official inventories,” said Fischer, head of the California Greenhouse Gas Emissions Measurement Project (CALGEM) at Berkeley Lab. “Even if we made emissions from livestock several times higher than inventory estimates would suggest for the southwest, you still don’t get enough to cover what’s actually being observed. That’s why it looks like oil and gas are likely responsible for a large part of the remainder.”

These maps show (A) two-year averaged methane emissions found in the PNAS study, (B) methane emissions in the EDGAR inventory and (C) the difference between the two. The largest differences are in Texas and California.Images courtesy of Lawrence Berkeley National LaboratoryMethane is a short-lived greenhouse gas but which traps about 70 times more heat than carbon dioxide per unit mass when averaged over a 20-year timescale. It stays in the atmosphere about 10 years, compared to carbon dioxide, which stays in the atmosphere about 100 years.

The methodology in both studies is often called a “top-down” approach, in contrast to the “bottom-up” approach used in emissions inventories, such as by EPA and EDGAR. The latter method is essentially an accounting of all known sources of emissions that have previously been measured; it involves collecting samples or calculating the emissions from livestock ranches, waste water, agriculture, landfills, coal mines and petroleum and natural gas facilities.

In the inverse modeling approach, the scientists take atmospheric samples collected either from airplane flights or towers, then compare those highly accurate measurements with a transport model of large-scale airflows over the earth with a high spatial and temporal resolution. This allows the scientists to track where the air came from.

In the PNAS study, the California data and analysis came from Fischer’s work using a 2,000-foot tower in Walnut Grove, California. Data from the south-central U.S. came from Biraud’s work at the DOE’s Atmospheric Radiation Measurement (ARM) Climate Research Facility in Oklahoma, which has been using aircrafts to collect atmospheric samples for more than 10 years.

The study used measurements over two years from 2007 to 2008, the early years of the natural gas boom in this country. It estimated methane emissions from Texas, Oklahoma and Kansas to be 2.7 higher than the EDGAR inventory. These three states contribute about 24 percent of total man-made sources of methane emission in the U.S.

To determine the source of the emissions, the researchers correlated concentrations of propane, a tracer gas associated with oil and gas activities, to concentrations of methane. “We found there was a spatial region where those correlations were very strong in the south-central region, which would be indicative of a petrochemical source of methane,” Fischer said. “Cows don’t produce propane; oil and gas does.”

The study estimated that methane emissions due to drilling, processing and refining oil and gas could be 4.9 times larger than the EDGAR inventory and twice as large for emissions from livestock.

In the JGR study, Fischer and Berkeley Lab colleague Seongeun Jeong took measurements from a network of five towers situated throughout California’s Central Valley. In a study last year based on measurements from just one of the towers, they found that methane emissions in California were 1.5 to 1.8 times greater than previous estimates. Repeating the study with five towers has confirmed their earlier results.

“The new study gives us better spatial resolution and better discrimination of emissions between different regions of California,” Jeong said. “It looks much clearer that we were underestimating emissions from livestock, but there is some possibility that some part of the southern San Joaquin Valley may have a significant contribution from oil and gas, though that has not been verified yet.”

Currently they are working on getting better data sources from urban areas and also on a project looking specifically at the extent to which oil and gas production and distribution may be contributing to methane emissions in California.

As for how to obtain more accurate estimates of methane emissions, Fischer says the ideal solution would be to incorporate both bottom-up and top-down approaches, as each has its advantages.

“From the perspective of greenhouse forcing, what we care about is what’s in the atmosphere. I think these top-down estimates are more likely to be reflective of the sum of all emissions,” Fischer said. “We should identify gaps in both methods, and fill the gaps to harmonize things so we come up with a consensus. From that consensus we can have a plan for reducing emissions over time.”

DOE flights were supported by the DOE Office of Science.

For a full list of co-authors of the JGR paper, go here. The research was supported by the California Air Resources Board and the California Energy Commission.