Good News: Methane Leaks are Easy to Fix. Bad News: There’s a lot of them

March 15, 2017


Purdue researchers flew an airborne chemistry laboratory over natural gas-fueled power plants and refineries to measure greenhouse gases. They found that although these facilities are much better for the environment than coal or oil-fueled plants, the actual amount of pollutants produced is as much as 42 times higher than estimates by the Environmental Protection Agency. Credit: Purdue University photo

Power plants that burn natural gas produce significantly less pollutants and greenhouse gases than coal-burning plants, according to current estimates of how much methane escapes from such power plants, as well as from oil refineries, and estimates could be off by a wide margin, a new Purdue University study finds.

For the past decade,  has been replacing coal as a fuel for electric . It’s become relatively inexpensive, and it’s much less damaging to the environment if – and it’s an important “if” – it doesn’t leak out of the system before it is burned to make power.

That’s because although burning natural gas is much cleaner than coal or oil,  (which is mostly what natural gas consists of) has the potential to be even more damaging over the short term than coal or oil if it isn’t handled properly, says Paul Shepson, Purdue’s Jonathan Amy Distinguished Professor of Analytical and Atmospheric Chemistry.

“Methane is a 34 times more  than is carbon dioxide,” he says. “It’s a better fuel all around as long as you don’t spill it. But it doesn’t take much  to ruin your whole day if you care about climate change.”

The breaking point for natural gas leakage is about 3 percent. If more than that leaks, the fuel has a bigger climate effect than burning coal.

“The good news from our study is that while emissions are greater than anticipated, natural gas-burning power plants are still cleaner, relative to burning coal” Shepson says. Shepson said this pilot study found that the amount of methane escaping from the plants was only 0.3 percent on average.

Even taking into account previous estimates of methane leakage in the supply chain of 1.7 percent, the total methane emissions are still below the 3 percent threshold, the study found.

The study also found that methane emission rates were significantly higher than two sets of estimates reported by the Environmental Protection Agency; the EPA’s Greenhouse Gas Inventory of Emissions and Sinks estimated that total methane emissions from all U.S. refineries and natural  was negligible in 2014.

However, this study estimated that annual methane emissions may actually be 11-90 times higher for refineries and 2 -120 times higher for natural gas power plants than those calculated from data provided by facility operators and reported to the EPA’s Greenhouse Gas Reporting Program, and used in the Greenhouse Gas Inventory of Emissions and Sinks.

“There is much more methane being released into the atmosphere by leaky compressors, valves, and industrial hardware,” Shepson says. “But the good news here is that you can take a specialized infrared camera around the plant to find the leaks and then patch the them with a wad of bubblegum. I’m joking about that, of course, but the point is that it’s a relatively easy thing to fix.”

The study’s paper was released today by the journal Environmental Science & Technology, which is produced by the American Chemical Society.

The study conducted in collaboration with the New York-based Environmental Defense Fund, with funding provided by the Alfred P. Sloan Foundation.

Joseph Rudek, a lead senior scientist at the Environmental Defense Fund and a co-author on the paper, says that natural gas power plants and refineries could be a significantly unaccounted-for source of methane emissions. “More measurements are needed to better understand the methane emissions from these sectors.”

Steve Hamburg, chief scientist at the Environmental Defense Fund, says that the leaking methane will especially diminish the environmental effects of using natural gas over the first few critical decades.

“There is the capacity to cost-effectively reduce  associated with use and production of natural gas, so there’s no excuse for the waste and serious long-term impacts” he says.

The study was conducted using Purdue’s flying atmospheric chemistry laboratory, the Airborne Laboratory for Atmospheric Research, or ALAR. The ALAR is a modified Beechcraft 76 Duchess that flies at a height of 6,000 to 12,000 feet (2 to 4 kilometers) collecting air samples and conducting sophisticated measurements.

“ALAR is a unique machine, and it was created by combining three of Purdue’s major strengths: atmospheric sciences, analytical chemistry, and aviation technology,” Shepson says.

Shepson says the benefit of this research is that everyone involved will be able to improve the emission factor formulas used in calculating the amount of methane entering the atmosphere based on the total emissions of the plants, not just the amount going up the smokestacks.

“But the important overall message of the study is to say while natural gas power plants appear to provide a climate benefit, it can still be easily improved'” he says.


11 Responses to “Good News: Methane Leaks are Easy to Fix. Bad News: There’s a lot of them”

  1. What is the impact when the difference between the life of half life of of methane and and carbon dioxide in the atmosphere is also taken into consideration.

  2. Tom Bates Says:

    The study got its money from those who make a living claiming the sky is falling. If the sky were not falling they would actually have to take a job flipping burgers at McDonalds because that is all they are good for except for the lawyers who are all ambulance chasers.

    Even with all that baggage they could not come up with a number higher than 3 percent.

    What this doom and gloom BS neglects to mention is how many plants were looked at and why were they chosen. I bet if you look at that, the plants are the oldest and most busy plants so the chances of a leak are higher. I bet they only looked at plants on the EPA list which self reported the most leaks. The plants have a money incentive to fix leaks so they are unlikely to simply dump methane which they are paying for out the door.

    • Glenn Martin Says:

      I bet that you don’t know what you’re talking about. Why don’t you actually read the studies and see what they did rather than offer half-baked speculation the the studies are flawed in exotic ways? Why do you think they should have come up with a number higher than three percent? They took the carbon produced from burning coal and its’ warming effect and compared it to the volume of natural gas production and its’ warming effect and did basic arithmetic to find that a three percent leak of the gas would equal the warming from burning all that coal. It’s simple math. What’s your problem?

      • Tom Bates Says:

        You people pretend to know something and completely miss the point. What I said was they cherry picked plants to monitor that were larger emitters than the average plant and than claimed all the plants emitted at similar rates. That is not science, that is voodoo.

    • Paul Magnus Says:

      yeah. I bet Tom’s a dumb ass.

  3. indy222 Says:

    That “34x” figure is almost meaningless – it depends on the consistency of the leakage over time. When first emitted, methane is 120x that of CO2, and if the leak is a consistent 1.7%, that 1.7% is always 120x more powerful, plus the effect of the older methane partially oxidized.

    And that 1.7% was an estimate… by whom? The same industry people who lied about leakages earlier, by factors of 10x or more? I’m not reassured.

    Also, no mention of the albedo-enhancing cooling effect of the smog from coal, vs nat-gas.

  4. dumboldguy Says:

    Not clear on exactly what John Earl is asking. Methane is a more potent GHG than CO2 but is shorter-lived in the atmosphere. Any leakage eventually converts to CO2 and contributes to longer-term AGW. The methane leakage problem has not been studied adequately, and the numbers are still questionable, as indy222 points out.

    Take a look here:

    The “methane leak” I worry most about is the “methane bomb” that may erupt from melting subsea clathrates and permafrost in the arctic when we hit the tipping point. Not as easy to fix as the leaks from the power plants etc.

  5. andrewfez Says:

    No downstream leakage is relatively small compared to upstream leakage.

    Read the first several pages of this review (by the author who had the first peer reviewed study on fracking emissions) for an overview of the major emissions studies up to late 2015. I put a snippet of the shale gas section below, from the middle of the paper.

    The thing about the fracking industry is that the number of wells involved make it impossible to regulate. In WV the fracking industry racks up something like 2 violations per day, not that the DEP does anything about it.

    “A paper published by Schneising et al in the fall of 2014 used satellite data to assess global and regional trends in atmospheric methane between 2003 and 2012. Methane concentrations rose dramatically in the northern hemisphere, particularly after 2008. In a detailed comparison across the USA for the time periods 2006–2008 (before there was much shale gas or shale oil development) and 2009–2011 (after shale gas and oil production began in earnest), atmospheric methane concentrations rose dramatically in many of the major shale-producing regions. By evaluating trends in drilling and hydraulic fracturing activity, Schneising et al estimated methane emission rates of 9.5% (±7%) in terms of energy content during the 2009–2011 period for the two large shale regions – the Eagle Ford in Texas and the Bakken in North Dakota – where they felt most comfortable in estimating emissions. They reported similar methane emissions for the Marcellus shale, but with much greater uncertainty in the analysis of the satellite data because of sparser spacing of wells, the mountainous terrain, and the proximity of the region to the Great Lakes. For the Bakken, shale oil production was far greater than gas production during this time period, and the methane emissions may have been more associated with the oil production. However, natural gas was the dominant form of shale energy produced in the Eagle Ford formation between 2009 and 2011, contributing 75% of all shale energy with oil contributing 25%. For the Marcellus shale, virtually all shale energy production through 2011 came from shale gas and not oil. Therefore, it seems reasonable to attribute a methane emission rate of ∼9.5% to shale gas development in the Eagle Ford and Marcellus formations.

    The satellite methane emission estimate is largely for upstream emissions and does not fully account for downstream emissions during storage and delivery of gas to customers, which may on average add another 2.5% of methane emission. The conclusion is that shale gas development during the 2009–2011 period, on a full life cycle basis including storage and delivery to consumers, may have on average emitted 12% of the methane produced. This is more than twice what we had estimated for shale gas in our 2011 analysis, but the satellite-based estimate is based on more robust data and integrates across a period of 2 years. These shale gas emissions already may have a globally observable effect on methane in the atmosphere.

    The satellite-based estimate is ∼20-fold greater than the estimate presented by Allen et al, a study that worked closely with industry to measure emissions from various component processes of shale gas development. In my 2014 review, I suggested that the study by Allen et al may represent a best-case scenario for low emissions, given that measurements were made only at sites where industry allowed. Since then, two papers published in 2015 have indicated that in fact the data in the Allen et al’s paper may be flawed. Allen et al used a high-flow analyzer that employs two independent sensors, switching between a catalytic oxidation detector when methane levels are low and a thermal conductivity detector when methane concentrations are greater. Howard et al noted that the high-flow analyzer is prone to underestimating methane fluxes when switching between detectors. A follow-up paper by Howard et al carefully evaluated the use of a high-flow analyzer by Allen et al and concluded that “the data reported by Allen et al. (2013) suggest their study was plagued by such sensor failure”, and as a result “their study appears to have systematically underestimated emissions.” The sensor failure issue may well have affected other data reported by industry to the EPA and used by the EPA in their assessment of methane emissions, leading to serious underestimation.

    Several other recent studies have estimated upstream methane emissions from shale gas and other unconventional natural gas development (ie, from tight-sand formations) using more robust and more integrated measurement techniques such as airplane flyovers, but still with highly variable results. Estimates were ∼30% greater than the satellite-derived data for one gas field, were comparable in two other cases, were only about half as much for two sets of measurements in another gas field, and were substantially less in three other cases. Peischl et al have suggested that higher emissions are associated with wet-gas fields and lower emissions with dry-gas fields. Alternatively, the variation in emissions may simply reflect variance in space and/or in time: many of these studies were quite short in duration, for example, based on measurements made during airplane flyovers of just 1–2 days. It is also important to note that these emission estimates are given as percentages of the gas production rates. The activity of the natural gas industry and rates of production in various gas fields are quite variable in time, and some of the differences in percentage emission rates may reflect this variability. For instance, Caulton et al reported high emission rates in the southwestern Pennsylvania portion of the Marcellus shale based on a June 2012 flyover, while Peischl et al reported a very low percentage of emission rate in the northeastern Pennsylvania portion of the Marcellus shale from a July 2013 flyover. Between these two flights, gas drilling activity for shale gas fell by 64% due to low prices for gas, yet shale gas production remained high based on prior drilling and hydraulic fracturing. If methane emission is more related to drilling and hydraulic fracturing activity than to production, these rapid changes in activity may explain at least part of the differences between the two estimates for Marcellus shale. I therefore conclude that the satellite data provide the most robust estimates for upstream methane emissions from shale gas operations to date.”

  6. Paul Magnus Says:

    methane emissions are almost irrelevant. Theres too much GHG’s in the atmosphere already to burn anymore.

  7. Tom Bates Says:

    You folks really should read the actual paper, The paper itself is the usual trash science. They looked at three power plants and three refineries What is not clear is why they picked those six installations. I have suspicions they were ones reporting higher leaks so they knew any finding would make their numbers higher. That is bad science , voodoo science, something which passes for most of science in the climate sphere in this day and age. A real study would have taken a sample of the plants big enough to actually say something about the subject. Thousands of plants and six picked is voodoo statistics.

Leave a Reply to Tom Bates Cancel reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: