EPA confirms high Natural Gas leakage rates

From http://theenergycollective.com/david-lewis/48209/epa-confirms-high-natural-gas-leakage-rates , article by Dr. Robert Howarth.

The latest EPA study confirms that its original "seminal" study of methane leaks from natural gas use, i.e. "Methane Emissions from the Natural Gas Industry (GRI/EPA 1996) was in error.  The GRI/EPA 1996 study was the holy grail.  The IPCC used GRI/EPA numbers when it assessed the climate impact of gas.  

The old figures drastically underestimate the climate impact of the use of natural gas. 

This confirms the work of Dr. Robert Howarth, the Cornell professor whose work I described in two recent posts (first, here, and later, here) in a most authoritative way. 

The original GRI/EPA study was regarded as so good it served as "the basis for most CH4  [methane] estimates from natural gas systems".  These include:  the main EPA Inventory of US GHG Emissions and Sinks, the American Petroleum Institute (API) Compendium, the studies done by the Interstate Natural Gas Association of America, the protocols developed by the California Climate Action Registry, and "many of the emission factors included in the Intergovernmental Panel on Climate Change Guidelines for National Greenhouse Gas Inventories." 

The EPA page that refers to the study is here.  The study in question is: "Technical Support Document:  Petroleum and Natural Gas Systems". 

Gory details follow.  (Noises of busy newsroom in background) 

Actually this isn't some news flash.  EPA posted the study on a webpage that looks like it was created in November 2010.  The document I dug out was an obscure looking link only a demented wonk would click on, late at night, near the bottom of the page.  How could something this old, stuffed here, have news in it?  Gas gets such a free ride from the usual suspects who scrutinize this type of thing something like this can be hidden in broad daylight.  I quote: 

"Although the GRI/EPA Study has been the cornerstone for estimating CH4 emissions from the natural gas industry to date, the data on which the study is based are now over a decade and a half old and in some cases (e.g., wells, compressors), not always reflective of current conditions in the United States. In recognition of the fact that existing methane emission factors were becoming quickly outdated, in 2007 EPA funded a 4-year cooperative agreement with UT Austin to support research and, as appropriate, measurement studies to update selected CH4 emission factors from the 1996 GRI study"

So it isn't even a final paper from a completed study yet.  They didn't get enough funds to completely revamp GRI/EPA 1996, even though they know its data is invalid.  They are now in the process of "evaluating the most efficient use of the remaining resources". 

But the paper used new data on methane leakage its authors believe supercede the old GRI/EPA numbers.  There are drastic changes. 

Howarth and the EPA are in the process of changing how the world views natural gas.  

On Table 1, note the Revised Emissions Factors.  One is more than 8,000 times higher than the old one.  Its the figure for new shale gas well completions :

The explanatory note under Table 2 was the clue for me.  I wanted to put the data into the basic form Howarth uses when he assesses how bad methane emissions from gas are.  I.e. what percentage of the overall production was leaked to the atmosphere as methane?

In that explanatory note I saw that for 2006, they are breaking out the data for just the gas industry that isn't separated in the Table.  So they say "the natural gas industry emitted 261 MMTCO2e of CH4 in 2006.

A  GAO report I've been studying had a conversion factor to convert MMTCO2e to billions of cubic feet of natural gas.  Total US gas production for 2006 is available from EIA in cubic feet.  The conversion figure GAO-11-34 published in 2010 page 41 is".4045 million metric tons of carbon dioxide equivalent per billion cubic feet of vented natural gas".   So I took the 261 million metric tons of CO2e, divided by .4045 and got 645 billion cubic feet.  I took this 645 Bcf and divided by the 2006 marketed production of around 20 trillion cubic feet.  This gave me:

 3.25 % of US natural gas production leaks into the atmosphere as methane gas. 

The industry has been saying the leaks are tiny, less than a fraction of 1%. 

Howarth wrote in an email to me that his figures were much higher than old EPA data: "at least 1.9% and perhap 5.6% or greater"  Howarth's to be published paper will have more detail.  Howarth was saying EPA was low and he had a number of solid studies showing why.  Now EPA is saying they agree - their own figures are much higher than old EPA data.  This EPA data is in the range of gas is worse than coal.  David Hone said on the Natural Gas Webinar recently aired by The Energy Collective that leaks in Europe are essentially "zero".  Hone called Howarth's work "unimpressive".   

This is a big can of worms. 

And, as I analyse this new EPA document I see there is potential for the number to be greater.

E.g.:  EPAs new reporting rule allows an estimated 15% of emissions to go unreported - it isn't clear whether they then jack up or intend to jack up what's reported by 15%.  "Engineering" calculations such as some the EPA still seem to rely on have proven to be in error compared to what actual measurement of what's in the air surrounding facilities shows.  And I didn't include US gas flaring CO2 because it was too late at night.  I just wanted to calculate long enough to get the drift of what EPA was saying. 

One reason I explained how I derived my 3.25% figure up there is because it seemed odd I have to calculate at all.  I thought they hired the people preparing these reports to do the calculating, so people like me won't misunderstand what they've discovered.  It was the same thing with the GAO study on venting and flaring I just studied.  If I want data in a form that is meaningful to compare the climate impact of gas to coal, I have to calculate.  Everyone involved understands that there are some who see climate as serious, even though the denial campaign has been a splendid success. If I was guessing, I'd say if an agency expresses itself too clearly about what it has discovered about US gas, the report is taken by industry to be antagonistic.)

But the fact that the EPA now estimates methane venting to be this high is the end of the fantasy world for all those who were saying gas is "green".  Its one thing to say some professor must be off his head.  This is the agency that came up with the reference data everyone was using.  EPA clearly states the reference data was bad all along.  

The gas industry can't clean up their act overnight, and the fact that it is this bad at this moment speaks volumes about what their attitude to climate science has been and is now.  They believe greenhouse gas emissions don't matter.  Its been economical for them to just fill the atmosphere with methane rather thanmake even more money than they were making already by collecting it and selling it.

What proves this is the way BP cleaned up its US operation in the San Juan Basin starting just after their CEO declared he believed climate was a problem during the runup to Kyoto.  BP and Royal Dutch/Shell, #2 and #3 biggest non state sponsored oil and gas companies in the world, split away from the Exxon-Mobil denial campaign in a break historians will see as significant.  BP's CEO, then John Browne, announced BP would no longer deny,  in 1997 in this way:  "over time we can move towards the elimination of emissions from our own operations and a substantial reduction in the emissions which come from the use of our products".  The word came down from BP's senior management starting from then on in a way that simply did not happen with Exxon-Mobil and most of the rest of the US oil and gas industry:  BP started to do something about their own emissions.  BP made money doing something about methane.  BP CEO Haywardin 2007 called on world governments to put a price on carbon, saying "although the challenges of climate change are grave, they can be solved.  Human ingenuity knows no bounds - and human kind will solve its biggest challenge to date".  

Basically, I'm more cynical than the next guy.  But it is obvious.  If US Big Oil would throw in the towel on the denial campaign, everyone would feel better and we could move on.  We could aim for stabilization.  General Accounting Office data shows BP's words were translated into action in the San Juan Basin, and the same dataset proves US industry could care less. 

If gas industry people want to know how bad this leak problem is, they can ask BP, who are in a position to say how much of this can be eliminated, how long it will take, and what it will cost. 

Clearly, the use of natural gas in the entire world up to now has been worse than coal. 

E.g.:  in 2004 the GAO accepted the US gas industry line that leaks were lower than 1%, and looked out at the mayhem going on in the rest of the world (Russia announced it was not flaring at all even as NOAA satellite data proved they were doing "significant" flaring) and in comparison estimated world leakage at 3%.  Maybe world data was actually beyond 6%.  Why is 10% out of the question?  This industry is using natural gas in areas where there is no electric grid as a power source, not by burning it, but as a substitute for compressed air.

In the US, past gas use must have been worse than coal and may even now be worse than coal.  It may add up that the historic use of gas in the US to the present has been worse for the climate than if coal had been substituted all along, even if the climate impact is viewed over 100 years.

1. EPA/GRI (1996) Methane Emissions from the Natural Gas Industry. Prepared by Harrison, M., T. Shires, J. Wessels, and R. Cowgill, eds., Radian International LLC for National Risk Management Research Laboratory, Air Pollution Prevention and Control Division, Research Triangle Park, NC. EPA-600/R-96-080a.

2 EPA (1996) Methane Emissions from the U.S. Petroleum Industry (Draft). Prepared by Radian. U.S. Environmental Protection Agency. June 1996.