Natural Gas Setbacks – Water Contamination and Escaped Gases

Natural gas would like to be bridge fuel as we deal with oil shortages, but it keeps running into obstacles.

A big obstacle is the fact that the price is now too low, relative to what it costs to extract the natural gas. Arthur Berman has shown based on his detailed analysis of drilling data that shale gas seems to need a much higher price than today’s $4 per thousand cubic feet to be profitable. I have shown something similar, looking at aggregate drilling costs. However, if the prices go up enough to be profitable for producers, natural gas will not look like nearly as good a “deal” for the homeowner.

But today I would like to talk about two new setbacks:

  1. An EPA ruling against Range Resources regarding the contamination of two wells in Texas’ Barnett Shale.
  2. An EPA analysis that says escaping greenhouse gases are more of a problem than previously assumed, particularly for well completions and hydraulic fracturing after completion.

The latter analysis is only an interim report, but adds further fuel to the debate about how “green” natural gas is. Indicated green house gas emissions based seem to be as high as coal emissions, although this is not explicitly stated in the report. There are no doubt steps that can be taken to reduce these emissions, but if the report is correct, without change, the indication is that natural gas is not very “green”.

EPA ruling against Range Resources

Tuesday afternoon, the EPA ruled that Range Resources has contaminated two drinking wells in the Barnett Shale area in Texas. The wells were contaminated with both natural gas and the carcinogen benzene. Hydrofracturing was used near the wells, but according to the Wall Street Journal, it is not known whether it was the hydrofracturing itself that caused the contamination. The EPA reported that there is a risk to the homes of explosion or fire.

Range Resources has been ordered to take steps to protect homeowners. They have also been ordered to investigate the problem.

Hydrofracturing has been controversial, but this is the first time that the EPA has issued a finding of this type. Without such a ruling, it is easy to say that such events are highly unlikely. But a single, well-publicized counter-example is a problem, almost like an oil spill. Once that it is clear that contamination can happen, it is likely to be much more difficult to get approvals to drill, for example, in the Marcellus shale. The chance of aquifer contamination is just too high.

Added note: I talked to Arthur Berman about this, and he pointed out a couple of things that make it very unlikely that this is really the result of contamination from natural gas. For one thing, benzene is a by product of liquid petroleum, not natural gas. The area where the contamination was found was a source of oil, before natural gas was found there. There are other issues as well about the geology of the area. So it seems likely that when this is fully investigated, it will be found to have nothing to do with natural gas fracking. We probably should not jump to conclusions.

Interim EPA analysis regarding natural gas emissions

There has been considerable discussion over the past year about the possibility of natural gas leakage being greater than previously estimated. Chris Vernon talked a little about the issue of Natural Gas perhaps not being so green back in June. He indicated that based on a 1997 EPA analysis, the leak rate was 1.4% ± 0.5%. The original analysis by EPA/GRI claiming low natural gas emissions (on which most recent analyses are based) was released back in 1996.

Now, the EPA has released an undated document called Background Technical Support Document – Petroleum and Natural Gas Industry. The document is listed on this EPA website, last updated November 30. It is not a full analysis; instead, it is more of an interim report. It says:

The EPA/GRI study used the best available data and somewhat restricted knowledge of industry practices at the time to provide estimates of emissions from each source in the various segments of the natural gas industry. In addition, this study was conducted at a time when CH4 emissions were not a significant concern in the discussion about GHG emissions. Over the years, new data and increased knowledge of industry operations and practices have highlighted the fact that emissions estimates from the EPA/GRI study are outdated and potentially understated for some emissions sources. (Emphasis added.)

This sound s a lot like the 1996 study was wrong.

This analysis has some very different numbers, particularly for unconventional gas completions and workovers:

EPA comparison of prior and updated emissions factors

EPA comparison of prior and updated emissions factors

If one looks further in the “Background Technical Support Document,” one finds that the emission factor for Unconventional Well Completions is 9,175 thousand cubic feet per completion, and the emission factor for workovers involving hydraulic refracturing is assumed to be the same.

In developing these figures, the EPA indicates that it included as unconventional natural gas only shale gas and coal methane wells. It does not include tight gas wells, and because of this, is said to be an understatement.

One question that the interim analysis does not have good information on is whether natural gas emissions during unconventional natural gas completions and workovers are flared rather than vented.  EPA’s analysis assumes that they are vented, except in Wyoming, where regulations require that emissions be flared or captured. This assumption would seem to work in the direction of overstating emissions. It seems as though that at least some producers would be flaring emissions or capturing them. Furthermore, it seems that those who are not currently doing so could make a change fairly easily, if new regulations were put in place.

The EPA shows this summary of its indications, including the high emissions from new unconventional gas wells:

David Lewis at the Energy Collective calculates based on Table 2 data that natural gas emissions based on this analysis average 3.25% natural gas production. At this level, the green house gas impact of natural gas would be similar to that of coal.

Clearly more work needs to be done, to firm up precisely what the current level of natural gas emissions is. But if there is a chance that unconventional wells are venting large amounts of natural gas when hydraulic fracturing is used, regulations need to be put into place to prevent this from taking place. Table 1 would suggest that even conventional well completions could significantly reduce natural gas emissions.

The current analysis suggests that the EPA is producing some of the high global warming gas indications that other researchers have found. This is another setback that the natural gas industry is likely to need to deal with. It may be that changed rules can eliminate much of the problem, but without further analysis, one doesn’t know.

About Gail Tverberg

My name is Gail Tverberg. I am an actuary interested in finite world issues - oil depletion, natural gas depletion, water shortages, and climate change. Oil limits look very different from what most expect, with high prices leading to recession, and low prices leading to inadequate supply.
This entry was posted in Alternatives to Oil and tagged , , . Bookmark the permalink.

15 Responses to Natural Gas Setbacks – Water Contamination and Escaped Gases

  1. majorian says:

    If the US burns 23 Tcf of gas then it releases about 1220 million tons of tCO2 equivalent directly.

    http://www.epa.gov/climatechange/emissions/downloads10/US-GHG-Inventory-2010_Chapter3-Energy.pdf

    The old EPA total added an additional 201.8 million tons for a previous grand total of ~1422 mmtCO2 now increased to 1438mmtCO2 or a 1% increase which isn’t too serious.
    Coal is still much worse at 2067mmtCO2 and coal mining also releases a lot of methane
    459 mmtCO2.
    The EPA doc above says fossil fuel burning still represented 97% of the emissions from energy production(Table 3.2).

    Wasn’t just a few years ago we were talking about 300 Tcf, a 15 year US supply but USGS experts are producing estimates like 1670 Tcf. I’ll be interested in seeing the next USGS assessment.

    It appears to me that the US/Canada still has very large fossil reserves though it is deficient in its main source of energy, oil. What stands out the most is the high rate of per capita energy use, double that of Europe/Japan and probably 6 times the global average.
    Most other countries lack ANY appreciable fossil reserves.
    IMO, it is simply criminal to waste the common inheritence of humanity on a lifestyle based on infinite energy.

    • I think that this new analysis (which seems to be from November 2010) indicates quite a bit more leakage that was assumed in the EPA study you link to. It is a little confusing (intentionally?) because of the different units used. A note just below Table 2 in the new analysis says:

      After revising the U.S. GHG Inventory emissions estimates for the sources listed in Table 1, total equipment leak and vented CH4 and CO2 emissions from the petroleum and natural gas industry were 317 million metric tons of CO2 equivalent (MMTCO2e) in 2006. Of this total, the natural gas industry emitted 261 MMTCO2e of CH4 and 28.50 MMTCO2e of CO2 in 2006. Total CH4 and CO2 emissions from the petroleum industry in 2006 were 27.74 MMTCO2e and 0.29 MMTCO2e respectively.

      If the conversion factor from metric tons of carbon dioxide equivalent to cubic feet is 0.4045 (used in this GAO study), the 261 metric tons of CO2equivalent CH4 is equivalent to 645 billion cubic feet of natural gas. Compared to about 20 trillion cubic feet of production in 2006, this is a little over 3%. I am not certain how the additional 28.4 metric tons of CO2 emitted by the natural gas industry (mentioned in the quoted note) is to be treated. That would add further to the impact it would seem.

      • majorian says:

        I use .05307 tons of CO2 per 1000 cubic feet of natural gas burnt–23E12 cf x .05307mtCO2/1000 cf = 1.221 billion tons which roughly agrees with EPA Table referenced above.

        Also EPA gives .005 metric tons of CO2 per therm(100 cu.ft. of gas) so .5307 per 1000 cu.ft. is close.
        http://www.epa.gov/cleanenergy/energy-resources/refs.html

        • majorian says:

          That’s .05307 not .5307. Oops.

          • I was just looking at Table 2 more closely. The first column is the amounts used in the GHG Inventory Estimate you mention. The next column is the revised estimate for the year listed. The four items listed add up to 317.4 MMTCO2e in the last column, which corresponds to the sum given in the paragraph I referenced earlier as the total for the oil and gas industry. The sum of the first column (that is, used in the GHG Inventory Estimate) is 201.8, so presumably the difference is 115.6 MMTCO2e. Converting this to natural gas gives 286 billion cubic feet of natural gas. So this would suggest that the GHG inventory was using a higher base number of natural gas emissions than would be implied by the 1996 analysis (but not as high as the new analysis would show).

            The new analysis could still mean that the amount of natural gas that is escaping is quite high in relationship to the amount of natural gas produced. This would bring the CO2e emissions up quite a bit for the natural gas portion, even though the total for the oil and gas industry may not rise by a large percentage.

            Your conversion factors are for natural gas that has been burned. The problem is that the allegation is that the natural gas that is escaping is natural gas that hasn’t been burned–it just has been vented.

        • majorian says:

          Good catch!
          Oops again 1221+317=1538/1422 = 1.08, an 8% increase due to revised EPA figure.
          However if you’re figuring leakage backwards you need to account for the fact that
          methane has 25 times the global warming potential of CO2. If we take 317 mmt CO2 equivalent as escaped into the atmosphere
          I would take (317mmt x 1000)/(.05307 x 25)=239 billion cu. ft. of natural gas while the world produces about 106,000 billion cubic feet of natural gas so leakage is only 0.22%.
          One reason the natural gas industry leaks CO2 is because natural gas wells produce CO2 and methane.
          Algeria’s Al Sahal field is ~10% CO2 as I remember. Australia’s Gorgon field and Some North Sea gas also haves a heavy CO2 load. This CO2 has to be removed to bring the product, pipeline natural gas up to commercial specifications and meet its property heating value. Landfill gas for example is 50% CO2, 50% CH4.

      • majorian says:

        Yikes I did it again!
        239 billion US leakage / 23000 billion of US gas consumption is 1%, not 0.2%.

  2. majorian says:

    Gail,
    You seem to be up on the latest trend in climate mitigation ‘suggestions’—control methane instead of CO2 directly.
    Gavin Schmidt debunks it at Realclimate.

    http://www.realclimate.org/index.php/archives/2010/12/losing-time-not-buying-time/#more-5494

    Reducing total CO2 emissions is the key.

  3. marty schoffstall says:

    Even if they are equivalent in Carbon. There are other differences between coal and NG generation right? Living in the coal country of Pennsylvania, you have particulate matter, mercury, sulfur, and tailings as issues.

    • majorian says:

      Coal certainly is dirtier than natural gas and it produces more CO2 per btu of energy(103.63 kg per mmbtu coal versus 54.1 kg per mmbtu natural gas). The amount of solid waste(ash) from coal is 5-10% by weight or more.
      The US gets 21% of its energy from coal(21 quads) and 25% of its energy from natural gas(23 quads). The US has reserves of +5000 quads of coal(250 years) and 1670 quads of natural gas if you include shale gas(72 years).
      If you choose to move from coal to natural gas you’ll draw down that resource faster, you’ll pay more($1 per mmBtu coal versus $5 per mmBtu natural gas) and of course our infrastructure isn’t designed to produce and deliver double the gas we use now.
      40% of all energy goes for generating electricity which requires cheap fuel as thermal plants waste 2/3rds to 1/2 of their energy due to thermodynamics of heat engines. Electricity consumption is expected to grow 1% per year in the future(we do love our toys) so it is inevitable that cheap coal will be with us for the next 100 years. Where do you think the electricity for those great oil saving EVs will come from?
      Our best bet IMHO is to pay the 30% energy penalty of CCS coal and bury the CO2/sulfur/mercury in vast underground saline aquifers(the estimated geological capacity–1200 billion tons CO2 is +4 times all reserves of US fossil fuel).

      • marty schoffstall says:

        What I find interesting about electricity generation are the per state import/export numbers. On the east coast you have huge surpluses of electricity generated by West Virginia and Pennsylvania, and huge deficits in say Mass.

        While I think ultimately that Shale gas will be a flop and we’ll see an increases in price for NG, I’d be interested in a policy discussion on state based environmental laws and NIMBY.

        I realize that historically generating power where the coal made sense economically and possibly even enivornmentally (moving those lumps is not very efficient). But…. if you really believe in the cheap shale gas world, which is much more efficiently moved, I’d be interested in seeing the consumer/generation paired geographically within environmental law domains.

        Of course my greatest interest is seeing the diffusion of generation, down to individual rooftops and yards, but that is a whole other can of worms.

    • I agree. Mercury is a real problem for coal, but so are the other things.

  4. I talked to Jane Van Ryan at API. She is going to get some of the API people to look into the questions about the new ERA study. What is the situation now on venting of natural gas when fracking is done, for example?

  5. willem Post says:

    There is a lot of back and forth with numbers, etc. Confusion abounds.
    Someone, please make a spreadsheet type table with notes, conversion factors (shown how you arrive at the conversion factor), and calculations showing the old EPA numbers and the new numbers.

    • Hopefully I will get some things back from API this week too, that may illuminate what is going on as well. Maybe at that point I can either do something more, or get someone else to help me on it.

Comments are closed.