In April 2011, Robert W. Howarth and two other professors from Cornell published a study in which they concluded that shale gas has a higher greenhouse gas ("GHG") footprint than coal.  Earlier this month, a different group of Cornell professors that included Lawrence Cathles published a study in which they conclude that Howarth's analysis is "seriously flawed" and that shale gas has a GHG footprint that is only one-third to one-half that of coal.  Now, Howarth and his original collaborators have responded with a paper that defends their original study.  So, how do Howarth and his co-authors respond to the criticisms leveled by Cathles?

One of Cathles' primary criticisms is that Howarth "significantly overestimate[]" the emissions of natural gas that occur during shale gas extraction.  Cathles asserted that a large portion of Howarth's overestimation results from his assumption that companies always vent the natural gas that accompanies water to the surface during flowback (see the January 16, 2012 post in the Oil & Gas Law Blog for a detailed discussion of what this assumption is all about and for more details regarding the Cathles article). 

Cathles states that, despite Howarth's assumption that companies always vent, the reality is that often companies do not vent during flowback.  Howarth concedes in his new paper that companies do not always vent, but he cites an EPA estimate that companies vent 85% of the time.  Howarth states that he could reduce his estimate of emissions by 15% to account for the fact that companies do not always vent, but given other uncertainties in the available data, he sees no reason to make that correction.

There is, however, a more significant problem with Howarth's assumption that companies always vent during flowback.  The EPA has published regulations that generally would prohibit venting altogether, and those regulations are scheduled to become final on April 3, 2012, a mere ten weeks from now.  Howarth's new paper acknowledges those proposed regulations, and the fact that the EPA estimates those regulations will cut emissions of natural gas during flowback by 95%. 

Howarth then states, however, that the proposed regulations will only require recovery of the natural gas when a pipeline connection is available.  Howarth's statement is true, but potentially misleading, because even when a pipeline connection is not available the regulations generally will prohibit venting.  In those circumstances, companies will be required to flare the natural gas unless doing so would present a safety risk, and the products of such flaring have a substantially lower GHG footprint than the natural gas that is flared.  Observers expect that the EPA's proposed regulations will go into effect as planned.  If that happens, Howarth's assumption that companies always vent will be a serious flaw in his analysis.

Cathles also criticized Howarth for his assumption that natural gas flows to the surface throughout flowback at the same rate that it flows after flowback is complete.  Cathles states that this assumption leads to an overestimation of gas flow because water that is present during flowback depresses the flow rate of natural gas, particularly at the beginning of flowback.  Howarth concedes that water significantly restricts the flow of natural gas in the initial portion of flowback, and even that the fluid flowing to the surface is all water at the very beginning of flowback.  He suggests that his assumption is justified because natural gas flows freely by the end of the flowback period.    

In addition, Cathles and his colleagues state that Howarth's study overestimated the amount of natural gas that leaks during storage, transmission, and distribution of the gas to market.   Howarth acknowledges that the estimates of leakage rates he used in his study are much higher than the EPA's estimates of leakage rates, but he states that he thinks the EPA's estimates are too low.  He asserts that the EPA's estimate of leakage rates are too low because the estimates are based on studies conducted at "model" facilities that he implies were younger than the facilities that often are used for natural gas storage and distribution.  

Cathles asserted that another flaw in Howarth's analysis is that he fails to account for the fact that natural gas-fired power plants are more efficient at converting heat energy to electricity than coal-fired plants.  Howarth's reply is that most natural gas is used for generating heat, rather than in generating electricity, and therefore it is appropriate to ignore the difference in efficiency between gas-fired and coal-fired power plants.  Howarth's response may be valid to the extent someone wants an estimate of the life cycle GHG footprint of shale gas when it is used for generating heat.  But a major issue that has been raised in public discussions is how the GHG footprints of natural gas and coal compare when they are used as fuels for the generation of electrical power, and for that comparison, an accurate consideration of the differences in power plant efficiencies is essential.  

Finally, Cathles and his colleagues stated that Howarth erred by using a 20-year time horizon rather than a 100-year time horizon.  This issues arises because a comparison of the relative GHG footprints of coal and shale gas requires consideration of both carbon dioxide and methane.  This requires selection of a particular time horizon because methane has a stronger GHG footprint than carbon dioxide, but methane breaks down in the atmosphere over time, whereas carbon dioxide accumulates in the atmosphere.  Howarth chose a 20-year time horizon for the main comparisons he made in his study, but Cathles and his colleagues state that a 100-year time horizon is more appropriate. 

Howarth concedes that researchers "quite commonly us[e] only the 100-year time frame."  Nevertheless, Howarth defends his use of a shorter time frame.  He states that some studies have estimated that the earth is about 18 yeas away from a "tipping point" in which rising temperatures would cause significant methane release from the melting of permafrost, which could reinforce a trend toward global warming.  Howarth states that this makes a short time horizon critical, even though the GHG footprint of methane is considerably lower when looking at longer time horizons.

In their new paper, Howarth and his colleagues state that they "stand by" their prior "analysis and conclusions," and that they believe that "most" of Cahtles' criticisms "have little merit."  Howarth's reply provides some interesting information, though some of his rebuttals are unconvincing.