Author: Dr. Thorsten Warneke

Natural gas used for heat and power generation is generally regarded as climate friendlier than oil or coal. This is definitely true when only carbon dioxide emissions during the burning process are considered; but methane emissions during the complete well-to-consumer lifecycle, could potentially offset the climate benefit of natural gas. In order to fully assess the impact of natural gas on climate, a comprehensive set of measurements is needed, which does not currently exist. Ideally, the measurements would not only quantify the overall emissions, but would also identify the strength of the individual sources. The recent article by Allen et al. (2013) focuses on methane emissions during the pre-production and production stages of hydraulically fractured shale gas wells. It makes an important contribution towards a better understanding of methane emissions from natural gas, specifically for shale gas.

Methane emissions from natural gas in the U.S. and the contribution of the study by Allen et al.

The study was published in PNAS on 16 September, 2013.

According to the US Environmental Protection Agency (EPA, 2013), in 2011 natural gas systems represented the largest anthropogenic source category for methane emissions in the United States, responsible for emissions of 6893 Gg*. This amounts to about 25 % of total anthropogenic methane emissions in the US in 2011. Over the complete well-to-consumer lifecycle, the highest methane emissions from natural gas occur during the field production stage (EPA-estimate: 2545 Gg in 2011). This field production stage is targeted by Allen et al. (2013). Allen et al. (2013) mainly advance an understanding of methane emissions from completion flowbacks, equipment leaks, and pneumatic pumps and controllers. The main findings of Allen et al. (2013) are:

1. Methane emissions from well completion flowbacks are lower than expected. This is based on methane measurements from 27 events. The results show a high variability among these events, ranging from 0.01 to 17 Mg* per event, but all measurements are significantly lower than the estimate by the EPA (2013) of 81 Mg methane per event. The potential emissions from investigated wells range from 0.2 Mg to more than 1 Gg of methane. The measurements by Allen et al. showed that the actual methane emissions from all 27 completions were 98 % less than the potential emissions. This demonstrates that good practices can achieve low emissions from this important and potentially high source category.
2. The lower emissions from well completion flowbacks are partly counterbalanced by equipment leaks. In particular, the measurement of methane emissions from pneumatic devices show 60 % higher values than estimated by EPA (2013). This highlights a source where emissions could be reduced in the future.
3. Overall the emissions measured in this study are low compared with other estimates. To compare with the EPA (2013), Allen et al. assumed that the emission factors from their work for completion flowbacks, equipment leaks, and pneumatic devices, were representative, and they calculated total annual emissions from these source categories to be 957 Gg/yr (uncertainty ±200 Gg/yr). This is slightly lower than the estimate by the EPA (2013).

Representativeness of the study and implications for the greenhouse gas balance of natural gas

The study by Allen et al. advances the understanding of methane emissions from natural gas production. It is an important contribution towards the better understanding of methane emissions from natural gas, but for an overall assessment of the greenhouse gas balance, additional studies are needed for the following reasons: 

1. The source categories addressed by Allen et al. represent potentially very strong emissions sources in the shale gas production chain. The emissions from these sources are controversial and the measurements by Allen et al. provide the most comprehensive dataset that currently exists. To ascertain the overall climate balance, natural gas methane emissions from all stages in the well-to-consumer lifecycle need to be quantified. Despite their importance, however, the source categories addressed by Allen et al. make up roughly only 17 % of the total emissions from natural gas systems in the U.S. (EPA-estimate, EPA, 2013).
2. The U.S. natural gas system encompasses hundreds of thousands of wells (EPA, 2013). It is questionable if the measurements by Allen et al. can be regarded as representative for all these sites. The companies involved in this study were aware of the measurements and most likely applied good practices. Nevertheless, the measurements definitely show that emissions from the investigated source categories do not have to be high, if good practices are applied.

*Gg = Gigagram = 1000 tonne
*Mg = Megagram = 1 tonne

Note: Correction of the conflict of interest statement

About 3 weeks after publication of the article "Measurements of methane emissions at natural gas production sites in the United States" the authors corrected their conflict of interest statement. The new statement shows links between the authors and the gas industry. The correction of the conflict of interest statement is available here.