Life Cycle CO2 Emissions: Pipeline Gas vs. LNG and the Climate Benefits of LNG Relative to Coal

Life Cycle CO2 Emissions: Pipeline Gas vs. LNG and the Climate Benefits of LNG Relative to Coal

This analysis comes mainly from a Wood Mackenzie report mainly addressing upstream CO2 emissions. Basically the conclusion is that average full life-cycle CO2 emissions for LNG are about 11-12.5 % higher than the avg. for pipeline gas. The analysis does not include so-called fugitive methane emissions from leaks but other analysis from the Pace Global/Centre for Liquefied Natural Gas paper includes total GHG analysis. The vast part of the increased emissions is fuel use – the fuel used in liquefaction being the highest, followed by the LNG burned in transport via LNG tankers. 

If one were to do a strictly upstream comparison then LNG would be even higher in emissions and as the paper states could even be comparable to Canadian tar sands projects in terms of ‘emissions intensity’ or ‘emissions per barrel of oil equivalent.’ However, this is misleading. Such analyses take into account the high energy density of crude oil. Even so, the typical LNG project would be at the low end of any tar sands project so still significantly below the average tar sands project. Pipeline gas life cycle emissions would be half of the lowest emitting tar sands project and typically less than 20% of the avg. tar sands project. The graph given could be manipulated by anti-gas interests to show (quite erroneously) that LNG emissions are comparable to tar sands emissions. In that sense I am not sure why the authors chose to compare with tar sands projects. Life-cycle emissions for avg. global LNG are still around 62% of what they would be for a modern coal plant and about 33% of what they would be for the average global coal plant so there is still a very clear climate benefit. Downstream emissions for LNG would typically be about 5% less than for pipeline gas since pipeline gas typically does not strip and vent CO2 as LNG processing does, so that the CO2 in the pipeline gas stream (global avg. of about 5%) comes out in combustion at the gas-burning plant. 

It should perhaps be noted that CO2 in pipeline gas from the main U.S. shale gas fields is quite variable but for the biggest gas play, the Marcellus Shale, is typically lower than 1% and as low as 0.1%. Haynesville Shale gas can be up to about 5% CO2 but most U.S. shale gas plays seem to be below about 2.5%. Thus, U.S. current LNG exports from Sabine Pass are likely to have significantly less CO2 than the global avg. and those from the small Cove Point LNG export terminal under construction will be very low since they will be mostly or entirely from the Marcellus. Thus CO2 from stripping and venting for LNG or venting at combustion for pipeline gas for Marcellus gas would be less than 20% of the global average. Thus the gas quality of the Marcellus in terms of CO2 content is very good and far exceeds the global average. That means Marcellus gas is about 3.2-4.5% less in life cycle CO2 emissions than typical global avg. gas which would put U.S. LNG predominantly from the Marcellus at 58% life cycle emissions of a modern coal plant rather than the 62% quoted above. The Wood Mackenzie report does not appear to have used any U.S. LNG since it is relatively new on the scene.

The report does break down the emissions into component processes. LNG burned in shipping is typically about three times gas burned for compression to move it through the pipeline. In terms of life-cycle emissions that is about 0.4% for compression and 1.2% for shipping. Shipping varies according to distance traveled. Liquefaction contributes about 6.6% to life-cycle emissions, typically burning 7-9% of the feed gas. Regasification emits only about 5% of what liquefaction emits so is negligible in comparison, far less than a tenth of a percent of life cycle emissions.

The Wood Mackenzie paper also mentions that if carbon pricing exceeds about $30 per tonne then CCS for LNG projects could become economically viable. Currently, only 2 of 64 global LNG projects utilize CCS and the reasons they do are both carbon pricing and very high CO2 content in the gas relative to other projects. The two projects are Statoil’s long-time Snohvit project in the Norwegian North Sea and Chevron’s Gorgon project in Western Australia which is set to become the largest CCS project in the world with 3.4 to 4 million tonnes of CO2 per year to be captured and sequestered. This is over five times the amount at Snohvit.

The bottom line of the Wood MacKenzie report is the same as the bottom line here: life cycle emissions from natural gas are typically less than half of modern efficient coal plants and a third of emissions from the average global coal plant which is neither efficient nor modern. The quite detailed Pace Global report available from the Centre for Liquefied Natural Gas referenced below compares total life cycle GHG emissions between coal and LNG. This includes fugitive methane estimates. The conclusion is similar or actually even more in favor of LNG relative to coal – that the highest emitting LNG scenario emits about half of the GHGs of the lowest emitting coal life cycle emissions. In some places the GHGs emitted from just mining and transporting coal are as much as the life cycle GHG emissions from LNG projects and significantly more than pipeline gas projects.

References:

Upstream Carbon Emissions: LNG vs. Pipeline Gas – by Wood Mackenzie, April 2017

Composition Variety Complicates Processing Plans for U.S. Shale Gas – by Keith Bullin and Peter Krouskop, Bryan Research and Engineering, Inc. (2008)

LNG and Coal Life Cycle Assessment of Greenhouse Gas Emissions – prepared for Center for Liquefied Natural Gas – by Pace Global, Oct. 2015