See the Tabbed Pages for links to video tutorials, and a linked list of post titles grouped by topic.

This blog is expressly directed to readers who do not have strong training or backgrounds in science, with the intent of helping them grasp the underpinnings of this important issue. I'm going to present an ongoing series of posts that will develop various aspects of the science of global warming, its causes and possible methods for minimizing its advance and overcoming at least partially its detrimental effects.

Each post will begin with a capsule summary. It will then proceed with captioned sections to amplify and justify the statements and conclusions of the summary. I'll present images and tables where helpful to develop a point, since "a picture is worth a thousand words".

Sunday, August 17, 2014

A New Analysis of the Keystone XL Pipeline: Global Emissions Would Increase

Summary.  Last year President Obama stated that approval of the Keystone XL pipeline, which would transport as much as 830,000 barrels of tar sands oil per day from Canada to the Gulf Coast, requires a finding that the pipeline be in the national interest.  As part of the approval process the U. S. Department of State released an environmental impact statement earlier this year.  Its findings and arguments suggest that the pipeline would not have a significant effect on the environment, including greenhouse gas emissions. 

In a new journal article, however, Erickson and Lazarus conduct a comprehensive lifecycle analysis of the effects of transporting the additional oil, pointing out shortcomings in the Department of State’s environmental analysis.  They find that when operating at full capacity, the XL pipeline would increase global oil consumption by 490,000 barrels per day as of 2020, leading to new greenhouse gas emissions that are four times higher than the environmental impact statement estimated, throughout the 50 year lifetime of the pipeline.

In an era when the world should be making every effort to reduce greenhouse gas emissions, the findings of Erickson and Lazarus clearly show that approval of the XL pipeline would increase emissions, and so would not be in the national interest.  President Obama should not approve its construction.

Introduction.  President Obama is weighing whether to approve construction of the proposed Keystone XL pipeline (XL) segment from the Canadian border to an existing pipeline terminal in Oklahoma.  At full capacity it is intended to carry 830,000 barrels of oil per day (bpd) of “heavy crude” oil, also called “tar sands” oil, from Alberta to refineries in the U. S. along the Gulf of Mexico.  The Department of State (DOS) is evaluating the application, since it involves international policy, to determine whether the project serves the national interest.  DOS weighs “a wide range of factors, including … environmental… and economic impacts; [and] foreign policy”, among others.  As part of its review  DOS issued its Final SupplementalEnvironmental Impact Statement (Final SEIS) concerning the application In January 2014.

The Final SEIS, as part of its lifecycle analysis of greenhouse gas (GHG) emissions, estimates that transporting and consuming 830,000 bpd would lead to emitting between 147 and 168 million metric tons of carbon dioxide equivalents (MTCO2e) per year throughout the operational lifetime of the pipeline, i.e., over several decades.  The analysis finds this is 1-27 MTCO2e higher than would be obtained by consuming a comparable amount of a reference crude oil such as Mexican Maya.

Erickson and Lazarus find that the Final SEIS radically underestimates the lifecycle emissions from operation of XL.  Peter Erickson and Michael Lazarus published their article, “Impact of the Keystone XL pipeline on global oil markets and greenhouse gas emissions”, in Nature Climate Change on Aug. 10, 2014.  The authors conducted a comprehensive economic analysis of the principal contributions to lifecycle GHG emissions (see Details at the end of this post).  It includes important considerations that they state were not apparent in the Final SEIS.  An important feature of their analysis evaluates  the decrease in the global price of oil as a result of adding the tar sands oil to world supply.  This would lead to higher consumption: “for every barrel of increased production, global oil consumption would increase 0.6 barrels”, corresponding to an increase in consumption by 490,000 bpd.  This in turn would lead to an increase in emissions, when the oil is burned, of as much as 110 MTCO2e annually, four times higher than the upper estimate presented by DOS in the Final SEIS (see Details; these upper estimates assume operation of XL at full capacity).  Erickson and Lazarus ascribe this serious discrepancy to inadequate evaluation in the Final SEIS of the economics of the global oil market.


In President Obama’s address at Georgetown University on climate change delivered on June 25, 2013, he stated that approval of the XL application required a finding that the pipeline was in the “nation’s interest.  And our national interest will be served only if this project does not significantly exacerbate the problem of carbon pollution.  [XL’s] effects … on our climate will be absolutely critical to determining whether” to approve the project.

Erickson and Lazarus have made a significant contribution to the task of assessing whether approval of XL would be in the nation’s interest.  They have pointed out two errors in the DOS Final SEIS which effectively show that the nation’s interest would not be met.

First, whereas the Final SEIS finds that the incremental increase in annual emissions rate would be at most 27 MTCO2e compared to a reference crude, the economic analysis presented by Erickson and Lazarus concludes that emissions would be about four-fold higher than the DOS estimate, when XL operates at its full capacity.  Second, the Final SEIS stated that production of tar sands oil would not be affected whether or not XL would be approved because alternative means of transporting the oil to the Gulf Coast would be used.  (This writer has commented on the moral implications of this argument here.)  In contrast, Erickson and Lazarus show by widely recognized economic methods that shipment of tar sands oil through XL would require expanded production, and that global consumption of oil would increase by 490,000 bpd when XL is operated at full capacity.

The Intergovernmental Panel on Climate Change has set forth the urgent imperative that the nations of the world have to agree on meaningful reductions of GHG emissions very soon.  Increased levels of GHGs in the atmosphere result directly in higher long-term global average temperatures, bringing physical damage, and ecological and societal harms with them.  Any project such as the XL pipeline contributes to increased GHG emissions throughout the operating lifetime of the project, which the present authors estimate at 50 years.  As a result XL clearly would not meet President Obama’s criterion for approval, namely that it be in the nation’s interest.  The President should not approve the XL pipeline project.

The Final SEIS evaluated lifecycle emissions from projected operation of XL, and compared the result to emissions expected from production of reference sources of heavy crude oil.  The analysis included extraction, processing, transportation, refining, and final uses including gasoline for fuel, as well as co-products such as petroleum coke.

Extraction of tar sands oil was acknowledged to emit about 17% more GHGs than occurs in extraction of crude oil in the U. S. as of 2005.  Overall, the annualized burden of GHG emissions from tar sands oil was found to be 147 to 168 MTCO2e for operation of XL at its full capacity of 830,000 bpd, compared to 124 to 149 MTCO2e for four reference crude oils.  The incremental lifecycle burden from use of tar sands oil was evaluated to be in the range 1.2 to 27.4 MTCO2e per year.  (The Final SEIS states that the broad range of these values arises because several reference crudes were used and because the result depends on which lifecycle study was used in the comparison.)

Erickson and Lazarus conducted a comprehensive economic lifecycle analysis of oil use and emissions arising from operation of XL, evaluated for the year 2020.  They  formulate the increment in yearly emissions over the projected lifetime of the pipeline as the total from a) those required during construction, b) those required to operate XL compared to pipeline shipment of reference crudes that would be displaced by sending tar sands crude through XL, and c) consumption of tar sands refined products compared to consumption of displaced reference crudes.  The authors agree with the Final SEIS that contributions a) and b) are trivial in comparison with c); they did not analyze those further. 

Algebraic rearrangement of their simple expression showed that a principal factor in the economic analysis is a ratio that expresses the extent to which expanding oil sands production may increase global oil consumption.  The authors state that the Final SEIS did not assess this contribution, and furthermore that it has not been treated adequately by others either.

The authors evaluated this ratio using methods of economic lifecycle analysis employed for other proposed fossil fuel extraction projects.  The principal contributions to evaluating the ratio are first, a projection of how worldwide prices for crude oil would decrease as the additional supply from operating XL would ramp up to its maximum capacity, 830,000 bpd; and second, how demand would be shifted to higher levels at any fixed price as shipment of tar sands oil increased up to the maximum.

Erickson and Lazarus conclude that net worldwide consumption of crude oil would increase by 490,000 bpd when XL operated to capacity over the 50 year-lifetime of the pipeline.  This translates to an incremental GHG emissions burden of 100-110 MTCO2e per year, an estimate that is about four times higher than the highest value of 27 MTCO2e per year estimated in the Final SEIS.  According to the authors, the “sole reason for this difference is that we account for the changes in global oil consumption resulting from increasing oil sands production levels, whereas the State Department does not…. Our simple model shows that, to the extent that Keystone XL leads to greater oil sands production, the pipeline's effect on oil prices could substantially increase its total GHG impact.”  (They also point out that the analysis used in the Final SEIS is proprietary and “is opaque with respect to key assumptions and features”.  In contrast, Erickson and Lazarus state that they use openly available peer-reviewed methods.  Also, their article itself underwent peer review prior to publication.)  They conclude that the Final SEIS issued by DOS has failed to consider the most important factor determining global consumption of crude oil: the increase in use due to increased supply and its effect leading to lower prices.

© 2014 Henry Auer


  1. Do you think that oil will not be burnt, because there is not a pipeline of course it will, it will be shipped somewhere by other means, probably more damaging and dangerous ways.

    1. The comment above by Anonymous makes the fallacious argument that if the U. S. doesn’t approve XL, other means to transport the oil will be found. It is wrong because it implies that regardless of its decision, the U. S. will be responsible for tar sands oil reaching the Gulf Coast refineries. This blurs policy requirements and evaluations of the national interest (see this post for details:
      Decisions regarding our evolving energy economy are not yes-or-no, black or white; the question is not simply whether to build XL or not. The energy economy is more like a zero-sum enterprise: in the face of worsening global warming, how can we make choices that minimize further warming?
      Thus we need to choose between extending use of fossil fuels (e.g. by building XL) or using that same capital to invest in reducing warming. This includes investment in renewable energy sources and enhancing energy efficiency. Rejecting XL should be accompanied by promoting the latter alternatives.