How to Deconstruct the Difficult Math of Keystone XL's Carbon Footprint

We may never know the Keystone XL's precise carbon footprint, but it's possible it will add up to a billion tons of CO2 over the lifetime of the project.

Activists protest the Keystone XL oil pipeline on Aug. 27, 2013. Credit: Ben Powless

It is not easy to measure the carbon footprint of the tar sands crude that would move through the proposed Keystone XL pipeline to refineries in the United States.

There is no question that the oil extracted from Canada's vast bitumen reserves is dirtier than most—heavy with carbon and emitting even more of the greenhouse gas carbon dioxide when it's produced or consumed.

That surplus carbon, and its implications for the fight against climate change, is one of the reasons for widespread opposition to the 830,000-barrel-a-day Keystone XL. And it is what led President Obama to declare last June that "the national interest will be served only if this project doesn't significantly exacerbate the problem of carbon pollution."

But how to put a number on the word "significantly"? That's tricky, because any such estimate is built on a matrix of assumptions, modeling methods and ever-shifting data about how the fuel is produced.

The consultants who prepared the final Keystone XL environmental impact statement, issued by the State Department on Jan. 31, seemed to contend that the carbon footprint number was essentially irrelevant anyway.

Whether or not the Keystone XL is built, they reasoned, Canada's relentless tar sands expansion would continue apace, and its oil would find its way to market one way or another. The implication of that—left unstated—was that the carbon in the oil would find its way to the atmosphere one way or another as well.

Even so, the authors included in their report detailed estimates of the carbon footprint of the amount of oil that could move through the Keystone XL, saying this analysis would provide important "context" to President Obama and Secretary of State John Kerry, who will ultimately decide the pipeline's fate. Those carbon footprint estimates found their way into the executive summary of the report, and they included a couple of numbers that environmentalists who oppose the pipeline were quick to cite.

One of those numbers is an estimate that the Keystone XL's dirtier tar sands fuel would emit 17 percent more carbon dioxide than the same amount of typical U.S. fuel supplies. The other is that the extra carbon dioxide hauled by the pipeline could surpass 27 million tons a year, when compared to the pollution from alternative fuels.

InsideClimate News cited both of those numbers in its spot coverage of the environmental impact statement, as did many other publications.

Both numbers are valid enough. But each is based on its own set of assumptions and separate calculations. And each of them is just one value from a range of values plucked from among many other estimates—all buried deep in the report's fine print.

In other words, the seemingly precise numbers are something between shorthand and a guesstimate. No one should bet the farm on their accuracy.

That's why, when asked after the report's release for the "bottom line" on whether the Keystone XL line would increase carbon emissions, Assistant Secretary of State Kerri-Ann Jones told reporters: "The specific answer you are looking for, that is not in this document."

"I know everyone wants certainty here," she added. "But this is about looking at technical data and seeing how you model it, and then trying to understand what might actually happen."

Where the Numbers Come From

The consultants who wrote the environmental impact statement did not base their estimates of the Keystone XL's carbon footprint on new calculations they did themselves. Instead, they reviewed the existing technical literature and focused on the work of a handful of labs and consultants who have expertise and special models for estimating the carbon footprints of various kinds of fuel, including tar sands.

None of their models are perfect; none of the experts have a clear crystal ball.

Recognizing the uncertainties, the State Department's contractors presented the various estimates side by side in one chapter and an appendix of the report, using them to build ranges of estimates for various scenarios.

From these collections of broad ranges of estimates, presto! the report's authors produced the numbers like the oft-cited 17 percent figure.

It's more art than science to settle on a single number.

Every one of the studies they looked at was built from its own set of measurements and calculations. One might have been based on data from 2005, another from more recent data. One might have compared tar sands to Mexican oil, one to Venezuelan. How much of the tar sands fuel was diluted bitumen, or dilbit? How much was upgraded to a synthetic crude? Different models made different assumptions. One might have counted the byproduct known as petcoke as though it was all burned, another as if it was all buried, a third as if it was exported—out of sight, out of mind.

The section of the environmental impact statement that explains its carbon footprint calculation discusses all this in detail. The differences between various methods of estimation go on and on for pages.

The Pipeline's Lifecycle Carbon Footprint

Whether the resulting numbers are too high, too low or Goldilocks perfect, however, they all point in a single direction: Keystone XL's fuel comes with a hefty surfeit of carbon.

Here is one of several key findings that the report said is "clearly supported" by its review: "The lifecycle carbon footprint for transportation fuels produced in U.S. refineries would increase if the project were approved."

Lifecycle analysis, or LCA, is what energy specialists call the task of looking at every phase of the fuel cycle and measuring its carbon emissions. The phases include: getting the oil out of the ground, cleaning and upgrading it, shipping it to refineries, the refining process and, ultimately, the combustion of refined products like gasoline in a car.

At its simplest, the lifecycle analysis conducted by the State Department's consultants was supposed to produce a reasonable estimate of the pipeline's carbon footprint.

To come up with their lifecycle estimate, the contractors relied mainly on the work of the National Energy Technology Laboratory (NETL), part of the U.S. Energy Department, and also on the work of two consulting groups, Jacobs and TIAX, which have worked extensively for the Canadians.

From the steamy bitumen fields to the smoking automotive tailpipes, the emissions from 830,000 barrels a day of tar sands fuel would amount to something between 147 and 168 million metric tons of carbon dioxide a year, they estimated.

By anyone's standards, those are meaningful amounts of greenhouse gases.

During five decades in service, the Keystone XL's deliveries would account for emissions of seven or eight billion tons of carbon dioxide. Cumulatively, over half a century, that adds up to more than what the whole United States dumps into the atmosphere in any single year.

Looked at another way, it's enough pollution to offset all the carbon dioxide emissions that will ever be avoided by the strict new automobile efficiency rules President Obama established early in his presidency. Those mileage standards are expected to prevent 6 billion tons of tailpipe CO2 emissions during the entire lifetimes of all the fuel-efficient cars that fall under the regulation.

But it may be an overstatement to claim that Keystone XL's contribution to global warming is such a huge number. Nobody thinks that if the Keystone XL line were plugged up, all those millions of cars would sputter to a stop, never to be fueled again.

We'd just burn other fuel, right?

The real point is this: The Keystone XL's fuel is dirtier than any other fuels that might tank up those cars in the future. It's the difference between tar sands and alternative fuels.

So the authors of the review moved on to the next step.

Calculating the 'Incremental' Emissions

How much of the emissions attributed to the Keystone XL can be considered "incremental" pollution, the amount of CO2 beyond what would enter the atmosphere if cleaner fuels were used instead?

That's a more complicated question, involving many assumptions about the ebb and flow of oil markets.

To address it, the analysts sought to compare the Keystone XL's lifecycle emissions to four alternatives, or "reference" crude oils, which the tar sands might displace.

They examined one alternative fuel type from Venezuela and one from Mexico, because both nations have supplied heavy oil to the Gulf Coast refineries over the years. They also looked at one light variety from Saudi Arabia, a so-called "balancing crude" that the oil cartel leaves in the ground when it wants to offset any glut. And finally, they considered a mixture of fuel representing the average consumption of crude oil by all American refineries, because booming production from around the U.S. oil patch can be viewed as competing with Canadian oil.

Depending which reference fuel was examined—and which of the various models were used to estimate emissions—the well-to-wheels annual carbon footprint from 830,000 barrels per day compared to the alternative fuels fell in a range between 124 and 159 million tons a year.

That's slightly lower than the range of 147 and 168 million metric tons that was found to be the Keystone's annual carbon footprint, and it suggests that alternative oils would indeed be cleaner.

But if you are trying to quantify the extra emissions that Keystone XL would deliver, how to compare the two ranges? You can't just subtract one from the other, the report pointed out.

Doing so, the State Department's contractors warned, would only lead to a flawed, apples-to-oranges result, because of the fact that each range encompasses multiple studies using multiple assumptions and methods.

Instead, the analysts broke the diverse conclusions down study by study. And what they were left with was a basket of footprint estimates as diverse as anything you can order from Harry and David.

Of all the various individual studies used to assemble the ranges, the single calculation that produced the lowest estimate of the Keystone XL's incremental pollution found its extra emissions to be just 1.3 million tons a year—equivalent of 271,000 cars, or less than half a coal fired power plant.

That individual low-end estimate comes from comparing the fuel from the tar sands to a heavy Venezuelan or Mexican crude. It was calculated using the Jacobs methodology.

Was it correct? Not necessarily.

Comparing tar sands fuel to a heavy Venezuelan or Mexican fuel, but using the methodology of the TIAX and NETL models instead of Jacobs, produced much higher estimates of tar sands' extra annual emissions: 17 million tons under TIAX, and 18 million tons under NETL.

So much for the low end. How about individual high-end estimates?

Of all the various model runs, the one with that arrived at the highest incremental carbon footprint came up with the figure of 27.4 million tons of extra carbon dioxide a year. That worst-case scenario found its way as the high number in a range in the executive summary of the State Department's report, and it became one of the instant take-away messages.

It would be equivalent to the annual emissions from 5.7 million cars, 1.4 million homes or about eight-coal fired power plants.

This single estimate comes from comparing the tar sands to Middle East sour, a relatively light crude. It was calculated using NETL's methodology.

But was it right? Again, not necessarily. Using the same reference fuel for comparison, but adopting the methods of TIAX instead of NETL, the estimate was 22 million tons. And using Jacobs, the answer came in at just 9 million tons a year.

So much information! But in boiling it all down and reporting simply that the range of estimates for the extra carbon footprint of the tar sands was between 1.3 and 27.4 million tons, the State Department's executive summary hid the complex reality of how frail these estimates can be.

There are just far too many uncertainties and far too many variables in any of these studies to pretend that the numbers are precise.

The same thing goes for the oft-quoted conclusion that tar sands is 17 percent dirtier than typical fuels consumed in America.Competing estimates based on the same NETL model found tar sands oil 13 percent dirtier than Mexican, 18 percent dirtier than Venezuelan and 19 percent dirtier than Saudi crude.

Still, despite the fact that it may never be possible to precisely determine the extra emissions from Keystone's oil, it is possible to reach a robust conclusion: Tar sands crude is dirtier than other fuels—even other heavy grades.

It seems safe to say that the fuel carried by the Keystone pipeline, by the time it is  burned, will probably emit tens of millions of tons of extra carbon dioxide each year compared to alternative fuels—an amount that could easily add up to a billion tons of CO2 over the lifetime of the pipeline.

Facebook Twitter Google Plus Email LinkedIn RSS RSS Instagram YouTube