CO2 from Canadian Oil Sands – Higher, not Horrifying

This post was written by James Coan, Research Associate at the Baker Institute Energy Forum

Environmental opposition to the Keystone XL pipeline that would bring crude oil to the U.S. from the Canadian oil sands has heated up in recent weeks as the decision for whether to approve the pipeline approaches. This past week saw the start of a rally in front of the White House in opposition to the pipeline, as well as a negative editorial in the New York Times and an article in Foreign Policy.

A major concern is that the oil sands are more CO2-intensive than the typical crude consumed in the U.S. because the production process requires more energy. This is absolutely true, but many opponents of oil sands take a misleading, narrow view of emissions. They tend to focus only on the production and refining of the fuel before it is ever burned in, say, a car or truck – yet it’s that burning in the gas tank that creates the majority (roughly 70-80 percent) of emissions.* For instance, according to the IEA’s World Energy Outlook 2010, CO2 emissions before combustion (known as “well-to-tank”) are about 100 kg CO2/barrel for Saudi crude, but burning that oil releases 420 kg.

The CO2 numbers sound frightening when only the production and refining are taken into account: Charles Homans in Foreign Policy cites the EPA that “the extraction process produces 82 percent more emissions than conventional oil drilling.” The EPA clearly states that this figure is before the fuel is ever burned. Yet once the oil is burned, a variety of sources say the total lifecycle impact of oil sands relative to the average crude used in the U.S. is much smaller, including the Council on Foreign Relations (17 percent higher emissions) and Cambridge Energy Research Associates (5-15 percent).

It’s the latter full lifecycle numbers that really matter. It would be ridiculous to say that only the manufacturing process matters when looking at the environmental impact of a car and not both its overall manufacturing and ongoing operations, but this is basically what those who point to the extraction process of oil sands are doing.

Still, even though the differences are modest, oil sands definitely do result in more CO2 emissions, and it is important to reflect on the importance of emitting up to 20 percent or so more per barrel. But from my perspective, this difference in CO2 emissions is too small to warrant blocking the pipeline. There are far larger sources of CO2, such as coal burning.

Say we assume that oil sands production eventually reaches 4-5 million b/d, as the EIA projects for the 2030s. The production and combustion of 4-5 million b/d of conventional oil would result in CO2 emissions that represent a little over 10% of current U.S. CO2 emissions. If oil sands CO2 emissions are on the high end at roughly 20% greater than average crude, replacing that average crude with oil sands would increase CO2 by roughly 2-3% relative to current U.S. CO2 emissions.**

After visiting the oil sands, it’s very likely that CO2 emissions of the oil sands will decline in the future, further reducing the amount of extra CO2. Mining, the more mature process that seems to have less opportunity for improvement, already has lower lifecycle CO2 emissions of about 580 kg/barrel according to the IEA, or about 10% higher than average oil used in the U.S. (which is more like 525 kg/barrel, according to the IEA and Council on Foreign Relations). Meanwhile, companies are aggressively innovating with the most common in-situ process, Steam-Assisted Gravity Drainage (SAGD), which currently leads to emissions 100 kg per barrel, or 18% percent higher than conventional sources. SAGD is responsible for about 40 percent of oil sands production and is expected to reach around 2015. It has only been commercial for a little over a decade, and companies are trying approaches such as the use of solvents to reduce the amount of steam – and thus, CO2 – needed to heat the subsurface and let the viscous oil sands flow.

Alberta is also beginning four pilot projects to test carbon capture and sequestration (CCS). Oil sands operations are not necessarily the best use for CCS, but Alberta is coal-intensive, so it may try to offset other emissions.

Besides, if the Keystone XL pipeline is blocked, it is very possible CO2 emissions would actually marginally increase. The Northern Gateway project is an attempt to bring crude oil to the coast of British Columbia to be refined in Asia, which would require tankers that require more energy than pipelines. Additionally, in-situ production can be profitable at $50/barrel WTI or less, providing an incentive if oil prices are high to move crude by any means possible including CO2-intensive truck or rail.

Pressure should continue on companies extracting oil sands to do what they can to reduce their CO2 emissions. But we must keep in mind that while oil sands are more CO2-intensive, the differences are quite modest.

* Using figures from the World Energy Outlook, for the best oil production, end use carbon accounts for 80.7% of emissions. For the worst case in situ mining, end use accounts for 67.7% of the total emissions.

**According to the EPA (both here and here), light-duty vehicles accounted for about 19% of U.S. CO2 emissions from 1990-2008, and roughly 9 million b/d were used in the late 2000s according to the EIA. The production would increase emissions by 25%, so total “well-to-wheels” CO2 emissions from that oil results in about 24% of U.S. CO2 emissions. Half of this amount (4-5 million b/d) is about 12% of emissions, and 20% of that figure is 2.4%.