Oil Going Up? Cost of the Tar Sands

In the final analysis, the price of oil has nowhere to go but up. The easiest, most profitable extractions of this precious commodity have already been done. Gone are the days when a little drilling would release a gusher. Take, for example, the tar sands, our latest and perhaps most costly extraction, as I pointed out in a recent post, Frankel and Rapier*.

Extracting bitumen will not be easy. In fact, as elsewhere, the easiest extraction may already have occurred. Eighty per cent of all tar sands bitumen lies too deep in the ground for open mining. The cost for deeper extraction may well exceed anything we might have expected, in terms of retrieval, water usage, and, of course, pollution.

Bitumen is truly messy stuff. Think of something akin to heavy molasses or soft tar. When it lies on the surface, you can push a finger into it. Walking on it, you may well leave a footprint. Bitumen has to be separated from the sands themselves. For each cubic meter of bitumen rescued, we have five cubic meters of toxic tailings. Already the tailing ponds are so vast that they can be seen from outer space. In fact,

The Syncrude Tailings Dam is deemed by the U.S. Department of the Interior to be the world’s largest dam by volume of construction material. The pond, built in 1973, covers 22 square kilometres and holds 540 million cubic metres of water, crud and sand. When China completes the Three Gorges Dam this year, Syncrude will surrender the record.

Nixen has mining operations south of Fort McMurray, the Long Lake project. A report from Nexin itself should give us pause:

Our estimates for performance and recoverable volumes for the Long Lake project are based primarily on our three well-pair SAGD pilot and industry performance from SAGD operations in like reservoirs in the McMurray formation in the Athabasca oil sands. Using this data, our assumptions included average well-pair productivity of 900 bbls/d of bitumen and a steam-to-oil ratio of 2.5. In order to test operating parameters, we built a three well-pair SAGD pilot, commenced steaming the reservoir in May 2003 and commenced production in September 2003. The pilot is currently producing at a rate of about 400 barrels of bitumen per day per producing well-pair and an average steam-to-oil ratio of about 4.2. Since September 2003, the pilot has recovered less than 3% of the original bitumen in place. While we expect actual performance to improve as we commence commercial operations where SAGD drilling has confirmed higher quality reservoir conditions, there can be no assurance that our SAGD operation will produce bitumen at the expected levels or steam-to-oil ratio. We have addressed this risk by drilling additional wells and adding steam generating capacity to allow for a steam-to-oil ratio of 3.3, however, if the assumed production rates or steam-to-oil ratio are not achieved, we might have to drill additional wells to maintain optimal production levels, construct additional steam generating capacity and/or purchase natural gas for additional steam generation. These could have a significant adverse impact on the future activities and economic return of the Long Lake project.

Initially, the steam-to-oil rato was to be 2.5: Two and half barrels of steam for every barrel of oil. Now it is 3.3 barrels of steam…and possibly rising. Steam requires energy; energy is a quantifiable cost. But energy costs in turn have there own costs.

In 2008, the Kyoto Protocol will come into effect in Canada. The Long Lake project does emit green house gases, possibly adding as much as $0.40 to each barrel of oil extracted.

But the real cost has yet to be fully assessed–and it might be staggering: Already there is substantial risk that ground water will become polluted. And those giant tailing ponds are not “settling” out as expected. Initially, everyone hoped that the tailings would settle into hard clay at the bottom of the ponds. Sorry. It isn’t happening.

As the team leader on the subject at Natural Resources Canada’s CANMET Energy Technology Centre in Devon, Alberta, Mikula calls the tailings waste problem a “frightening” and vexatious issue. Engineers originally thought that the tailings waste would quickly settle, leaving clear water on top. But that never happened, thanks to what Mikula calls “the bad behaviour of clays.” He suspects the waste won’t settle to solid form for thousands of years. “So something has to be done.”

One suggestion has been to use giant centrifuges to separate out the tailings.

A better solution might be a sort of “brute force” centrifugal approach, says Mikula. It involves spinning the material to create something dry and stackable that could eventually be reclaimed—while recovering water at the same time. Both Syncrude and Suncor have begun pilot projects. “We could reduce water usage from four barrels to two [per barrel of oil] or maybe less, which means less water withdrawn from the Athabasca,” says Mikula.

That certainly will not be cheap. The cost of tar sands oil keeps rising.

And I have yet to mention that water is not exactly an abundant commodity in the McHenry region. Even without peak oil, the cost of oil just keeps rising….

Flight to commodities? You have to be kidding if you think guys in back room betting parlors determine the cost of everything. I bet not one of them has visited a tailing pond.

*As one commentator observed, I inverted the conventional EROIE ratio in that post. An acronym describing my ratio should have been ERIOE. Nonetheless, all subsequent explanations hold true based on my use of the ratio.