Canada's Oil Sands - Part 2

This is a follow-up to Part 1, which tells about my recent trip to Canada's oil sands, on a trip sponsored by the American Petroleum Institute (API).

In Part 2 of this post, I provide some additional thoughts to help the reader come to his / her own conclusions about the future of the oil sands. I talk a little about how Canada's oil sands production fits in with its other sources of supply, and how this in turn relates to Canada's exports. I also look a little at some political issues and how these fit in with environmental issues. A closely related post is this recent post.

How much will oil sands production expand in the future?

There is no doubt that there is a huge amount of resource in place - between 1.7 and 2.5 trillion barrels, according to the Oil Sands Discovery Centre's Oil Sands Story. Of this, 173 billion barrels (about 10%) is considered producible with current technology at 2006 prices ($66 barrel for WTI). Production to date has been relatively low, though--only 1.2 million barrels a day in 2008, according to Canadian Association of Petroleum Producers (CAPP).

Sources of Canadian oil, including imports from the East Coast - Based on EIA and CAPP data. See this post.

A big part of the reason that Canada can export oil to the United States is the fact that it is importing oil on its East Coast for its own use. When one looks at oil sands in relationship to its other oil sources (including imports), oil sands oil is only about one quarter of the total. The non-oil sands portion is expected to decrease in the future, so a significant increase in oil sands production is needed simply to offset expected decreases elsewhere. See my earlier post about this.

Unless oil prices rise above today's level, and stay higher, it seems unlikely that there will be a major increase in oil sands production. Don Thompson from the Oil Developers Group told us that in order to justify new development, the cost of West Texas Intermediate (WTI) crude must be at least $80 a barrel. CERA indicates in a new report (free with registration) that more than 70% of proposed Oil Sands expansion projects were postponed, after the drop in oil prices in 2008. CERA quotes a needed WTI price of $65 to $85 barrel to justify Oil Sands expansion--but based on a rate of return of 10%--which is probably not high enough.

Even beyond high oil prices, a major ramp up would also require a huge amount of investment funds. For example, if one wanted to add 1 million barrels a day of upgraded oil by adding new mines and upgraders, it would require an investment of roughly $126 billion dollars, based on a CERA estimate of $126,000 per flowing barrel. It would also require a huge amount of credit availability and a veritable army of workers.

In the absence of a major improvement in technology (or perhaps even with one, since a big change in technology often takes a long time to implement), my estimate of the ramp up in oil sands production to 2020 is given below:

My rough estimate of Canada's future oil sands production, other oil production, and overseas imports to their East Coast as shown in this post.

I am forecasting that oil sands production will approximately double by 2020 (from 1.2 million barrels per day in 2008, to 2.4 million barrels a day in 2020). This forecast, in part, reflects Tony Eriksen's ("ace's") calculation that projects already approved and in construction are expected to bring production up to 1.94 million bpd. Growth beyond that is expected to be constrained for a number of reasons, including difficulty in obtaining sufficient investment funds, need for a high price of oil to justify new construction, difficulty in obtaining enough diluent, and the possibility that royalties will be higher as governments discover that oil companies are among the few companies from whom higher taxes might be extracted.

Note that even with this increase in oil sands production, the total supply of oil available to Canada (including imports) dips somewhat, and exports are likely to decline. My forecast for oil sands production to 2020 is similar to the average of CERA's two lower production forecasts, taken from the new CERA report.

CERA forecast of Oil Sands production, from report cited above.

Even in CERA's "Barreling Ahead" forecast, the ramped up production will barely make a dent in the four or six Saudi Arabias of new oil production that Fatih Birol of the IEA has said will be needed if new production is to offset declines.

How much Canadian oil will the United States import?

Canada's exports to the United States, if it exports everything that is left over to the United States

Based on the forecast I made for oil sands production, the amount of imports to the United States are likely to drop somewhat, even if the United States continues to receive virtually all of Canada's oil exports.

Will oil actually be exported to the US?

But how likely is this scenario? When one reads Canadian material, it becomes clear that many Canadians would very much like to diversify the countries it exports to, even if currently nearly all pipelines lead from Canada to the United States. With the US as its only export partner, Canada has little leverage in bargaining over price. Also, there are details such as the US charging less for gasoline than Canada--why should Canadians be taxed to keep their consumption down, and the same time Americans can buy Canadian oil and sell it for less? NAFTA limits changes right now, but over the long term, that can be changed.

Canada is already working on an approach that will allow it to export oil in directions other than the US, to help diversify its exports. The approach is a rail link that acts like a pipeline, and can be implemented quite quickly. China and others with few environmental concerns are particularly favored as export partners. Canadian National (CN) also has tracks going East, so this approach could also be used to bring oil from western Canada to eastern Canada.

CN's revolutionary pipeline on rails

The Canadian National Railway has developed a transformative strategy it calls the “Pipeline on Rail” which can move oil sands production quickly and cheaply to markets in North America or Asia. . .

CN could gear up its capacity to ship by rail up to four million barrels a day of oil at less cost and more quickly, bypassing the need to finance huge pipelines. By the end of this year, the company will be shipping 10,000 barrels daily from producers whose reserves are now stranded.

Oils sands: Canada to China, Japan, India not US

This project, in its early stages, will eliminate three barriers to the development of Canada’s vast oil sands: the cost, delays and financial risks involved in building multi-billion dollar pipelines; the politics of obstruction south of the border from environmentalists and the danger of selling oil to monopoly buyers in the U.S. which has, in the past, resulted in contracts being ripped up when times were tough.

It also allows Canada to decouple from the American economy when it comes to its most important commodity which is oil products. This is because all the oil sands production can be routed to the west coast for shipment to Asia or anywhere, thus avoiding monopoly pricing and bullying by the Americans. Besides that advantage, oil sands are the national trump card in the future and the American economy, now sputtering, will never be as robust as before.

China is interested in investing in the oil sands. Just today we read that China is paying C$1.9 billion for an oil sands investment.

Alberta will ramp up oil sands, to the best of its ability

Alberta is very dependent on the oil sands industry, both for jobs and revenue. A report by Canadian Energy Research Institute (CERI) indicates

. . .every dollar spent on big oil projects stokes nearly $2.50 in further spending in Alberta, plus another roughly $2.50 in the rest of Canada. In other words, a $10-billion decrease in energy spending translates into about $60-billion in lost economic activity across the country.

These revenues translate to tax revenues as well, so I find it difficult to believe that Alberta will cut back oil sands production voluntarily, or even limit new production. If the US wants to cut back on imports because it is unhappy about CO2 or other environmental issues, I don't see that as a huge problem to Canada. Canada will sell its oil elsewhere.

Potential Impact of Loss of Oil Sands Production on US

From the US point of view, the potential loss of oil sands oil could be problematic. Pipelines that feed the oil sands oil feed directly to the Midwest. If oil sands oil is cut off, the ones likely to be hit the hardest hit are Midwestern American farmers. We know from experience that when there is a shortage of oil, it is the folks at the end of the pipeline that are hit. If the US decides it doesn't want Canadian oil, or if Canada cuts us off, it is likely the folks at the end of the pipelines, away from Chicago and the Midwest that would be hit most--places like North and South Dakota.

We could theoretically import more oil (perhaps from OPEC) to one of our coasts, and pipe it to the Midwest. This might or might not be successful. We don't have a rationing plan to work around shortages of diesel fuel in the Midwest at this point--maybe we should be thinking about such a plan.

How big an issue is environmental considerations?

Oil sands oil has a number of environmental issues. The one which is most obvious is the clearing of forests and taking up layers of overburden, storing them for many years, and then putting them back. This is an issue primarily with mining, rather than in place (in situ) production. Mining also tends to produce tailings ponds that are polluted with bitumen which is missed in the extraction process and with naphthenic acid. There is concern that water from this tailings water will escape, or will harm birds that happen to land in it. Companies use scarecrows and noise makers to attempt to keep birds away, but sometimes these precautions fail. There is also concern about the amount of water use, especially if production of bitumen from the oil sands should scale up.

There are other concerns with the extracted sulphur, and whether it gets into the air, as sulphur dioxide gas. As far as I know, the biggest issue with this is with upgrading operations, where the sulphur is separated out from the bitumen, such as that of Syncrude.

There are also metals that are found with the oil sands oil. This can be both good and bad. Bad because they could potentially be another source of pollution, and good, if they can be used as a source of rare earth minerals. We read:

It has been found that the tailings are a unique and rich source of titanium (5-10 % on solids), zirconium (2-5 %) and iron and rare earth minerals (8-12 %). These valuable minerals are at present not being recovered.

The quote relates to a patent for a method of extraction of these minerals, by applying centrifuge methods to tailings ponds. I expect we will see more centrifuge methods used in the future, both because of the possibility of financial gain, and because of a desire to reduce the amount of water stored in tailing ponds.

In situ production has much less direct environmental issues than mining, because it disturbs the soil much less, and because it leaves some of the potential pollutants underground. In situ results in far less tailings ponds. Most of the problems from in situ mining result simply from the fact that it is a low EROEI process, and uses a lot of natural gas in its production. As a result, its carbon footprint is quite high.

While there are quite a few environmental issues, I think that what we hear may be exaggerated.

Scale-up Confusion

I think the issue of "scale up" is a matter of huge confusion to environmental groups. How much worse will the environmental impact be, 10 or 20 years from now, if oil sands production continues to grow as forecast? Will the scale up have 20 times the current environmental impact of the worst company, or will the impact be quite small, as production is shifted more and more to in situ (in place) facilities, that don't disturb the ground except to lay underground pipelines to heat the bitumen, and as new environmental laws begin to have more impact?

Suppose that production actually grows to the extent forecast by the Canadian Association of Petroleum Producers (CAPP) in its "growth case". (This is more of an increase than I am showing in my forecast earlier in the post.)

CAPP "Growth Case" forecast for Western Canada, including conventional as well as oil sands production, from June 2009 Report.

With this forecast, mining will roughly double between 2008 and 2025, and in place production will triple. Mining is the problematic one, in terms of land disturbance, water use, and tailing ponds. At worst, it would seem to me that there will be double the amount of pollution issues in 2025 that we have now, based on the forecast increase in mining. But even this estimate is high--companies will be getting rid of legacy technology that holds them back environmentally, and recent law changes (such as one passed in 2009 requiring much more rapid retirement of tailing ponds) will further reduce the impact of increased production.

The amount of surface area that is currently covered by oil sands mines amounts to about 200 square miles (518 square kilometers), according to the CERA report. This corresponds to a square 14 miles (23 kilometers) on its side, which is smaller than the footprint of many cities. What we are talking about, based on the CAPP mining forecasts, is, at worst, is doubling this footprint. The equivalent square would have an area of 400 square miles, so would measure 20 miles (32.2 kilometers) on a side.

Likely Area to be Disrupted by Mining

The confusion about scale up has led to very strange statements. For example, I recently got an e-mail from the Sierra Club that said (regarding the recent approval of the Alberta Clipper Pipeline):

This decision is deeply distressing because tar sands development in Alberta, Canada is creating an environmental catastrophe, with toxic tailings ponds so large they can be seen from space and plans to strip away the forests and peat lands in an area the size of Florida.

The entire area where oil sands deposits are located is the size of Florida. But the mines themselves are tiny in comparison--the size of a not very large city. Perhaps we are talking about doubling this by 2025. The tailing ponds are contained within this area. The percentage of this area covered by tailing ponds is now about 10%. This percentage may very well decrease, as new technology and new stricter laws are implemented.

When trees are cut, we were told that the trees are sold to the logging companies in the area, and thus offset other trees that would have been logged elsewhere. The peat moss is stored for later reuse, according to the Province of Alberta.

Carbon Dioxide Emissions

Another question is how to measure carbon emissions. Do you look at emissions during their whole life cycle, or only during extraction and refining? One would expect emissions to go up with the cost of producing oil (or decline in EROEI), and that is generally what a CERA study shows. The majority of the emissions relate to burning the oil that is produced, and these remain unchanged regardless of how low the EROEI of producing the oil is.

CERA's comparison of Greenhouse Gas emissions, on a well to wheel basis.

Also, in terms of absolute level, if one compares CO2 emissions from oil sands operations to those of coal fired power plants, the coal fired power plants seem to be a much bigger problem. If Alberta's oil sands emissions were doubled (as might possibly occur by 2025), they still would be less than the coal fired electricity emissions of many US states.

Comparison of Oil Sands Emissions with Coal Fired Power Plant Emissions, from Setting the Record Straight presentation by Don Thompson.

Athabasca River Pollution

Another source of confusion is pollution of the Athabasca River. The Athabasca River has been polluted for many years, and will continue to be polluted, because the river runs right through the oil sands area. In fact, the oil sands deposit was discovered because of water pollution.

Oil Sands Natural Leakage to Athabasca River.

The health of rivers in the area is being monitored by Regional Aquatics Monitoring Program (RAMP), a multi-stakeholder committee which includes representatives from oil companies, Ft. McMurray First Nation, Ft. McKay First Nation, Health Canada, Alberta Pacific Forest Industries, Fisheries and Oceans Canada, and many others.

Each year, RAMP issues a scientific report. The indications of these reports seem to show little problem with water pollution or overuse, except for a general mercury pollution problem, which I understand is quite widespread, extending throughout the US and Canada.

Air Pollution

With all of the pollution problems I had heard about, I assumed air pollution would be a real problem. Instead, the air was very clear. Don Thomson of the Oil Sands Developers Group told us the air in Ft. McMurray tests better than that in any other major city in Canada for air pollution, on almost any common pollutant. This may be in part because Ft. McMurray is several miles away from the mines.

I don't have many details on air pollution, except that this is one of the areas regulated by the province of Alberta. For example, Syncrude has undertaken a project called Syncrude's Emissions Reduction Project (SERP), designed to reduce stack emissions of sulphur compounds by 60% from current approved levels by 2011, under an agreement with the province.

Political Football Game

It seems to me that the Oil Sands are a huge political football game, and a good percentage of the environmental stories we hear are related to the posturing going on to win this football game.

Eastern Canada vs. Western Canada

First, Eastern Canada knows that it is likely to losing its imports from the East, due to peak oil. It needs the oil from Alberta, but as long as Alberta regulates the oil sands and exports the oil from the oil sands to the United States, (or to China), the East Coast is out of luck. So if Eastern Canada can show that Alberta is not doing a good enough job of regulating the oil sands, then it can perhaps get control through new Federal regulation of the oil sands. With that control, the Eastern part of Canada can be assured of getting oil. Also, if they want to raise taxes, it will be helpful to them. So we have reports like the one by Ecojustice, advocating additional Federal regulation.

US vs. Canada / China / Far East

Second, the US is the current recipient of the oil from the oil sands. If the United States can be convinced that we don't want it, because of environmental problems, that makes it all the easier for Eastern Canada to get it, or for Western Canada to export to China. Western Canada really would prefer not to sell to the United States, since the price it gets from the US is perceived to be not the best, and the US is viewed as a bully.

Consultants /Think Tanks / Environmental Reports

Because of the political situation, there is a big "market" for reports by consultants and think tanks that show huge environmental problems. There are also all kinds of ways one can legitimately show that there might be a problem. For example:

Undermining the Report Card: The Oil Sands Report Card by WWF and the Pembina Institute. This report compares the various oil sands projects against one another on a number of different variables, and makes statements such as, "If all companies had as low emissions as [lowest company], there would be a savings of __________ in emissions." That is interesting, but doesn't tell you whether there is a problem in the first place.

Statements like "Toxic tailing ponds filled with liquid mine wastes already cover more than 50 square kilometers," also from Oil Sands Report Card. Fifty square kilometers corresponds to a square a little over 7 kilometers (4.4 miles) on a side. Having so many tailings ponds is not great, but it is not as huge a problem as it is made out to be. In the future, the amount isn't likely to grow much, with more in-situ production, and with recent changes in regulations regarding tailing ponds that require much quicker ends to the ponds. Environmental Commissioner Renner told us that the oldest tailings pond (which is the one with by far the most leakage issues) is expected to be retired in 2010.

Photos with captions like, "Oil sands operations remove rivers, forests, and wetlands in order to access the oil sands beneath," also from Oil Sands Report Card. The angle of the picture makes it look like a huge area is involved, and the way the caption is worded makes it sound like there are no steps being taken to put the pieces back in place according to current views of best environmental practices.

Statements by aboriginal groups that they have some cancer above normal levels. Any of us who have studied probability know what the issue is here. There are all kinds of little aboriginal groups with population of around 1,000. So if you look at enough of these groups, some of the groups will have a handful of people with one or another kind of cancer. There is pollution in the area--there always has been, because the river naturally runs through the oil sands. The background pollution level may raise the probabilities of cancer a bit higher. But there doesn't seem to be any study showing an overall problem.

Not for Profit Issues

Not for Profit organizations are almost certainly under funding pressures. If they want donations, it is to their advantage to make things sound as dire as possible. If the leaders of the organizations truly don't understand the size of scale up issues, these organizations can easily make statements that inadvertently overstate how bad the future will be, relative to today.

Also, needless to say, their analyses are not of the nature of a cost/benefit analysis. They just tell you that there might be environmental costs involved, not that the oil eventually produced might provide benefits.

These organizations also don't tell you about the likely environmental consequences, if the replacement oil were extracted in a much more heavily populated area, such as one of the OPEC countries. The environmental impacts on the population would likely be worse--but a lot less reported, and a lot less regulated.

People trust Not for Profits. After all, they are "do good" organizations. A person might think that someone is looking over the shoulder of Not for Profits to see that what they say is adequately supported by facts, but I can't see that that is happening at all. This is complicated "stuff" to understand. If these organizations miss the boat, or slant things in a way to get more donations, no one is likely the wiser.

How the Politics Will Work Out

No one really knows the answer to this--but we certainly have a lot of people who are willing to take reports at face value, without considering the possibility that political maneuvering may be distorting what we see. Since peak oil is not in the newspapers, even Oil Drum readers assume "higher-ups" don't know about peak oil. That is simply not true. They know that imports may be declining in the future, and find Trojan horses in which to hide their causes.

There is one thing I am pretty sure of. No politician in Canada really has any intention of stopping oil sands production based on environmental concerns, although a few may want to make certain it doesn't grow too rapidly. Mostly what politicians want is the oil, or the profits from the oil, for themselves. If it takes an exaggeration or two about environmental issues to reach this end, so be it.

Note: Oil sands vs Tar sands. The province of Alberta and CAPP prefer oil sands, so that is the terminology I have used. Oil sands is also far more commonly used, according to Google.

Previous Oil Sands / Tar Sands articles.

Canada's Oil Sands - Part 1 - Gail the Actuary - Aug. 2009

World Oil Exports; US Oil Imports; and a Few Thoughts on Canada - Gail the Actuary - Aug. 2009

EROI Update: Preliminary Results using Toe-to-Heel Air Injection - Dave Murphy – March 2009

Unconventional Oil: Tar Sands and Shale Oil - EROI on the Web, Part 3 of 6- Charles Hall guest post – April 2008

Tar Sands: The Oil Junkie's Last Fix, Part 2 – Guest post by Chris Nelder – September 2007

Tar Sands: The Oil Junkie's Last Fix, Part 1 -
Guest post by Chris Nelder – August 2007

Extracting Heavy Oil: Using Toe to Heel Air Injection (THAI) – Gail the Actuary – August 2007

Canadian Oil Sands Production Update – Sam Foucher – Oct. 2006


You do a great job of showing just how complicated the whole oil sands show really is.

Most people seem to think of such things in terms of black and white with at most a few gray areas.

This one may have more gray than any either black or white given the political situation you have outlined.

I for one know almost nothing about Canadian politics even though I have tons of reading time(stuck in the house for now)and follow the energy news regularly.

Give the fact that most of us,collectively ,know only that Canada is at the north pole and that the cops wear red uniforms and are all naned Dudley Doright and that Dastardly Boris and Natasha are
after Rocky and Bullwinckle ...

You have done the readers of this site a great service.

Here is a map of the various companies' tar (if it smells like a duck...) sands holdings:

and superimposed in Google Earth:

A square about 67 miles on a side, or 4500 square miles.

Of course, just because there are holdings doesn't mean that all that area will get plowed up. But it is misleading to taut the tar sands vast bounty without considering all the land covering said bounty.

Anyway, it's covered with snow most of the year, not? Glad that Albertans are willing to sacrifice to keep the rest of us in BAU.

Tar sands sounds more accurate than oil sands, the latter diminishes the complexity of the actual oil production process and basically make it sounder "cleaner". What are the implications of the use of natural gas?

It is a little fuzzy how much of a limiting factor natural gas is.

If one is doing mining with an upgrader (as Syncrude does), the upgrader produces natural gas and other gasses that can be used as part of the energy requirement of mining. With some fine-tuning, this process could probably be adjusted so that all of the energy requirement could be met with self-produced petroleum products.

Mining with an upgrader also produces coke that is quite polluted. At this point, this material is just buried. There may be some possibility of cleaning this up for use (but this would be a pretty carbon intensive fuel.)

In situ production as it is currently done uses natural gas, but other approaches are being explored. One approach is burning some of the bitumen itself to provide the heat for melting the bitumen, for example with Toe to Heel Air Injection, a method currently being evaluated. Another approach under development uses electrical induction.

Between these various approaches, it seems likely that natural gas is not as much of a limiting factor as we have thought it to be. Availability of capital may be a more important limiting factor.

It is a little fuzzy how much of a limiting factor natural gas is.

It's only a limiting factor for those thinking inside the box. In reality, what they need is a source of process heat. They can get this from natural gas, they can get it from coal, they can get it by burning the heavy ends of the bitumen, or they can get it from nuclear reactors. The possibilities are many and diverse.

They burn natural gas because it's relatively cheap, easy to use, and available locally. If it ceases to be those things, they can switch (with a little difficulty) to something else.


Natural gas is limiting only so far as the current economics make it. There are PLENTY of alternative approaches that would/could/will be used if the price of Natural Gas increases. It is also worth noting that there is natural gas produced from the bitumen when it is exposed to heat. The in-situ mining process generates a fair bit of it.

There is also the syngas approach using the petroleum coke that has been used in one form or another for a hundred years. (adding water to carbon and generating H2 and ultimately CH4 which can be added to the bitumen to produce lighter crudes or used directly as fuel). You are correct that they are "carbon-intensive", but when "people are freezing in the dark" through lack of access to fossil fuels, they will be happy to have these processes available!!

Unfortunately, I think you are all too correct in all of your other (excellent and prescient) statements about the politics of the oil sands. However, I think I should add that Albertans naturally feel a certain affinity for Americans as they tend to have a "cowboy culture" themselves. I would think that it would be very very difficult for Albertans to consider "cutting off" the Americans from access to the Oil Sands. Also, the political difficulties in "re-negotiating NAFTA" are HUGE. I have trouble imaging that any Canadian government would seriously consider that. The economies of the two countries are just TOO intertwined. Of course, Albertans are business people too and you can be sure that China will be given the choice of importing Alberta oil as well!! I am sure that they will gladly accept. I would expect that Albertans would gladly use that leverage (the Chinese importation) to get a better deal from the US.

Technology improvements are already available (e.g. "wet combustion") that would allow the tailing pond waters to be combusted and the napthenic acids with them. It is just the current economics and conservatism of the industry that prevents their adoption.


Thanks for your perspectives. In some future world, if one wanted to put together an area with both oil and food, it seems like an area extending from the plains states up to Alberta would work. So maybe there is potential for some long-term synergy.

In the meantime, I don't know what the real direction of trade/free trade will be. With the recession, there has been pressure for more buy local legislation, even in the presence of NAFTA. This is a recent article about US--Canada relations:

To the North, Grumbling Over Trade

Before coming here on a visit in February, President Obama assured anxious Canadians in a television interview that the United States’ trading partners had nothing to fear from the “Buy America” provisions in federal stimulus bills.

But now that the stimulus money is flowing, many Canadian companies are finding that, despite the president’s assurances, the North American Free Trade Agreement does not in fact offer them much protection.

I don't know whether there will be any renegotiation of NAFTA. It may just erode around the edges, or be ignored in some contexts.

Experience is that the US will definitely ignore it if that benefits them. However, I recal there being a provision in NAFTA which requires Canada to continue exporting oil and gas to the US at at minimum the same percentage of production that existed when NAFTA was signed, regardless of other developments. I'll bet that's one provision which will be enforced.

It is a little fuzzy how much of a limiting factor natural gas is.

If one is doing mining with an upgrader (as Syncrude does), the upgrader produces natural gas and other gasses that can be used as part of the energy requirement of mining. With some fine-tuning, this process could probably be adjusted so that all of the energy requirement could be met with self-produced petroleum products..

It's worth remembering that natural gas has two purposes, at least as I understand it. One is to produce heat - something they should be able to burn the bitumen or crude produced, if it's really a positive EROEI. The second is as a source of hydrogen. You need hydrogen to turn long carbon chains into short carbon chains. The hydrogen can be produced in a reaction with water and coke/bitumen/oil but it's very carbon intensive. It also reduces the EROEI equation even more.

In situ production as it is currently done uses natural gas, but other approaches are being explored. One approach is burning some of the bitumen itself to provide the heat for melting the bitumen, for example with Toe to Heel Air Injection, a method currently being evaluated. Another approach under development uses electrical induction.

I'm quite disappointed that no mention is made of EROEI for in-situ production. It would surprise me if it even reaches unity if one counts the energy lost to combustion in the ground; it's entirely possible that more than one barrel of bitumen is burned in order to coax a barrel of bitumen to drain out.

Whether this is profitable or not depends not on whether EROEI is negative or not, but on the fact that the one-plus barrel of oil being lost to production doesn't have to be produced itself. IE, if it's cheap enough to drill holes and pump in air and burn the stuff in the ground it will be produced unless there's a large enough price attached to releasing carbon in the air.

As for using electricity, that's about the most hideous mis-appropriation of resources I can imagine. Generate one of the most flexible, non-polluting, efficient forms of energy - indeed, a form that can even be multiplied by using it in a heat pump - and using it to heat one of the dirtiest forms of energy so that it can be used at 10% efficiency in the transportation sector.... It may make some perverse economic sense, but only if you have a "stranded" or otherwise very low cost source of electricity. Nuclear wouldn't seem to qualify.


With respect to the need for natural gas, you have a good point about the natural gas for upgrading. If the end product can be exported as bitumen, or mixed with a diluent so that it can be exported, the need for the natural gas can be put off to the end user, so it is not really an Alberta natural gas issue. But it is a natural gas issue, nevertheless.

The EROI of in situ processes will vary. The SAGD process I saw claimed an EROI of 6, but it left out so much that it was clearly lower. One reader wrote to me and suggested 3 was a better estimate. Part of the problem is that the energy used in the calculation was only the natural gas portion; another problem was that the end product was only bitumen. If one looks at the greenhouse gas calculations of CERA (as Jeff Vail did in another comment), it looks like the ratio is closer to 2.

If one burns part of the bitumen, the ratio will come out lower. But as you say, this could still be economic, if there is a low enough cost applied to CO2.

Alberta says that 2.4% of the area in which oil sands deposits are located are suitable for mining. So of the area the size of the state of Florida, only 2.4% could be mined, if there were adequate investment. Some of it would be put back into a condition vaguely like its original state by the time other areas are dug up, so it is unlikely one could get 2.4% dug up at the same time, under the worst circumstances.

Florida has an area of 67,000 square miles, so an equivalent footprint on Alberta is much much larger that I show above (it would cover a fourth of it). I would assume that the holdings are where the most accessible muck is, so it is not unreasonable to assume that most of that might be dug at some point.

Mining can only be done in specific locations, where the material is close enough to the surface, thick enough, and (as a practical matter) of a high enough grade--10% or more concentration of bitumen. Other places are going to be either passed by all together, or extracted with in situ methods.

I don't know what percentage of the company holdings are suitable for mining, but I would be willing to bet it is considerably less than 100%.

Here are some updated maps on the oilsands and operators and if it's Mining/SAGD.

Oil Sands Publications and Maps

Alberta’s Oil Sands Leased Area - PDF

Alberta's Current and Proposed Oil Sands Projects - PDF

I notice there is a lot that is already under construction. That is why "ace" (Tony Eriksen) is seeing a fairly good increase, even without additional projects being approved, and I am following his lead in my forecast.

The mineable area is sort of a triangle, just north of Ft. McMurray (although even within that, some areas would be better than the rest). Otherwise, future production is expected to use some in situ approach.


Excellent post by the way!! (as always). The number I have heard about the amount of bitumen accessible by mining is about 10% of the total that would be extractable by other means. So the in-situ method is definitely better.

I should mention that there is an excellent "hybrid" technology called "OSUM" (OIl Sands Underground Mining".

It uses the stunningly easy idea of a hybrid between conventional coal mining and SAGD, already demonstrated in 1982 at AOSTRA (government test facility for oil sand technology). Basically:

1) Dig a coal mining type shaft at 15 degrees down below the oil sand deposit (e.g. 1000-2000 feet deep) - very cheap compared to conventional coal mining

2) Install a regular SAGD operation at the bitumen face with the steam being injected UP and the mobilized (low viscosity) bitumen flowing DOWN into a collection area

This has MANY advantages:

1) The footprint for environmental disturbance of the forest above is about 40 TIMES less than regular SAGD (and even less than Bitumen mining)

2) The amount of energy required (EROEI) to mobilize the bitumen is about 10-20% less

3) The temperature at the bottom of the inclined shaft (easy walk) is about 50 deg. F year round (shirtsleeve weather). As anyone who has worked in Northern Alberta during the winter can tell you, -40 F for months at a time is no picnic

4) It has the same advantages as regular SAGD in terms of leaving the heaviest bitumen in-place, but the environmental advantages are huge.

SO there are plenty of new approaches that would either completely eliminate the environmental concern (disturbance, not CO2) and have other advantages as well.


It seems like the big issue will be price. It has been my impression that for coal, underground mining is generally significantly more expensive than surface mining. The energy savings you refer to might fully offset this additional cost--so the new process might turn out to be cheaper.


You are correct about coal mining, but the kind of the digging that we are talking about here is much simpler than the average coal mine with deep vertical shafts, horizontal shafts, elevators, etc. This is simply a gentle 15 degree sloped tunnel about a mile or less long to access the bitumen deposit from about 10-30 feet below the injection point of the steam. That is very cheap compared to a regular coal mine. Given that the amount of disturbed area is about 40 times less, there are some significant environmental advantages!

When the price of oil rises again, this technology will be very competitive indeed. Although initial capital will always be a problem.



This article is a breath of fresh air. But just a couple of points.

1. The Opti Nexen Long Lake project is worth special study because they have gotten it as right as possible from an environmental and EROEI point of view. By gasifying the asphaltenes (15% of the bitumen) they get syngas for steam generation, hydrogen for their upgrader and electricity to power the site. They have a small footprint and yet are focused on ways to reduce CO2 emissions. This is capital intensive but with low operating costs per bbl of premium synthetic crude.

2. One gets most of the available bitumen/oil extracted by open cast mining. But that isn't a large part of the available resource Some 80% of the bitumen prone resource is only available by using in situ techniques but the SAGD and THAI methods cannot extract all the available bitumen - only a portion.

So if I beg or say "pretty please" could you possibly do a follow up article - or seek one from CAPP - on the way forward for environmentally responsible oil sands exploitation?

The greenies - unfortunately - tend to do more damage than good when they attack productive solutions.

Thanks again for a well-researched and well written article.

the SAGD and THAI methods cannot extract all the available bitumen - only a portion.

Yes but the proportion that SAGD can recover ranges up to 80%, according to some geologists I have talked to. It's extremely effective when used correctly.

They're still working on the technology, and it's getting better.

Thanks for the tip! I'll see what I can do--am not sure how much I can promise, though.

The greenies - unfortunately - tend to do more damage than good when they attack productive solutions.

I am not sure what you mean by the term 'greenie'. If you mean someone who believes that the long term survival of the human race depends upon our learning to live within the limits of the ecological community of which we are a part, then I will have to raise my hand and plead 'guilty'. Strangely enough I am also a 'mathie' who believes that 2+2=4. Go figure.

The problem with tar sands is that they are not a solution to our problems. They are at best a tool which can help us transition to a sustainable economic infrastructure. Unfortunately you cannot create a sustainable economic infrastructure unless you actually intend to do so. I see very little sign of such an intention forming in our society even among people who label themselves as 'green'. Yes, fuel from tar sands will be used to transport food, medicine, and essential building materials, but it will also be used for driving SUVs, for jet airplane tourism, for transporting plasma screen TVs, etc.

If we were really serious about solving our problems we would be shutting down wasteful and unecessary forms of production and using the saved resources to create sustainable infrastructure. Unfortunately such a course of action is unthinkable in an economic system who end-all and be-all is maintaining the short term 'health' of capital markets.

You can make the argument that another decade or so of healthy capital markets will allow alternate energy technology to advance further, so that when we are finally forced to deal with limits to growth we will have a better set of tools at our disposal. But it is possible that such extended sucesss of capital markets will just dig us deeper into an ecological hole, and make the ultimate crash even worse than it would have been otherwise.

I am not in favor of shutting down the scientific enterprise, but we are going to need profound social changes and not just clever technology if we are going to save ourselves from the dilemma which we have created.

Let's add a bit of perspective to the numbers:

The total oil sands areas of Alberta are about 54,000 square miles, so when they compare it to Florida, they must be talking about the land area of Florida, and ignoring the water area.

By comparison, the coal deposits of Alberta are about twice the size of Florida, including water, but nobody expects anybody to open a coal mine half the size of Texas.

The mineable area is about 2100 square miles, which is about 1/3 the size of Metropolitan Edmonton. Keep in mind that our modern cities with their sprawling suburbs are extremely large in area (and Edmonton is built on some of the best farmland in Canada, whereas the oil sands are under peat bogs and second-grade forest.)

They wouldn't mine the entire oil sands area at once. They would be involved in a continual process of opening one area for mining while reclaiming the previous one, and moving on to the one after that. This would involve stripping the topsoil from an area, mining out the oil sand, putting the cleaned sand back, replacing the topsoil, and revegitating the area.

The rest of the oil sands would have to be developed using in-situ methods, usually using dual injection/production wells. This involves less surface area than you might think, since they typically directionally drill up to 20 horizontal wells off a single gravel pad in the interests of reducing costs.

2100 square miles is larger than the land area of the State of Delaware. How big, exactly, is Edmonton?

According to Wikipedia, the Edmonton Capital Region is 3,636 square miles.

However, it is part of StatsCan Census Division 11, which is 6,083 square miles in area. That's probably where they got the "1/3 the size of metropolitan Edmonton" ratio from.

The land area of poor little Delaware is 1,954 square miles, so yes, Edmonton is much bigger than Delaware.

On the other hand, metropolitan Houston is 10,062 sq. mi., which is bigger than the land area of 9 states and almost 5 times the mineable area of the oil sands.

When you're in oil country you've got to think big.

First, good article again, Gail. Thanks.

I might also add that IF that entire area of "tar" sands had its "oil" extracted, the resulting production (1.7 trillion bbl OIP) would be on the order of 1.5x the amount of oil which has ever been produced by all producers in all countries since the beginning of oil production in Petrolia, Ontario in the last century. It's a very long-term project even to arrive at the proposed 173 billion bbl economically extractable amount, over 170 years at eg. double present rates.

That really helps put it in context. Thanks. Do you have more information on the EROEI? I noticed your chart on the embodied CO2 for various oil resources, which is closely related. Do you think the ratio of CO2 produced in mining to the CO2 produced in finished product consumption is close how others would define the resource EROI?

I am sure you saw my comments in my first post about EROEI. Syncrude claims an EROI of 6:1. I would presume this ratio reflects the processes all the way to upgraded Synthetic Crude Oil (SCO), but I don't know how wide their boundaries in calculating the EROI is. Syncrude pays high salaries to its workers, but part of the reason for its high salaries is the fact that everything (food, clothing, cars, etc.) needs to be shipped from long distances.

Surmont also claimed a 6:1 EROI ratio, but that was only looking at their natural gas use for SAGD, and the end product was only bitumen which was then mixed with SCO to produce a salable product. It would seem like getting to a similar end product, and including other fuels, it would be lower.

Then the two data sets conflict, don't they? If the ratio of process to product CO2 is only ~4:1 and they claim an EROI of 6:1 someone is messing with one of the two sets of numbers and forgot to mess with the other, or something. EROI is such a complex calculation, and there are so many ways to define it. Isn't it surprising that the specific calculation method isn't generally mentioned when people claim EROI's for things?

I take it you are trying to use the CO2 amounts to approximate EROI. I thought it was interesting that using CO2, SAGD seems to have a lower EROI than mining, and that was the impression I got the from the quoted EROI numbers.

For mining, the 6:1 EROI is close to an EROI for making SCO. For Surmont using SAGD, the EROI I was given was 6:1 but only included the natural gas used to make bitumen. There was a double shortfall--other energy products and the additional processing to make the equivalent of SCO.

To get to gasoline or diesel, both would need to undergo additional steps, even beyond SCO. They would need to go to a refinery and have additional processing done. I would presume the CO2 calculation would include these extra steps.

But there are differences in the calculation too--even with these adjustments, it wouldn't match exactly.

Precisely, it's all a mishmash of different people calculating different things. So we clearly don't have or don't know about the definitions for EROI measures that would clear up the mess. I've tried very hard to contribute to that conversation, offering a simple whole system EROI strategy to include the energy equivalent for all the business and technology costs of resource extraction, taken to a comparable end use equivalence. I didn't get anyone to offer their suggestions so didn't go beyond concept. I'd be glad to help others working on the problem with the systems theory if you know of anyone.

Here's the reddit links for this post (we appreciate your helping us spread our work around, both in this post and any of our other work--if you want to submit something yourself to another site, etc., that isn't already here--feel free, just leave it as a reply to this comment, please so folks can find it.):

Find us on twitter:

Find us on facebook and linkedin as well:

Thanks again. Feel free to submit things yourself using the share this button on our articles as well to places like stumbleupon, metafilter, or other link farms yourself--we appreciate it!

So if Eastern Canada can show that Alberta is not doing a good enough job of regulating the oil sands, then it can perhaps get control through new Federal regulation of the oil sands.

Wow, that'd be my cue to leave. Provincial control of resources is an important and defining feature of the Canadian constitution, and people here get very, very upset when there is talk of increased federal powers. Alberta has its own office in Washington, apart from the Canadian embassy, and embarks on its own trade missions to Asia. Things would get ugly here pretty fast, and I can't say I'd want to stay.

  • The National Energy Program is not remembered fondly in Alberta, and I'm sure it would renew calls for separation if it were resurrected in any similar form.

    I raised some of the Canadian issues in an earlier post and got quite a bit of feedback from Canadian readers. One reader said:

    Dale is quite correct. Much of the west has gas & oil supplies including the North. When supplies become tight in the East, the government of the day will approach the west for access the supplies. Canada is a nation of regions and depending on the party in power and the methodology of how they do this..could cause the country to literally break into pieces. Western Canada has long suffered through "Western Alienation". Many of the decisions made by the government of the day benefit Central Canada but completely ignore the needs of Western Canada. The recent auto bailout is an example. The auto industry is located in Ontario, a province with the largest number of federal ridings. The government of the day was very fast to bailout the auto industry, but people in the farming, meat, forestry and mining industry were essentially left to fend for themselves or received scraps.

    In the future, I can see a showdown happening in Canada. The West versus East..and I suspect the outcome will be a nation that literally self destructs. It's common to see the provincial flag in Alberta which represents a sort of independent streak in Alberta. You can find the same thing in Newfoundland, an Island province in the East which also has oil. The Premier of that province has gone head to head with the government of the day..going so far as asking the citizens to NOT vote for the party in power..thought the man running that province is part of the same party.

    Funny how no-one ever mentions that from the beginning Ontario has always had to support all the regions at one time or another but w never hear a squeak of thanks. Newfoundland especially, what a joke.

    Funny how no-one ever mentions that from the beginning Ontario has always had to support all the regions at one time or another

    All the regions except Alberta - and that's the crux of the problem: Ontario has never supported Alberta financially, especially not during the 1930's when the provincial government and most of the local governments went bankrupt, and the federal and Ontario governments refused to bail them out.

    You can disagree, but the Alberta government has run the numbers to determine who was kidding whom, and it was Ontarioans kidding themselves.

    In retrospect, it was probably a mistake not to bail Alberta out during the depression, but then nobody at the time realized Alberta had so much oil.

    And now that Ontario is getting aid from the regions, the argument is starting to go over badly in other areas as well.

    Ok, granted. Probably "never Alberta" as you say. But i'd suggest that if you were in charge of the Federal govt. in 1933 and took a look out at the dryland farm areas of the west, you'd probably have made the same decision, eg. it would be simply throwing scarce money away to provide farmers in those regions support which would keep them farming on land which simply didn't even appear to exist anymore. Better they should move elsewhere.

    There's a lot more hatred toward Ontario in Alberta than any history can justify, and I've personally decided its a simple combination of jealousy and an inferiority complex. I have two brothers in Alberta with whom I've simply resolved to not discuss anything remotely political/economic/etc, else the discussion simply goes nuts. And they grew up in Ontario.

    I'm not a Conservative or a particularly patriotic Albertan, but even I can see the animosity is deeper then trivial jealousy. There are serious issues with balance of power between the east and the west and no serious effort to correct them ever occurs. If tougher times are ahead, these must be resolved.

    Let me put this East/West thing in perspective.

    Let's say you have two brothers who are dirt farmers, and you own a Cadillac dealership. They come to you and say "Can you lend us some money to keep our farm going? The bank won't lend us any, but we'll pay you back when we can".

    And you tell them, "No, that would just be throwing money away. You're wasting your time farming, but I'll tell you what - You leave the farm, and I'll give you jobs sweeping the floor of my Cadillac dealership, and a place to sleep next to the furnace".

    But they say thanks, but they prefer farming. So they scrape by year after year, and every so often they ask for money to buy the kids shoes. But you remind them you don't give charity, but there's a job waiting for them sweeping the floor of your Cadillac dealership, and you'd pay them enough to buy shoes for maybe half the kids.

    And then one day you read that there's been a massive oil strike, and most of it is under your brothers' farm. So you go to visit them, and they're standing in front of their new mansions, lighting big cigars with $100 bills.

    So you tell them that it's disgusting the way they're throwing their wealth around, and they have a moral obligation to give you half of their money. They ask you why, and you say it's because you're their brother, it's unfair that they're now much richer than you, and you once offered to help them out by letting them sweep the floor of your Cadillac dealership.

    Now, I don't know your brothers, but I'm betting their reaction would largely be in words of four letters, with numerous scatological and sexual references.


    Now isn't that a cutsey analogy. I'm sure if wasted two minutes i could come up with a similar one which would support the opposite view.

    Fat chance, Dude.

    I usually live on the west coast. A few years ago I cycled across Canada - In Alberta I started noticing 'I hate Ontario' bumper stickers, leaving Manitoba there is one 2 lane highway and a single track for about 100 km. Then two parallel roads for the next 1000km or so. I'm not aware of a pipeline, and didn't notice any power lines.

    Around Sault St Maire we entered eastern civilization which was pretty consistent to New Brunswick. Nova Scotia and Newfoundland felt a bit different again.

    My take on this - people in British Columbia know who the Canadian Prime Minister the same way we know the American guy, what's his name, and the Australian guy with the glasses. Frankly, it has no significant meaning to our lives day-to-day. Across the mountains the people of the prairie recognize the federal government but have been disgusted with it for the past 30 years (that I've been around).

    Canada is a political entity, but there are significant physical boundaries at the Rockies, Lake Superior, and the Gulf of St Lawrence. These are already psychological divisions, and could easily become national.

    Quebec is the obvious example - they've been pissed off since the Plains of Abraham thing. Every ten years they have a little hissy fit and say they're going to hold their breathe until they turn blue. I find it amusing that they never threaten to leave while the economy is down...

    Quebec is the obvious example - they've been pissed off since the Plains of Abraham thing.

    Well, as a simple conservative Ontario person who's never owned a bank (or Cadillac dealership) in my long life, I'm starting to appreciate the common sense reasonableness of our neighbors in Quebec more than most Albertans. Much better manners also, eg. I can't recall the last time anyone from Quebec stated on a public forum that they wished Toronto would drop off into Lake Ontario killing all citizens.

    Here is a figure showing how the stuff makes it into the US:

    How critical are the incoming (red) diluent pipelines to scaling?

    I don't think the need for diluent is a huge issue in the long term.

    If companies use mining with upgraders, the end product is Synthetic Crude Oil (SCO), which is the product other producers use to dilute their bitumen with. So with enough mining with upgraders, there theoretically would be no need for additional diluent from outside sources.

    Another point that has been made is that diluent is only needed for pipeline transport. If bitumen is transported by rail cars, the cars can be heated (at least for loading /unloading). So there is no real need for diluent.

    I don't think it is practical to add upgraders to in situ facilities, to eliminate the need for diluent. My impression is that in situ operations tend to be smaller. The Surmont facility was built in such a way that all the buildings and equipment could easily be moved to another location, when the bitumen had been steamed from the current location. Upgrading equipment would be less portable.

    Looking at the EIA Company Level Imports data a bit closer I see something puzzling: overseas shipments of very small quantities, say 150 kb in March '08. Using the 7.5 bo = 1 DWT figure from Oil 101, this equates to a ship of 20 DWT, which is absurd; another shipment in the same month/port/company/API/sulfur content offloaded 240 kb, so these must have been batches from the one tanker. This makes things a bit confusing when trying to estimate how many ships are pulling into a state in a given month; or, in the case of WA, how much is from pipelines versus ships. I assume nothing enters WA state from Canada via ship but have no way of ascertaining this from the data; might have to call or email someone from BP.

    Anyway, that big tanker shown delivering oil to California on the CAPP or CERA map above is just as much strictly on the drawing board as the pipeline on the BC coast exporting crude/importing Russian dilutent. This big boy would deliver VLCC loads to the Port of Long Beach, at least in theory. Certainly the crude will follow where the demand is. I'll look into this further but the only Canadian cargo of crude to enter CA in March '08 was 399 kb = Panamax tanker of 53.2 dwt. Maybe it was a larger boat that had multiple stops to make. On an annual basis Canada is behind Saudi Arabia, Iraq and Equador as a supplier of exports to PADD 5.

    Perhaps the TMX pipeline can be yet again expanded if PNW demand is on the rise; or the Kitimat port can service the Pueget Sound area with smaller tankers, as well as exporting with VLCCs to the Far East. All of this is dependent on where demand is heading.

    I'm wondering about the Russian diluent. It seems like it would be easier/cheaper to upgrade some of the bitumen in the area, and use it. Also, the long term supply would be better assured.

    Russia has the biggest natural gas reserves in the world, so it follows that they would have the biggest production of condensate for diluent.

    This sounds like something the Chinese would do: buy Russian diluent, ship it to the West Coast of Canada, pipeline it to the oil sands. Then, mix it with bitumen (dilbit), pipeline the dilbit to the West Coast, and ship it to China. Chinese refinery processes it and supplies gasoline and diesel to the Chinese market.

    Makes sense to me. Keeps your tankers full both ways.

    Russian diluent(oil) is a pipedream. Russia's Far East(Sakhalin) oil has a pipeline NOW, going to China.
    More like Alaskan oil.
    I see Sarah Palin's gas pipeline(from my house!)
    Who draws these maps?

    Canada tar sands will never produce that much by Hubbert's linearization; if tar sands have produced 5 Gb so far and there is 200 Gb and the parameter is say 9% then

    P=Q*(1-Q/URR)*k or P=5*(1-5/200)*.09=.44 Gb per year(today) so
    the peak production rate for tar sands will be P=100*(1-100/200)*.09= 4.5 Gb per year or 12.3 mbpd.

    This is TERRIBLE news!
    Tar sands cannot save Team America which guzzles 20.7 mbpd!

    Time for plan B.

    I enjoy you comments, and style, a great deal. Don't fade.....

    We are down in New Zealand, so I might be able to see you from my house using ionospheric bounce.

    At night, of course.

    I gotta blog which is largely NZ topical, but you might empathise with....

    cheers, Murray Grimwood

    ps: if u r ever down this way, there's somewhere to stay (and the red wine is passable)

    I am not convinced it necessarily follows the Hubbert curve. The limiting factor is probably investment capital (but could be something else). I would expect production to be closer to flat for a long period, as one might expect if not a lot of capital is being added.

    The amount of production depends partly on technology and partly on how the world around the oil sands holds up after peak oil. If the rest of the world gets too disrupted (can't send enough food up to Ft. McMurray, for example), then production will stop.

    Forget Alaskan oil. It's mostly gone. It's only a matter of time before they shut down the Trans-Alaska Pipeline and strand the remaining oil on the North Slope.

    Plan A may not work for the U.S. but Canada has only 1/9 of the U.S. population.

    12.3 million bpd is nearly 4.5 billion barrels per year. At $70/bbl, that would bring in $315 billion per year. Since Canada has 33 million people, that would be about $9,500 per Canadian. If the price doubled to $140/bbl (remember 2008), it would be $19,000 per year for every man, woman and child in the country. And thank you for buying Canadian.

    However, the proven reserves of 173 billion barrels would last only 38 years at that rate, so most likely they'll keep it to a much lower level. I doubt they'll ever let production go over 5 million bpd.

    So that's Plan A for Canada. The U.S. will have to find its Plan B somewhere else.

    However, the proven reserves of 173 billion barrels would last only 38 years at that rate, so most likely they'll keep it to a much lower level. I doubt they'll ever let production go over 5 million bpd.

    That's a Hubbert's Linearization so you'd see a 'bell' curve
    peaking at 4.5 Gb in about 2070 and drying up completely around 2130.
    I don't believe you will have a bell shaped curve. You'll have a quicker rise but post peak it will become a plateau before a symetrical decline. If the tar sand peak occurs in 2020 at 1.8 Gb/a(5 mbpd) then the plateau lasts +90 years.

    The only partial US plan B I can see would be an oil shale crash program, with 20 years to ramp up to 1.8 Gb/a.? to last another +200 years.

    But why would you Western Canadian sheikhs want to share your tar sand profits with Ontario? You could have over $250,000 for every West Canada man woman and child!

    I am having a hard time seeing oil shale provide an EROI of more than 1.0.

    Part of the problem is just the need for water. I contacted Shell recently, and they said that they thought they would need 3 barrels of water for each barrel of oil produced.

    If they also use electricity, that will add a whole new set of problems. (Coal from Wyoming, etc.)

    With an upgrader, they can perhaps use part of the oil produced instead of electricity, but how little will be left when the process is finished?

    Shell says they need 12 GW power plants running 8760 hours per year to produce 1 mbpd or .365 Gb/year.

    12GW x 8760h =105 Twh of electricity(.35 quads). If the electricity is supplied by inefficient(30%) coal or nuke plants then we need 1.2 quads of coal or nuke primary energy; .35/.3=1.2

    You also don't include associated natural gas(1/3 of total energy released is combustible gas) that is created in Shell's method, which could be used to generate much if not most of the power.

    The amount of coal required would be 52,500,000 tons. Wyoming currently mines 460 million tons of coal per year.

    A barrel of oil is 5.8E6 Btus so 1 mbpd(.365Gb/yr) is 2.1 E15 Btus a year (2.1 quads).

    So 2.1/1.2= a EROEI of 1.75, not 1.0.

    Actually this is an IDEAL application for stranded wind. Wyoming has 900
    Twh of wind energy potential according to the AWEA charts. I don't know if you've been to southern Wyoming but the area is has continuous strong winds thru South Pass very close to the Colorado oil shale deposits near Parachute.

    Imagine the entire area being covered with 36000 1 MW wind turbines covering 4000 square miles. It would produce probably around 108 Twh of electricity with zero carbon emissions; 36000MW x 3000hr =108 Twh

    Now it is true that wind is intermittent but the oil shale material is actually an ideal insulating material so electrically generated heat will not readily dissipate into the surrounding rocks.(Fire bricks are used to store heat in night-set back electric resistive furnaces).

    So in this case the EROI would be be quite high probably around 10 if you believe that the EROI of wind is 18.

    Basically, this is a no-brainer and if GHG legislation passes I believe that investors will happily invest $60 billion dollars($1M per MW wind plus transmission lines) to get carbon credits and a stake in the last great oil reserve on earth(600 Gb).

    You say that 3 barrels of water will be needed per barrel of shale oil, so 3 x .365 Gb oil is 1.1 Gb of water is required or 46 G gallons of water or 6.1 billion cubic feet or 140,000 acre feet.

    As I pointed out before the Glenn Canyon Dam/Lake Powell loses 700,000 acre feet per year in seepage and evaporation. It produces 3.2 Twh of electricity per year. Ecologists say it is ruining the Grand Canyon.

    I am suggesting that the dam should be removed and water better used for Colorado oil shale rather than be evaporated in Lake Powell so rich guys can motor about in a remote wilderness.

    As far as I can see no upgrader is needed for shale oil( no upgraders existed in 1905 when oil shale was being mined and retorted in Scotland), where as it is obvious that a diluent is required to pump tar thru a pipeline.

    I get the impression that the people at ASPO are either stupid or
    deceitful on the topic of oil shale.



    It IS practical to build upgraders for in-situ projects! There were several such projects in advanced planning stages when the bottom fell out of the credit markets. Upgraders can be simple affairs (just thermal cracking without hydrogen and dumping the pet coke) all the way up to full refineries with a full stream of petrochemicals being produced along with synthetic crude oil. Most of them were planned for the Edmonton area. A small pipeline from the in-situ site to Edmonton is sufficient to transfer the non-upgraded bitumen to Edmonton and then the full upgrading could occur there. A company could own the upgrader and in-situ facilities or just the upgrader.

    It all boils down to capital. If Peak Oil is as bad as we think, there will be more upgraders built in the near future in Edmonton to augment the existing ones.


    Thanks for the tip! I learn something new every day.

    What's immediately obvious from this web page (Alberta Govt. PAR prices by month) is that oil sands product is extremely sensitive to the monthly market price for light sweet crude. The oil sands producers are paid WTI light market price less 3 flat rate overhead costs (shipping from field to upgrader, upgrading, shipping from upgrader to market hub). The flat rate overheads don't change with market price, and are each significantly above Cdn$10/bbl meaning if the oil price drops below about US$35 to $40 / bbl, the extracters are making nothing to pay their costs of operation or investment (assuming they're producing Ultra Heavy grade).

    I think the document you link to has to do with royalties, which are indeed linked to the price of crude.

    Synthetic Crude Oil (SCO), which is a light sweet crude oil made by Syncrude and by other companies with upgraders is a traded commodity that sells for a price similar to that of West Texas Intermediate (WTI).

    Bitumen can be diluted with various diluents. Surmont diluted their bitumen 50% /50% with SCO. The resulting heavy crude oil is marketed as Western Canadian Select Blend. The relativity of this product to WTI can vary. According to this recent article:

    High prices for heavy oil defy logic

    Usually, the Canadian benchmark heavy crude, Western Canadian Select Blend, is priced about 30 per cent lower than West Texas Intermediate crude, the US light crude benchmark. But last month, the gap narrowed to less than 15 per cent.

    Elsewhere in the article, it says that the smaller discount is likely to persist, since there is a shortage of other heavy crude for various reasons (Mexican production declining, for example).

    if the oil price drops below about US$35 to $40 / bbl, the extracters are making nothing to pay their costs of operation or investment

    Those are actually the parameters for calculating Alberta government royalties. The government does not let producers calculate their own costs, but gives them a fixed number to use.

    What it is really saying is that if the price drops below $35-$40/bbl, the royalty rates drop to zero. That affects the government more than it does the companies, many of whom are making money on the refining and marketing end of the business.

    Hi Gail, interesting and important series.

    A few comments:

    I too would like a clarification on the EROEI numbers for oil from tar sands in Alberta. For a while numbers of as low as 3:1 were kicking around. Now the popular number seems to be 6:1 but keep in mind that this is official industry figure which could very well be short of reality. When you consider the incredible energy cost of surface mining, retorting and reclamation and the stated need of prices of 66 to 80 dollars a barrel needed to make profit, I wonder if the 6:1 EROEI is reasonable. Underground technology has got to be very expensive too because of the vast amounts of energy needed to heat large quantities of sand to a temperature needed to get the keratin to flow. I assume the 66 to 80 dollar figure is meant to apply to under surface extraction too as it is a given fact that the industry will switch more and more to the underground technique.

    So if Eastern Canada can show that Alberta is not doing a good enough job of regulating the oil sands, then it can perhaps get control through new Federal regulation of the oil sands. With that control, the Eastern part of Canada can be assured of getting oil.

    When I was a student at University of Alberta, gasoline was dirt cheap in Alberta with its local oil production. The University of Alberta had a surplus of about 200 million dollars in its coffers due to oil revenue. The federal government didn't think it fair that the east of Canada should pay much higher prices (even though Alberta had to put up with the environmental degradation, etc.) for gasoline and forced prices in Alberta to be essentially on par with the east.

    Also when the national health plan was put in place (a very good thing for the nation and its people - take note America) they forced Alberta to take part. As background, Alberta already had an excellent "universal health plan" in place within the province paid for partly by oil proceeds. Albertan's were not very happy with that change either as it adversely affected their pocket books.

    My point being that there is plenty of precedence for national government forcing major issues on provinces. Thus there is no question that the quoted action (above) must be taken seriously and that the "nationalization" of a province's out put could happen.

    The federal government didn't think it fair that the east of Canada should pay much higher prices (even though Alberta had to put up with the environmental degradation, etc.) for gasoline and forced prices in Alberta to be essentially on par with the east.

    I don't think any federal government initiative has anything to do with fuel prices in Alberta, other than the GST universal value added tax. It's more likely to do with local refinery supply and demand issues. Fuel prices in Canada are not regulated AFAIK, and are set by the retailing and wholesaling companies in a theoretically competitive manner.

    Alberta already had an excellent "universal health plan" in place within the province

    Did that plan you refer to precede T. C. Douglas's introduction of single-payer medical insurance in Saskatchewan? I'd like to learn more about the history of the Alberta plan. Have any references? I note from this website,$File/FACT_SHEET.PDF?OpenElement

    1935 - The Alberta Government supports and agrees to subsidize a prepayment
    insurance program in 1935 and again in 1942

    1948 – Alberta sets up Medical Services (Alberta) Incorporated (MSI) offering medical,
    surgical and obstetrical services to subscribers

    1969 – The Health Care Insurance Act passes introducing compulsory federal medicare
    to the provinces - MSI ceases operations on June 30 - AMA and CPSA become independent of each other

    Sounds like you're referring to MSI, but note "to subscribers" implies that there will be some group of people not covered, payments required, and perhaps procedures not covered by "medical, surgical and obstetrical services".

    With respect to EROI, I think there are a couple of factors involved.

    First is that the process is getting better over time. Canadians have been working on the process for 40 years. There is no issue of finding the material, at this point. The point is efficiently processing it, and processing is getting better. One booklet put out by the province of Alberta states, "Per barrel of oil, carbon dioxide emissions have been reduced by 45% since 1990." Companies are constantly looking for ways of doing things more efficiently. Also, if way of extracting "rare earths" from tailings is developed, there would be the possibility of getting two products for the effort of one. So it is not unreasonable for the EROI to go up over time--in fact, that is what one would expect.

    The other side is that EROI leaves out a lot of indirect energy. People's salaries don't come into play, even though energy is quite a bit of what salaries go to pay for. It is necessary to borrow huge amount of money to fund these projects, and the borrowing costs are used to pay for the energy costs of a lot of financial executives, plus profits that eventually go to buy energy products for many of the recipients. But from an EROI point of view, this has no cost.

    You expect the EROI to go up, but isn't that only possible while the highly concentrated pockets are being exploited? The EROI will move down as soon as lesser grade deposits have to be extracted. This is what is happening in the oil industry for some time already with ever deeper wells at more difficult locations, longer pipelines, heavy oil and the exploitation of the Canadian tar sands...

    I am not sure how it will work out in practice. The total resource in place is huge, and the technology is evolving rapidly.

    There are several variables that are important--resource concentration, depth of the resource, and thickness of the resource are ones I am aware of. The projects right now were probably chosen to be in as close to sweet spots as possible.

    I would hope with the amount of the resource, we are not getting too close to the difficult to produce stuff very quickly. If the 173 billion barrel amount was picked to be easily producible, we could be looking at flat to increasing EROI for quite a while.

    I certainly do NOT accept the 6:1 EROEI number!! The "standard" steam/oil ratio for SAGD is at least 3.0. That means the use of about 1.0MMBTU of natural gas just to heat and extract 1 BBL of bitumen from the ground. Since 1BBL of Bitumen has about 6.0 MMBTU of energy is 100% of the barrel were usable, that means about a 6:1 EROEI IF there were no other energy costs involved!!! (which is certainly NOT true). So in addition the energy cost of extracting the bitumen from the ground, there is the energy cost for upgrading (significant especially if you want high grades of oil with lots of hydrogen added, which costs MORE natural gas). Another energy cost is the pumping of the water, the cleaning of the water, the pumping of the upgraded bitumen and finally, the loss of the energy content from the "bottoms" of the bitumen barrel. There is too much carbon in bitumen (about 85%) compared to regular crude oil so you have to add hydrogen or get rid of carbon (pet coke) or both. Another way to look at it is that 15-20% of the barrel of bitumen (the high carbon, high molecular weight bottoms with most of the metals and the sulfur) is not usable, so that effectively gets rid of more EROEI. The most advanced schemes (like Nexen/Opti-Long Lake) are using the water shift reaction to add the hydrogen from water the "bottoms" and capturing most of that lost "bottoms". HOWEVER, there is another energy penalty and the net gain is mostly financial (producing more barrels of syncrude per unit time) and not EROEI. I would estimate a REAL EROEI of about 3:1 for the SAGD process. I am reasonably certain that the Bitumen mining operations are not more efficient.

    I am NOT saying that the EROEI can never improve. It probably will because the SAGD process is being improved continually. However, the higher grade deposits are near the surface and the deeper, less concentrated deposits will require more steam and more steam losses, so the technology improvements will probably allow lower grade bitumen deposits (oil sand) to be accessed, but will probably not improve EROEI substantially (a well known Peak Oil phenomenon to anyone who reads TOD). I would strongly suggest that anyone citing an EROEI of 6:1 for the Oil sands is not including all of the losses, but just the extraction process itself.


    Thanks for the in-dept look at oil sands, Gail.

    I was struck by the implicit net-energy data in one of the graphics in your post:

    No idea where they (CERA, the source listed in the caption) get their data, or what is included. However, if we *assume* it's accurate, then it seems like a decent proxy for EROEI: the energy required to produce given oil as a portion of the energy of that oil produced should be *roughly* comparable to the CO2 produced by each, no?

    If that's accurate, then it looks like a rough net-energy of 4-5 for Saudi Light and less than 2 for Canadian oil sands. That seems like too little difference from Saudi Light to oil sands to me, but I'd be curious to hear others' thoughts (or if anyone has a report from CERA that discusses their methodology here)...

    If all studies used same boundaries your estimate might be close, but:
    a)much of saudi investment was made long ago, whereas tar sands needs to be built now - ergo there is the fixed vs marginal EROI issue again
    b)measuring the CO2 alone wouldnt account for say 10x more metal and equipment (and the wide boundary emergy costs of that equipment) for the more industrial oil sources if it was only measuring emissions at the plant. I assume they (CERA) looked just at direct energy and didn't include the carbon sink loss from land use change - equivalent land in KSA wouldn't sequester much carbon - but I couldn't find the Sierra Club report.

    Still your observation is interesting in that we have no EROI #s for Saudi (no data) - Charlie Hall has suggested it was at one time over 1,000:1 - we just don't know. What is relevant is what it is now, and going forward.

    Here is similar GHG graphic for Coal-to-liquids:

    It really makes me wonder where CERA got their numbers, and what methodology they used. Part of me thinks that this must be little more than conjectcure (especially since all the boundary issues with EROEI also appear here, and I doubt CERA magically solved them without telling us about it). Either way, I can't see how CERA would consciously publish something that implies a net energy of less than 2 for canadian oil sands--that's tantamount to an admission that oil sands are actually a means of converting other forms of energy to liquid fuel, not an energy source. Of course, as Gail points out below, we don't really know what other gasses they've included here that skew the calculations. Gail: was there a larger report where you found this graph that might explain the context better?

    This is all part of the CERA report that is free with registration. I don't remember that the report gave a lot of information. It covered a lot of topics besides global warming gasses.

    The report lists the 11 publicly available reports (e.g. Suncor, Syncrude, other corporate sustainability reports) that it averaged to reach this data. While it does make clear that they're comparing "apples to apples" (same boundary, presumably), and that this is "well to retail gas pump," it doesn't make clear where exactly the "well" boundary is drawn. E.g., does it include cost of establishing water supplies, does it include cost of durable equipment, etc. Finally, it appears that this metric also cuts off at delivery to the retail pump--meaning that remediation costs are not included.

    Still, for those interested, there are additional interesting graphs on pages III-4 and III-5 of the report. They show, essentially, that oil sands have about 2 times the "well to retail pump" greenhouse gas emissions as Saudi light or Brent. IF it's reasonable to use greenhouse gas emissions as a proxy, that means roughly 1/2 the EROEI.

    oil sands have about 2 times the "well to retail pump" greenhouse gas emissions as Saudi light or Brent

    I'm surprised at your conclusion, considering this statement is contained in the executive summary.

    Growth in the Canadian Oil Sands: Finding the New Balance (Executive Summary)

    The well-to-wheels basis includes GHG emissions from oil extraction, processing, distribution, through to the combustion of the refined products, such as gasoline and the resulting emissions that exit through the tailpipe. On this basis total GHG emissions from oil sands are approximately 5 to 15 percent higher than the
    average crude oil consumed in the United States. That is, about 5 to 15 percent more carbon dioxide (CO2) in total is released into the atmosphere as a result of using oil sands instead of an “average” crude oil.

    I think Jeff made a mistake in what he is saying. I think he is only looking at the incremental piece added for production and refining (striped part), not the whole well to wheels comparison.

    Also, Jeff's statement is more or less a "worst to best" comparison, while the paragraph at the beginning of the book is a "worst to average comparison.

    So I think the CERA statement is correct (assuming their calculations are correct).

    That 5 to 15% higher applies to "well to wheels," not "well to retail pump," and compares oil sands to "average american crude," not Saudi light or Brent.

    They make two sets of measurements. The first is "well to wheel" which includes the gasses produced from burning the fuel. The second is "well to retail pump," which gets right up to where the fuel is burned but doesn't actually include the gasses from burning the fuel. They also are stating that oil sands are 5 to 15% higher than the average crude consumed in the US--which includes much oil sands and heavy crude. My comparison was to Saudi light or Brent.

    If you look at the report on page III-4 (I think, could be III-5), you'll see the "well to retail pump" graph that I'm refering to. They show around "75" for emissions to produce Saudi light well to retail pump, and around "150" for the mid-range for oil sands.

    I can't comment on CERA's motive here, but if you ask me it's a bit silly to headline with the "well to wheels" number because that figure is dominated by an irrelevant value--the "well to retail pump" figure is a better measure of the greenhouse gas differences. It's a bit like saying the cost of buying an F150 and driving it 1000 miles was only 5% higher in summer 2008 than it was in summer 2009 when what we're really trying to compare is the change in gas prices...

    I think the issue is if you are going to decide to extract and refine oil, you are also going to burn it, so it is fairer in deciding whether or not to extract a particular type of oil to look at the well to wheels life cycle. The Sierra Club looks at some piece of the process (extraction (or extraction/refining)) and makes comparisons on this basis. The Sierra Club in the same e-mail that I quoted from earlier said:

    In addition, greenhouse gas emissions from tar sands production are three to five times that of conventional crude oil.

    As far as I can see, it is impossible to reproduce those numbers from CERA numbers, using just the well to wheels numbers. The most you get, comparing the figures you get is 150 / 75 = 2.

    I tried to see if I could find what reports the Sierra Club numbers were based on, but I couldn't figure it out--the links they provided also lacked detail. As best I can figure out, they looked at CO2, methane, and some other greenhouse gasses, and somehow got that range, by looking at them separately for some portion of the process. But even that wouldn't make sense.

    CERA shows footnotes, showing what publicly filed reports their data comes from, so that seems to make a little more sense.

    Does anyone know where there is a link to the report supporting the Sierra Club numbers?

    I couldn't find the Sierra document online. I did have an old Pembina report, Oil Sands Fever, which has a section starting on page 19 claiming three times the GHG impact from oil sands as conventional. Their source (#107 in footnotes) is:

    Average intensity for conventional oil derived from
    information provided in Canadian Association of Petroleum
    Producers, A National Inventory of Greenhouse Gas (GHG),
    Criteria Air Contaminant (CAC) and Hydrogen Sulphide
    (H2S) Emissions by the Upstream Oil and Gas Industry (2004

    Table 2. GHG Emissions For Year 2000 From Subject Oil
    and Gas Activity by Province/Territory, Pg. 19, Volume 1.
    Average intensity for oil sands SCO was calculated as the
    average of operating surface mining operation intensities
    (Suncor, Syncrude, Shell – 2004 data from respective
    Sustainability Reports) and the “SAGD to SCO”
    emission factor provided by Len Flint, Lenef Consulting
    (personal communication).

    It makes sense this is the research that the Sierra quote came from (same figure 300%). Its a very detailed report and I don't have any time to peruse it - help welcomed. But if this is the source, I wonder how the technology has improved in past decade.

    Thanks. At least it is half of the Sierra Club numbers--the 300% figure. The 500% figure is totally mystifying.

    And this is an old report, based on old technology.


    I looked at the documents you referenced, and tried to see if I could find any others. I couldn't. I think the old Pembina Oil Sands Fever report is the most recent one available. As you recall, the Sierra Club statement was

    In addition, greenhouse gas emissions from tar sands production are three to five times that of conventional crude oil.

    The issues I see are

    1. Based on the labeling on the graph of the OIl Sands Fever report, the green house graph comparison relates only to oil production --not refining, or transport --or anything else. For Syncrude, I presume production is defined as all of the way through upgrading, since this is the data is given in the sustainability reports. So this is being compared to the simplest oil production, and a high ratio being derived.

    2. Nowhere in the material that the Sierra Club links to (or in other Pembina Institute material) can I find any documentation that suggests a factor of five is correct. The highest factor that there is really any documentation for is the factor of three, and that seems to come from the Oil Sands Fever report linked to above, and footnot 107 that you cited.

    3. I am wondering if the Oil Sands Fever report is making an apples to oranges comparison. Footnote above says that a comparison of numbers from the individual company sustainability reports was made to numbers from Table 2 of the CAPP report referenced above. When I look at my copy of Syncrude's 2007 sustainability report, a foonote says "Emission calculations for the purpose of federal and regulatory reporting will differ, as certain sources of emissions are excluded."

    4. The CAPP report referenced uses 2000 data, which is at this point quite out of date. Even the 2004 sustainability reports referenced are getting out of date.

    5. The Oil Sands Fever report has no data at all on SAGD. Instead, according to Footnote 107, "the 'SAGD to SCO' emission factor provided by Len Flint, Lenef Consulting (personal communication)." This is a rather "iffy" source for numbers that are crucial to the calculation. The name Lenef Consulting seems to be derived from Len Flint's name. The consulting firm has no supported website.

    6. Table 2 of the CAPP report is a table of emissions by province for various activities. To me, it doesn't seem reasonable as a basis as comparison, since it is not on an emissions per barrel basis. It is hard for me to see that it would be helpful at all, without other data which is not referenced in the footnote.

    I can understand that if you take a small enough slice of front end costs and compare them, maybe you can get a high ratio. I don't think it is a very good way of making a comparison, and the documentation leaves me wondering whether even this was done very well, or if the numbers are still relevant now. And I still have not found a reference for the supposed five times number.

    Gail and Nate: I want to thank you for all the effort you've put in to clearify the issue. I think the positions you've arrive at are rational and supportable. So IMHO claiming that the product of the oil sands is incredibly more damaging to the environment than eg. the product of Venezuela or California, as the Sierra Club appears to be doing, is simply not supportable with any facts you experts could find, which position co-incides with my subjectively derived conclusion. I note that much of the "light" crude marketed by S. Arabia has already been upgraded from heavy crude by in-country upgraders, much of the crude exported by Venezuela is nearly as "heavy" (eg. long-chain) as the raw product from the oil sands, and the future of crude oil worldwide is much heavier crudes from all suppliers.

    In essence, if the US Gulf Coast doesn't get the heavy oil it specializes in from Alberta, it will get it from the many other sources of heavy crude, and the result will be of no practical difference.

    no EROI #s for Saudi (no data)

    That's the problem with Saudi data - there isn't any. They're exempt from Kyoto Protocol limits, they aren't required to monitor greenhouse gases, and they don't report the underlying data on their oil production.

    And they consider oil production data a state secret, so it's not safe to go snooping around their oil fields looking for better information. You might get a midnight visit from the secret police.

    So anything involving energy or emissions calculations about Arabian oil would have to be considered speculative, at best.

    I don't know if it is material to the calculations, but these are total greenhouse gasses, so they are putting in other gasses besides CO2 in this calculation. Vented natural gas comes out particularly badly for example.

    We are used to seeing net energy calculations on a less complete basis. So I suppose it is theoretically possible that a factor of 2 is just barely acceptable (but does explain the problem with getting a high enough price.)

    Here's another report on Life Cycle Analysis of North American and Imported Crude Oils

    Emissions from oil sands comparable to other crude oils

    Calgary... Two independent studies have found direct emissions from producing, transporting and refining oil sands crude are in the same range as those of the other crudes refined in the United States.

    The Life-Cycle Analysis of North American and Imported Crude Oils is based on two independent studies that comprise the first robust comparison of domestic, imported and oil sands crude processes in U.S. refineries. The research, conducted over the past year by U.S.-based consulting companies Jacobs Consultancy and TIAX LLC, was funded by the Alberta Energy Research Institute (AERI).

    Full report is here:

    Thanks! For what it is worth, this is a summary slide from a presentation you link to. There were two different Life Cycle Analyses (LCAs) done by two different consulting firms, and this is a slide summarizing the results.

    The slide makes it appear that oil sands oil is not much worse than the other types of oil, over the life cycle.

    Any recent news on building a nuclear plant for the oil sands? It does make a lot of sense, while nuclear is still expensive for electricity, I bet you it's pretty competitive when it comes to thermal heat. I mean hell you can't (I don't think) have supercritical nuclear plants, meaning they can't really get above 33% efficient for electricity. It'd save a hell of a lot of natural gas for, well, anything else.

    Just today we read that China is paying C$1.9 billion for an oil sands investment.

    And that's part of an ongoing strategy: China is taking advantage of the current global economic downturn (and American inattention) to buy up all of the available oil reserves of the world.

    I don't think most Americans realize what is going on. When the American economy bounces back and the U.S. goes looking for new sources of oil to supply its growing economy, China is going to own all of it.

    In this case, PetroChina paid about 60 cents (US) per barrel to buy up controlling interest in about 5 billion barrels of Canadian non-conventional oil reserves. That's more than one years oil consumption for the U.S. (or two years for China) in one deal. There have been several other deals like that recently, involving Africa, Iraq, and other places.

    Of course, they'll have to pay $50-$60/bbl to produce the stuff and get it to China, but I think they think that's a reasonable price to pay for energy security.

    War is an extension of diplomacy.

    War is an extension of diplomacy.

    It's a lot cheaper to just buy things. The U.S. spent about $600 billion on the war in Irag, and how much oil did it get for that money - none.

    On the other hand, China just walked into Switzerland with a wad of money and bought a bunch of Kurdish oil (controlled by a Swiss company) without anybody getting shot. It's important to have quick access to those Swiss bank accounts when doing this kind of thing, you know.

    It's a lot cheaper to just buy things.

    Yeah, but a perceived willingness to engage in war might help those things stay bought. Sooner or later someone who "sold" an asset to China will try to confiscate it back again, in the tried-and-true tradition of the so-called 'third world'. Will China respond robustly, or will they simply roll over and surrender as politically-correct Europeans and Americans often do nowadays? Stay tuned.

    I would say later rather than sooner. The current owners of North America would sell the entire place lock stock and barrel to China if the price was right and the sheeple could be kept complacent. IMO, unlike China, this place doesn't have anyone in a postion of effective power who is actually looking out for the overall economy or society.

    Cheaper to buy for now,yes.

    But in the longer run,business and diplomacy will move on to the natural extension of war.

    For the time being we are blindered and hobbled by our own squeamishness at the sight of dead bodies on television and paralyzed by our own good nature and sense of fair play-we have managed to go to war over oil with hardly a dead American body in sight-the loss of a local man is big news in small town America-we BUY the oil we are fighting for,in addition to trying to help the locals there live a little better.

    Before too long the gloves will come off and the knives will come out.Of course we will continue here in America to dress our actions in whatever fig leaves we can.Perhaps the West Europeans will do the same.Unless they resort to nukes,thet're finished militarily. All of them together don't have enough coal ,iron ,and oil to fit out one good sized army and move it a thousand miles and keep it in the field for a year.

    It is not likely the Russians or the Chinese will prove to be so squeamish,and we have the recent historical examples of Germany and Japan to consider,not to mention the mustier examples of the English,the Spanish,the French.....The US too,but we were too isolated or too young to get in many good licks,excepting against the Mexicans and our own original citizens.

    The Chinese are biding thier time and when the time is right we will find that thier position is strong indeed.We have never hesitated to throw our wieght around.They won't hesitate either,once the balance of power shifts in thier favor.

    It's hard to keep track, but it appears that over the last few years the Chinese have locked up well over $50 billion worth of oil worldwide. This does not include deals done under the table with countries that don't like to talk about what they're doing.

    None of this involves any American company, and it has all been done without launching a single air strike or driving a single tank onto another country's territory.

    On the other hand, the United States' major concern is that the Chinese keep buying Treasury Bills. Otherwise there's no way to finance the economic recovery program. Unfortunately, when the economy does recover, Americans may find that there's no oil available to keep it going.


    Thanks for your insights. I hadn't figured out the quantity.

    It seems like there is such a belief that wind turbines / local gardens / electric cars / technology in general will save us that we don't need to worry about fuel for farm machinery and other practical things. If infinite investment funds were available, and if we could compress all of the needed changes into a year or two's time, this might really work. I would like to hedge my bets a bit though, and make certain we have some supply for essential machinery--as the Chinese seem to be doing.

    It seems like there is such a belief that wind turbines / local gardens / electric cars / technology in general will save us that we don't need to worry about fuel for farm machinery and other practical things.

    We have decades of our own production of oil left for things like farm machinery.

    We certainly don't have decades left to maintain our perceived need to commute in single passenger SUV's and transport the average bit of food 2,000 miles.

    The tar sands will ease the crunch so very slightly, and that at a significant cost to our environment. What I find depressing about this thread is that people acknowledge the extra CO2 produced through the production of tar sands without seeming to realize that it's a problem.


    Correction: 5 billion barrels is over 1 years IMPORTS for the U.S., or over 8 months total consumption. For China, it's over two years total consumption.

    With respect to the use of EROEI as a figure of merit for the productivity of energy production I would like to point out once more that the economics of energy production cannot be understood without accounting for the opportunity cost of the non-energy production resources (e.g. labor, farm land, freshwater) consumed during the production process. These opportunity costs have a first order effect on the productivity of the energy producing process, not a second order affect. They can never be neglected.

    It is really the labor, capital, freshwater, etc. intensity of net energy production that determine the economics of energy production from tar sands. The input energy can be made a proxy for these other inputs under specific assumptions. Specifically it is possible to show that the productivity of different energy producing processes is proportional to Net Energy/Input Energy (NEROEI=EROEI-1) under the assumption that the opportunity costs (hereinafter referred to as OC) of the non-energy inputs is linearly proportional to the input energy. However, it can easily be shown that strict linearity between the input energy and OC does not hold so that caution must be used in using NEROEI as figure of merit for energy productivity.

    Let me illustrate these principles with some concrete numbers. Energy costs are most easily understood if we think in terms of batch processing where the batch size is the gross output of energy. Consider an energy producing processes with batch size of 0.63GJ and an input energy of 0.03GJ. The net energy output per batch is 0.60GJ and NEROEI=20. Five batches will produce 3.0GJ of energy which the average weekly per capita energy consumption of Denmark. I do not choose Danish levels of energy consumption for any particular reason; I just want to give a sense of scale to the energy production process.

    To illustrate the importance of non-energy inputs I will consider a case where labor is the limiting non-energy resource input. Suppose that when 1 hour of labor is combined with 0.03GJ of energy an output of 0.63GJ results (NEROEI=20). Under this assumption it requires five hours of labor to produce a week's worth of energy for one person. Given the fact that a substantial fraction of the population is not working on a full time basis this energy is fairly expensive in labor terms.

    Now suppose our energy source runs dry and we are forced to switch to a second source for which the combination of 1 hour of labor with 0.03GJ of energy produces only 0.33GJ (NEROEI=10). Now it takes 10 hours of labor to produce one week's worth of energy for one person. This represents a substantial increase in cost.

    If the alternate energy source combined 1 hour of labor with 0.03GJ of energy and produced only 0.06GJ (NEROEI=1) then 100 hours of labor would be required to produce a week's worth of energy. This energy source is too expensive to be able to produce 3.0GJ/person, and its use would require a substantial reduction in total energy consumption even if energy from this source was available in great abundance.

    Note that the labor productivity scales with NEROEI. This scaling results from the assumption that the ratio of the labor input to the energy input is a constant. If this ratio changes then NEROEI alone cannot be used to compare productivity between different energy sources. For example if an energy source exists for which 2 hours of labor combined with 0.03GJ produces 0.63GJ (NEROEI=20) then NEROEI alone does not determine the productivity of this process relative to the three processes considered previously.

    I have been considering energy production processes limited by a single non-energy related production input just for conceptual simplicity. In the real world the opportunity costs of multiple resource inputs would have to be summed up. For example biofuel production might involve labor, land, and irrigation water inputs. However, the same concept of productivity of the energy producing process can be applied. I define the Energy Productivity Ratio (EPR) as:

    EPR = Net Energy / OC

    where OC is the sum of all the opportunity costs of the required non-energy resource inputs. Suppose that OC is proportional to the input energy:

    OC = c×Input Energy

    In this case we find:

    EPR = (1/c)×NEROEI

    NEROEI can be used as a figure of merit for energy productivity only if c is constant for the various energy producing processes being considered. I do not know how to test whether or not c is approximately constant for various real energy sources. However, it is clear than c cannot be a universal constant since it can vary for the same energy source depending on the overall scale of energy production.

    For example consider electricity generation from concentrated solar power using water cooled steam turbines. For relatively small amounts of generation the opportunity cost of water consumption may be fairly low, but if the industry is scaled up to terawatt size then this opportunity cost could increase by a large amount. If you use NEROEI as a figure of merit and ignore the increase of water use opportunity costs with system size you will reach incorrect conclusions about the economic potential of this energy source.

    Thanks for your insights! Inputs other than energy are a huge consideration in the oil sands.

    Capital is a huge consideration on these new projects, especially with the credit crunch.

    Labor is a limiting issue, too. I gave an example of adding 1 million bpd of mining production, which in some sense is not a huge amount, compared to the world's needed. But it one attempted to do this all at one time, the number of workers required would be difficult to find, and there would certainly be no place to house them or for their children to go to school.


    let me check my understanding.

    This amounts to saying three things: (1) that EROEI is limiting only if there are no other, tighter, limits.

    No argument with that. No single number will adequately represent the merit of any real-world investment decision. There is no single recipe that can be followed every time by investors.

    (2) that which input into an investment turns out to be the limiting one depends on scale.

    No argument about that, either. It seems intuitively obvious ... but may be hard to prove.

    (3) that scale, the rate of energy return, (and therefore also the return on capital invested, labour invested, land invested, etc) is ultimately limited by the maximum available rate of application of the limiting input.

    ("Input" is intended widely, including, for example, ecosystem services of chemical pollution neutralisation and physical system services of heat dispersal.)

    And again, no argument with that. A nice symmetry with (2).

    Minor nitpick ... your solar thermal example is a little too simplistic. If water availability is limiting, an alternative design that uses dry cooling (basically, big car radiators) would be considered instead, in order to achieve the desired scale. The EROEI and money return on investment may be slightly lower than if water were not limiting, but still better than any alternative investment.

    The point being, it's often possible to "rescue" an investment calculation by slight modifications in technology selection. ... The devil is in the details, there are no right and wrong answers, etc. etc. ;-)


    A couple of points.

    Your statement

    (1) that EROEI is limiting only if there are no other, tighter, limits.

    is not correct. EROEI (or NEROEI, which I believe is a better parameter) can never be used as a figure of merit for energy production without reference to the required inputs of other resources. To demonstrate the truth of this statement consider the case where the inputs of non-energy resources are trivially small. I call this case the magic wand limit of energy production.

    Suppose that instead of building, maintaining, and operating machinery to produce energy, I can just wave a magic wand over a batch energy and 'pouf' the energy disappears and an instant later a larger batch of energy materializes in its place. I will suppose that the amount of time required to work this spell is small, but finite. And I will further suppose that the amount of time required is proportional to the input energy. To be specific suppose that 1 second of magical labor is combined with 0.03GJ of energy to produce 0.63GJ of energy (NEROEI=20). In effect I have reduced the ratio of labor to input energy by a factor of 3600 compared to the example given in my original post.

    What happens to the economics of energy production? It now takes me only 5 seconds to produce a week's worth of energy (3.0GJ). This energy is effectively free. The difference between spending five seconds a week to get all of the energy I need and having someone deliver it to my door free of charge is negligible.

    Now suppose my magic wand breaks, and I make a new wand which is not as powerful as the old wand. With my new wand I combine 1 second of labor with 0.03GJ of energy and produce 0.33GJ of energy (NEROEI=10). It now takes me 10 seconds to produce 3.0GJ. The extra five seconds is a nit. My total productivity will not be significantly effected.

    Even if my new wand combines 1 second of labor with 0.03GJ energy to produce 0.06GJ (NEROEI=1) the required 100 seconds per week will have almost no effect on my economic production.

    In the limit in which the inputs of non-energy resources are so small that they make no difference, the value of NEROEI makes no difference either.

    Notice that I am assuming that our energy appetite is limited and that we could be satisfied with 3.0GJ per person per week. If, on the other hand, our demand for energy is infinitely elastic, so that the cheaper energy becomes the more we consume, world without end amen, and if greatly increased energy use does not create negative externalities, then in the presence of such ultra-cheap energy sources we would go on an energy production frenzy and stop only when labor became a limiting factor of production. In that limit the same considerations as in my original post would apply, only now the energy scale would be three and a half orders of magnitude higher.

    In reality, if such ultra-cheap energy sources existed negative externalities would probably limit energy use rather than the direct cost of energy production. Fortunately (or unfortunately depending on your point of view) I do not think that the ecological dilemma of ultra-cheap energy is one that you or I will have to face.

    With respect to CSP you are right that my example was over simplified in order to make a point about the importance of the total resource inputs to the energy production process. Air cooling will work, but will alter the economics of CSP to some degree since the up front capital cost (and presumably the input energy) will increase, and the operating efficiency will decrease in some degree.

    The comment about investing 126 billion dollar into a 1 million barrel per day production facility made me do a quick calculation on whether it is economic to burn this oil in electricity plants. I know you can manipulate these numbers each way, but still, lets see:

    What if we invested that money in purchasing solar panels. Several manufacturers are already capable of producing solar panels for less then 1 dollar per Wp (like the American First Solar) so lets say we use this money to build a 126 billion dollar solar belt across the south of the USA. Lets assume: in southern regions 1 kWp can produce 2000 kW/year, one liter WTI contains 10 kWh of electricity. With these numbers a the solar plant would produce 345 million kWh per day (on average ofcourse) and the oil produces 1590 million kWh per day, a 4.5x difference.

    Now, solar panels provide electricity for at least 30 years (most panels already have a power guarantee of >80% of initial power after 25 years). But what about the 1 million barrel/day production line? How long will that be able to sustain that amount? Maybe 10 years max? If so then the oil production line only produces 1.5x the amount of energy per dollar then an equally expensive solar farm. Isn't that a tremendously low factor for such a hi density fuel that only has to be pumped or dug up? It makes me think that the margins on oil are rapidly turning towards solar..

    But what about the 1 million barrel/day production line? How long will that be able to sustain that amount? Maybe 10 years max?

    No, and that's the key to the importance of the oil sands. 1 million bpd is 365 million barrels per year. If the total resource (producible at today's prices using today's technology) is conservatively estimated at 173 billion barrels, it will last for 474 years at that rate.

    Most likely the production rate will grow higher than that. If you assume a rate of 5 million bpd, that would cut the lifetime to 95 years. However, 95 years gives ample time for somebody to work on a method of producing the remainder of the 1.7 trillion barrels of resource.

    One of the issues with fuel is built infrastructure. No matter how much electricity we have, the vast majority of our cars, trucks, and a wide range of equipment will not run on it. So adding electricity is not equivalent, no matter how many electric cars we might have. That is a cost that has to be built into the equation, and it is not clear we have all the materials.

    Also, the $1 figure for PV is not delivered and installed. The PV also has to be maintained.

    An oil sands mine I think has a life of something like 40 or 50 years, but may vary with the size of the mine. Suncor is still going after 40+ years in the same location, and Syncrude still had time left.

    But your point is a valid one. We are getting to a pretty low EROI for oil, if this level of investment is required.

    With so much electricity to flow from this hypothetically plant, peak oil wouldn't that make a quicker shift towards electric transportation? But that's a whole different question, I was just looking at electricity production in an oil plant vs solar. Materials aren't a problem if you look at frameless thin film and wasn't silicon the most abundant material in the earths crust? Besides, those oil pipelines, heavy machinery, refineries, power plants also require a huge amount of (high quality) material?

    Also, the $1 figure for PV is not delivered and installed. The PV also has to be maintained.

    True, but with 126 billion to spend the price per Wp can be halved which would leave room for that. Note that with the current panel production (a few gigawatt per year) the 1$ per Wp is already reached. Regarding maintenance: this also counts for the bitumen production line in an inhospitable place, pipelines, refineries etc. So who can tell which is more expensive.

    An oil sands mine I think has a life of something like 40 or 50 years, but may vary with the size of the mine. Suncor is still going after 40+ years in the same location, and Syncrude still had time left.

    Well if you add 1 million barrels per day then that's a whole different ball game then what Suncor was doing 20 years ago. Do the numbers from that situation count for the new situation?

    Another option could be to compare it to CSP which more equal characteristics to an oil fueled electricity plant (i.e. 24/7 production). I have no figures about cost per installed Wp, but it could well be even lower then solar panels if you're talking about installing a few thousand multi-MW versions.


    thank you very much for this series. Wonderfully clear exposition of the details at just the right level for me, and minimal speculation. Much appreciated!

    My take-away points were these.

    First, there is plenty of resource. If 10% is producible with prices at $66, then presumably 20% or more is producible with a price floor somewhere over $100. This does not bode well for climate change (although coal is by far a bigger problem).

    Second, the amount of capital investment is large. If the world wants 10 million barrels per day in 2020, it is looking at an investment of at least 1.3 trillion 2006 dollars - and probably much more than that, given that the most easily produced oil has already been claimed. Investment on that scale will absorb a noticeable fraction of the world's industrial output, assuming it turns out to be possible.

    Gail, excellent, again.

    So...a suggestion would be now to turn your attention to the Venezuelan Tar Sands and the Orinico Oil Belt which seems to have a similar number of reserves with similar problems.


    Gail, Thanks for posts on the Canadian oil situation. I live in Nova Scotia, where we use imported oil, and so am very interested in Alberta's production and exports.

    Could I ask you for the numbers that underlie the graphs on Canada's oil supply and exports? I can't pull the numbers off the graphs easily:

    Canadian conventional production mbbl/day
    Canadian oil sands production
    Canadian imports on the East Coast
    Canadian consumption
    Canadian exports to USA

    I am having a protracted peak-oil conversation with my Member of the Legislative Assembly and I would like to have these numbers to pass on to him (and my neighbours).

    Thanks . . . Alan

    Could I ask you for the numbers that underlie the graphs on Canada's oil supply and exports?

    That sort of information is available on the National Energy Board Web site:

    Alberta's pipelines don't reach as far as Nova Scotia, but there is oil available from Newfoundland. However, as far as I know, most of Newfoundland's oil is exported, too.

    Basically, I got total data (oil production, oil consumption, exports to the US, and net exports) from the US EIA, and used CAPP data to break out oil sands production from conventional. Canadian imports on the East Coast were "backed into". If I knew how much was accounted for all uses, and there was inadequate production, the rest must be imports.

    Regarding US EIA data-- The EIA has a data series for Canada here. If you click on Download full data series, it will give you an xls exhibit that is easier to work with.

    A breakdown of Canadian data can be gotten from this source.

    Where production is in cubic meters per year, it can be converted to barrels per day using factors at the front of the Statistical Handbook.

    Hope this helps. There may be a direct source of imports, but I didn't have it. If Canada also exports, to say, Mexico, my import number might be a little low.

    Over 99% of Canadian exports go to the United States, so U.S. EIA import data and Canadian NEB export data should match.

    Should, but probably won't. There are some differences in definitions and reporting practices.

    Gail, I've got to say I'm terribly disappointed with this article.

    Is TOD now a part of the "Drill Baby Drill" crowd?

    You get invited on a junket by the dirtiest energy providers in the world and come back to report how badly misunderstood they are, how all the criticism of them is misplaced.

    I see so much hand waving here it's just a blur:

    Undermining the Report Card: The Oil Sands Report Card by WWF and the Pembina Institute. This report compares the various oil sands projects against one another on a number of different variables, and makes statements such as, "If all companies had as low emissions as [lowest company], there would be a savings of __________ in emissions." That is interesting, but doesn't tell you whether there is a problem in the first place.

    So comparing how bad one is compared to another doesn't tell us what "bad" means. And we can't figure out whether the whole process is bad or not?

    Statements like "Toxic tailing ponds filled with liquid mine wastes already cover more than 50 square kilometers," also from Oil Sands Report Card. Fifty square kilometers corresponds to a square a little over 7 kilometers (4.4 miles) on a side. Having so many tailings ponds is not great, but it is not as huge a problem as it is made out to be. In the future, the amount isn't likely to grow much, with more in-situ production, and with recent changes in regulations regarding tailing ponds that require much quicker ends to the ponds. Environmental Commissioner Renner told us that the oldest tailings pond (which is the one with by far the most leakage issues) is expected to be retired in 2010.

    It's OK to have a toxic waste site that's only 7 Km on a side if it's made out to be worse than that?

    What planet are you from? What are the consequences of an earthen dam failure? How much of the toxins leach into the water? What's your confidence level that this is ever going to be properly remediated?

    Do you have any idea how bad the track record of mining companies is at cleaning up their polluted messes, even in countries like the US with relatively strict environmental laws? Are MTR coal mines restored nicely or do the companies do the absolute minimum in search of maximum profit, to the point of deliberately breaking the laws if the fines + legal fees are less expensive than following the law?

    Photos with captions like, "Oil sands operations remove rivers, forests, and wetlands in order to access the oil sands beneath," also from Oil Sands Report Card. The angle of the picture makes it look like a huge area is involved, and the way the caption is worded makes it sound like there are no steps being taken to put the pieces back in place according to current views of best environmental practices.

    Of course, the nice Canadian companies are not like mining companies in Montana, or Kentucky. And we've already established that 50 square kilometers isn't really "big", so photos that exaggerate for effect are misleading.

    Statements by aboriginal groups that they have some cancer above normal levels. Any of us who have studied probability know what the issue is here. There are all kinds of little aboriginal groups with population of around 1,000. So if you look at enough of these groups, some of the groups will have a handful of people with one or another kind of cancer. There is pollution in the area--there always has been, because the river naturally runs through the oil sands. The background pollution level may raise the probabilities of cancer a bit higher. But there doesn't seem to be any study showing an overall problem.

    My goodness, what lame hand-waving this is. "We all know what the problem is here" "We've always had pollution, people have always gotten cancer here, so obviously there is no risk that the tailing ponds are leaking or that air pollution causes a problem"

    And, to sum it up "But there doesn't seem to be any study showing an overall problem".

    Well, I feel better then. As long as there don't seem to be any problems.

    No, I haven't presented any counter evidence. That's not my point, and I haven't the time to do more than complain, and perhaps that's unfair.

    Indeed, it's grossly unfair to you, Gail, if the point of the article were to report to TOD readers what the tar sands companies PR companies are publishing about tar sands operations. If that was the point, then you're spot on.

    If, OTOH, the point was to objectively determine anything about any of the controversial issues of mining the tar sands, then I've got to point out that you failed quite badly at critically investigating any of what they presented to you.


    No, I haven't presented any counter evidence. That's not my point, and I haven't the time to do more than complain, and perhaps that's unfair.
    Not only is it unfair, David, it's uninformative.

    Thanks all above for the great article and commentaries. If I've understand it all correctly the oil-sands won't do much to save the bau-system anyway. So we remain privileged to live in the most interesting of times.

    Thanks Gail, a useful article.
    I should stop complaining about lack of scientific method and submit something myself.