Older Archives

How long will petroleum resources last?

November 23rd, 2008 by Roberto F. Aguilera, International Institute for Applied Systems Analysis

The quantity of available conventional petroleum is greater than often assumed, given the tendency to overlook unassessed areas and future reserve growth. Furthermore, volumes of unconventional resources are even more abundant than conventional resources. As a result, conventional and unconventional resources combined are likely to last far longer than many now expect.

Some energy analysts are concerned that the world will soon face a global crisis due to dwindling petroleum resources and a peak in oil production. To shed light on the subject, we have assessed the threat that depletion poses to the availability of petroleum resources by estimating cumulative availability curves for conventional petroleum (oil, gas, and natural gas liquids) and for three unconventional sources of liquids (heavy oil, oil sands, and oil shale).

Our analysis extends the important study published in 2000 by the U.S. Geological Survey on this topic by taking account of (1) conventional petroleum resources from provinces not assessed by the Survey or other organizations, (2) future reserve growth, (3) unconventional sources of liquids, and (4) production costs.

The figure below shows the cumulative availability curve for conventional petroleum,
heavy oil, oil sands, and oil shale. The curve has three relatively wide blocks, each representing an unconventional source of liquid. The combined future volumes of conventional petroleum, heavy oil, oil sands and oil shale total 29.9 trillion BOE.

Figure 6

Global Cumulative Long Run Availability Curve for Conventional Petroleum and Unconventional Sources of Liquids Including Heavy Oil, Oil Sands and Oil Shale

The life expectancies for any particular energy resource depends on three factors—its future volumes, its current production, and the growth over time of its production. In the case of conventional petroleum, for example, we calculated that with production growth of 2% a year (which is somewhat above the average annual growth in production over the past several decades) future volumes from assessed provinces, assuming no future reserve growth, would last for 47 years. Adding in future volumes from unassessed provinces increases this figure to 55 years and considering future reserve growth pushes it to 70 years.

The table below shows that the life expectancy of 70 years increases to 87, 104, and 132 years when future volumes from heavy oil, then oil sands, and finally oil shale are taken into account. If we consider all three unconventional resources, but a future growth rate of 0%, the life expectancy of 132 years increases to 651 years. Alternatively, the table shows that assuming 5% future production growth reduces the life expectancy to 70 years.

Life Expectancies

Life Expectancies

In conclusion, even if conventional petroleum peaks, as some predict will happen soon, this does not necessarily mean a ‘hard landing’ with sharply increasing prices and declining consumption. Rather society may rely increasingly on unconventional resources, including heavy oil, oil sands, and oil shale as well as non-fossil sources of energy. Should non-fossil fuels become more important, some of the higher cost fossil fuels including conventional petroleum resources may never need to be exploited. These implications suggest that expectations of a sharp rise in real oil prices over the next decade or two may be misplaced. Indeed, a case can be made that the high prices observed over the past several years for conventional oil are not sustainable given the available conventional and unconventional energy resources that can be exploited at costs below current market prices and substantially below mid-2008 prices.

Roberto F. Aguilera (IIASA), Roderick G. Eggert (Colorado School of Mines), Gustavo Lagos C.C. (Catholic University of Chile), and John E. Tilton (Colorado School of Mines and Catholic University of Chile)

P.S. More on this research is available from our forthcoming paper, “Depletion and the Future Availability of Petroleum Resources” (The Energy Journal, Vol 30, No. 1, 2009).

Send this to a friend

6 Responses to “How long will petroleum resources last?”

  1. Steven Stoft Says:

    This is a good antidote to peak-oil silliness. But plotting, for example, all tar sands at the average cost of producing tar sands is extremely misleading. Since conventional oil is plotted in a sensible manner viewers will tend to think the same has been done for tar sands. The result of this, and of other optimistic assumptions is to mislead as much on the optimistic side as peak-oil proponents mislead on the pessimistic side. Perhaps that is the point of this note, and the reader is intended to take a mental average.

    Note that according to this graph the world still has twice as much oil left (2 T BOE) at a cost of under $2 / bbl as it has used in total so far. This implies that the competitive price of oil is at most $5 / bbl, while peak oil analysis suggests it is over $200 / bbl. My guess is the oil markets are smarter than either group.

    Just as peak-oil analysis leads to the conclusion that climate change is not a problem because we will run out of CO2, this analysis should state the obvious. If we do manage to use these resources, the earth is doomed. For a more sensible, though less detailed analysis see Carbonomics at

  2. John Busby Says:

    The Albertan Energy Resources Conservation Board publishes report st98 annually, which in 2008 claimed that 315 Gb of bitumen is recoverable from 1712 Gb of tarsands “in place”. However, for its extraction and conversion to synthetic crude oil (SCO), around 30% of its heating value is lost, mostly in the natural gas needed for steam raising, electricity generation and for hydrogen production for hydrogenation.

    If say 10% of the North American natural gas reseves were to be dedicated to SCO production only 15 Gb of the “recoverable” bitumen would be upgraded. In recognition of this problem. there is a proposal to build a nuclear power plant to provide steam and hydrogen. An allied problem is the amount of process water needed and also the permanent scarring of the landscape.

    North America is now a major importer of LNG and it is questionable whether it is worthwhile using ever more expensive, partially imported gas for tarsands recovery. If a nuclear power plant is built in Alberta, it would be of more use generating electricity!

    Around half of the recovered bitumen is diluted with pentanes and piped South of the border to US refineries, where it is blended with other refinery feedstock.

    The other major problem is that only 6% of the tarsands is mineable, the remainder requiring more sophisticated measures to extract the deeper lying bitumen by for example steam-assisted gravity drainage (SAGD). So far the more readily accessible mineable deposits have provided the majority of the production.

    So what is claimed to be the equivalent of Saudi Arabia is in fact likely to provide but 6 months of the world’s crude oil production.

  3. Hugh Sharman Says:

    Oh dear! Which planet are these authors living on? How many times does one have to remind energy optimists that “peak oil” is not about stocks but flows. None (or very few) of us “peak oilers” deny that there is a huge hydrocarbon resource still to be extracted. May be as large as that claimed in these “lala land” graphs. But the capital is not being spent to mainain these flows faster than depletion is reducing them. Only Saudi Arabia has sufficient spare production capacity to extend production flows. But why should it open these up simply to bring the market value of oil down? That would be quite irrational from a business point of view.

    Even the IEA, after a decade of wildly optimistic future hydrocarbon price predictions is facing up to the fact that future hydrocarbon resources will be both progressively scarcer and more expensive.

    For example, the last time I saw tar sand extraction costs at a notional $25/b was 10 years ago. The estimated extraction cost for tar sand is now $80 – 100/b. The figure for oil shale is probably unknowable because the energy investment to extract useful HCs from shale may exceed the energy won. Unless, of course, one simply burns oil shale as if it were a particularly poor quality coal as they still do in Estonia.

    Most large reservoirs of conventional oil are depleting much more rapidly than new production capacity is being brought on line. The IEA’s latest estimate is higher than CERA’s 2007 estimate of 4.5% per year. The depletion rate of conventional reserves exceeds 300,000 b/d per month which means that simply to stay on plateau at 87 million bopd, we need to be bringing on a new large oil field every month.

    We are not doing this.

    The current economic crisis is undoubtedly slowing down the annual global increase in demand and this may be causing marketing problems and thus temporary stock increases and price falls. However, new investment in production has also slowed so that when and if demand picks up again, we will see an unholy price spike and thus further economic crises in the years just head.

    By the way, LNG will be in permanent shortage for a decade, simply because the investments in many plant liquefaction plants never got past Final Investment Decision (FID). So yes, we will see shale gas reversing years of declining North American gas production but for a much higher price than the current $7 per mmBtu. But the rest of the World will likely see gas and oil prices converging above $15 per mmBtu and rising steadily despite the resulting financial crises.

    I wish it were otherwise! I really do.

  4. John Busby Says:

    BP’s chief economist, Christof Rühl has denied “peak oil” in an interview with EurActiv, while BP’s Statistical Review 2008 shows a global oil peak in 2006.

    See my rejoinder on his views

    In the same statistical review a peak in Saudi Arabian oil production is shown in 2005, see

    If its chief economist interprets figures in a contrary manner to those plotted from BP’s statistical review, he needs to explain why!

  5. Carlos Ferreira Says:

    Unfortunately, I am forced to disagree with the authors. They fail to appreciate two important factors:

    1. how is the cost of oil calculated.
    2. that not all oil is the same.

    Authors (for example, Hanley, Hogren and White, “Environmental Economics: in Theory and Practice”, 2007; Palgrave Macmillan) define the price of any non-renewable resource as a sum of the shadow price (a measure of uncertainty concerning the size of economically recoverable reserves) and the marginal cost of extraction. The author’s discussion addresses the former, not the later.
    2. Not all oil resources are the same. Even conventional fields have different characteristics, with differing marginal costs of extraction. After a well peaks, the marginal cost of extraction increases because of the diminishing pressure, leading to more energy being needed to sustain (a decreasing) output. So, as many wells peak, the average marginal cost of extraction increases, taking prices with it. Extraction costs from unconventional sources are higher than conventional sources – that is the reason why they are unconventional and have only recently come online.
    The Marginal cost of extraction is measured in the energy needed to produce oil. In the end, the energy output for a well might be inferior to the energy input needed to produce oil. When there’s little or no energy gain in the extraction, why extract?

    Also, extracting oil from unconventional sources is an extremely CO2-intensive activity.

  6. swami krishnananda Says:

    can we not save oil?? if oil reserves get exhausted what next? solar energy? wind energy? any other source is future bleak with out crude oil. Have the world leaders thought of it?

Leave a Reply