Nature Climate Change | Climate Feedback

Energy vs Climate: A surprising confluence of goals

Guest Commentary by Andrew Dessler

There is an emerging view among some experts that recoverable fossil-fuel reserves are far smaller than previously thought. If so, the Intergovernmental Panel on Climate Change’s (IPCC) highest emissions scenarios could be unrealistically high, thus limiting the worst-case climate change during the 21st century. This view of a constrained fossil-fuel supply points to a potential convergence of thinking about policies and actions needed to address the seemingly divergent problems of energy supply and climate change.

<img alt=“oil_plot.jpg” src=“” width=“400” height=“211” border=“0” align=“right” hspace=“10px”//>

One of the standard techniques for estimating future oil production is the method pioneered by M. King Hubbert. Briefly, Hubbert’s theory involves fitting historical production data to a logistics curve; extrapolation of the curve allows estimation of future production. Hubbert successfully predicted domestic U.S. oil production would peak in the early 1970s, and his theory predicts that world oil production should be peaking about now, which production figures seems to be confirming. Application of Hubbert’s theory to world-oil production suggests that total future recoverable conventional oil is ~1.2 trillion barrels or 7 Zeta Joules (ZJ). Figure 1 shows oil consumption from emissions scenarios produced by the IPCC, along with this estimate based on Hubbert’s method. All IPCC scenarios assume integrated oil production greater than the Hubbert estimate of recoverable oil.

For coal, the canonical wisdom is that there are hundreds of years of availability. However, several recent analyses have cast doubt on this. A recent NRC report on U.S. coal availability concluded, “… there is probably sufficient coal to meet the nation’s needs for more than 100 years at current rates of consumption. However, it is not possible to confirm the often-quoted assertion that there is a sufficient supply of coal for the next 250 years.” Prof. David Rutledge of Caltech has applied a Hubbert-like analysis to coal production data and concluded that the availability of coal is overstimated. The Energy Watch Group, a German think tank, has performed a detailed country-by-country analysis of coal reserves. Integrating their estimate of future coal production, we get 11 ZJ for world coal production from 2008 to 2100 — about one-ninth of the coal reserves cited by the IPCC (Table 4.2 of the fourth assessment report). As is the case for oil, the IPCC’s scenarios’ projected coal consumption generally exceeds this estimate of available resources.

If these analyses of oil and coal production are correct, then the IPCC’s high emissions scenarios are not achievable. Using these new and lower estimates, Rutledge estimates that combustion of all remaining conventional oil, gas, and coal reserves would produce an atmospheric concentration of CO2 of 470 ppmv in 2100, near the stabilization target that many climatologists argue we must achieve in order to head off the worst-case climate impacts. Kharecha and Hansen recently concluded that atmospheric CO2 could be kept from exceeding about 450 ppm if we consume all of our reserves of conventional oil and gas but restrict coal usage to amounts similar to the recent availability estimates of the Energy Watch Group and Rutledge.

There are, of course, enormous reserves of unconventional sources of fossil fuels, such as tar sands or oil shale — reserves of oil shale alone could exceed 1 trillion barrels. However, the issue with unconventional sources is not the size of the reserves, but the rate at which they can be produced. Shale, for example, needs to be heated to release the oil, and one estimate suggests that it takes 12 GW of power to produce one million barrels of oil per day from shale. Additionally, oil production from shale requires significant quantities of water, which is problematic because oil shale resides mostly in the water-stressed western U.S. These and other resource requirements limit production rates from shale to, at most, a few million barrels per day — not an insignificant amount, but also not enough to meaningfully affect emerging constraints in our supply of fossil fuels.

Our goal as a society should be to adopt policies that head off, to the maximum extent possible, both climate-change and energy-supply risks. First and foremost, we need better estimates of our reserves of oil and coal. The NRC report on coal, for example, was essentially a plea for better information on coal availability. The situation is not much better for oil. Saudi Arabia, for example, claims reserves of a quarter of a trillion barrels. There is some evidence that these estimates are inflated, and Saudi Arabia will not provide relevant data to back up their high reserve estimate.

Given the high stakes involved, this lack of information is simply unacceptable. The world therefore needs an IPCC-like assessment of our fossil fuel reserves. While some countries may balk at providing the relevant data needed to verify their reserve estimates, the world must compile the necessary information to assess this risk. Second, we should recognize that solutions to the problems of energy supply and climate change are, to a great degree, aligned. Solutions such as energy efficiency, the cheapest and most cost effective “new” source of energy, and renewable energy sources, such as solar, wind, and nuclear, can help solve both energy supply and climate problems.

A few solutions, however, address only one problem and not the other, and these policy pathways should be avoided. For example, coal combined with carbon sequestration, where CO2 from coal combustion is sequestered rather than released to the atmosphere, is a poor choice if coal is less abundant than expected. Geoengineering, active manipulation to bring about a cooler global climate, is also a poor choice because it does nothing to address energy availability.

Many people are rightfully concerned about emissions of carbon dioxide doubling, tripling, or even quadrupling this century. There is however another risk we should worry about: namely, that recoverable fossil fuel reserves are far smaller than previously thought. If these new estimates of fossil-fuel availability turn out to be correct, then reducing our carbon emissions could be easy — because we will be out of fossil fuels. Constraints on energy supply could, like climate change, challenge our very way of life and would not be good news. We must therefore take the opportunity we have now to work toward a common goal of energy sufficiency and climate-change mitigation through development of suitable alternative energy sources that address both problems.

Acknowledgments: I thank Drs. Dave Rutledge, Werner Zittel, Alex Dessler, and Hans Juvkam-Wold for their help on this project.

Figure caption: Cumulative oil production between 2008 and 2100 from the IPCC emissions scenarios, in units of ZJ. The bars show the spread over all scenarios in each scenario family; medians emissions are the tick marks and the marker scenario is the black dot. The dotted line is the Hubbert estimate of oil production. IPCC values are from Fig. 4-8 of ref. 11, adjusted by subtracting 2.8 ZJ from the scenarios to account for production from 1990-2008.

Andrew Dessler is a professor in the Dept. of Atmospheric Sciences at Texas A&M University and a contributor to Grist.


  1. Report this comment

    Olive Heffernan said:

    But what about forcing from other sources such as land use change? As we are unlikely to able to completely halt deforestation, it would seem highly improbable that we will only reach atmospheric concentrations of 470ppm by 2100. I understand you are saying that is only the contribution from exploiting conventional fossil fuel reserves, but we can’t take these things in isolation…

    Also on your suggestion to look to policies other than CCS…my concern here is that if we continue to accumulate CO2 in the atmosphere, we’ll end up with increasingly acidic seas, and other impacts besides warming.

    Do we have enough information to say that stabilising atmopheric CO2 concentrations at 470ppm would be enough to prevent such impacts? I don’t think so. A recent study by Coa and Caldeira in GRL shows the potential for major damage to at least some ocean ecosystems at atmospheric CO2 stabilization levels as low as 450 ppm. ppm

    So, even if conventional fossil fuel reserves are alot smaller than expected, we’d be smart to implement CCS.

    I completely agree about the need for a global inventory of fossil fuel reserves – any thoughts on who should take the lead on this?

  2. Report this comment

    Andrew Dessler said:

    Clearly, Rutledge’s estimate of 470 ppm does not include deforestation, just as it does not include unconventional sources of fossil fuels. That said, I find it unlikely that we would be able to get to the highest emissions scenarios without conventional fossil fuels.

    You make a good point about carbon capture. The issue is really one on economics. If we have to invest an enormous amount of money in infrastructure for carbon capture, then carbon capture only makes sense if there is a certain amount of carbon to be sequestered. If not, then it probably makes sense to reduce carbon emissions by other means, such as switching to other energy sources.

    Also, I want to be careful and emphasize that I am not arguing for any particular carbon dioxide stabilization. The value of 470 ppm is simply one estimate of atmospheric concentrations if we burned all conventional fossil fuels.

    Finally, I do not have any specific suggestions about who should lead an assessment of fossil fuel inventories. There are several world organizations that focus on energy, such as the International Energy Agency, that could do it. And there is always the United Nations.

  3. Report this comment

    Karen Nyhus said:

    Do either of these analyses include methane hydrates? While they are currently hard to recover (being mostly in solid form under shelf and slope seabeds, and thus in situations vulnerable to slope failure), they are located in continental margins worldwide, and constitute about double the global store of coal, oil, and (non-hydrate) natural gas combined: 10,000 GT. I don’t think LNG can replace oil as a transport fuel, but the temptation to mine it for existing NG uses strikes me as large, especially given the installation of LNG infrastructure globally over the last decade. My concern is not just extending the fossil fuel era, but also the possibility of increasing methane emissions through the mining process itself. See press release on a 2004 risk analysis report at

  4. Report this comment

    G.R.L. Cowan said:

    This — “carbon capture only makes sense if there is a certain amount of carbon to be sequestered” — seems to ignore the demonstrated possibility of sequestering carbon that was emitted years ago. Strewn pulverized alkaline earth silicates increase entropy by taking down atmospheric CO2, and thus are the more sensible method than the ones that now seem to be mostly discussed, which seek to prevent both that entropy increase and the earlier one of the CO2’s dispersal.

  5. Report this comment

    David B. Benson said:

    Carbon removal via enhanced mineral weathering appears to be one of the least expensive way to ‘retire’ unwanted carbon from the active carbon cycle.

    Olivine weathering:

    Peridotite weathering:

    “Rocks Could Be Harnessed To Sponge Vast Amounts Of Carbon Dioxide From Air”:

  6. Report this comment

    Milan said:

    If you listen to the speeches being made by presidential candidates in the United States, you constantly hear two ideas equated that are really quite independent: ‘energy security’ and climate change mitigation. The former has to do with being able to access different kinds of energy (natural gas, transportation fuels, electricity) in a manner consistent with the national interest of a particular state. The latter is about reducing the amount of greenhouse gasses emitted in the course of generating and using that energy.

    Some policies do achieve both goals: most notably, building renewable energy systems and the infrastructure that supports them. When the United Kingdom builds offshore wind farms, it serves both to reduce dependence on hydrocarbon imports from Russia and elsewhere and to reduce the link between British energy production and greenhouse gasses. Arguably, building new nuclear plants also serves both aims (though it has other associated problems).

    There are plenty of policies that serve energy security without helping the problem of climate change at all. Indeed, many probably exacerbate it. A key example is Canada’s oil sands: they reduce North American dependence on oil imports, but at a very considerable climatic and ecological cost. Corn ethanol is probably an example of the same phenomenon, given all the emissions associated with intensive and mechanized modern farming.


    Energy security and climate change

    Coal and climate change

  7. Report this comment

    Andrew Dessler said:

    Here are a few thoughts on the above comments:

    1) methane hydrates: There are very few suggestions that we try to use these as energy sources (although I saw a news report about it a few days ago). In any event, like shale, the main issue is not size, but how fast we could obtain it.

    2) air capture vs. carbon sequestration: My point here is that whether CCS makes sense depends on the fixed costs of the infrastructure - if they are high, then it will only make financial sense if you’re going to remove a lot of C from the atmosphere. Thus, the amount of available fossil fuels does determine whether we proceed down that path. This goes for removing C from the atmosphere, too.

    3) national security vs. climate change: I agree with Milan on his assessment. I would add that, if fossil fuels are beginning to run out, then our only option is to adopt non-emitting sources, which would tend to align these two goals.

  8. Report this comment

    Mark Duffett said:

    Inventories of fossil fuel reserves can only ever be a minimum estimate of the total amount that’s actually there. For decades now, analysts from the Club of Rome to Paul Ehrlich have been making themselves (or at least left themselves open to) look silly by failing to adequately account for an entire industry called ‘exploration’ dedicated to finding more of the stuff.

    In the case of oil, inventories of reserves may actually be starting to asymptote towards the total available resource, because it’s valuable enough to be transported to markets from just about anywhere on Earth. But that’s not the case for coal. Its current economics (the need to be close to either major consumers or a port) dictate that vast areas of the planet remain unexploited and unexplored. Inventories of current coal reserves are thus likely to greatly underestimate the amount that may potentially be extracted.

    For similar reasons, I’m not sure that Hubbert-type analysis is applicable here. The power of Hubbert is not so much in modelling the total resource amount (integrating the production curve), rather in predicting the timing of the peak production rate. Indeed, one reason why Hubbert’s technique is so powerful is that the latter is not as sensitive as you might think to the former.

  9. Report this comment

    Alexander said:

    Wind power is not a solution.

    The whole truth about wind turbines is never told by lobbyists and governments.

    How could the very weak and extremely unreliable initial energy source of a wind turbine ever produce a steady power of any significance?

    Please think!

    And read: “Wind energy- the whole truth” at: