19 May 2011

The Science and Politics of Wedges

To follow up on my post yesterday on the so-called "stabilization wedges" of Rob Socolow and Steve Pacala, I thought that it would be useful to revisit the substantive reasons why it is that the wedges analysis has beem so misleading in the climate debate. I thought it also would be useful to add a perspective on why it is that some parts of the environmental community go on the attack when it comes to those who have criticized the wedges.

NYU physicist Marty Hoffert provides a concise explanation as to the problems with the approach, when it is used to imply that we need 7 (or now 8) wedges to "solve" climate change for the next 50 years (emphasis added):
Pacala and Socolow (8) analyzed a scenario that envisioned stabilizing atmospheric concentrations of CO2 at 500 ppm within 50 years. They found that reaching that goal required the deployment of seven existing or nearly existing groups of technologies, such as more fuel-efficient vehicles, to remove seven “wedges” of predicted future emissions (the wedge image coming from the shape created by graphing each increment of avoided future emissions). Those seven wedges, each of which represents 25 gigatons of avoided carbon emissions by 2054, are cited by some as sufficient to “solve” climate change for 50 years (9).

Unfortunately, the original wedges approach greatly underestimates needed reductions. In part, that is because Pacala and Socolow built their scenario on a business as usual (BAU) emissions baseline based on assumptions that do not appear to be coming true. For instance, the scenario assumes that a shift in the mix of fossil fuels will reduce the amount of carbon released per unit of energy. This carbon-to-energy ratio did decline during prior shifts from coal to oil, and then from oil to natural gas. Now, however, the ratio is increasing as natural gas and oil approach peak production, coal production rises, and new coal-fi red power plants are built in China, India, and the United States (10).

The enormous challenge of making the transition to carbon-neutral power sources becomes even clearer when emissions-reduction scenarios are based on arguably more realistic baselines, such as the Intergovernmental Panel on Climate Change’s “frozen technology” scenario ( 11, 12). Capturing all alternate energy technologies, including those assumed within this BAU scenario, means that a total of ~18 of Pacala and Socolow’s wedges would be needed to curb emissions (13) (see the figure). And to keep future warming below 2°C . . . an additional 7 wedges of emissions reductions would be needed— for a total of 25 wedges (see the figure).
The total is even more than 25 wedges if you want to avoid using the oceans as a store of carbon dioxide or reduce emissions below 2010 levels.. The numbers that Hoffert presents in his perspective are the same as those that I present in The Climate Fix, under a similar analysis.

What does a "wedge" mean in more intuitive terms?  Jom Romm helps to explain:
. . . [one wedge] by 2050 would require adding globally, an average of 17 [nuclear] plants each year, while building an average of 9 plants a year to replace those that will be retired, for a total of one nuclear plant every two weeks for four decades — plus 10 Yucca Mountains to store the waste.
Romm thinks we need about 14 "wedges" of effort to stabilize carbon dioxide concentrations at 450 ppm, and if one wedge implies a need for 26 nuclear plants per year, then 14 wedges implies 26 * 14 = 364 plants per year, or the equivalent effort of one nuclear power plant per day of carbon-free energy.  If you accept Hoffert's arguments, then using Romm's conversion you'd need 26 * 25 = 650 nuclear plants worth of carbon free energy, or closer to 2 per day.

This task -- at either Hoffert's or Romm's level of effort -- is clearly impossible with today's technology.

One might think that the modern environmental movement is adept enough at simple math to accept this message and thus proceed to advocate policies consistent with our lack of technological capabilities, such as calling for a much greater commitment to innovation. While some have, of course, the loudest, most well funded and arguably most influential parts of the movement have strenuously resisted the notion that we do not have the technology needed to rapidly decarbonize our economy, preferring to hold on to the myth that -- in the words of the original "wedges" paper -- "Humanity can solve the carbon and climate problem in the first half of this century simply by scaling up what we already know how to do."  I discuss this myth at length in Chapter 2 of The Climate Fix.

What explains the environmental community's strict adherence to bad math and flawed policy?

In an overlooked part of Matt Nisbet's recent report titled Climate Shift (see his Chapter 2), Nisbet explained that the American environmental community decided to unite under a common set of strategies that were expressed in a  2007 report called Design to Win (PDF, I earlier critiqued the report as "Doomed to Fail").   This common approach would allow efforts to be coordinated and reinforcing. Central to the approach was the so-called wedges and the misleading notion that we have all the technology that we need, meaning that the challenge was one of shaping political will and public opinion. On the scientific advisory committee of the DTW report was Robert Socolow.

The 2007 Design to Win report explains that to 2030, 80% of emissions reductions could be achieved with existing technologies and at little or no cost (the figure is from the report):
The good news is that we already have the technology and know-how to achieve these carbon reductions – often at a cost-savings. Design to Win’s synthesis of the latest scientific and economic analyses, including the Stern Review, Vattenfall climate abatement map prepared by McKinsey & Company, and reports by the Intergovernmental Panel on Climate Change, concluded that about 80 percent of the needed mitigation – 25 gigatons of carbon – can be achieved with existing technologies (Figure 4). The key lies in rapidly deploying such technologies in our power plants, buildings, factories and vehicles, and improving land management practices.
The focus on deployment, often to derision if not the exclusion of the need for innovation, is still central to environmental messaging, even as the math of emissions reductions would seem obvious and the policy centerpiece of DTW -- cap-and-trade -- has failed comprehensively.

What explains the adherence to bad ideas in the form of bad policy?  I'm not entirely sure but it just so happens that groups such as the Center for American Process have been funded under the Design to Win strategy to spread its message. Apparently that includes a healthy dose of efforts to delegitmize alternative points of view and to poison what otherwise might be characterized as a healthy public debate over policy options.  To the extent that these efforts succeed, climate policy and the broader environmental movement suffer.


  1. Have these scenarios taken into consideration the population growth and that close to a billion people today are without electricity? Just to connect these people on the grid will mean around 3 billion new consumers of electricity by 2050

  2. -1-Pirate

    In general, no. See


  3. I'll be a contrarian on Business as Usual Scenario's.

    It's only May and the Chinese are rationing electricity already.

    Business as Usual Scenario's generally assume an inexhaustible supply of inexpensively extractable coal. It's questionable whether such quantities of inexpensively extractable coal exist based on China's struggle to control coal prices.

  4. Roger: The prospect of building on the order of one or two nuclear plants per day worth of carbon-free electric generating capacity is certainly daunting, but should be put in the perspective that BAU entails building an equivalent fossil-fired capacity. Hoffert's graph clearly shows that if we don't build clean energy, then BAU entails building 25 wedges of new fossil energy.

    How much more impossible is it to build one or two 1000 MW nuclear plants every day than to build one or two 1000 MW coal plants every day? How many new coal mines does that entail? How does that compare to mining, refining, enriching, and shipping uranium?

    Choices are clearer when we lay all the alternatives on the table rather than presenting more details about one alternative than about the other(s).

    Innovation is necessary, but we also need policy to bridge the time between today and whenever the innovation pays off. That was Socolow's intent---a stop gap, not a cure. And yes, whether we shift to nuclear, renewable, or CCS, it will be more expensive than BAU, at least for the next several decades, and the right policy debate should be what price we're willing to pay---something you discuss early in TCF.

    In this spirit, my take on the wedges concept is not to ask a binary question---will it or won't it solve the problem for 50 years?---but how much time will a realistically achievable number of wedges buy us to perform the necessary innovation? If I look at the concept that way, it seems a useful one, although very different from the way it's often presented (although you could legitimately criticize this for the difficult details I hide in the fuzzy terms "realistically achievable" and "necessary innovation").

  5. -4-Jonathan Gilligan

    Thanks for this comment.

    Yes indeed the world needs more energy, much more of it. And right now is it easier to build carbon intensive energy supply than carbon free? I think that the answer is obviously yes. But is a post to come shortly, even meeting current demands with fossil is problematic. Decarbonizing energy supply is about much more than just carbon dioxide.

    I'm not sure what you BAU comment means (sorry), but BAU will require a rapid decarbonization if we are to meet demand. The more innovation we see, the better we'll be able to meet that demand, and at a lower cost.

    The wedges are not so much a policy proposal but rather a measuring stick. And if you get your measurements correct then whatever job you are doing is going to be more effective. the problem with the wedges is not conceptual but how they have been used to underestimate the challenge.


  6. "groups such as the Center for American Process have been funded under the Design to Win strategy to spread its message. Apparently that includes a healthy dose of efforts to delegitmize alternative points of view and to poison what otherwise might be characterized as a healthy public debate over policy options."

    Global warming is all about the politics. And the strategy you summarize in the quote above is simply the standard political strategy employed by the same political people on every other political issue and in every election.

  7. @Roger: "I'm not sure what you BAU comment means."

    By BAU, I mean the blue line on Hoffert's graph, which Hoffert describes as extrapolated recent growth in fossil fuel emissions.

    When Hoffert writes that we'd need 25 wedges to phase out fossil fuel emissions in the next 50 years, he's taking a baseline in which there are 25 wedges worth of new fossil capacity coming on line (that's the area between the blue BAU line and the orange, "fossil fuel tech ages out" line).

    Regarding your last para: Correct. Wedges are a tool for policy analysis (or measurement), not a policy proposal. I like them because when used properly, they can shift people's focus from big grand-slam policies and technologies that aspire to solve the whole problem at once and can get people thinking about finding a variety of smaller technology and policy approaches, each of which gets us a couple of steps down the long road. However, that's not the way most people have used the wedges and I share your distaste for claims that completely decarbonizing the energy supply would be painless.

  8. -7-JOnathan Gilligan

    Gotcha, thanks.

    Yes, Hoffert argues that we should expect CO2 emissions to increase by closer to 3% per year rather than 1.5%. His view is shared by Ross Garnaut and less illustrious scholars (like me;-) and the lower number is held by IPCC, IEA etc.

    As I write in TCF, relying on Smil, you can't put much faith in energy forecasts, from whatever source. I don't. The question here is how you want to be wrong. Seems to me that effective policy is robust to uncertainties, so if you expect and plan for the 3%/yr then you are covered for the 1.5%/yr but not vice versa. This is the essence of the problem with how the wedges have been used -- they are used to actually close down options.

    And, oh yeah, if you plan successfully for 3%/yr you might also get expanded energy access, increased security and lower costs. I think that you and I agree on this point ;-)