Friday, November 06, 2009

Roger Pielke Jr. on climate change mitigation

Yesterday I heard Roger Pielke Jr. speak twice at Arizona State University, first in a talk to the Consortium for Science, Policy and Outcomes (CSPO) on climate change mitigation, and second in a class lecture on climate change adaptation. This post is about the former.

His talk was entitled "The Simple Math of Emissions Reduction," and began with a quote from Steve Raymer of Oxford University:
Wicked Problems
have Clumsy Solutions
requiring Uncomfortable Knowledge
which he then followed up with a slide on "Where I stand," which included the following bullet points (nearly, but probably not exactly verbatim):
  • Strong advocate for mitigation and adaptation policies
  • Continuing increase in atmospheric CO2 could pose large risks, as described by IPCC
  • Stabilizing concentrations at low levels can’t succeed if we underestimate the challenge (and we have)
  • Mitigation action will not result from elimination of all scientific uncertainty
  • Poisonous politics of the climate debate serves to limit a broader discussion of options
  • Ultimately technological innovation will precede political action, not vice versa
Regarding the IPCC, he says he has no debate with working group I on the science, some disagreements with working group II on impacts, adaptation, and vulnerability, and lots of debate with working group III on economics and mitigation, which this talk covers.

His slide for the outline of his talk looked like this:
  • Understanding the mitigation challenge
  • Where do emissions come from?
  • Decarbonization
  • The UK as a cautionary tale for U.S. policymakers
  • The U.S. situation and Waxman-Markey/Boxer-Kerry
  • How things might be different
Understanding the mitigation challenge

Although climate change involves other greenhouse gases besides CO2, he focused on CO2 and in this part of the talk gave a summary of CO2 accumulation in the atmosphere as a stock and flow problem, using a bathtub analogy. The inflow of CO2 into the atmosphere is like water pouring out of the faucet, there's outflow going out the drain, and the water in the tub is the accumulated CO2 in the atmosphere. The inflow is about 9 GtC (gigatons of carbon) per year and growing, and expected to hit 12 GtC per year by 2030. The current stock is a concentration of about 390 parts per million (ppm), increasing by 2-3 ppm/year. And the outflow is a natural removal of about 4 GtC/year. To stop the stock increase, the amount going in has to equal the amount going out. If we reach an 80% reduction in emissions by 2050, that is expected to limit the stock to 450 ppm.

Emissions have been growing faster than expected by the IPCC in 2000, with a 3.3% average increase per year between 2000 and 2007. While the economic slump has reduced emissions in 2009, it's expected that recovery and continued growth in emissions will occur.

Where do emissions come from?

Pielke used the following four lines to identify policy-relevant variables:
engage in economic activity that
uses energy
from carbon-emitting generation
The associated variables:
Population (P)
GDP per capita (GDP/P)
Energy intensity of the economy (Total Energy (TE)/GDP)
Carbon intensity of energy (C/TE)
The total carbon emissions = P * GDP/P * TE/GDP * C/TE.

This formula is known as the "Kaya Identity."

The policy tools available to reduce emissions by affecting these variables are: (1) population management to end up with fewer people, (2) limit the generation of wealth to have a smaller economy, (3) do the same or more with less energy by increasing efficiency, and (4) switch energy sources to generate energy with less emissions.

And that's it. Cap-and-trade, carbon taxes, etc. are designed to influence these variables.

Pielke then combined the first two variables (P * GDP/P) to get GDP, and the second two (TE/GDP * C/TE) he identified as Technology.

He argued that reducing GDP or GDP growth is not a policy option, so Technology is the only real policy option. Regarding the former point, he put up a graph very much like the graph of world income, and observed that the Millennium Development Goals are all about pushing the people below $10/day--80% of the world's population--on that graph to the right. Even if all of the OECD nations were removed from the graph, there would still be a push to increase the GDP for the remainder and there would still be growing emissions.

He quoted Gwyn Prins regarding the G8 Summit to point out how policy makers are conflicted--they had a morning session on how to reduce gas prices for economic benefit, and an afternoon session on how to increase gas prices for climate change mitigation.

With this kind of a conflict, Pielke said, policy makers will choose GDP growth over climate change.

So that leaves Technology as an option, and he turned to the topic of decarbonization.


Pielke put up a graph of CO2 emissions per $1,000 of GDP over time globally, which showed that there has been a steady improvement of efficiency. In 2006, emissions were 29.12 GtC, divided by $47.267 trillion of GDP, gives 0.62 tons of CO2 per $1,000 GDP. In 1980, that was above 0.90 tons of CO2 per $1,000 GDP.

Overall emissions track GDP, and the global economy has become more and more carbon intensive.

He looked at carbon dioxide per GDP (using purchasing power parity (PPP) for comparison between countries) for four different countries, Japan, Germany, U.S., and China (that's ordered from most to least efficient). Japan hasn't changed much over time, but is very carbon efficient (below 0.50 tons of CO2 per $1,000 GDP). Germany and the U.S. are about the same slightly above 0.50 tons of CO2 per $1,000 GDP, and both have improved similarly over time. China has gotten worse from 2002-2006 and is at about 0.75 tons of CO2 per $1,000 GDP.

He put up a slide of the EU-15 countries decarbonization rates pre- and post-Kyoto Protocol, and though there was a gap between them, the slopes appeared to be comparable. For the first ten years of Kyoto, then, he said, there's no evidence of any improvement in the background rate of decarbonization. The pre-Kyoto rate was from above 0.55 tons of CO2 per $1,000 GDP to about 0.50 tons of CO2 per $1,000 GDP. The post-Kyoto rates went from about 0.50 tons of CO2 per $1,000 GDP to below 0.45 tons of CO2 per $1,000 GDP.

At this point, Clark Miller (head of my program in Human and Social Dimensions of Science and Technology) pointed out that given Japan, there is no reason to assume that there should have been a continuing downward trend at all, but Pielke reiterated that since the slopes appeared to be the same there's no evidence that Kyoto made a difference.

The UK as a cautionary tale for U.S. policymakers

Pielke identified the emissions targets of the UK Climate Change Act of 2008:

Average annual reductions of 2.8% from 2007 to 2020, to reach 42% below 1990 levels by 2020.

Average annual reductions of 3.5% from 2020, to reach 80% below 1990 levels by 2050.

The former target of 42% below 1990 levels is contingent upon COP15 reaching an agreement this December; otherwise the unilateral target is 34% below 1990 levels.

Pielke showed a graph of the historical rate of decarbonization for the UK economy, and compared it to graphs of manufacturing output and manufacturing employment, observing that the success of decarbonization of the UK economy from 1980-2006 has been due primarily to offshoring of manufacturing, something that's not sustainable--once they reach zero, there's nowhere further down to go.

He then used France as a point of comparison, since it has the lowest CO2/GDP output of any developed country, due to its use of nuclear power for most of its energy--it's at 0.30 tons of CO2 per $1,000 GDP, and a lot of that is emissions from gasoline consumption for transportation.

It took France about 22 years, from 1984-2006, to get its emissions to that rate.

For the UK to hit its 2020 target, it needs to improve to France's rate in the next five years, by 2015. That means building 30 new nuclear power plants and reducing the equivalent coal and gas generation; Pielke said he would "go out on a limb" and say that this won't happen.

That will only get them 1/3 of the way to their 2020 goals.

The UK plan calls for putting 1.7 million electric cars on the road by 2020, which means doubling the current rate of auto sales and selling only electric cars.

For the entire world to reach France's level of efficiency by 2015 would require a couple of thousand nuclear power plants.

The U.S. situation and Waxman-Markey/Boxer-Kerry

The U.S., said Pielke, has had one of the highest rates of sustained decarbonization, from 1980-2006, going from over 1.00 tons of CO2 per $1,000 GDP to the current level of about 0.50 tons of CO2 per $1,000 GDP.

The Waxman-Markey target is an 80% reduction by 2050, not quite as radical as the UK.
The Boxer-Kerry target is a 17% reduction by 2020.

Pielke broke down the current U.S. energy supply by source in quadrillions of BTUs (quads), and pointed out that he got all of his data from the EIA and encouraged people to look it up for themselves:
Petroleum: 37.1
Natural gas: 23.8
Coal: 22.5
Renewable: 7.3
Nuclear: 8.5
Total energy was about 99.2 quads in 2007, of which 83.4 came from coal, natural gas, and petroleum.

Emissions by source:
Coal: 95 MMt CO2/quad
Natural gas: 55 MMt CO2/quad
Petroleum: 68 MMt CO2/quad
Multiply those by the amount of energy produced by each source and add them up:
95 * 22.5 + 55 * 23.8 + 68 * 37.1 = 5,969 MMt CO2
The actual total emissions were at about 5,979, so the above back-of-the-envelope calculation was pretty close.

In 2009, U.S. energy consumption will be about 108.6 quads, of which 21 quads will come from renewables and nuclear (40% growth from 2007), which leaves 87.2 quads from fossil fuels, a 4.6% increase from 2007.

If we substituted natural gas for all coal, then our 2020 emissions would be 5,300 MMt CO2, higher than the 2020 target and 12% below 2005, and would still lock us into a carbon intensive future.

In order to meet targets, we need to reduce coal consumption by 40%, or 11 quads, and replace that with renewables plus nuclear, plus an additional 3.8 quads of growth by 2020.

One quad equals about 15 nuclear plants, so 14.8 quads means building 222 new nuclear plants (on top of the 104 that are currently in the U.S.).

Or, alternatively, assuming 100 concentrated solar power installations * 30 MW peak per quad, 1,480 such installations for 14.8 quads, or one online every two days until 2020.

Or, assuming 37,500 * 80 kW peak wind turbines per quad, 555,000 such wind turbines for 14.8 quads, or one 150-turbine wind farm brought online daily until 2020.

To reach these targets with wind and solar would require increasing them by a factor of 37 by 2020; Obama has promised only a tripling.

Could we meet the targets by increasing efficiency of our energy consumption? We would have to reduce total energy consumption to 85.5 quads by 2020 (rather than 108.6), about equal to U.S. energy consumption in 1992, when the U.S. economy was 35% smaller than in 2007. That would be improving efficiency by about a third.

How fast can decarbonization occur? We don't know, because no one has really set out to intentionally do that. Historical rates have been 1-2% per year by developed countries; for short periods, some countries have exceeded 2% per year. Japan, from 1981-1986, improved by over 4% per year.

Pielke argued that these targets are not feasible targets in the U.S. or UK, and so policy makers are adding safety valves, offsets, and other mechanisms to allow some manipulation to give the appearance of success. Achieving 80% reduction in global emissions by 2050 requires > 5% decarbonization per year.

The problem, Pielke argued, is that the policy logic of targets and timetables is backwards, and we should focus on improving efficiency and decarbonization rather than emissions targets.

How things might be different

Pielke's suggested alternative strategy was presented in a slide something like this:
  • Focus policy on decarbonization of the economy (not simply emissions)
  • Efficiency gains (follow the Japanese model, “frontrunner program” by industry, look at best performer and set it as regulatory standard)
  • Expand carbon free energy (low carbon tax, other policies--subsidies, regulation, etc.)
  • Innovation-focused investments
  • To create ever advancing frontier of potential efficiency gains
  • Air capture backstop
  • Adaptation
The Japanese "frontrunner" program was where the government went industry by industry, identified the most efficient company in each industry, and set regulations to make that company the baseline standard for the other companies to meet.

Pielke argued that there should be a carbon tax of, say, $5/ton (or whatever is the "highest price politically possible"), with the collected funds (that would raise about $700B/year) used to promote innovation in energy efficiency.

If we find that we're stabilizing at 635 ppm, we may want to "brute force" some removal of carbon from the atmosphere (e.g., geoengineering).

In the Q&A session, Clark Miller questioned Pielke about the impossibility of replacing our energy infrastructure quickly--if it costs $2.61B for a 1400 MW nuclear plant, we'd need 65 of them (fewer than Pielke's number, he assumed smaller plants) at a cost of $260B. Since there is capital floating around causing asset bubbles in the trillions, and the energy industry is expected to become a $15T industry, surely there would be some drive to build them if they're going to become profitable. (Not to mention peak oil as a driver.) He agreed that it would take longer to construct these, but asked what the upshot would be if this was done by, say, 2075.

Also in the Q&A, Pielke pointed out that in a previous presentation of this talk, a philosophy professor had suggested that the population variable could be affected by handing out cyanide pills. (Or by promoting the growth of the Church of Euthanasia.) What I didn't mention above was that Pielke also briefly discussed improvements to human lifespan, and in his other talk (summary to come), he talked about how the IPCC's projections assume that we will not try to eradicate malaria...

ADDENDUM (November 7, 2009): I've seen estimates that U.S. carbon emissions will be about 6% lower in 2009 as a result of the recession, which amounts to considerable progress towards the Boxer-Kerry target. Projections of an economic recovery in 2010 strike me as overly optimistic; in my opinion there's a strong possibility that we haven't hit bottom yet and there's worse to come. Still, though, I think Pielke's probably right that energy consumption will go right back up again unless the recession becomes a depression and results in significant changes in consumption habits.

My summary of Pielke's lecture on climate change adaptation is here.

ADDENDUM (November 9, 2009): It should be noted that Roger Pielke, Jr. is a somewhat controversial figure in the climate change debate, and believed by many in the climate change blogosophere to be in the climate change skeptic camp, or to be biased towards them in terms of where he levels his criticisms. A post titled "Who Framed Roger Pielke?" from the Only In It For the Gold blog links to a number of opinions expressing these views.

UPDATE (February 5, 2010): A post titled "The Honest Joker" at Rabett Run critiques Pielke Jr.'s stance as an "honest broker" as a sham.

UPDATE (August 28, 2010): A talk by Pielke that appears to have some similarity to this one may be found here.

UPDATE (July 13, 2014): An updated version of the information in this talk is Ch. 3 of Pielke Jr.'s book, The Climate Fix (2010, Basic Books).


EliRabett said...

Not sure that you want comments or not, but a good place to start analyzing the talk is to realize that greenhouse gases accumulate, so there is great virtue in reducing or at least limiting current emissions by substitution and conservation.

Roger ignores this issue. He is not only assuming that a miracle will occur, but that it will occur on schedule.

Lippard said...

Comments are welcome and appreciated.

What exactly is the issue that he ignored? He did discuss GHG accumulation, and his argument in the last part of the talk expressed skepticism about the likelihood of substitution occurring on the timeline necessary to reach proposed emissions targets. He didn't argue that we shouldn't do it--his argument for "decarbonization" is explicitly an argument that we *should* replace our existing energy infrastructure with non-CO2-emitting infrastructure.

You're right that he didn't say much about conservation--he assumed continuing growth in energy demand.

It's not clear to me that he's assuming a well-timed miracle. Can you elaborate?