RealClimate logo

Note 3/23/2021: we had a few hiccups with comments after moving the site to https/SSL. Hopefully they're fixed now. Please let us know if there are remaining issues.

An open letter to Steve Levitt

Filed under: — raypierre @ 29 October 2009

Dear Mr. Levitt,

The problem of global warming is so big that solving it will require creative thinking from many disciplines. Economists have much to contribute to this effort, particularly with regard to the question of how various means of putting a price on carbon emissions may alter human behavior. Some of the lines of thinking in your first book, Freakonomics, could well have had a bearing on this issue, if brought to bear on the carbon emissions problem. I have very much enjoyed and benefited from the growing collaborations between Geosciences and the Economics department here at the University of Chicago, and had hoped someday to have the pleasure of making your acquaintance. It is more in disappointment than anger that I am writing to you now.

I am addressing this to you rather than your journalist-coauthor because one has become all too accustomed to tendentious screeds from media personalities (think Glenn Beck) with a reckless disregard for the truth. However, if it has come to pass that we can’t expect the William B. Ogden Distinguished Service Professor (and Clark Medalist to boot) at a top-rated department of a respected university to think clearly and honestly with numbers, we are indeed in a sad way.

By now there have been many detailed dissections of everything that is wrong with the treatment of climate in Superfreakonomics , but what has been lost amidst all that extensive discussion is how really simple it would have been to get this stuff right. The problem wasn’t necessarily that you talked to the wrong experts or talked to too few of them. The problem was that you failed to do the most elementary thinking needed to see if what they were saying (or what you thought they were saying) in fact made any sense. If you were stupid, it wouldn’t be so bad to have messed up such elementary reasoning, but I don’t by any means think you are stupid. That makes the failure to do the thinking all the more disappointing. I will take Nathan Myhrvold’s claim about solar cells, which you quoted prominently in your book, as an example.

As quoted by you, Mr. Myhrvold claimed, in effect, that it was pointless to try to solve global warming by building solar cells, because they are black and absorb all the solar energy that hits them, but convert only some 12% to electricity while radiating the rest as heat, warming the planet. Now, maybe you were dazzled by Mr Myhrvold’s brilliance, but don’t we try to teach our students to think for themselves? Let’s go through the arithmetic step by step and see how it comes out. It’s not hard.

Let’s do the thought experiment of building a solar array to generate the entire world’s present electricity consumption, and see what the extra absorption of sunlight by the array does to climate. First we need to find the electricity consumption. Just do a Google search on “World electricity consumption” and here you are:


Now, that’s the total electric energy consumed during the year, and you can turn that into the rate of energy consumption (measured in Watts, just like the world was one big light bulb) by dividing kilowatt hours by the number of hours in a year, and multiplying by 1000 to convert kilowatts into watts. The answer is two trillion Watts, in round numbers. How much area of solar cells do you need to generate this? On average, about 200 Watts falls on each square meter of Earth’s surface, but you might preferentially put your cells in sunnier, clearer places, so let’s call it 250 Watts per square meter. With a 15% efficiency, which is middling for present technology the area you need is

2 trillion Watts/(.15 X 250. Watts per square meter)

or 53,333 square kilometers. That’s a square 231 kilometers on a side, or about the size of a single cell of a typical general circulation model grid box. If we put it on the globe, it looks like this:


So already you should be beginning to suspect that this is a pretty trivial part of the Earth’s surface, and maybe unlikely to have much of an effect on the overall absorbed sunlight. In fact, it’s only 0.01% of the Earth’s surface. The numbers I used to do this calculation can all be found in Wikipedia, or even in a good paperbound World Almanac.

But we should go further, and look at the actual amount of extra solar energy absorbed. As many reviewers of Superfreakonomics have noted, solar cells aren’t actually black, but that’s not the main issue. For the sake of argument, let’s just assume they absorb all the sunlight that falls on them. In my business, we call that “zero albedo” (i.e. zero reflectivity). As many commentators also noted, the albedo of real solar cells is no lower than materials like roofs that they are often placed on, so that solar cells don’t necessarily increase absorbed solar energy at all. Let’s ignore that, though. After all, you might want to put your solar cells in the desert, and you might try to cool the planet by painting your roof white. The albedo of desert sand can also be found easily by doing a Google search on “Albedo Sahara Desert,” for example. Here’s what you get:


So, let’s say that sand has a 50% albedo. That means that each square meter of black solar cell absorbs an extra 125 Watts that otherwise would have been reflected by the sand (i.e. 50% of the 250 Watts per square meter of sunlight). Multiplying by the area of solar cell, we get 6.66 trillion Watts.

That 6.66 trillion Watts is the “waste heat” that is a byproduct of generating electricity by using solar cells. All means of generating electricity involve waste heat, and fossil fuels are not an exception. A typical coal-fired power plant only is around 33% efficient, so you would need to release 6 trillion Watts of heat to burn the coal to make our 2 trillion Watts of electricity. That makes the waste heat of solar cells vs. coal basically a wash, and we could stop right there, but let’s continue our exercise in thinking with numbers anyway.

Wherever it comes from, waste heat is not usually taken into account in global climate calculations for the simple reason that it is utterly trivial in comparison to the heat trapped by the carbon dioxide that is released when you burn fossil fuels to supply energy. For example, that 6 trillion Watts of waste heat from coal burning would amount to only 0.012 Watts per square meter of the Earth’s surface. Without even thinking very hard, you can realize that this is a tiny number compared to the heat-trapping effect of CO2. As a general point of reference, the extra heat trapped by CO2 at the point where you’ve burned enough coal to double the atmospheric CO2 concentration is about 4 Watts per square meter of the Earth’s surface — over 300 times the effect of the waste heat.

The “4 Watts per square meter” statistic gives us an easy point of reference because it is available from any number of easily accessible sources, such as the IPCC Technical Summary or David Archer’s basic textbook that came out of our “Global Warming for Poets” core course. Another simple way to grasp the insignificance of the waste heat effect is to turn it into a temperature change using the standard climate sensitivity of 1 degree C of warming for each 2 Watts per square meter of heat added to the energy budget of the planet (this sensitivity factor also being readily available from sources like the ones I just pointed out). That gives us a warming of 0.006 degrees C for the waste heat from coal burning, and much less for the incremental heat from switching to solar cells. It doesn’t take a lot of thinking to realize that this is a trivial number compared to the magnitude of warming expected from a doubling of CO2.

With just a little more calculation, it’s possible to do a more precise and informative comparison. For coal-fired generation,each kilowatt-hour produced results in emissions of about a quarter kilogram of carbon into the atmosphere in the form of carbon dioxide. For our 16.83 trillion kilowatt-hours of electricity produced each year, we then would emit 4.2 trillion kilograms of carbon, i.e. 4.2 gigatonnes each year. Unlike energy, carbon dioxide accumulates in the atmosphere, and builds up year after year. It is only slowly removed by absorption into the ocean, over hundreds to thousands of years. After a hundred years, 420 gigatonnes will have been emitted, and if half that remains in the atmosphere (remember, rough estimates suffice to make the point here) the atmospheric stock of CO2 carbon will increase by 210 gigatonnes, or 30% of the pre-industrial atmospheric stock of about 700 gigatonnes of carbon. To get the heat trapped by CO2 from that amount of increase, we need to reach all the way back into middle-school math and use the awesome tool of logarithms; the number is

(4 Watts per square meter) X log2(1.3)

or 1.5 Watts per square meter. In other words, by the time a hundred years have passed, the heat trapped each year from the CO2 emitted by using coal instead of solar energy to produce electricity is 125 times the effect of the fossil fuel waste heat. And remember that the incremental waste heat from switching to solar cells is even smaller than the fossil fuel waste heat. What’s more, because each passing year sees more CO2 accumulate in the atmosphere, the heat trapping by CO2 continues to go up, while the effect of the waste heat from the fossil fuels or solar cells needed to produce a given amount of electricity stays fixed. Another way of putting it is that the climate effect from the waste heat produced by any kind of power plant is a one-off thing that you incur when you build the plant, whereas the warming effect of the CO2 produced by fossil fuel plants continues to accumulate year after year. The warming effect of the CO2 is a legacy that will continue for many centuries after the coal has run out and the ruins of the power plant are moldering away.

Note that you don’t actually have to wait a hundred years to see the benefit of switching to solar cells. The same arithmetic shows that even at the end of the very first year of operation, the CO2 emissions prevented by the solar array would have trapped 0.017 Watts per square meter if released into the atmosphere. So, at the end of the first year you already come out ahead even if you neglect the waste heat that would have been emitted by burning fossil fuels instead.

So, the bottom line here is that the heat-trapping effect of CO2 is the 800-pound gorilla in climate change. In comparison, waste heat is a trivial contribution to global warming whether the waste heat comes from solar cells or from fossil fuels. Moreover, the incremental waste heat from switching from coal to solar is an even more trivial number, even if you allow for some improvement in the efficiency of coal-fired power plants and ignore any possible improvements in the efficiency of solar cells. So: trivial,trivial trivial. Simple, isn’t it?

By the way, the issue of whether waste heat is an important factor in global warming is one of the questions most commonly asked by students who are first learning about energy budgets and climate change. So, there are no shortage of places where you can learn about this sort of thing. For example, a simple Google search on the words “Global Warming Waste Heat” turns up several pages of accurate references explaining the issue in elementary terms for beginners. Including this article from Wikipedia:


A more substantive (though in the end almost equally trivial) issue is the carbon emitted in the course of manufacturing solar cells, but that is not the matter at hand here. The point here is that really simple arithmetic, which you could not be bothered to do, would have been enough to tell you that the claim that the blackness of solar cells makes solar energy pointless is complete and utter nonsense. I don’t think you would have accepted such laziness and sloppiness in a term paper from one of your students, so why do you accept it from yourself? What does the failure to do such basic thinking with numbers say about the extent to which anything you write can be trusted? How do you think it reflects on the profession of economics when a member of that profession — somebody who that profession seems to esteem highly — publicly and noisily shows that he cannot be bothered to do simple arithmetic and elementary background reading? Not even for a subject of such paramount importance as global warming.

And it’s not as if the “black solar cell” gaffe was the only bit of academic malpractice in your book: among other things, the presentation of aerosol geoengineering as a harmless and cheap quick fix for global warming ignored a great deal of accessible and readily available material on the severe risks involved, as Gavin noted in his recent post. The fault here is not that you dared to advocate geoengineering as a solution. There is a broad spectrum of opinion among scientists about the amount of aerosol geoengineering research that is justified, but very few scientists think of it as anything but a desperate last-ditch attempt, or at best a strategy to be used in extreme moderation as part of a basket of strategies dominated by emissions reductions. You owed it to your readers to present a fair picture of the consequences of geoengineering, but chose not to do so.

May I suggest that if you should happen to need some friendly help next time you take on the topic of climate change, or would like to have a chat about why aerosol geoengineering might not be a cure-all, or just need a critical but informed opponent to bounce ideas off of, you don’t have to go very far. For example…


But given the way Superfreakonomics mangled Ken Caldeira’s rather nuanced views on geoengineering, let’s keep it off the record, eh?

Your colleague,

Raymond T. Pierrehumbert
Louis Block Professor in the Geophysical Sciences
The University of Chicago

807 Responses to “An open letter to Steve Levitt”

  1. 151
    jeffro says:

    I don’t know why I wasn’t aware of this blog before, but you can count me among your regular readers now. Just really excellent. Thanks for doing this.

    [Response: Welcome aboard! ]

  2. 152
    Hank Roberts says:

    Pough (do you mean Poe?)
    IF not, you should use the search function for the word, starting at the top.

  3. 153
    Eachran says:

    Well done raypierre.

    I remember the demolition job you did on Allegre some years ago, which I have since used on a number of occasions to dissuade people I know to have anything to do with the gentleman. His name recently came up as a possible adviser in Mr Sarkozy’s last mini reshuffle, fortunately M. Allegre was left in the pack (deck ? for USians).

    Mr Levitt should really die of shame for his latest effort.

    An economist friend of mine, who teaches at a UK business school, recommended, after it was reviewed in the UK press, that I read the chapter. I got hold of a version through Mr Connelly’s site (I think, or maybe a link from there). My friend asked me what I thought of it and I said that I thought that the science was wrong, the economics was junk, it was badly written and assuming he correctly reported the views of his interviewees (which I doubted) I didn’t think much of Mr Gates’ crowd either. He agreed with me.

    Mr Hogan’s point in 71 about ignoring fundamental physical relationships is well made.

    Unfortunately many professional economists tend to ignore physical and behavioural relationships. It is probably this which has led many to fail to forecast the financial mess, and for poor advice to be given on the so-called costs of mitigating and adapting to global heating (after a number of years reading your site and others, I am well off the alarm-o-meter). Mitigation, if started early enough (including today), costs nothing : it will have a non-negative impact on living standards in developed countries.

  4. 154
    David Miller says:

    Regarding #127

    Jim, there are several other carbon sequestration plans I’ve seen that make at least as much sense as liming the oceans. The advantage of liming the ocean is that it increases pH some, though that change can’t last until atmospheric levels of CO2 decline.

    The terrestrial equivalent of liming the ocean is spreading of olivine on land or pumping air through a variety of minerals which will react with CO2. See for a good rundown on different sinks.

    One sink I’m rapidly growing fond of is the production of biochar to be added to agricultural land. It acts as a nutrient sponge, lowering nutrient runoff and providing an ideal home for the biota that make soil more productive. It appears to have the potential to sequester carbon for centuries to millenia, reduces fertilizer and water requirements for crops, probably makes food grown on biochar amended soil more nutritious, and increases the amount of carbon pulled from the air by the crops. A very good thing all around. Biochar can be produced from wood waste (think pine bark beetle damage before forest fires get it), agricultural “waste” like corn stover and wheat straw, or municipal waste (leaves, grass clippings, pet waste).

  5. 155
    Eli Rabett says:

    If you include cogeneration (using the heat) natural gas systems can get above 75% (seem to recall that about 55% of that is in the form of electricity.

  6. 156
    Peter T says:

    I would not want to put anyone off reading Mr Levitt’s first book, Freakonomics, which I understand is quite entertaining. It may also be right about claims it makes in areas where I have no knowledge (such as professional sumo-wrestling). However, a quick skim showed it to be very shallow, not to say wrong, on a few things where I do have some knowledge, such as the economics and sociology of drug-dealing. I would advise, if you quote it at parties, to check the credentials of your listeners carefully first.

  7. 157
    George Blahusiak says:

    I hadn’t intended to go into this much detail, see below, but circumstances have overtaken me.

    World energy consumption per capita (US DOE[?]) 150 kw-hrs/day

    150 kw-hrs/day/person x 365 days/year = 54,750 kw-hrs/year per person

    54,750 kw-hrs/person/year x 6.5e9 people = 355,875,000,000,000 kw-hrs

    355.875 trillion kw-hrs/year x 2.388 (1.022 growth per annum over 40 years)

    849,829,500,000,000 kw-hrs/year(2050)

    If everyone consumes at US/Australia rates, 250 kw-hrs/day/person (A), multiply by 1.66, giving total energy consumption 1,415.8 trillion kw-hrs by 2050

    Some of this will be renewable, of course, and perhaps some of will be non-carbon producing, eg., French nuclear reactors, but the remainder is considerably more than the 16.83 trillion kw-hrs/year quoted in this rebuttal

    In this instance, using the numbers above, total area required would be 84 times larger, i.e., 4,479,972 sq kms.

    Pedants can argue all day about the relative carbon emissions of diesel, or petrol, or gas, or whatever, and the albedo of sand versus rooftops. I do not intend to do that here.

    What seems clear is that Professor Pierrehumbert has made the same mistake he accuses his colleague of, over simplification and not addressing the numbers. The problem isn’t simple, as he appears to state, and as soon as everyone recognizes that one basic fact the better off we will all be.

    [Response: No, you have made the mistake of only looking at the number that supports your preconceived notion and ignoring any number that upsets your world view. If you scale up the electricity usage, you increase the warming due to the albedo effect of the solar cells, but you also increase the warming due to the effect of the CO2 released burning coal to make the same amount of energy. The latter is still overwhelmingly dominant. It is true that under exponential growth you eventually reach the point where waste heat even from carbon free energy becomes an issue (think about Trantor) but we’re nowhere near that point yet. Further, you can do the arithmetic to show that if you reach the point where waste heat becomes a significant warming effect on the global scale, then your energy has to be coming from some source other than fossil fuels anyway. (Hint: there are only about 5000 GTonnes in carbon in coal resources. Make a reasonable supposition of the time span over which that is all burned up and converted to energy. Clathrates are a harder target to estimate, but try assuming clathrate carbon is 10X coal and see what you get) –raypierre]

  8. 158
    George Blahusiak says:

    Re: 8 – Funny you should say that. I looked up sea level rise in Wiki and found that almost all of sea level rise is (allegedly) due to thermal expansion, 3 mm (more or less) in the last year alone. Yet I looked up JPL and found that sea level temps haven’t changed during the last 4-5 years. Someone not checking their facts?

    [Response: This is getting off-topic, and moreover is a pretty egregious misreading of the facts. Take it somewhere else. –raypierre]

  9. 159
    Bruce Tabor says:

    jonathan at 129:
    “I saw that Professor Levitt used the word “blasphemy” in his weird response. If you’re at all familiar with the internet crank/crackpot scale, you know that’s like a plus 10 because only a crank portrays his work and his views as being persecuted by some religious orthodoxy.

    I’m sure it’s some side effect of being physically located so close to the theology department ;-)

  10. 160
    Bruce Tabor says:

    raypierre and gavin,
    From raypierre’s response to Levitt at 47:

    “Now regarding geoengineering, your “global warming quiz” in the NYT Freak blog had the main danger of geoengineering laid out right in front of you, but you failed to see it. Namely, you see the rapid recovery time after Pinatubo as evidence that geoengineering is harmless because you can reverse it any time. However, the bigger implication of that fact is that if you rely on geoengineering to allow the economy to burn up all the coal and raise the atmospheric CO2 to high levels, then if you ever have to stop, you are hit with the full effects of maybe a century or more of global warming practically all at once.”

    Is this not itself confusing “stock” and “flows”? What I mean is that the climate is in large part determined by the surface ocean heat content. That is why if we stopped adding CO2 to the atmosphere immediately we would still endure an additional 0.6 C warming as the planet settled into heat balance. So sulphate aerosols would buy time to the extent that they turned off the heating of the oceans. If removed, we would then resume heating but from a lower ocean heat content. This heating would be faster due to greater CO2, but that is not the same as being “hit with the full effects of maybe a century or more of global warming practically all at once”.

    That said I think this form of geoengineering would be a disaster for other reasons. It would likely produce catastrophic changes in weather patterns by reducing thermal gradients and destroy much of the ocean’s ecosystem as acidification would be unchecked.

  11. 161
    HappySkeptic says:

    Keep this kind of thing up. The most powerful arguments you guys can make are ones that lay people can understand themselves (as long as they remember some high-school science and a bit of logic).

  12. 162
    ubrew12 says:

    Aerosol cooling would ruin crop yields through lowered sunlight, destroy forests via acid rain, kill people outright who are sensitive to emphysema. That’s why these pollutants were limited by the Clean Air Act. I’m not completely against it, but there are consequences beyond merely dimming the planet to combate AGW.

    Per the fossil energy cost of manufacturing solar cells, it seems to me that one could design a solar cell manufacturing plant, self-contained, that did little more than take in sand and raw iron/Cu ore, take in sunlight, and spit out solar panels. Locate it in the desert. So now your solar panels take zero fossil energy to manufacture.

  13. 163
    Andrew Hobbs says:

    George Blahusiak #157

    Where did you get your figures.

    Total world energy production is about 5 x 10^20 Joules per year.(1)
    This is equivalent to 1.39 x 10^14 KWH per year. About 80 to 90% is derived from fossil fuel. ie about 1.2 x 10^14 KWH.
    With a population of 6.5 billion this equates to 18460 KWH per capita per year from fossil fuel, which is equal to 50.5 KWH per day per capita from fossil fuel(58 KWH for all energy production). Definitely not 150 KWH per day as you suggested. This factor alone cuts your result by a factor of three.

    You then compounded the error.

    Professor Pierrehumbert was addressing the specific point with regard to total current annual electricity production which, most recently,
    was calculated to be 16.8 trillion KWH for 2008(2). At an efficiency of about 35%, electricity production represents about 40% of total fossil fuel use.

    You based your final figure on the current (incorrect) energy production and assumed that the world needs to produce that amount of energy from solar power. However the amount of energy quoted is the total thermal energy produced by burning fossil fuel; not the amount of useful energy used. For example electricity production is only about 35% efficient, so for the 40% of the energy used in electricity production you only need to produce 35% of that amount by solar panels to completely replace 40% of total current fossil fuel use. (One could probably make a similar argument regarding replacement by solar power of fossil fuel use in even less efficient internal combustion engines etc). Another factor of 3 at least.

    In any case a back of the envelope calculation might have suggested a problem. Your “849,829,500,000,000 kw-hrs/year(2050)” might need 4 – 5 million square Km of solar panels, but it would need a formidable amount of fossil fuel too. At a rate of 250 grams of Carbon per KWH, (35% efficiency) this equates to about 212 GT of carbon (784 GT CO2) released into the atmosphere per year, almost 30 times the current rate. Not impossible, but atmospheric concentrations of CO2 would be increasing at rather more than the current 2 ppm per year. What one could call an interesting scenario.

    And, as for unhindered growth as understood by some economists, at current population growth rates, the earth will be 2 metres deep in humanity by 2600, and by 4000 we will be expanding out into space at the rate of about 1 km per hour.

    (1)”Energy – Consumption’, Statistical Review of World Energy 2009, BP.

    (2) CIA World Fact Book.

  14. 164

    John Burgy (137),
    Well said. We should be careful not to “make more enemies than friends” in the way we respond to others (including ‘the enemy’, if there even is one).
    (I guess I’m at risk being called a left wing pacifist having said these things more than once lately, but so be it)

  15. 165
    kejr says:

    Re 157 George Blahusiak says:
    1 November 2009 at 1:46 AM

    Why would you assume the world would necessarily follow US & Australia’s rather excessive and inefficient energy use? (Not sure many countries would support law banning use of washing line in garden to dry clothes forcing community to use tumble dryers and the like) And given there is a problem, support / progress to make buildings better insulated etc reduces energy consumption.

  16. 166
    Ike Solem says:

    You have to factor in local solar radiation as well as economics when thinking about green & white roofs and solar roofs, as well as local water issues. Here’s the U.S. solar irradiation map:

    Now, the same point made about the black solar roofs absorbing heat also applies to the white roofs – how much roof space is there, and how does it compare to the size of the necessary photovoltaic space? In other words, if we paint all the roofs white, and then subtract the space needed for the photovoltaic array, what % remains white? What then is the radiative effect on a global scale?

    I’m guessing the analysis would show that painting roofs white has a similarly minimal effect when it comes to cooling the planet – it’s more of an economic argument aimed at reducing air conditioning costs in the summer. This is also what raypierre says:

    “But, for either green roofs or white roofs, the big climate impact comes from the reductions in air conditioning needed, which reduces the CO2 emitted. – raypierre (#50)”

    You can add that a solar-powered roof combined with good building design can actually eliminate all fossil CO2 emissions related to air conditioning. It’s clearly the best choice for reducing fossil CO2. The only argument against it is the relatively high up-front cost, similar to that of a new automobile.

    However, if the argument is economic, what’s a better deal: setting up a solar PV roof under a financing arrangement that allows for a 10 year repayment (as with vehicles) and eliminates your utility bill payments, or painting your roof white? Paint is cheaper than solar panels, but if you own your own power station, and it runs on free fuel, that’s a better long term payoff. Of course, this argument is more convincing in Arizona than in upstate New York, because there’s more free sunlight in Arizona (where keeping a green roof would be implausible due to water limits).

    It would be nice to see the DOE Secretary, who is in charge of the planning for the nation’s energy future (along with the many DOE National Labs, such as LLNL & LANL), work this question out on a blackboard in front of the cameras:

    What are the effects of white vs. solar roofs, from the global radiative perspective, the energy-to-CO2 ratio perspective, and the economic perspective?

    Maybe Steve Levitt and the main go-to think tank on electricity for the U.S. press, the Electric Power Research Institute, could do the same.

  17. 167

    #108 BPL:

    You can also add these to your list of climate sensitivity values and sources:

    Knutti and Hegerl (2008) and IPCC (2007) conclude a climate sensitivity value of about 3 oC, with a likely range of about 2 – 4.5 oC.

    Knutti, R. & Hegerl, G. (2008). The equilibrium sensitivity of the earth’s temperature to radiation changes. Nature Geoscience, (1), 735 – 743.

  18. 168

    We have now entered the era of snarly wiki science.

    It’s about time. The planetary and space sciences are your friends.

  19. 169
    EL says:

    162. In addition,

    “355.875 trillion kw-hrs/year x 2.388 (1.022 growth per annum over 40 years)”

    A 1.022 growth rate per year does not double in 40 years; instead, it doubles in 68.1686 years; Thus, 534 instead of the 849.

  20. 170
    Jim Galasyn says:

    Australia oil well catches fire, after the fourth attempt to cap the well failed.

    A director of the company, Jose Martins, said the only way to stop the fire was to plug the leak.

    “The measures which we have been able to take so far can only mitigate the fire. They will not stop the fire.

    “The best way to stop the fire is to complete the well-kill and stop the flow of gas and oil at the surface from the H-1 well, cutting off the fuel source for the fire.”

    This thing just goes from bad to worse.

  21. 171

    Burgy, when your friends are idiots who fail to listen to reason and admit their mistakes, they are your enemies.

    These people are are not amenable to reason and they are not our friend, and thus should be treated with the respect that they deserve – none, thus the sarcasm.

    Even satire doesn’t seem to work on these people, and honestly, Burgy, you are indeed one of these people.

  22. 172
    slow to follow says:

    “World energy consumption per capita (US DOE[?]) 150 kw-hrs/day”

    GB – Please check this sum –

    150kWh/day/person means:

    150/24 kWh/h/person = 6.25kWh/h/person = 6.25kW/person continuous consumption worldwide

    Is the input data correct?
    Is the maths correct?
    Is the answer reasonable?
    What can you cross reference it to as a check and balance?

  23. 173
    Matt R. says:

    Nice post. I was lucky to be able to attend some talks on geoengineering at the KITP in Santa Barbara last year, which were fascinating (I’m a physicist, but have only an amateur’s interest in climate science). Ray, I seem to remember the punchline of your talk on “Damocles Earth” being that in a simplified model, tuning the sulfate aerosol concentration to cool the tropics to a reasonable temperature led to an ice age on most of the planet. Has there been more work along these lines? It seems intuitive to me — sulfates give you one knob to turn, but there’s a lot more to climate than just “average global temperature”, and it wouldn’t be too surprising if lowering the average temperature with this strange cocktail of CO2 and sulfates in the atmosphere led to vastly different climate patterns. This seems like it could potentially be a very strong argument against this sort of geoengineering strategy, but the counterarguments I’ve seen cropping up lately have been mostly about the geopolitical problems or ocean acidification than the effects of climate itself….

  24. 174
    Hank Roberts says:

    Burgy, distinguish the people posting comments from the Contributors, capital-C, listed in the sidebar and the About info.

    They don’t keep this place polite, though it seems to me they do work hard to keep pointless anger and flaming out of the discussion.

    There are places where screaming and abusing people is effective educationally. Well, one place — Army drill sergeants indoctrinating new inductees apparently do it quite effectively.

    But — imagine if they tried it using text messaging only.

    That’s what you get here sometimes, people who imagine they can be effectively convincing using drill-sergeant abusive behavior via text message.

    If you can sit back and find it amusing, it works better. If you’re a naive reader — especially a new one — and you read that kind of stuff, you don’t have the context to know who’s typing, what the history is, nor whether the person being abused ‘deserves it’ or not. So it’s a detriment.

    But — look at CA or WTF and see the same kind of stuff done by true believers there who are sure they’re right, abusing people they’re sure are idiots. It happens; if your ‘friends’ don’t see it there and only see it here, then they’re supplying the emotional tone themselves based on their own beliefs.

    All there is here is ASCII text. The feelings are supplied by the reader.

  25. 175
    Cherenkov says:

    231 Kilometers on a side provided you have no access between cells and no repair access roads.

    Because you will need access, I dare say that this number is a bit misleading. I would put the size at least larger by a factor of four.

    The next question is, how fast will it need to grow to keep up with human needs?

    Then, I must ask, why do we keep destroying the planet to obtain planet-destroying paradigms?

    Why not use this intellectual wattage to figure out how to get rid of the need for planet-destroying paradigms?

    Why must we keep using?

    This is not solvable using scientists and engineers. It is like using dynamite to prevent the production of dynamite factories. Any solution they offer will suffer the inevitable cycle of unintended consequences: first, invent a technology that is poorly understood (Everything is poorly understood), second, commercialize it and create a financial incentive to keep it going even when it is discovered to be just another planet-destroying paradigm, create a techno-fix to remedy the first cock-up, commercialize it to create a financial incentive to keep it going when it is discovered to have its own problems, create a techno-fix for the techno-fix, commercialize it….rinse and repeat.

    Technology will solve the problem of technology. Nature will solve the problem of technology. It is called die-off. The only question is will we decide to destroy the planet in an attempt to save it.

    I’m guessing the answer is a resounding yes.

  26. 176
    Sene Bauman says:


    First, my compliments, truly outstanding. You are my idol :).

    I am curious though, why you limited your discussion to photovoltaics and no mention of concentrator systems, such as Sterling Energy Systems use of sterling engines, or even solar towers? I believe Sterling claims they can produce at about 6 cents per kilowatt-hour now. Was your intent to focus on distributed, rather than on concentrated sources or was this simply intended to provide but one example out of many possible options in the solar arsenal?

  27. 177
  28. 178
    Chris Colose says:

    Matt (173)

    You are quite right! Even if you somehow keep pace with the rising CO2 by raising the albedo of the planet (e.g., with sulphate aerosols) the average global temperature can stay roughly constant with much larger regional changes in precipitation or temperature. You can imagine some of the political tension that would be raised between countires, if say, China or India or some other country were to have an anomalous drought or other storm event. It would become that much easier to blame whoever was controlling the knob on climate. This is just one of the many problems of Solar Radiation Management outlined in Alan Robock’s 20 Reasons Geoengineering may be a bad idea.

    Really, the whole raising of the albedo thought experiments (in their many forms) are all unconvincing. They cannot tackle the problem of ocean acidification, they all must be managed on timescales similar to that of the perturbed CO2 residence time in the atmosphere (which is not practical), they all involve a lot of regional climate responses not necessarily correlating with the global mean, you have the ‘unknown unknowns’ component, and many other reasons. The climate system is not a game where you can just turn different variables up and down simultaneously and hope to get the pre-industrial (or current) climate regime as a result.

  29. 179
    Brian Brademeyer says:

    EL (169)

    The growth rate is not 1.022 per year, but 2.2% per year. Doubles in 32 years.

  30. 180
    Beast Of Bodmin says:


    “Levitt’s style of research emphasizes asking questions to understand the economics of real-life issues.”


  31. 181
    Greg says:

    @47: “Blasphemy”? Still going with the “religion” meme, I see, Steve. Religious faith is “belief without evidence”. The only people practising that around here are you and Myhrvold.

  32. 182
    Ike Solem says:

    Bruce Tabor, you’re getting some things backwards here, but so is Levitt:

    “What I mean is that the climate is in large part determined by the surface ocean heat content…”

    That is wrong – the climate (assuming a given land-ocean configuration) is mainly determined by the long-term atmospheric forcing, and the surface ocean responds to that forcing. If the atmosphere is warming, the ocean lags but eventually heats up – and if the atmosphere cools off due to a draw-down of CO2, the ocean’s heat lingers on for some time, until cooling sets in. Don’t confuse response and forcing.

    The ocean acts as a thermal buffer for changes in atmospheric forcing, especially the deep ocean, which is distinct from the surface mixed layer. The thermal buffering also creates a CO2 buffer effect – and that, in turn, also helps explain CO2 patterns in glacial cycles – CO2 increases initially lag behind the temperature increase, because the ocean outgases CO2 as it warms (think of a thawing freezer full of food – first it warms up, then it begins to smell).

    Levitt and others are trying to use this fact to argue for ocean-based geoengineering based on pumping warm water down below the thermocline, or vice versa. However, their argument is without much physical basis. To understand why, see Mark Denny’s 2008 Princeton textbook”>How the ocean works: an introduction to oceanography”, pg170:

    “…With this background, we now can tackle our original question: how much energy is required to increase the depth of the thermocline by mixing water up from below?….Let’s suppose we stir the ocean’s surface, and as a result, high-density water from a thin stratum of the lower layer is mixed with the water of the upper layer… After this mixing, the surface mixed layer is thicker than before and the ocean has a new potential energy. In effect, some of the dense water from the lower layer has been lifted upward, and that requires an increase in the potential energy of the system.”

    Now, we can start to consider surface ocean cooling via cross-thermocline pumping – for some details, see Levitt as paraphrased by Newsweek in their review of Superfreakonomics:

    “Large, floating cylinders in the Gulf of Mexico could cheaply push warm surface water deeper, cooling the ocean and disrupting the conditions in which hurricanes are born…”

    The energy calculation diagram for a simple case is found in Denny 2008, pg 170. Any responsible geoengineering scheme would begin there, but reliable answers require full scale models. If someone wanted to build a skyscraper, and offered to do the necessary calculations on a napkin, would you be impressed or skeptical? However, there aren’t even rough calculations – Levitt is “not a scientist” – so who did he rely on for this disinformation? Hard to say.

    For a good rebuttal to the claim, see Scientific American, Oct 23, 2009:

    “If you take 20 gigawatts of heat away from surface, you think that has to cool it, but that is not necessarily true,” Smyth says. “What it’s actually going to do is raise the base of the mixed layer. If the base is at 50 meters, and you pump away the upper meter of the ocean, the mixed layer will then extend down to 49 meters. It’s not that the 20 gigawatts disappear into thin air. It’s just that it’s not doing anything useful in terms of changing sea-surface temperature.”

    It is true that a shallower mixed layer can inhibit hurricanes – the strongest ones form over deep warm cores – but what about the engineering scale needed? How many pumps per square kilometer? You also have your mechanical engineering issues, ocean currents, replacement costs…

    Thus, geoengineering schemes based on pumping water across the thermocline appear to be little more than arguments for maintaining the status quo. Like the other ocean-based bio-geoengineering scheme, iron fertilization, the whole thing falls apart when examined closely.

    Similarly, no quasiperiodic-or-chaotic oceanic fluctuations (ENSO, for example) will have much effect on the long-term atmospheric forcing induced by fossil CO2. Ocean heat content continues to rise, as the models predicted – and yes, there are fluctuations, but that’s expected:

    P.S. In terms of “fair and balanced” media coverage and book reviews, try searching Google News for mentions of “Climate Cover Up” (32) vs. “Superfreakonomics” (826)…

  33. 183
    Deech56 says:

    Science writer Carl Zimmer (The Loom)labels this post as his entry to the mythical category “The Best Pwnage of 2009”.

  34. 184
    Halldór Björnsson says:

    The world total energy use quoted above (@163) 5E20 J per year is roughly the same amount of energy as the globe receives from the sun in one hour.

    If all this energy is eventually turnes into waste heat it still is tiny in comparison with the numbers in the annual energy balance of the globe. (There are 8760 hours per year).

    The fact that we are using about one hour of total solar energy per year has however, implications for the Trantor limit that (see Raypierre reply @157). With reasonable efficiency assumptions and growth rate for energy use you can start estimating long time would pass until when human energy usage becomes a significant fraction of total solar.

    [If the energy use grows by 1.5% per year, the doubling time of energy use is slightly less than 50 years]

  35. 185
    EL says:


    His exact quote: (1.022 growth per annum over 40 years)

    0.o 1.022 growth = 102.2% growth or 1.022% growth

    I see what you are saying though. Apparently he is attempting to state the units are 100% and they grow 2.2%.

  36. 186

    This is not solvable using scientists and engineers.

    Actually it is, but since that is roughly equivalent to evolution, you have to keep going. That involves embracing science and technology, as opposed to Hollywood CGI movies and religion, moving off the planet, and not looking back. Humans and in particular Americans, judged by the behavior of NASA these last four years, and the US government the last eight years, apparently aren’t even ready to take the first steps of that process in any substantive or rational way.

    Nine billion people all demanding and striving for a quality of life of 350 million in America alone? Give me a break. Develop low Earth orbit or die. That’s your immediate choice. Nobody can predict how this will play out, but unless you come up with multiple Manhattan style projects in both condensed matter and quantum physics (not high energy or nuclear physics or fusion crap – real everyday physics from absolute zero to 1100 K or so) AND low cost earth to low Earth orbit transportation and LEO development (RLVs, closed ecological life support systems, Earth observation and observatories) then you simply don’t have a chance here. The choice is very stark and it is yours alone, but most Americans are so dumbed down by substandard education and the main stream media now they just drink the koolaide and believe any old crap. Don’t even get me started. You’re breaking up, I can’t read you … over.

  37. 187
    Bruce Tabor says:

    Thanks Ike Solem at 182,

    I’m quite happy to get things backwards or plain wrong, and be corrected. I find I learn more that way and (hopefully) am wiser the next time around.

    I’ve not read Superfreakonomics (and won’t bother), but I have read Freakonomics and as a statistician (and engineer) I found the arguments facile – often lacking insight into the complexity of the issues. It’s as if an economist suddenly discovered you can use statistics to study subjects other than economics – when other researchers (such as epidemiologists, social scientists etc.) have been doing this for generations. But Levitt ignores the shortcomings and inbuilt biases in such analyses. (I guess not questioning the assumptions of your model is stock in trade for an economist.)

    My issue was not with geoengineering to stop hurricanes, but with the injection of SO2 into the stratosphere, which was obliquely referred to in Raypierre’s reference to Pinatubo.

    Let me put the issue in more stark terms. To me the phrase, “…if you rely on geoengineering to allow the economy to burn up all the coal and raise the atmospheric CO2 to high levels, then if you ever have to stop, you are hit with the full effects of maybe a century or more of global warming practically all at once” implies the following (to me at least).

    Say humanity immediately began injecting enough SO2 into the stratosphere to cancel any net forcing due to GHGs. Say we did this until 2100 and continued business as usual (BAU) with CO2 emissions. In contrast, say that if we had not done this the world in 2100 would be on average 6 degrees hotter. Our clever use of SO2 injection has avoided 6 degrees of warming and saved a lot of sea level rise! (At enormous cost to ocean biodiversity and important weather systems like the monsoons and probably many other “unknown unknowns”.)

    Now the plain reading of the phrase, “…you are hit with the full effects of maybe a century or more of global warming practically all at once” used by Raypierre (a similar expression by Gavin in an earlier post) would suggest to me that if we suddenly stopped the SO2 injection, we would instantly go into a world 6 degrees hotter that practically identical in climate to a world where we had followed BAU with no geoengineering.

    My contention is that, yes, we would suddenly be exposed to warming at a much greater rate than in 2009, as the forcing would be so much greater. But I also contend that it would still take some time – not “all at once”, but perhaps decades – to get to a +6 degrees world.

    You are virtually saying this yourself: “…the climate…is mainly determined by the long-term atmospheric forcing, and the surface ocean responds to that forcing.” That is it is not the instantaneous forcing – the heat flow – but its integral – the heat stock or energy. The main place that heat energy is stored on the planet (over the medium term) is in the surface ocean – the ocean heat content. Thanks for the graphic.

    My final contention is that the ocean limits the effect of a transient forcing on the climate (it must be heated to drive change climate), but also is the main source of the enormous momentum in climate change, which is why the world would continue warming even if we add no more CO2, and why if you suddenly switch off geoengineering you don’t go through a step change to a new climate state. I still takes time.

    I have absolutely no dispute with your arguments against Levitt’s hurricane mitigation proposal. Nor do I dispute Raypierre and Gavin’s criticisms of Levitt’s book. Raypierre’s analysis was very informative and insightful. Thanks Raypierre. And I am not a proponent of this form (SO2 injection) of geoengineering.

  38. 188
    JCH says:

    I’m a big fan of Ludiwici clay roof tiles. They are making a barrel style that come in a glaze of semi-gloss matte white. Maintaining it with a garden hose in gleaming matte white for a few centuries would be very easy. Surely matte white, in Texas anyway, is greener than green. If not, would a mix of light beige and tans and grays be better compromise for both summer and winter? That way I could have a scratched shake look, which is what I prefer looks wise.

  39. 189
    gregory says:

    In the most literal sense, highly concentrated nonrenewable energy has shaped today’s economy. In order to maintain the existing institutional superstructure, we would have to continue to rely on a highly concentrated flow of energy through the system. Solar energy, however, is not concentrated like nonrenewable energy. because solar radiation is diffuse, it must be concentrated to do work. Since the laws of thermodynamics tell us that work can only be performed when there is a temperature difference between two places, and since solar energy falls essentially equally on each square foot of land in any given geographical area, the solar flow must be collected. If electricity is desired, the stored solar energy must be transformed from one state into another. The nature of the flow and the economies of scale indicate that in order to run our current industrial superstructure we would need to cover between 10 and 20 percent of the total U.S. land area with various types of solar collectors. Manhattan daily consumes more than six times the energy that could be provided by a 100 percent efficient collector of all the solar flow that falls on the city. To power New York City through various solar techniques, an area many times the city’s mass would have to be given over entirely to solar collectors. While New York is obviously unique in its consumption of energy, other major urban areas will be subject to similar strictures in the solar age.

    The sheer size of the solar infrastructure that would have to be erected to maintain society is mind-boggling. So too is the amount of time and labor that would be required to build and sustain it.

    Although the evidence suggests otherwise, let us assume that new collecting techniques could be discovered that would allow us to very efficiently concentrate the flow of solar radiation far beyond the capacity now available or even deemed conceivable by many engineers. If this remarkably efficient recovery process were somehow possible, we could then support an urbanized, industrial-technological society through solar flow. But what would be the result? Simply this: we would continue to witness the exponential increase of entropy here on earth as solar energy is used to convert more and more of our limited terrestrial energy resources (matter) into the production process, transforming them from a usable to an unusable state. It is not, then, just the form of energy a society uses that is critical; it is also the amount of energy. If solar energy actually could flow in highly concentrated forms for industrial use, we would experience many of the same economic and social dislocations that result from our high energy use now. That is because the use of solar energy cannot be divorced from the stock of fixed terrestrial matter that it interacts on and converts. In living and an industrial processes, solar energy must always be combined with other terrestrial resources in order to produce a product. That conversion process always results in the further dissipation of the fixed stock of terrestrial resources on the planet.

    Nicholas Georgescu-Roegen of Harvard University pointed out the obvious flaw in current approaches to the harnessing of solar energy:

    The truth is that any presently feasible recipe for the direct use of solar energy is a “parasite,” as it were, of the current technology, based mainly on fossil fuels. All the necessary equipment (including the collectors) are produced by recipes based on sources of energy other than the sun’s. And it goes without saying that, like all parasites, any solar technology based on the present feasible recipes would subsist only as long as its “host” survives….The intensity of solar radiation reaching the ground level being extremely weak, a large material scaffold is needed for its collection….It is highly plausible that the difficulty may not be superable at all, given that the intensity of solar radiation is a cosmological constant beyond our control.

  40. 190

    Oh the Super-Schadenfruende!
    Now the chapter on prostitution is starting to kick up some dust. This in the Guardian.

    Good news, ladies. You, too, can make millions by
    charging for sex! And you’ll just have a slam-bang, gee-
    golly splendiferous time doing it, too — at least if
    you absolutely adore the sort of men who pay for it. Be
    warned, however: Disliking those men will consign you to
    the minimum-wage ranks of sex professionals, forever
    longing for the big bucks you could be earning, had you
    only an appropriately chipper attitude.

    Such is the advice of Steven Levitt and Stephen Dubner,
    of Freakonomics fame…. — snip–

  41. 191
    Mike says:

    Your article states that solar panels can produce 15% X 250 watts per square meter = 37.5 watts per square meter.

    How did you arrive at this amount? My work with solar panels at 33 degrees North between February and June generated about 5 watts per square meter.

    Is your 37.5 watts verified by any real world numbers?

  42. 192
    Patrick 027 says:

    Re 175 Cherenkov says:

    “231 Kilometers on a side provided you have no access between cells and no repair access roads”

    The area between the actual collectors would not generally also experience reduced albedo, and might actually have an enhanced albedo if so decided. What matters is the albedo of the solar panels or other devices, and the insolation that falls on them.

  43. 193
    Richard says:

    I find it disturbing that other people have to lie when it comes to the fate of the planet.

  44. 194
    Rene says:

    There seems to be a constant confusion here between the terms
    ‘denialist’ – someone who denies AGW is a problem – and
    ‘skeptic’ – someone who questions or doubts whether it is.

  45. 195


    The nature of the flow and the economies of scale indicate that in order to run our current industrial superstructure we would need to cover between 10 and 20 percent of the total U.S. land area with various types of solar collectors.

    US energy consumption was 1.05 x 10^20 joules in 2005. This translates to a mean power of 3.30 x 10^12 watts. Mean sunlight absorbed by the climate system is 237 watts per square meter. Therefore, at 10% conversion efficiency, you would need 1.40 x 10^10 square meters of solar cells (an area 119 km on a side). The area of the United States is 9.63 x 10^12 square meters. The fraction needed is therefore 0.0015 or 0.15%. Your estimate is too high by two orders of magnitude, a factor of about 100 (67-133).

  46. 196
    Donald Oats says:

    Re #186: Putting people into space is an old idea that also fails the energy analysis test, or what we used to call “running the numbers” on a problem. If you take this blog article as an exemplar on how to do simplified accounting of energy, then it just doesn’t work out.
    A discussion of Gerard O’Neill’s “Islands in Space” may be found in Robert Park’s book “Superstition: Belief in the Age of Science”, Chapter 12. He also gives a description of one of his physics examples for students, that of getting humans to another star. To quote a little bit:
    “On the first day of class, I ask how many believe that someday in the future humans will be able to travel to another star and its planets. Weaned on Str Trek, most of them—sometimes all of them—raise their hands. So we set aside a few minutes of every class period to plan the trip. Each class period I ask for volunteers to come to the next class with number we will need for the next step in planning the mission, such as:
    * How far is it to the nearest star?
    * How long would they be willing to spend traveling?
    * How fast must they travel to make the round trip in that time?
    * How many people should be in the crew?
    * How big must the spacecraft be?
    * What would they need to take along?
    Finally, as we approach the end of the semester, we agree on a conservative estimate of the total mass of the spacecraft. I tell them to come to the next class with a number for the annual energy consumption of all humans on Earth to use as a reference point. We are now ready for the final calculation. A few in the class begin to twitter, having figured out where we’ve headed, they’ve already done the calculation. Using simple Newtonian mechanics, we calculate the the energy required to accelerate a spacecraft of that size to the velocity needed to make the trip in that period—one-half the mass times the velocity squared—and compare that with annual human energy consumption on the entire planet. The energy needed for a trip to the nearest star in the lifetime of a human—by now they’re all tittering—is many thousands of times greater than all the energy that is expended on Earth in a year. So great, in fact, that there is no point in quibbling over the assumptions. It’s just out of the question.”

    Great book. I recommend it.

  47. 197
    Mark says:

    “In the most literal sense, highly concentrated nonrenewable energy has shaped today’s economy. In order to maintain the existing institutional superstructure, we would have to continue to rely on a highly concentrated flow of energy through the system. Solar energy, however, is not concentrated like nonrenewable energy”

    Gregory, today we have technology.

    Cut out the 19th century thinking and get into the year 2000.

    (PS our electrical grid is a concentrated form of energy. And we can connect many solar plants to it.)

  48. 198
    Mark says:

    “231 Kilometers on a side provided you have no access between cells and no repair access roads. ”

    Who says that the tiles can’t be conveniently off the ground so you can repair from underneath?

    Or do you prefer to think up problems and hope nobody thinks whether they exist?

  49. 199
    Fred Magyar says:

    Wow! That take down certainly ties if not tops this one of an evolutionary biologist, Dr. Lenski vs. a creationist.

    Unfortunately the uncritical and lazy thinkers of the world far outnumber scientists and educated laymen…

    Go Ray! You rock!

  50. 200
    Mark says:

    “‘denialist’ – someone who denies AGW is a problem – and
    ’skeptic’ – someone who questions or doubts whether it is.”

    Incorrect and you have just demonstrated WHY you and others are called denialist rather than skeptic.

    A sceptic is someone who questions an assumption NO MATTER WHERE IT IS ON THE CLAIM OF AGW.

    And the confusion is the deniers like you who have got this wrong and therefore erroneously justify your label to yourself as “skeptic” where it really IS “denialist”.