Artificial Volcanoes

I’m a bit late to this party, but must report:  We’ve known for decades how to fix global warming cheaply!  Many fear devastation if warming continues, while others fear impoverishment from quitting carbon.  But we can cancel the entire warming effect for less than one part in ten thousand of world product!  No new tech is needed, nor much preparation to start fast or stop suddenly if problems appear.  We just copy the cooling effect already seen in volcanoes:  For a few $B/yr or less we take a tiny fraction of the SO2 humans already emit and put it way up in the stratosphere, e.g., via Naval guns, where it blocks sunlight.   

A burst of news about a year ago included this in Time and this in Nature.  It isn’t a perfect fix:

Carbon dioxide does more than just warm – it also acidifies the ocean. Even if the warming effects of ever-increasing carbon dioxide could be cancelled out, the effects on corals, shellfish and eventually the entire marine food web would still be disastrous. … The pattern of warming expected from carbon dioxide, and the pattern of cooling expected from aerosols, would differ in both space and time. … Volcanic eruptions … seem to have an unfortunate side effect; the 1783 Laki eruption in Iceland, for instance, weakened the Indian monsoon and cut rains in the Sahel, in Africa, to boot.

It would also increase ozone depletion a bit.  But these seem minor compared with dire warnings on warming.  And people seem to have unfairly lumped this very solid approach with far more speculative approaches, e.g., orbiting shades or iron ocean seeding.  Volcano shading is all well-understood physics and chemistry!  Worse, it faces serious ideological opposition and academic indifference:

Much of the climate community still views the idea with deep suspicion or outright hostility. Geoengineering, many say, is a way to feed society’s addiction to fossil fuels. "It’s like a junkie figuring out new ways of stealing from his children," … This new interest in geoengineering was set off by … Paul Crutzen … in August 2006. [His] article contained relatively little that wasn’t already in the literature … but it had a major impact because of who was saying it. … "Nobelist and general environmental worrier Paul Crutzen – someone who showed the world the risks of ozone depletion very early on." … It was for exactly this reason that Crutzen’s colleague Andreae urged him not to publish. …
Lowell Wood … is a forceful spokesman for extreme ideas, most notoriously the proposed X-ray laser … of Ronald Reagan’s Star Wars programme.  In the 1990s, he had become enamoured of radiation management, as had his mentor, Edward Teller, Livermore’s hydrogen-bomb-begetting eminence gris. If geoengineering had not already had a bad name among climate scientists concerned about the environment, Teller’s championing of the idea in the pages of the Wall Street Journal would have won it one. …
A confident understanding of geoengineering’s promises and problems would require years of dedicated work from groups all over the world, an effort comparable to that reflected in the IPCC’s massive reports on the natural science of climate change. And even that, say critics, would not be enough. … The very thing that motivates people like Crutzen to study geoengineering – the risk of large surprises that require immediate action – leads others to see the whole idea as fundamentally unworkable … it is easily argued that betting the monsoon on the ability of models to accurately capture such subtleties would require a foolhardy level of trust.  …
[NAS pres.] Ralph Cicerone … singles [Angel’s paper on orbital sunshades] out for praise for the painstakingly careful way it was done. … For him and many others, such academic excellence is the main point of publishing research on geoengineering.  For these researchers, the aim is not to find feasible solutions but to do good science that provides a standard against which to judge the less good, or flatly foolish, schemes that might otherwise accrete around the idea. … [Most say] models of geoengineering’s benefits need to be a lot more accurate than models of the harm that will be done in its absence. … "the role of a geoscientist is to understand nature, not to change it." Climate scientists … [are] happy to advocate … massive changes in technology, in geopolitics, in social norms … Not changes in the workings of the stratosphere. Not changes in the natural.

There are risks from changing economies as well as from changing stratospheres, and we understand the later far better than the former.  If we cared more about global warming than symbolic gestures for planetary purity, we’d likely accept the risks and start artificial volcanoes as soon as we thought warming was hurting us.   

From a lunch with Lane Lee, arranged by Bryan Caplan.  An authoritative tech review is here.

Added:  I said it was a cheap solution, not a permanent and complete solution – I even included quotes making it clear this can’t exactly and forever counter arbitrary CO2 emmisions.  But since interest rates are positive, the longer we can delay pain the cheaper is that pain, and this looks it can delay pain well. 

Added 11July: Scott Barrett has a good review.  HT to Pedro Linares.  Amazingly, this whole discussion goes way back:

In 1965, the President’s Science Advisory Committee warned in a report called Restoring the Quality of Our Environment that increases in atmospheric carbon dioxide due to the burning of fossil fuels would modify the earth’s heat balance to such an extent that harmful changes in climate could occur.  This report is now widely cited as the first official statement on “global warming.”  But the committee also recommended geoengineering options.  “The possibilities of deliberately bringing about countervailing climatic changes . . . need to be thoroughly explored,” it said.  As an illustration, it pointed out that, in a warming world, the earth’s solar reflectivity could be increased by dispersing buoyant reflective particles over large areas of the tropical sea at an annual cost, not considered excessive, of about $500 million.

GD Star Rating
Tagged as:
Trackback URL:
  • mitchell porter

    How well does it scale? Stephen Schwartz says that an exponential increase in tropospheric aerosols would be needed to counterbalance a linear increase in greenhouse gases…

  • One of the things I like about the Angel Fleet proposal is that it has relatively predictable consequences compared to other forms of geoengineering. I would have thought spending a lot more money in return for a substantially reduced existential risk would be a clear win.

  • Caledonian

    It’s nice to know that, however many world-altering consequences of our own thoughtless actions there are, there will always be people willing to implement quick and easy solutions that will eliminate the problems without requiring us to change the lifestyles to which we have become accustomed.

  • Ashwin

    When an activity such as this concerns the global consensus, I am not sure whether it can ever be implemented in practice. For example, would India be prepared to face the risk of a weakened monsoon considering the side effects of the 1783 Laki eruption. Many other countries would resist such geo-engineering on similar grounds.

  • Ben Jones

    Caledonian – for better or worse, that’s the way things seem to be. Which would you say is the more likely of the following two scenarios:

    a) A technology-based solution to the problems of warming/climate change.
    b) A policy-based solution, whereby people voluntarily change their ways of life the world across, and their governments do all they can to meet targets like Kyoto.

    I know where my hopes lie. A feature of previous technological advances is that they tend to make traditionally arduous tasks both quick and easy. That’s not a bad thing, it’s a good thing. Fire up the giant lasers!

  • Ian C.

    If people really thought for 1 second that the seas were going to rise up and submerge great cities like New York they would fire canisters of SO2 in to the atmosphere without a second thought. But I fear there are a lot of people (big government types) who view global warming not as a mere technical problem for the engineers, but as an excuse for wide ranging social change.

  • Actually I’d like to congratulate Caledonian for a post so delicately ambiguous between for and against that it can genuinely be read either way. Well played. 🙂

  • I think there’s a widespread misconception that the economy is a machine; it’s human-created, so we must understand it and be able to control it, right? So messing with the economy is less dangerous than messing with nature, right?

    Just look at the words used to describe the economy:
    “… Bernanke will have big shoes to fill following Greenspan, who earned the nickname of the Maestro for his steering of the economy”
    “the United States is the growth engine for the world economy”

    Steering, engine– it’s like the economy is a car we can drive wherever we like. It’s the central-planning myth, deeply embedded into our language and culture.

    The economy isn’t a machine– it’s more like an ecosystem, with adaptation, growth, evolution and natural selection. I’m actually optimistic that this view is gaining traction; I think younger people are more likely these days to subscribe to the spontaneous-order, non-hierarchical view of the world. Now if we could just get schools to stop teaching economics as a machine made up of mathematical equations…

  • @Robin:

    Why not just cap-n-trade? I have a preference for market-based solutions to these changes; consumers & corporations can then choose as they please. Why don’t you also prefer markets in this case? The NOX and SOX market works splendidly, why wouldn’t a well-designed carbon market?

  • that’s fascinating.

    i suspect that there is an incentive scheme in place that makes climate scientists want to promote government interventionist policies.

    would a future market help us around this problem? people bet and if the volcano proposition is favored, we do it? could a bet be set up where it’s legal on the volcano question? could the volcano program be unilaterally done?

  • The observation that we can counteract global warming with aerosol releases is unfortunately far less helpful than it first appears, because it confuses the casual name for the problem it proposes to solve, “global warming”, with the actual problem, which is accelerating anthropogenic change in the surface and near-surface processes of the earth.

    Specifically, while appropriately calibrated aerosol releases will perfectly cancel global mean surface temperature change, it will not in general cancel such changes at any given point in space. Rather, some places will continue to experience warming while others experience cooling. This is obvious. Perhaps less widely understood is the cumulative nature of the problem. If we continue to emit carbon even at a steady non-increasing rate, the perturbation to the preindustrial state will grow. The postulated geoengineering response will therefore also need to grow, and thus the local perturbations will also continue to grow. If the unconstrained “business-as-usual” scenarios are roughly correct, the perturbations would not only grow but would accelerate. Because the atmosphere and oceans are fluids, local responses may change abruptly in many regions, making planning and infrastructure deployment increasingly difficult and expensive, and dramatically increasing stresses on natural ecological systems.

    Still, it would help a bit, right? At least we could avoid one of the biggest issues, rising sea level? Well, that would depend in large measure on whether the poles or the troipics were the regions most effectively cooled by the process. Unfortunately, it may be noted that the volcanoes most effective at cooling the globe are the tropical ones, which may lead to the insight that the cooling process is equatorially centered. This in fact is corroborated by numerical experiments.

    Also, please note that CO2 emissions directly reduce the viability of the oceans even in the absence of climate change.

    The only forms of geoengineering worth considering for the carbon problem are the ones that remove carbon from the environment and sequester it. Alternatives may be palliatives (though in the present case perhaps not even that) like treating an infection with a narcotic, but in the long run only adds more problems, i.e., more massive ill-considered human interventions in the biogeochemistry of the only planet, insofar as we know, capable of supporting significant biotic processes.

  • Scott Wood

    I, too, shudder at the notion of geoengineering. Although, I suspect that it’s easier to implement than social engineering.

    The “problem” with cap and trade is the cap. Other than the influence on global warming, capping CO2 emissions makes us poorer than we would be without capping CO2 emissions. Perhaps significantly poorer. If we can accomplish the same goal without the cap, then we should at least look into it.

  • Mitchell and Michael, see my added note to the post.

    Ashwin, no one asked for global consensus permission to cause the problem, and I don’t see why fixing the problem should be held to higher consensus standard.

    Frelkins, the issue is which solution is cheaper. Markets could be used to implement any solution.

    Mike, yes of course prediction markets could help.

  • “But since interest rates are positive, the longer we can delay pain the cheaper is that pain”

    This does not follow even stipulating the dubious presumption that there is a net short-term benefit from this sort of thing. This is because the size of the mess we have to clean up increases. This is a fundamental flaw in the way economists tend to approach the greenhouse gas problem. To a good first order, the problem is cumulative. The more carbon we spill into the atmosphere, the more we will have to mop up.

    There is also a self-fulfilling optimism here. How do we know that interest rates will continue to be positive now that circumstances are changing and the world is better modeled as finite than as infinite? The presumption is that the longer we wait he better off we will be in dealing with problems. But in fact as we spend down our patrimony of cheap fuel, the presumption of greater capacity to achieve certain tasks in the future seems to amount to an unwarranted extrapolation. I am confident that our skills will not substantially decline on the time scale of interest, but it is far from clear that the energy resources will continue to be available.

    Your comment to Ashwin about the sort of consensus required to “fix the problem” seems to completely miss my previous point which you claim to have acknowledged, which is that this approach does not solve the problem. In fact it only solves the diagnostic we use to measure the problem! The problem is anthropogenic global change, not increasing mean global surface temperature.

  • Michael, as long as the rate at which delayed pain increases is less than the interest rate we win. If you want to be sure interest rates stay positive, let’s set up very long term inflation-adjusted bond markets and check there.

  • Again, even presuming that any pain whatever is deferred by this approach, sure. However, the accumulating carbon is linear even at constant emission rates, and the deferred cost is strongly adversely nonlinear in the temperature; consider that a temperature increase barely an order of magnitude larger than the worst contemplated case would destroy all vertebrate life. To compensate, the future prognosis has to be not only very optimistic, but dramatically accelratingly so.

    While I am very much in favor of taking any way out we can find, the universe is not obligated to provide an easy escape from this situation.

    As for your suggestion to set up very long-term bond markets, how would I benefit from my expectation, presuming that it is correct, that these are far worse bets than the marketplace believes? By what mechanism will such a market cause the replacement of cheap, dirty, exhaustible energy with abundant cheap, clean, inexhaustible energy?

    Economic optimism strikes me as embracing bias, not overcoming it, but perhaps we can have a fruitful disagreement.

  • Tim Tyler

    The unstated premise of this post is that we need to cool the planet down. However, it seems that heating the planet up and melting the ice caps would increase the habitable zone, reduce the risk of reglaciation, and generally be a boon to planet. So: what is the basis for the whole idea?

  • consider that a temperature increase barely an order of magnitude larger than the worst contemplated case would destroy all vertebrate life

    I’m sorry, was this an argument that, if things get ten times worse than anyone has contemplated, we all die? What probability would you assign to “ten times worse than the worst case scenario”? (I am also curious about the probability of an outcome ten times better than the best case scenario.)

  • Roland

    I read that if China stopped burning coal in it’s massive powerplants the global temperature would IMMEDIATELY INCREASE. Why? Because those same powerplants output enormouse quantities of SO2(or some other S-based gas) which acts as a coolant. The problem with SO2 is that it doesn’t stay in the atmosphere for that long whereas the CO2 stays for centuries.

    This is a really scary fact.

  • “I’m sorry, was this an argument that, if things get ten times worse than anyone has contemplated, we all die?”

    It was an assertion of that fact, which was then used to demonstrate that the cost function is adversely nonlinear. Obviously if we boil the oceans we might as well be on Venus as on Earth.

    I actually don’t believe that a change of 85C is required to kill the great bulk of us; 8.5 C may be enough to do that. The point is more that 85 C is much worse than ten times as bad as 8.5 C. I mention it to make it plausible to the reader that 8.5 C could well be much worse than ten times as bad as 0.85 C, as is generally believed by people who think about such things seriously.

    Much of the foolishness we see among economists on this issue seems to stem from some blind application of linear extrapolation. Even if 0.85 degree of warming is on balance slightly beneficial, as some would argue, that doesn’t mean that 8.5 degrees of warming will be wonderful, and obviously 85 degrees is not a good plan. Yet we see estimates of the effects of rapid change treated as equivalent to a multiple of a small quasi-equilibrium shift in climate. This would seem to introduce more than a little bias.

  • Fly

    Roland: “I read that if China stopped burning coal in it’s massive powerplants the global temperature would IMMEDIATELY INCREASE. Why? Because those same powerplants output enormouse quantities of SO2(or some other S-based gas) which acts as a coolant.”

    The particulates produced by the dirty burning of coal block sunlight. They also act as seeds for cloud formation which then block ten times as much sunlight as the particulates themselves. So releasing particulates into the upper atmosphere should cool the air much as a volcanic eruption does.

    Here is a NewScientist article on the topic:

    Modeling cloud effects is difficult. High atmosphere clouds at the equator reflect incoming sunlight and are usually cold so they don’t trap surface heat. Low atmosphere clouds that are almost ground surface temperature act as a heat blanket, keeping the earth warm. Thus maximum cooling occurs with cold, high atmosphere equatorial clouds during the day time and no clouds at the poles or at night. If scientists could control cloud formation by seeding or targeted application of energy then it might be possible to control regional weather and global climate.

  • @Scott Wood

    “capping CO2 emissions makes us poorer than we would be without capping CO2 emissions. Perhaps significantly poorer.”

    Got any real evidence for that? I was told recently about evidence that the economic hit from cap-n-trade could be less than a 1.8% decrease in growth. Some really smart guys at a recent conference in San Diego however argued to me that actually new green technologies would create more wealth long-term than the initial cost.

    But cap-n-trade is already coming to North America via state & regional initiatives. In the Northeast, RGGI will be starting very soon.

  • Let me venture a prediction that as climate change gets worse, SO2 cannons and similar proposals will be discussed more and more seriously; they seem politically impossible now but won’t necessarily remain that way.

    If I were convinced that the tangible benefits of SO2 cannons outweighed the tangible drawbacks, then I’d have no further objection based on what Robin calls “planetary purity.” My interest is solely in survival.

    However, the fundamental objection (which Michael Tobis also pointed out) is that SO2 cannons could at best buy time, and that after that time had elapsed, the problem would be even worse than before. So yes, the fact that the future is discounted relative to the present militates in favor of SO2 cannons, but the same fact also militates in favor of spending your monthly paycheck on a heroin fix, rather than checking into rehab! In both cases, the question is whether the discounting rate exceeds the making-the-problem-worse rate. I’m convincable on SO2 cannons, but this is the point on which I’d need to be convinced.

    (Note that the answer already seems non-obvious to me assuming a more-or-less stable discounting rate, but that like Michael, I’m also not convinced that the discounting rate will be stable in a world exhausting many of its natural resources.)

  • Scott, I meant market interest rates, not some abstract discount rate. A personal discount rate that matched current interest rates would not favor a heroin fix. If you thought speculators were confident now that resources will later be exhausted, then interest rates now, regarding those distant future times, would already embody the interest-rate-changing effect you envision.

    Freklins, a cap and trade with the cap set at the right level would make us richer relative to doing nothing, but poorer relative to doing something even more effective like artificial volcanos.

  • Increasing the amount of sulfur dioxide in the atmosphere will worsen the already quite-serious acid rain problem. Combined with the oceanic acidity from elevated levels of carbon dioxide, ocean ecosystems would be devastated. And the death of the oceans might conceivably affect our comfort and well-being.

    We’d better find a quick and simple solution to that problem soon – or at least once the effects of our previous solution become inconvenient.

  • Robin, suppose the human race continues in its current form (no singularity or or other major changes), and that population remains stable at ~6.5 billion. At what point in the future would you consent to have humanity obliterated, if it meant that a billion dollars could be added to the world’s GDP right now?

    If we value each human life at roughly $20 million (which I read somewhere is a standard figure in risk assessment, and is higher than the figures cited by Steven Landsburg as accepted by economists), and assume an interest rate of 3%, then by my calculation you should consent to have humanity obliterated in the year 2640 in return for a billion dollars. Yet that strikes me as considerably too soon: I would think the answer (if finite) would have to be in the billions of years (say, almost until the Sun goes cold), which corresponds to an interest rate extremely close to zero.

    So, where did I go wrong? I can think of several possibilities:

    (1) Maybe I should bite the bullet, and accept that $1 billion withheld from GDP today is just as bad as the death of every human being 630 years in the future.

    (2) Maybe the market interest rate isn’t the right discount rate to use when we’re talking about the fate of humanity, and we should instead use a much lower rate.

    (3) Maybe human lives become exponentially more valuable with time (as Landsburg argues).

    (4) Maybe my assumption that the population will remain stable and humanity will continue basically as it is, is inconsistent with the assumption of a continuing non-negligible positive market interest rate.

    (5) Maybe a probability close to unity of humanity going extinct at some point in the next thousand years, can’t be assigned a monetary value at all (as a limiting case, what amount of money would we accept in return for humanity going extinct right now?).

    Now here’s the crucial point: arguments (2), (3), (4), and (5) all seem to me like serious impediments to any argument for SO2 cannons based on market interest rates. (2) and (3) would clearly change the expected-utility calculation. If (4) holds, then given that I think economic and population growth will run up against fundamental resource limits very soon, the conclusion I draw is that the market interest rate must go close to zero or become negative. (In principle, maybe I could leverage that belief to get rich short-selling 200-year bonds — but alas, I don’t expect to be around to collect my earnings.) If (5) holds, then given that I do see Venus-like runaway warming as a significant existential risk, any proposal that mitigates global warming in the short term at the cost of increasing that risk by ε in the long term, must pass a burden that scales much more than linearly with ε.

    The bottom line: I can easily imagine circumstances in which SO2 cannons would be justified; for example, if we judged that we just needed to buy enough time until we could solve the climate crisis using improved technologies, or if the short-term consequences of climate change were so severe that they needed to be dealt with before anyone could even contemplate longer-term solutions. On the other hand, if we believe that we can use SO2 cannons to put off addressing the underlying causes of climate change indefinitely, racking up a greater and greater existential risk to the human species as we do so, and justify that by an appeal to market interest rates, then I don’t see why we shouldn’t also accept my “Year 2640 Bargain.”

  • Scott, I posted in January about what interest rates say about when it is worth investing in the future. Bottom line: interest rates suggest we today in fact care so little about the distant future we might well accept the sad deal you suggest. At least we might if it were framed in such a way as to avoid symbolic overtones that we do care more about.

  • A marketing remark:

    I think you meant to say “global sunscreen” not “artificial volcanoes”.

  • mitchell porter

    Following up my own comment, in Crutzen’s paper “Albedo Enhancement by Stratospheric Sulfur Injections” I find: “the great advantage of placing reflective particles in the stratosphere is their long residence time of about 1–2 years, compared to a week in the troposphere. Thus, much less sulfur, only a few percent, would be required in the stratosphere to achieve similar cooling as the tropospheric sulfate aerosol”.

    The core technical document for the current G8 discussion of climate mitigation options – which in turn is likely to be a major input into the diplomatic process negotiating the post-2012 successor to the Kyoto protocol – is the IEA’s “Energy Technology Perspectives 2008”. Along with various emissions reduction scenarios, it contains a baseline scenario for economic growth and greenhouse emissions in the absence of any attempt at reduction (it goes to 550 ppm CO2 by 2050), originally described in another IEA publication, “World Economic Outlook 2007”. While there are reasons to doubt the accuracy of the baseline scenario (oil prices in 2050 are $65/barrel, for example), given its centrality to the current political process, it would certainly be of interest to see a geoengineering scenario mapped out, complete with costings, which assumes both the baseline greenhouse gas emissions and enough sulfate aerosol geoengineering to keep temperatures within what is regarded as a safe range. (There could be multiple subscenarios: 2 degree rise, 1 degree rise, no rise in temperature.) To really round out the model, one would also need assumptions about total oil and coal reserves, to identify the point in the future at which aerosol geoengineering would definitely be no longer necessary because the fossil fuels had truly run out.

    It would take a bit of work, but constructing and costing such “baseline geoengineering scenarios” would do a lot to solidify this discussion.

  • Alex Wilson

    I would like to see some justification for this assertion:

    ‘There are risks from changing economies as well as from changing stratospheres, and we understand the later far better than the former’

    It is far from a consensus view.

  • Alex, I’m an economist with a strong physics training, so I feel expert enough to make the claim that we understand economies less than stratospheres. Who says otherwise?

    Eliezer, good suggestion.

  • While I personally hold to the position that he doubts, Alex Wilson’s question interests me as well.

    Certainly as someone who has spent more time thinking about climatology than about economics, my neural network is more elaborated in that regard, while probably a larger population has the opposite experience. However, this in itself is not decisive; the fact that a topic is of greater interest in no way implies that it contains deeper knowledge. Choose a major religion for which you have little sympathy and consider that more people spend more time thinking about it than economics, for instance.

    I am very familiar with people considering my field to be empty or vapid. I can assert with absolute confidence that it is not; that it is a branch of physics, indeed mostly of classical physics, and that some of the greatest minds of our time and the recent past have been dedicated to working out the detailed implications of general and powerful principles of physics, and while their program is not complete and may never be, considerable progress has ensued. However, you have not met the people I have met, so I realize sadly that I can’t just expect you to take my word for it.

    Meanwhile, my impression of economic theory is that it shifts with the winds, that the practical importance of any particular economic theory is limited in place and time, that economists do not understand how to study complex systems by formalizing the domain of applicability of rigorous approximations (i.e., dimensional analysis; what is the economic equivalent of a Froude number or a Peclet number or a Rossby number, pray tell?), that they are obsessed by a sterile and mathematically demanding limiting case of general equilibrium with no practical implications, that they are opposed in principle to empirical test, that they are divided into schools of thought of which the currently ascendant one (Keynesians in my college days, monetarists now) brazenly claim final victory, and that there is little way of deciding the validity of any opinion besides bluster.

    In other words, I think there is no comparison between the maturity of the fields. Nevertheless, economics seems to dominate thinking about planetary sustainability, even though sustainability is not even a formal construct within the dominant schools of economics.

    So I see it very much from how Alex among others sees it. On the other hand, it is not hard to argue that my peculiar observational perch might skew my opinions. So I am very interested in understanding how one might go about judging the relative intellectual merits of two very distinct research communities.

    I do not believe that any field deserves a free pass. The issues are of too great consequence to allow this, as is often pointed out. Unfortunately, we lack mechanisms for fair evaluation of the matter. In the case of planetary sustainability, it becomes a matter of great importance.

    Discussions as to how to evaluate the merits of an intellectual field from outside its boundaries, while unfamiliar, are, I think, essential under present circumstances.

  • Bottom line: interest rates suggest we today in fact care so little about the distant future we might well accept the sad deal you suggest. At least we might if it were framed in such a way as to avoid symbolic overtones that we do care more about.

    OK, so by the lights of economic theory, humans act inconsistently in having an interest rate well above zero, and yet not being willing (as I assume most wouldn’t) to accept $1B pumped into the economy today in exchange for humans being obliterated 600 years from now, were the question “framed” in such a stark, explicit way. Supposing I accepted that, it still leaves open the question of whether it’s the interest rates (at least as extrapolated that far into the future) or the refusal to accept the Faustian bargain that’s irrational. Between the two, I would have to say the interest rates.

  • Tim Tyler

    Re: I feel expert enough to make the claim that we understand economies less than stratospheres. Who says otherwise?

    The scope of those systems seems poorly delimited, and the metric by which understanding is measured has not been defined. The claim seems too vague to come down on one side or the other.

  • Alex Wilson

    I can believe that the physical effect of pumping SO2 into the atmosphere is well understood as a first order effect. And in expressing the options as ‘changing economies’ versus ‘changing the stratosphere’ we are, by implication, suggesting that the economy is a complex system with unknown dynamics and that the stratosphere is a simple physical system, isolated and with well-understood dynamics.

    However the stratosphere is not an isolated model physical system but a part of a dynamic system of considerably greater complexity than any economy. Ignoring this complexity seems to me to be at the heart of many of our existing environmental problems. I imagine similar first-order assumptions accompanied the release of the first CFCs into the atmosphere, and who could have imagined, a hundred years ago that changes in CO2 concentrations (inert, harmless, naturally occurring) would have such a large effect on the global climate.

    It may well be that SO2 cannons are the solution to the climate problem, but to implement such a solution immediately because we believe that we understand the implications of making a change to the biosphere more than a change to the economy seems to fly in the face of evidence and to be hubristic in the extreme.

  • Scott Aaronson,

    Great example about “1 billion withheld from GDP today is just as bad as the death of every human being 630 years in the future.” I’m going to use it the next time I teach about discounting in intermediate microeconomics.

  • Douglas Knight

    Scott Aaronson,
    Another possibility you don’t seem to address is uncertainty as a possible wedge between discount and interest rate. I don’t think uncertainty enters the interest rate in a particularly rational matter, but I think it’s the only way to get people to use anything like a reasonable amount of uncertainty.

  • Scott, this question comes up often. When we have many well-integrated diverse actions which are consistent with each other and some set of values, and a much smaller set of mostly speak-acts which on their face suggest a very different set of values, I’m tempted to call the first what they really want and the second what they tell themselves and others they want. We have lots of actions that integrate with our interest rates, but see almost no actual choices remotely similar to killing off humanity for a small reward today – just a few passionate words said by people talking about that choice.

  • Robin, I think it’s possible to look at the same facts and draw a very different prescriptive moral: namely, that the “many well-integrated diverse actions” consistent with a large interest rate actually amount to suicidal shortsightedness and stupidity, while the “much smaller set of mostly speak-acts” urging us to sustain a habitable planet are what an extraterrestrial visitor might see as sanity.

    James, please let me know what your students make of the example! E.g., “$1 billion withheld from GDP today is just as bad as the death of every human being 630 years in the future? wow, who’da thunk?!” Or, “if that’s so, then to hell with microeconomics, at least as applied to this sort of problem.”

  • Scott, can you suggest a general way to tell which are our “sane” values when we see a conflict between a large set of well-integrated acts whose values differ from a small set of mostly speech-acts?

  • steven

    For one thing, if 6 billion humans die at the same time, that’s more than 6 billion times as bad as when a single person dies (if the future of the human species has positive expected value), so that’s not the right comparison.

    I think the idea is that human lives do get exponentially more expensive as the economy grows, but that an extra $1 billion if invested grows even (exponentially) faster than that, so in 630 years it would end up being a stupendous amount of money that could save (or compensate for the end of) a huge amount of lives. So I can totally believe the result if you add the stipulation that the future deaths are individual deaths that don’t add up to an existential risk.

    Of course if there’s the possibility of immortality, then statistical value of life estimates are probably too low; you might have to look instead at how much it costs to save an extra life through public policy (traffic safety or whatever).

    I don’t really understand the point about local change rather than global warming being the problem; surely if delta-temperature without geoengineering is 2C, 4C, 6C in different places, and geoengineering subtracts 4C everywhere so it’s -2C, 0C, 2C, then you’ve solved at least 2/3rds (if not all) of the problem?

  • steven

    BTW, even if you don’t buy the “it could be invested” argument for a discount rate (as you shouldn’t to the extent that the money gets consumed rather than invested), there are still the possibilities of human extinction, technical fixes, and anything else that might cause us to stop caring about the problem. I would be very surprised if these didn’t add up to a pseudo-discount rate of at least 50% over the relevant time frame.

  • Scott Aaronson et all: I think the only valid comparison would be to a specific threat or set of threats *that have a high probability* of wiping out humanity, *not* to the abstract “wiping out humanity”. Further more, the harm can justifiably taken to be accelerating as the deaths become large relative to all of humanity.

    Also, the reason (a reason) for discounting is that wealth today enables wealth tomorrow as it gets invested, and then the fruits reinvested and so on.

    Finally, in case it wasn’t clear, the “$1 billion” refers to “$1 billion worth of real resources” (which could be labor or metals or …), NOT printing out new bills.

    With all of that in mind, I don’t see the problem with reasoning that we should accept the offer:

    “I will give you $1 billion of real resources, but it will cause thousands of giant meteors to hit every part of the earth in 630 years.”

    The compounded effects of the resources now, will suffice, within 630 years, of granting humanity sufficient capabilities to make those meteors no more than a nuisance.

  • Robin, the general rule I’d propose is to imagine yourself as an alien zoologist, who surveys various civilizations to find out which values influence their actions, and then checks back a while later to see which of the civilizations still exist. If you find that civilizations that adopt certain values are almost always extinct the next time you check, whereas civilizations that reject those values survive, then while it might be an interesting exercise to figure out just what was wrong with the values, I claim it would be obvious to you that something must have been wrong with them.

    Why does this help? Because as products of natural selection, we have severe biases toward the short term over the long term, and toward our own interests and those of our local group over the interests of the species. (I call these “biases” because they still seem to influence our thinking, even when we claim to be maximizing a species-wide utility function, and might even sincerely believe we are.) It takes a conscious effort to see things from a long-term species perspective, and we need all the help we can get — whether from gadflies like Bertrand Russell and Carl Sagan, or from simple cognitive tricks like the alien zoologist.

    Moreover, even when we pay lip service to the alien zoologist’s perspective — as, of course, large numbers of people do — our own actions are exceedingly unlikely to reflect that perspective very well, for well-understood Tragedy-of-the-Commons reasons.

    Because of the two considerations above, even if a “large set of well-integrated acts” conflicts with the alien zoologist’s perspective, while only a small set of mostly speech-acts agrees with it, I find almost no reason to doubt that the alien’s perspective is ultimately the correct one.

  • steven

    Scott, as far as I know global warming is not a serious human extinction risk according to mainstream science or anything like it; see for example .

  • The compounded effects of the resources now, will suffice, within 630 years, of granting humanity sufficient capabilities to make those meteors no more than a nuisance.

    Thanks, Silas! That’s a clear empirical prediction, and is precisely what I dispute. So maybe it’s not even so important to see how the two opposing empirical beliefs translate into opposing theories of discount rates, the monetary value of human life, etc. Maybe we should do what Eliezer says, treat beliefs as anticipation-controllers, and just directly discuss the empirical question: would a billion dollars (or 2 cents; it doesn’t actually make a difference) injected into the economy right now, grant humanity sufficient capabilities to overcome climate change, meteors, and other existential risks centuries in the future?

  • Scott, it sounds like you are saying we our “sane” and “correct” values are to all work together to help each other and to help the very long term survival of our species. Maybe this is what we should morally do, but it seems to me quite possible that what we actually want differs from this ideal. Someone can quite knowingly and coherently want to just watch TV, instead of doing what they would admit are more uplifting activities like reading Shakespeare or jogging to save Nigerian children.

  • Robin, maybe I misread your post, but you seemed to be advocating the use of SO2 cannons to delay pain, as something good and unfairly maligned. Isn’t some notion of “should” or “ought” or “moral” implicit in that very act? What point is there in exhorting people to watch even more TV, or put off addressing problems even longer, or do other things that are already consistent with their basest natures?

  • Steven, what makes you think geoengineering will cause a uniform decline in temperatures?

    In fact, the evidence we have now pretty strongly suggests that it won’t. Unfortunately it seems to most effectively cool the places that are less effectively warmed.

    Again, there is a lot more to climate than the temperature at any point on the surface. The fluid dynamics of the whole system depends sensitively not only on temperatures but also on temperature gradients. Both greenhouse warming and aerosol cooling reduce the equator-to-pole gradient, favoring small scale structures over large scale structures and favoring expanded subtropical dry regions and more vigorous but more local convective events.

    Again, this is the main problem with calling the issue “global warming”; it leaves people the idea that it can be solved by an equivalent global cooling. It can’t. We need to stabilize the radiative properties of the atmosphere, let the system equilibrate, and the ourselves adapt to the new equilibrium. The longer the delay the more extreme the challenges that await; the disequilibrium becomes larger, the uncertainties become larger, and the number of system modes that are engaged becomes potentially larger and more severe.

  • Steven, I have a great deal of admiration for Drs. Annan and Hargreaves, whom I have met and find to be delightful company as well. I am quite convinced by their argument as well as others’ that the extreme sensitivity as proclaimed by some is not plausible and that the temperature sensitivity of the atmosphere/ocean system to CO2 doubling lies close to and probably slightly under 3 C.

    This in no way invalidates Hansen’s arguments about ice sheet collapse, for one example.

    It also does not account for couplings within carbon geochemistry, wherein increased temperatures themselves cause increased carbon emissions. Several candidate processes exist, and indeed it’s hard to explain the glacial cycle without such feedbacks. Such phenomena are not included in the Annan/Hargreaves work nor in the work they criticize.

    It is also important to understand that, even if carbon feedbacks are small, nothing guarantees we will come to our senses at 560 ppm any more than we seem to be doing so at 400 or 450. Two doublings leaves us within range of a 6 C perturbation, and three brings us up to 9 C. Which amounts to double that in continental interiors on some time scales, which is no laughing matter.

    I don’t think this is an extinction event for us, but other species are already not as lucky. (It’s important in this regard to note that unchecked ocean acidification from CO2 may kill most vertebrate ocean life.) Consider that we may need to stop thinking about the triumphal march of human progress to the exclusion of putting any attention into avoiding a massive involuntary population collapse.

  • Prof. David Archer of the University of Chicago department of Geosciences is of the opinion that contemporary global warming left unchecked is in fact likely to set of a series of events leading to the relatively sudden release of seabed methane clathrates some thousands of years hence, possibly enough to trigger a much larger global warming event. He does raise the ethical implications of this scenario when he discusses it. So the 630 year question is not entirely a hypothetical.

  • mcow

    The biggest problem I have with ideas that involve blocking out the sun is that sunlight and heat are not exactly interchangeable. For example, the rate at which water evaporates is affected by temperature, but the effect of direct exposure to sunlight is actually greater. Pan evaporation measurements have already shown a decreasing trend in the rate of evaporation, despite global warming, due to global dimming.

    Evaporation is no trivial matter. It is, for example, fundamentally necessary to thermohaline circulation. THC is, incidentally, kind of a big deal. Among many other things, it is believed to be important in regulating the amount of carbon dioxide in the atmosphere. So large-scale and intentional global dimming could actually make the problem worse.

  • Scott, I presume that my exhortation will at best move people’s values only a bit, and that my values are not in fact that different from most others’ values – if others lie to themselves about what they want I probably do this as well. In general I want to advocate policy that gets people what they want, and in the case I considered in my post, how long to delay pain, I’m comfortable with that. I’m less comfortable with people getting what they want when what they want is existential disaster, but hey I’m just one person and so my wants aren’t likely to make that much difference in the final outcome.

    Many seem to be saying that disaster seems to follow from continued exponential growth in CO2, so we must stop at some point, so we must stop now. This last step in the argument just doesn’t follow.

  • Tim Tyler

    Re: thousands of years hence

    We’ll have AI and nanotechnology within 50 years. That will make climate change into an irrelevant storm in a teacup.

  • mitchell porter

    Tim Tyler: “We’ll have AI and nanotechnology within 50 years. That will make climate change into an irrelevant storm in a teacup.”

    I wrote about this issue recently. It could even be the subject of a post here: what is the rational way to approach problems of unsustainability if you expect a Singularity? The answer I proposed is essentially to compartmentalize: treat sustainability as a matter of mundane quantifiable governance, like macroeconomics, and treat the Singularity as a highly important contingency that can’t be timed, like a natural disaster. I would still defend that as a first approximation, but clearly the interaction can be more complex: if you really do think that the Singularity will almost certainly happen within 50 years, then you won’t care about environmental changes “thousands of years hence”, or even those slated for the second half of this century. In general, expectation of a near-term Singularity should skew preferences towards adaptation rather than mitigation.

  • @Scott Aaronson: Glad to be of assistance! Now, all we have to do is find some alien race capable of giving us $1 billion worth of precious metals in exchange for using our planet as target practice in 630 years…

    Anyway, FWIW, I don’t even think humanity needs the gain in real wealth today to be capable of handling a mega-meteor shower in that many years. But it’s not like we can test this any time soon.

    However, I think now you would have to revise your position. Earlier, you found it flat out incomprehensible to value life in such a way as to make the previously discussed tradeoffs. Now, I think you would agree it reduces to the more comprehensible question of how humanity’s ability to handle specific catastrophes will change.

  • Many seem to be saying that disaster seems to follow from continued exponential growth in CO2, so we must stop at some point, so we must stop now. This last step in the argument just doesn’t follow.

    I agree, and would replace your last step by: “…and almost certainly the sooner we stop, the less disastrous.” (Which also doesn’t follow from the preceding two steps, but which I think is correct — hence the “and.”)

  • Regarding Tim’s and Mitchell’s point, this strikes me as no more sane than any other form of superstitious last days romanticism.

    There are few if any singularities in nature. It is far more likely that we are reaching the inflection point on a sigmoid.

    Regardless of the plausibility of such an outcome, it seems to me very much more responsible to presume the more modest one and risk being happily surprised than the other way around.

    I don’t actually relish the idea of your Singularity, though. I have no idea why you would want to encourage such a thing.

  • superdonk

    I think SO2 cannons would be useful because it might give us more time to observe the effect of CO2 on the climate and will let us make a more informed adjustment. If climate sensitivity is high then we will have to fire them quickly and we will be not much worse off than if we hadn’t waited and ignored the SO2 option. However, if climate sensitivity is low then we will make a much better adjustment.

  • See my added to the post.

  • Increase in atmospheric carbon dioxide is undoubtedly increasing climate warmth. However I suspect that an even greater affect on warmth is the baring of soil by increase in annual crop acreage, roads, buildings, grazing, and desertification. You may see an article that discusses this in more detail in

  • Ken Ray

    Not being an earth scientist and certainly having no detailed knowledge of the causes and effects of induced geological activity I am very wary of such a solution. I think there are several arguments against such an approach:
    1. It is not permanent. As we geometrically increase our CO2 emissions, we would have to geometrically increase the volume of our induced geologic activity.
    2. There is not way to know (as far as I know) whether the induced geologic activity will affect any particular part of the earth. It is therefore likely that such activity will adversely affect significant portions of the globe as has happened during major volcanic eruption events such as Krakatoa or Mount Saint Helens. (The only events which I am even passing familiar with).
    3. The volume of such activity would be enormous, since even Krakatoa does not seem to have had any lasting effect. This in itself would significantly affect some portion of the globe chosen for this activity.
    4. Other products of geologic activity, such as heavy metals, other particulates and steam would seem to have separate adverse effects that in turn would have to be mitigated.
    5. As per Number 1 above, even the successful use of such a technology would not obviate the necessity to reduce carbon emissions in the long term. The argument that postponing the reduction of such emissions simply pushes the much larger cost and environmental impact off to the future where it will most likely be too large to handle. (If it is not already).
    6. Messing with the earth in this way is like handling a ticking time bomb. The results, in terms of environmental and economic disruption, could be catastrophic and I have no confidence that they can be predicted. (This objection could be reduced or removed over time though.)
    7. The ozone depletion angle is worrying. It may be minor compared to the effects of global warming, but other solutions are or may be available that do not exacerbate this problem. (I am also not sure that this would be minor given Number 1 above.)
    8. Use of this “solution” would deflect us from the real solution, which is reduction of CO2 emissions.
    I am sure that others would have better and more informed opinions, but just my two cents.

  • frelkins

    The American Geophysical Union yesterday seriously considered artificial volcanoes. The proposal was countered by the idea of robot-driven ghost ships to fire sea salt into the air for a somewhat similar reflective effect.

    But artificial volcanoes appears to be the current leading scientific proposal:

    “Sulfur injection into the stratosphere is considered to be the leading candidate for geoengineering, since nature has done this many times via volcanic eruptions, and we have some idea of what to expect.”

    Wunder Blog

  • Nuclear power.

  • artificial volcanoes appears to be the current leading scientific proposal

    In what way is scattering precious sunlight into space supposedly “scientific” – when much of the planet is still a sterile, frozen, icy wasteland for 365 days a year?

  • Pingback: Overcoming Bias : CO2 Warming Looks Real()

  • Pingback: Overcoming Bias : Space Towers()

  • Tim Tyler

    We’re in the middle of an ice age – and you want to cool the planet?!? Won’t that substantially increase winter deaths?