Stephen Schneider, "Climate Forecast Uncertainties: Causes, Magnitudes and Policy Implications” - 7th Annual Kendall Memorial Lecture MIT
PRINN: Welcome you all to the seventh annual Henry Kendall Lecture. Before we begin the main lecture, itself, it's become a tradition to give out an award that is given out annually called the Global Habitat Longevity Award. It was donated or available through a kind donation of Frederick Middleton, who, unfortunately, can't be here today.
And it is given in the general area of global change but, particularly, looking at the evolutionary processes that have occurred on our planet over geologic time. And the winner this year is Dan Rothman from each department. And he is being recognized for his work on the Proterozoic era carbon cycle through to the younger times.
The Proterozoic goes back to a billion years ago. So this is quite different than what Steve Schneider will tell us about for the next 100 or two years. This is going back to a billion years. But Dan has brought some novel statistical approaches to the study of time series of various proxies of the carbon cycle going back in time and has mightily pleased the geologists, who are quite impressed with you there.
So you've obviously made a significant contribution. So it's my pleasure to ask you to just come forward and accept a little certificate. And there's a little letter here from a couple of higher ups. There's no check in here yet. You'll apparently get an account. It does come with an award of $25,000 for research. And I'm sure that Dan will make very good use of it.
So as I mentioned, this is the seventh annual Henry Kendall Lecture. And just let me introduce myself. I'm Ron Prinn. I direct MIT Center for Global Change Science and usually end up introducing this lecture, because, in fact, it was a conversation with me and the donor, who's Kurt Gottfried, at Cornell University that set up this lecture in honor of Henry.
As most of you, of course, know, Henry won the Nobel Prize in physics. But he also had a deep and abiding interest in the environment and many environmental issues. And, in fact, when the Center for Global Change Science was set up in 1990, he came up, as he always looks a little disheveled, roughed up, and said, this is a great event, because finally, people were getting together who should in order to think about how the climate system works, and not just how the atmosphere by itself or the ocean by itself, and so on.
And, of course, as you know, as most of you know, Henry died tragically in an accident. But he is suddenly remembered for his contributions to physics, but also, well-remembered for his contributions to environmental issues. He set up this declaration that I think you've all picked up a copy of, a rather famous declaration.
And I think it tells you his vision. It was signed by a very large number of people, including a large number of other Nobel laureates. And it's basically a philosophy and status philosophy, whether it's certainly I share of need to be a steward to the planet we're on and not just gather all of the things from it without thinking about its sustenance going into the future.
I would normally say more about Henry. But this year's lecturer knew Henry well. And I think in his talk will fold in some memories of interactions with Henry Kendall in his lecture. So the seventh lecture this year, it's my considerable pleasure to introduce Professor Steve Schneider from Stanford University.
Steve and I, in fact, go back a long time. I think we first met, probably, in about 1971 or '72 at about to where I think we both got our degree in the same year. And we've known each other since then and have had a lot of really enjoyable interactions. It is my job to read out a few formal things about Steve's career, and so let me do that.
He is currently the Melvin and Joan Lane professor for interdisciplinary environmental studies at Stanford. He's also a professor of biological sciences and a professor of civil and environmental engineering. He co-directs the Center for Environmental Science and Policy at Stanford. And he's a senior fellow in the Woods Institute at that university-- received his PhD in Mechanical Engineering and Plasma Physics from Columbia University. And that was right in 1971.
His early work was studying the role of greenhouse gases and aerosols at NASA's Goddard Institute for Space Studies. And that's when I first met him. And he was working with a wonderful character named Ishtiaq Rasool. Some of you may remember him. He was a member of the scientific staff of NCAR. That's where he moved from GIS to NCAR in 1973-- was there through 1996 where he co-founded the climate project at NCAR.
No surprise, given the topic of his lecture, his research focuses on climate change science, but also, in integrated assessment of ecological and economic impacts of climate change, while he's worked a lot on identifying viable climate policies in recent years and technological solutions-- consulted with federal agencies and the Hill, the agents and the House and the Senate through many, many administrations, beginning the Nixon and going through the Carter and Reagan and Clinton and Bush administrations.
I won't tell you which ones he enjoyed most. But he's actively involved in the IPCC. And he's currently a coordinating lead author of one of the main chapters in working group two that is titled, Assessing Key Vulnerabilities and the Risk from Climate Change.
As many of you know, the IPCC summaries from the working group one and working group tow have just come out-- one of them on February 2, and the other one about two weeks ago, and are available on the IPCC web if you wish to consult them.
Steve has won many awards-- the American Association for the Advancement of Science, Westinghouse award. He's won a MacArthur Fellowship in 1992. He's elected to membership in the National Academy of Sciences in 2002. Many other honors. And some of them are written down in the invitation you would have received.
He's the founder and still the editor of the Interdisciplinary Journal Climatic Change that I can vouch as a excellent journal, having published a lot of stuff in it myself. But it's a tough referee system that they have. It took a year to get through. It was a mega paper. But it did take a year to get through the system.
He's authored and co-authored hundreds of papers and so on and is very well-known, as I say, not just in the science, but particularly in recent years in trying to bridge over into the economics and policy areas. So with that, as I preamble, Steve, welcome to MIT. And we're all looking forward to your lecture.
SCHNEIDER: Thank you. Thanks very much for that, Ron. Oh, it is on, okay. And it's just a pleasure to be back and to see old friends and old mentors and, in fact, sitting and talking with Ed Lorenz. Just today, your name came up. I didn't tell you this yet when I was meeting with about 20 graduate students. And we were talking about everything from climate science and some impacts.
But a lot of people were asking policy-driven questions and coupling of economics and technology and so forth. And I just dawned on me why I was sitting in the very building that I first came to in probably 1974 when you had two students-- Alan Roebuck and another one-- that you asked me to help you to work with that the conversation that we had crossing so many disciplines with competence in the multiple disciplines would have been virtually inconceivable in 1974.
Although, those two students actually were thinking about it but didn't think about it out loud. That always had academic freedom as a driving force. And then I thought that when I was back again, maybe, in the late '80s, we did have conversations like that, but probably two students. Now, there are 20.
So I think it's a compliment to you that you've taken not just the good basic science, but that the integration of that and to either looking for emergent properties of coupled systems, the intellectual side, or looking to try to deal with problems where they are across multiple positions is great. So it was really terrific fun for me, because it's that next generation of students, of which I see many, will have to figure out how to fix the problem that my generation is leaving you.
Okay, in that sense, let me talk a bit first about Henry Kendall. And it's probably in there, so I can read it quickly. But just a few descriptions for those who may not know. I had the pleasure and the privilege of working with him on a number of occasions. I'll talk about one in a minute.
So he was the recipient of the Nobel Prize, as we heard. And the citation was for pioneering investigations concerning deep inelastic scattering of electrons on protons and bound neutrons, which have been of essential importance to the development of the quark model in particle physics.
I'm going to show you later a quote from a White House conference that has absolutely nothing to do with any of this, which has to do with the breadth of his capacity not only accomplished as a physicist and a skilled mountaineer and photographer, but one of the founding members of the Union of Concerned Scientists and active throughout his life to advance the causes of UCS, particularly, interested in energy and nuclear proliferation.
In '97, Kendall and this lecturer co-appeared before an East Room White House conference kicked off the Clinton-Gore administration's public campaign for US participation in the Kyoto Protocol. So there's the photo op. You might even recognize a few colleagues and others. It had a heavy dependence on Nobel laureates. I never quite understood that, but then a few other people. And then it had one really good movie star. And--
SCHNEIDER: With what?
AUDIENCE: You wanted to say if people [INAUDIBLE].
SCHNEIDER: Al Gore. I forgot the name of the doctor-- Les-something. There's like with this Henry. Jane Letrinko, Cheri Rowland, John Holdren, the guy in the suit, me, the guy in the light suit-- we were dressed down for this down-dressed behavior-- Mario Molina. And then there's this guy who also is well-known for some things.
Okay, now, he was terrific. So now, let's get to Henry. I know these quotes, because I happen to have had the transcripts. So I can give them to you since I was there. So here we were in the East Room. And this was the first public event in which the president and vice president wanted to introduce the public to the debate about the Kyoto Protocol.
Just to tell you how successful it was, that day, the US Senate passed-- sent to the Senate resolution called Byrd-Hegel, 95 to 0, saying that they wouldn't sign on any protocol that didn't include a meaningful participation of developing countries-- China, India, and so forth-- while forgetting to mention that we were a factor of 10 per capita energy more using than they were. They left that little detail out.
And so there was no naivete on the part of the president and the vice president about what the task was going to be like. And I remember thinking, it's great they're in it, but why is this July of '97 with the protocol to be negotiated in December? It should have been July of '94, because it takes a long time to get people used to a problem. But they did it.
Anyhow, so this was from Henry. "It's the bottom tier of the developing nations that will get hit directly by shortages helping to generate increasing numbers of hungry migrants, environmental refugees streaming across national borders. The result is that no nation will be sheltered from dislocation of food supplies, altered trade balances, freshwater difficulties."
By the way, that's very similar to what you just heard two weeks ago from the release of the IPCC Working Group Two Report, except this happened to have been 10 years earlier. "And even though the wealthy nations will remain able to feed themselves, no nation will escape the troubles from pressures on the food supply."
And then he out of line I just loved. And so did Clinton. He repeated it. "Mr. President, it's not the case that one end of a boat can sink!" As in we're okay, and they're not living in the same world with dependents and your systems and vulnerability to all kinds of chicanery.
And he framed it just to give you the rest of the sentence, and we'll move on. He framed it. So there it is. So let me say this is a developing problem in national security, which, now, again, I just heard six retired military brass frame it as. It took a long time, but he was 10 years ahead again. And we have to deal with it."
The best way to deal with it-- and I'm going to come back to this philosophy in a minute I guess near the middle of talk-- the best way to deal with problems of this sort is to stop them at the outset and not wait until the results are presented to us and great troubles have developed.
In other words, a statement of the precautionary principle rather than a principle of let's get high confidence degrees of certainty-- that I will argue, which this precautionary principle is at variance with the views of a lot of other people. And I'll contrast that later in the current IPCC.
Okay, so, again, it's a thrill for me, and thank you very much, Ron, for the invitation to remember Henry, who is a one of a kind. And the mold broke when he died. Okay, time to talk about my chapter and other things. The chapter has a direct remit from the government's-- the 120 governments-- give us what we call the PAO, the hallowed PAO, as somebody calls it, which is our plenary agreed outline.
It's the bullet points upon which you must put in your chapter or discussions of all those issues. And the discussions have to be based upon references in the literature. And our job is not just simply to be librarians and cite what's out there. They don't need us for that.
It's to make judgments about the relative quality of the literature and then to assign confidence to various conclusions, because that's what's difficult to do in the political media debate with 20-second sound bites of end of the world and good for you advocates when I will tell you my own prejudice is at end of the world and good for you, the two lowest probability outcomes in the debate.
And, therefore, that's recognized. And governments do actually want to have a more credible set of assessments in the US. It's the National Research Council. Australia CSIRO does it. The UK has other things. But those aren't credible outside of those countries or other countries like them. So the IPCC was created for the purpose of engaging, literally, the whole international community.
So the chapter includes in its plenary agreed outline the science of Article II. Article II is the UN Framework Convention about what is dangerous. So I thought I'd start with a very large trillion-dollar scale industry, Munich Re, the reinsurance company-- the ones who have to pay the families of the 40,000 people who died in euro heat wave in 2003, and companies lost in floods and the Katrina victims.
So they, obviously, not, surprisingly, think that the climate change experiment, as Roger Revelle called it once-- the Roger Revelle and Hunt [? Suise, ?] the great geophysical experiment we were performing. They think it's dangerous. And to their financial interest, they're probably right.
But what does it mean? Well, here's the UN Framework Convention language-- don't worry, won't read it all-- just a couple of key points. Stabilization of greenhouse gas concentrations in the atmosphere at a level we would prevent. And here's the famous phrase-- dangerous anthropogenic interference with the climate system.
Did it define that? Not quite. A level should be achieved within a time frame. So they recognize that transients matter. When somebody says, well, how much climate change you worried about? I said, well, I would rather have 3 degrees in 200 years and 2 degrees in 50, because the adaptive capacity of social systems and natural systems depends on rates.
So it's not a simple answer. You have to look at the system, the thresholds that matter, the rates at which they are, and your capacity. So they got that right, which was pretty good for 189 countries signed this-- second-most signed and ratified treaty. And by the way, President Bush, one, signed it, and the US Senate ratified it. This is the law of the land-- something I wish another Bush might remember.
What else? To allow ecosystems to adapt naturally, to ensure food production is not threatened, and enable economic development to proceed in a sustainable manner. The last one is very interesting, because you can interpret it two ways. It's very hard to sustain your economic development if you're in a coastal or small island state, and you're underwater, or you're about to be.
So you could talk about damages to the system and say that that wouldn't be sustainable. On the other hand, if you're a state entirely dependent on coal burning, and you haven't figured out how to sequester the CO2 underground or switch cost effectively to other systems, then you're sustainable development could be hurt by the action.
So, therefore, the balance-- and this is why it's such an interesting politically difficult issue at an international scale-- is balancing damages against cost of fixing them and not necessarily, John, in a cost benefit framework. But certainly, intellectually, you have to look at the wide ranges.
But the other thing that matters is that the damages that will occur are highly differential. For example, melting arctic sea ice will cause the damage to the shipping industry of probably saving them $50 billion a year and having shorter routes. But it will wreck the culture of the Inuits, which is established over 5,000 years.
How do you weigh that or destroy the polar bear ecosystem? How do you weigh those incomparable monetary metrics? Well, you don't. You have a political argument about the relative importance of those things. So therefore, what we have to do when we discuss this issue is to frame it in the multiple metrics that people care about, and later on, the be make a political decision at an international level about the relative importance of those.
So analysis is essential. But at the same time, the analysis can't aggregate everything all at once first without controversies. That's basically I've given you what our chapter has to point out. And how we do it will come out,
Okay, so first of all, who decides what is dangerous? When we were first asked by IPCC to do this, we said the obvious. We can't tell you what's dangerous. Dangerous is a value judgment about the acceptability of risks. And they said, well, you're the experts, who said, well, we're the experts in what can happen and what are the odds, risks.
We're not the experts in what to do. But what we can do is we can survey the literature, see what different stakeholders, different nations, different sectors, different groups are worried about in terms of thresholds that if exceeded would be, in their opinion, dangerous. We can give you criteria for what vulnerabilities are key. But then it's your job to picket.
So so far, we've had a reasonably, occasionally, uneasy balance with governments. And they did just approve it. So I guess that metric worked. But the key is, it's not a scientific judgment to decide dangerous. But it's not a lay judgment to decide risk. So we're all needed in this equation.
Okay, so there's many ways to define it. And it's, ultimately, as I just said, not a scientific choice. And now, we have another problem. We have the problem that everything has uncertainty. There's uncertainty in how many people in the world-- standards of living they'll demand, technologies we'll use, how policies will form just from the perception that there is a dangerous outcome, which leads to a different policy that then messes up your original calculation.
All of that is part of the story. And there's uncertainty in all the biogeophysical stuff, as well. So, therefore, you're not going to do an optimization with about seven median values of important parameters and get an answer that's honest. What you're going to do is put in bell curves on all of them. And then you end up the problem of, well, how do you do that when you don't have enough data to do it? So that's where I'm going to take us.
So it started out with best guesses. So you had a best guess for climate, for people, and for emissions, and for climate sensitivity, how much it warms up if you doubled CO2. That's climate sensitivity. And then you have best guesses for what the damages are. And you can come up with a number.
In that framework, you can optimize-- how much does it costs to fix the problem, versus how much it cost if you don't. But we know that's not right. So we started working on ranges. And the famous Charney Report-- as long as we're in MIT, how can I not mention that-- which first define climate sensitivity as a range-- 1.5 to 4.5 C.
And I don't remember what that was supposed to include. I think it was a 1090 range, but I can't recall. And what it was was a subjective assessment but an expert assessment, looking at what was available and then guessing what was missing in the models and trying to account for it.
And what I liked so much about that report wasn't the number-- which lived for 25 years, by the way, until the current report-- but that they explained the process. And that's really important. And they were very open about that. But, of course, what we really need is not just ranges. If we're going to do risk management, we need PDFs. Now, you all know how easy it is to talk about something like a PDF on the evening news or in front of Congress. They all know what a PDF is.
Yeah, of course. No, sorry guys. That's not the PDF we're talking about. We're talking about this one. So here is a PDF. So it's a probability density function. And I'm glad in this group I don't have to explain it, because if I do public talks, and I show these pictures, it's about six minutes to explain this. And I have to tell people, well, there's a 10% chance if you look at the area under the curve, that's called the CDF-- don't worry what that means.
So we have a 10% chance of a tenth percentile, which means 90% chance it's more than one, 10% it's less. Here, you are out here with an unbounded tail that goes out for a while without much probability. But you have a 10% chance it's more than four. Ice age interglacial cycle differences in much less time. That's one that even the outgoing-- former outgoing-- CEO of Exxon wouldn't recommend.
So you end up with these catastrophic outcomes and these, well, we probably would rather not have it, but we could manage that a lot better. And they're all out there. Now, this particular one was for one scenario for a probability density function of climate sensitivity, which I think we stole from your paper, Chris and your group's paper.
And we have the standard mantra-- please don't take numerical model-dependent results literally-- please take framework seriously. So this is a framework that we're trying to talk about. And I still remember when I first showed this a while back-- I don't know. John, you may have been in the room. It may have been at EMF-- and somebody started to complain. And we're having salmon for dinner and started to cling-- come on, Steve.
You only think there's a 10% chance of really catastrophic outcomes. You want to alter the basis of industrial civilization, deny the Chinese and the Indians the chance to catch up with the Victorian Industrial Revolution. And I remember saying, well, we're eating salmon, Joe, although, I think it was Rob. And I said, there's only a 10% chance of salmonella in the salmon. Of course, you're going to eat it, right.
And it led to a little thing that a little decision-- analytic elicitation-- I have now done with all groups of reasonable size. Why not with this one? So 10% thought was way too low a probability for significant action that cost trillions of dollars. And, of course, these are not trivial events either. They're just not as bad as those.
So I then said, okay, how many people in the room-- let's do it with this crowd too-- let's see if how you come out in this little informal survey-- how many people have had a significant house fire? Hmm, couple. This is good crowd, because normally, it's a little higher than that. In Berkeley, it was a lot, but that's special case.
I asked that question in Berkeley and defeated my entire point. But it's somewhere around 1%. Insurance guys tell me that. Now, how many of you have fire insurance? You fools. It's much less-- much less than 10%. So it's around 1%. So when somebody tells me that 10% is too small a number, what I keep coming back and saying, what are you talking about?
It depends upon the problem. It depends upon your degree of risk aversion. It depends upon what's important. I mean, what do you think the probability was of a whole bunch of threats to our security for which we have a RAND Corporation analyzing them all the time. And I happen to not mind my tax money going into that. I think we should do that.
And almost every corporation is worried about downside risk. Oh, my god, if we do that, we don't want to get wiped out, and they hedge. So that's called risk management. And Henry was way ahead of the curve on realizing that this problem was going to have inherent uncertainty for a long time, and that it was basically going to be a risk management problem, not a problem of optimization.
So that's, of course, the lesson. And I can't help but shamelessly come to the house-- the house of Vanna White. I mean, to the place where-- first guy who showed this to me was Mark Webster. But I don't know which of you got the idea first but congratulations to the group, Ron. It's a great thing.
And I show the regular audiences a PDF. And I said, that's a PDF you can understand. Yeah, they understand, and they get it. And out here, oh, my god. Out here, well, we've warmed ups 0.75. And we've had intensified hurricanes-- right, Kerry? It's still going to stay with that? Yeah, he's nodding, yeah.
And so even the small one is not trivial. But it's a heck of a lot less serious in terms of the number of systems and the degree to which the systems are damaged. Then out here and out there, you're looking at ice age interglacial cycle temperatures that are 10 to 100 times faster.
So this is a nice way to get people to think. I think this was your baseline case that was approximately a tripling of CO2-- something like that in 2,100. All right, so when you talk about uncertainty, you got to talk about probability. So let's get on with that. So how do we do our analysis in the future? What are the future implications of population, affluence, and technology, growth projections?
That's the old Holdren-Ehrlich IPAT-- impact population times affluence times technology. I was amazed once when a petroleum engineer told me, oh, that's just a model. I said, no, no, it's an identity. It's true by definition. It just doesn't tell you anything until you know that there's interactions among all those terms. Now, you need a model for that. But just the multiplier is true by definition.
So how do you scientifically, quote, unquote, "analyze" the future? Because the questions that we're asked in our plenty agreed outline are, what's going to happen? Not just what happened and how confident are you, but what do you think will happen? And how confident are you? And what does it mean? So we have to be doing projections.
I kind of liked that one. This is one of those tabloid things. And about every five years, I find it again. They bring it back out, figuring the ones, if they could read, who read it the first time will forget. But, actually, aside of the joke, there's something very real. So in order to do the future, you need one of these. And Nostradamus's crystal ball is clear.
Unfortunately-- and that's why we're all here employed, right-- the good news, it keeps us working. The bad news is I wish we knew the answers better is our crystal balls look a little more like that one. I found that one in a pawn shop, and I immediately photographed it, because that's how I feel about it.
It's not all cloudy. You can see through some of it. And some parts, though, here's cloud feedback in here. And in there, we've probably got ready to forcing. We've got that one down better, but not aerosols. So there are components of things that we know pretty well. We know that reefs are in pretty big trouble above 2 degrees warming, maybe, even above one on some of them.
We know that there's a large threat to species. But we don't know how long that commitment will take, whether if we had a transient, they could be recovered. So there are large components of the problem still pretty opaque-- other parts that are pretty well-established. So then people say, well, what about the future?
You guys are going to be assigning likelihoods. But likelihood sound like frequencies. And frequencies should owe our observations. So, of course, I love doing this with students, right. What's the probability this is a head? Is this the head on the Swiss coin or a tail? I just had to give a talk there, and I quickly pulled out the coin I happened to have in my pocket.
Now that you have these cameras, you can sit there and get it in your PowerPoint in five minutes. And, of course, somebody said, a half. And I said, how do you know? And they said, two sides. And the usual mantra I do with kids is, so, you're a theorist. Supposing you weren't capable of understanding the theory of coinness, could you figure it out? And they say, sure, you flip it.
I do this with high school kids. They get it too. So you have head and tail. And you check the box. And you get through. And that's the best tradition of science-- go out and get the data. So now, of course, I'm going to now trick them a little. What's the frequency data on the future? Well, how about none?
So how can we make probabilistic statements? That's where the cloud, partly cloudy crystal ball comes along. The way we make statements in the future, as we construct a model, which we hope characterizes enough processes, that when you push it under alternative scenarios of growth or activity, you can start differentiating a high emission scenario from a low.
You can look at what it means. And you're going to be honest and open that it's not going to be a perfect replica of reality. But the degree of confidence you have is going to depend upon how well it does on current conditions. So how do you assess-- I'll go through this fast. Everybody here knows this. But this is the framework that we have to deal with when you're trying to talk to decision makers.
It's not always an oxymoron. There are honest decision makers out there. But they need to be led through this, because they hear so much nonsense-- oh, it's too uncertain. You can't do probabilities in the future. Sure, you can, but they're subjective. But they're expert at the same time. And you have to do mix systems.
If the systems aren't constructed based upon good data and theory, then your confidence in them is very low. But just because it is, you're not certain that in the future under different conditions that the way you derived your equations will still be true. Well, we believe that they'll sure be true for Newton's laws.
But we're not sure that the parameterization of clouds inside them that we derive from current conditions will absolutely be true in the future. So we have to spread details of our PDF as we get further out to be able to represent it. So how do you do it? And I try to tell that we have no replicate earths. You can't run 100 earths, pollute half of them, and see what happens.
It's one place. No-- but we do have Mars and Venus to test process. So people say, aha, Mars-- thin atmosphere, weak greenhouse effect, cold. Venus-- thick atmosphere, strong greenhouse effect, hot. Earth-- medium right in the middle, which Lynn Margulis once called planetary Goldilocks Theory, right-- too hot, too cold, just right.
The trouble is it doesn't help you much. We already knew from experiments in labs what heat trapping was. It's nice to have it work at a large scale. And a planet Peter used to do stuff like that years ago. But the truth is, what makes the Earth's climate sensitivity uncertain, a factor of three or four, is hydrologic cycle-- its cloud feedback and ice, and other internal dynamics unfortunately unique to the earth.
So we've got a little bit of help out there, but not a lot. We don't have any clinical trials I can't resist. But that's another talk for another day about doctors' clinical trials and cancer treatments. I have a book, The Patient from Hell. It's about the application of Bayesian methods as opposed to clinical trials to rare diseases, which I had. And I still have, but it's 5 and 1/2 years later.
And I'm still here when I should have been dead with a 95% chance. because if you can't cure it, what I said-- very simple-- precautionary principle-- Henry Kendall-- if you can't cure it, manage it. So you can't get rid of it-- don't wait for it to come back. Let's use low dose prophylactic immunotherapy and keep the numbers of the cells down.
I don't care if I have cancer. I care if I have a trillion cells-- latter dead. Million cells, nothing-- just potential-- keep it down. So finally we got that, and it works so far. Okay, you never know. The cyst can mutate, and the drug won't work. Then you have to redeal with it. But I don't have any higher probability of that than any of you.
But what I keep saying, back to your wheel of fortune, Ron, is that in life, you can never control where the ball lands or where the fortune wheel goes. And our job is to make the really lousy slots narrower and the really good ones fatter. That's all we can do. That doesn't mean bad things won't happen to good people. It just happens less.
And this is, again, the whole philosophy that we try to argue when we talk about this. So you don't have clinical trials and replicate earths, but we have analogies. Volcanic dust fails. We test those. If you have an economic model, you want to run an on price shocks.
You want to get any magic spike of forcing and response that, although, not a perfect analogy to the global change problem, at least, gives you some capacity to test the model sensitivity, because when you build a model and drive it on the basis of present data, that doesn't validate it. You need to have external additional things. And that's standard. So we have stuff to do.
And, of course, we have our virtual earths. Our model's in the major effort, which a lot of which goes on here. And in other places is testing to assess confidence, because confidence is essential. It's become essential in the IPCC. Richard Moss and I wrote the guidance paper and uncertainty for the third assessment report, trying to really get people to force them when they use words like likely to have a probability scale, because there's been a lot.
Granger Morgan's work has shown that people have an order of magnitude difference in their mental conception of the probability of a likely. We said, well, come on. At least, let's be on the same page. And IPCC's been very good about that. And governments now have-- first, they were skeptical. And now, they really like this, because they realize ordering their risks requires how confident we are in the science.
Okay, so here we go-- models and data. So this is the famous hockey stick. I'm not going to argue about how wavy the handle is. It's probably wavier than that. And this is out from the noise. And also, I'm not going to argue it now. But almost undoubtedly, the last three or four decades are. But I wanted to show process.
So we have a fan of uncertainty here. What are those colored lines? Those are scenarios with the awful names. The top one-- A1FI. A1 is a globalized world of mobilized capital. We use all the available cheap energy, coal, other fossil fuels without any tailpipe dumping fee.
One of my colleagues called it the world of business as usual greed. It's a very rich world, and population stabilizes under $10 million. The lowest one is B1. That's a world of egalitarian sharing where new technologies are shared. That's, by the way, the lowest on the one-- notice it's curved over to the right.
That means it has stabilized by the end of the century. It is a doubling of CO2. And this one is still going up. That means we're going into tripling. It's tripling over here, and it's been toppling out there. So that's a pretty scary one. And that, by the way, is not the highest if we use tar sands and oil shales.
And this isn't the lowest if we have a breakthrough in deep Earth sequestration or a nanotube solar, or we have a world depression. Let's not root for that one. That's not the way to solve the climate problem. And then there's a technology one I like a lot. That's where you spend 1% of GDP.
So here's fan of uncertainty-- one human behavior. What about these vertical bars? That's fan of uncertainty too. That's the dynamics of the climate system they represent. So the bottom of that bar means you've got a lot of emissions. But you have a model with relatively low sensitivity to them. Up there, you've got a model with high.
This is the third assessment report, 2001. About 1.4 to 5.8 was the range. Again, it's your wheel. And they did not put a probability function through there. You had probability functions, because you had divisions.
So there are many of my colleagues who would argue that that was irresponsible, because you can't assign probabilities to future social events. I don't happen to agree with that. I think you can. But they're subjective. And you have to have lower or medium confidence in them. But you certainly can.
They didn't assign probabilities. They just gave you the range. Okay, let's unpack that a bit. So here's model one. This is what I'm calling the well-calibrated range. So this is your low climate sensitivity model, PCM, and NCARs. Here's model two, a medium one, another medium one-- model four, maybe GFDL.
I mean, one of the bottles-- and it's how much you warm up if you doubled CO2. And so you have a range. And that was the 1 and 1/2 to 4 and 1/2 in the turney range. And I'll show you later what's just come out from the new report. And I'll show you what they proposed versus what emerged.
Well, everybody knows that processes are left out, because the partial differential equations can't be solved by known calculus. So you have to disparatize. You've got grain sizes that are much larger than individual clouds. We've got parameterization, the usual story. Economists call it reduced form where you're wearing an economic model.
You have an agent who perfectly foresees the future and who then maximizes utility, which is typically the log of consumption. And this agent is called representative-- those with lots of studies that it isn't representative. It's representative of some firms. And you can argue that indefinitely. And we don't need to argue that here.
I have no problem with anybody who makes assumptions, only a problem when you call them high confidence. And they, in fact, all models are, of course, is the logical consequences of explicit assumptions, which, in my opinion, beats out handwaving any time.
But when you start doing that, you've got to make alternative assumptions that different groups believe and show the alternative consequences. So anyhow, we know processes are missing, so the judge range is larger. And when Granger Morgan and David Keith did a decision analytics survey, they got climate sensitivity range that was larger than the models, which is what you'd expect.
And cognitive psychologists, like Kahneman and Turske. You say wait a minute. Even new scientists are suffering from inherent bias and from anchoring, which means you see what's out there. And from availability, what did you read last week? And that we too have prejudice. And that even though with that prejudice would come out in our judge ranges. But, probably, the uncertainty is even larger than we think.
So the full range, which we won't know for a long time, is even bigger. That's the context, though, in which we operate. Most of what's in IPCC is in here. Recognize this? Yeah, you must. That's the work. This is in the forest. They're all paper. Here's PDFs of climate sensitivity-- one with a uniform prior-- one with an expert prior. That was Morgan and Keith.
I won't go over it, other than to say it's very typical. It's got a long right-hand tail. There's another one from Andrew Nova and Shlessinger, which also has a long right-hand tail with a bump. Physically, where did that come from? Where did they get 3, 4, 5, 6 degrees warming from doubling of CO2, because we know with pretty high precision how much heat trapping we've had from the CO2, methane, and other gases emitted.
That's pretty well-controlled. But we also have aerosols, hazes, which reflect sunlight and cool. And some of them, even warm, the soot ones. And when they get mixed up in clouds, they can change the amount of drops. You change the amount of drops-- you change the reflectivity of the cloud, depending on the color of the particle.
And therefore, this is called an indirect aerosol effect. And the direct CO2 and greenhouse gas is what-- 2 to 3 watts per square meter, pretty well-constrained, whereas, the aerosol effects run from minus 1/2 a watt to minus 1.5. Well, how would that make the long tail? Because if the indirect effects were very large, that means that the world is only experiencing 1, 1 and 1/2 watts in the last 100 years.
So when you look at the warming, and you scale it, you'd actually have much higher climate sensitivity, because you reflected it away. So I've heard people say, oh, isn't that good? So why don't we just run those hazes out there to offset the global warming? I said, wait a minute-- I don't know that acid rain and air pollution that sends kids to the hospital with asthma is a good way to do that.
And in fact, the economists have shown us that in every society as it got richer, we cleaned up the air, not because of climate, but because of health. And that's also happening in China and India right now. And I think it will continue. So we can't rule out that nasty tail, because we don't know, currently, the indirect effects well enough to know how to back it out.
Okay, now, update the AR4. That's the assessment report 4, which, the previous one was TAR, third assessment report. So why AR4? Why not FAR? Well, because that was the first, and that will be the fifth also. So this was the acronym. And there's Martin Manning. And he and I were in India at a meeting recently for the synthesis.
And he's a bird photographer, and I'm a twitcher. I go count them. So we went out and had a good time. Anyhow, let's go back again. I showed you this picture. There's the third assessment report, 1.4 to 5.8. So now, a similar study six years later was put by working group one to the plenary.
The plenary is when 100 governments get together-- 400 representatives-- and word for word, approve the summary for policymakers by deciding what to take from the chapters. It can't change the chapters. But they decide what they think is of significance, and what you'd expect fossil fuel-dependent states try to spin the report down in seriousness.
Small island states, coastal dwellers, or others who have concerns where they think that they are a victim of the impacts try to spin it up. And then the scientist's job is not to let that happen. And in the end, you have to negotiate. You don't have a report. And the process amazingly works. But it works very, very frustratingly.
So here was the old one. Now, what is the new one look like? So this, I've blocked out something. This is what was proposed by working group one-- fourth assessment report in its plenary to the 100 governments. I apologize for the lousy graphics, because as we just heard, they're not available yet in the good form. I just got that from a preliminary one.
So here's the same thing, except they don't run the hockey stick back. They just run the present observations. This one is an unimaginable miracle, which is that concentration stay fixed now for the century-- not emissions, because if emissions are fixed, you're still going to double just in 150 years.
So concentration stay fixed. It still warms up another 6, 7/10 of a degree-- doubles the warming we've already had. Then here are three scenarios-- b1, the egalitarian sharing, a balanced one and a2, which is a really nasty world. It's regionalized-- not much globalization. It's got a lot of population.
Remember the top one from the previous one? A1FI , the fossil intensive. Right now, that looks like current business as usual. It was not in there. Well, what would have happened if governments didn't force this on working group one?
Governments insisted that the plenary put in A1FI . So they added that. And then they put in these bars. But these bars are different. These are calculated in models. What these bars represent-- and I applaud working group one from Duke for doing that, even though some of the scientists, they were uneasy about it-- is they recognize from the literature that there is a medium confidence that the ecosystems of the world, which are currently believed, not the cutting down of tree part.
Given that you've got a certain stature of forest, that they will absorb some of the CO2 emitted into the atmosphere, thereby, being a negative feedback on concentrations to stabilize. But as you get hotter, you decompose organic matter in the soils at a much faster rate. And you start threatening, switching from sink to source. And you start damaging forests.
And so the literature is suggesting somewhere between a couple of degrees and four. You're going to switch the ecosystems-- I don't know, Jim, what do you think? From sink to source. So they built into the top of these bars for the warmer scenarios, an added half to 1 degree that's meant to represent that positive feedback. I think that's completely legitimate, appropriate risk management. And they explained that.
They were initially very reluctant. I don't quite understand why maybe you know, Peter. I just get nervous about being subjective. On the other hand, how can you not tell that to governments? Because part of the risk decisions they have to make includes that, so good for them for doing it. And good for the governments for saying you got to do it, because that's the authority that governments have.
So this is the table. Don't worry, I won't make you see it. 1.1 to 6.4 is the new range. But I want you to see this-- below is 1.1. If they had not put in A1FI, the top would have been 5.4. I could see the headlines-- IPCC reduces seriousness of global warming. The third assessment report was 1.4 to 5.8. the fourth was 1.1 to 5.4.
That's the reason the governments insisted on A1FI, because they wanted to make it comparable to the third assessment report, so you're comparing apples with apples. It's still not completely comparable, because the top number includes that extra bump. That was not included in the tar and wasn't included in the models from the decomposition or of organic matter and/or the shrinking of forests.
And I think it's good that they did that. It's good that they were clear. But in the world out there, people don't read the fine print, don't understand it. And, therefore, when you're doing a series of reports that have noncomparability, it's very easy for people to misinterpret that, as the science having gone up or down, when, in fact, it's just a different framing.
And that's all part of what we have to do. This is also from the AR4. And this is what I call the Sheared Blue Tornado. And I like this figure a lot. I'm afraid we'll have to spend a minute or two on it. What it says is-- so here's today-- 300 and whatever it is-- parts per million-- global mean temperature increase around 0.75.
These are projections forward in time. So here is CL2 stabilization levels. So what is that? Let me see-- my next one should have that-- no, uh, oh-- how come it's missing? Well, it's missing a slide, or it's not finding it. I'll explain it then. All right, what doubling of preindustrial is from 280 to 55560? So that's what this line is.
So this is a doubling. So that CL2 is doubling of preindustrial. I'd put it in some slides that actually show you that and label it-- sorry about that. And they said it was likely based upon their reading the literature. This is the first significant change in the Chinese 1 and 1/2 to 4 and 1/2.
They now, called likely 2 to 4 and 1/2. Okay, likely is 66%. That means you've got it if it's symmetrical. And nobody knows that-- 17% on either side, right? 17 and 17 is 34 plus 66. Last time I checked, it's the whole thing. So that's what they did, so 4 and 1/2 to 2.
But then they assessing literature, particularly, looking at paleoclimate and some other multiple runs from the head center where they did a Monte Carlo with different values of parameters. They said it was very unlikely-- that's the phrase up here-- that the climate sensitivity is less than 1 and 1/2.
So there we are-- 10%, because IPCC defined very unlikely as less than 10%. That's defy. That was from the guidance paper that we forced on them that everybody's bought. And the 7% was a residual. So here we are-- 17% chance that it's beyond that.
Oh, okay, so here's the language. Unlikely means less than 33, more than 10-- very unlikely, less than 10. So now, here's the tornado again. Okay, I did get it in. I just put it in the wrong place-- apologize. So it's missing something. So this is the likely climate sensitivity range.
This arrow should be going over to here. That's what happens when you do something at 3 o'clock. So there it was. So there is the likely climate sensitivity range for 2 times CO2 550 PPM. There it is. I've got it. At least, I got that in right. And there's the bar across there. But what about that? Remember the language for less than 33? Unlikely?
Where did that come from? That approved language is unlikely. So I asked my friends and colleagues, where did you get cannot be excluded? That is not in the guidance paper. It's true by definition. I can't exclude an asteroid collision with a 10 kilometer asteroid. And I can't exclude that global warming is here.
What does it mean? And I found out the reason they didn't call it unlikely-- at least, that's what I was told. Somebody correct me if it's wrong-- was, because they didn't want people playing in the shadowy tails of the probability distribution that Henry Kendall told them they're supposed to, because they don't like being away from good constraining data.
I sympathize. On the other hand, risk managers-- certainly in the military and health and in business-- that's exactly what they want to know. So I'm going to switch to a little sports humor here. I can't help it. I was in Colorado, as you said, for 20 years. I still Bronco fan.
So this was Denver's defensive end when they were really good defensively. And they were the best defense in the league in the middle of the season. And they were about to play Indianapolis, who was the best offense. So everybody was hyping up this game.
So they asked the defensive end what was going to happen-- predict the point total on Sunday, because Denver had apparently the second-lowest point total in the history of the NFL for the first six games.
Okay, and he said, I'm not predicting nothing. All I'm going to predict is a good game. And I'm hoping the Broncos come out on top. And then he said, if I make predictions, and it goes the opposite way, then I'll look like a horse's fanny.
Well, Henry, I ain't trying to look like a horse's fanny right now. In other words, this very good defensive end should be in the risk management business, because he has Type I error aversion-- now, explain that. Okay, so economists long ago-- I may have learned this from you, John, but it 2,900 years ago. I'd hate to tell you-- two ways-- it's false negative, false positive. That's what it really is.
So Type I is we're not sure, but we have concern. We alert society. They invest money for something. It doesn't happen. And the ultimate example of that is if you're an earthquake forecaster in LA, you see some chance that in the next two weeks, some of your indicators tell you there could be a major earthquake.
If you announce that 200 people are going to get killed in car crashes, there's going to be 1,000 staying behind who are loot. The National Guard is going to arrest and shoot some. You're going to have a horrible event. So I don't want to risk that. I'm like the defensive end or working group one.
I'm Type I error averse. I don't want to give you a wrong forecast-- don't say anything. Probably the forecast will be false. I think this is backwards. Let's get it right. So the decision-- in a Type I is the policy response follows.
Okay, the forecast is false. That's right. It's right. You make a Type I error. You gave a forecast-- didn't happen. All those people died, leaving the town. And we had the horrible riot, Okay, with the remainders. So you've squandered resources and wasted stuff.
And if the forecast is true, well, it's a good thing. Supposing you didn't want to take the risk, and you don't want to get blamed for all that side effect. And it happened. Then you have unmitigated damages, and 10,000 people die. And there's no right or wrong answer.
Type I, Type II is not about science. It's about personal values on risk aversion. Are you precautionary like Henry? Or are you precautionary on the forecast? And so scientists are innately tend to be Type I error averse because we don't like to say stuff when we're really not very confident in it.
And society and many people who face risks like to be Type II error averse. So what's our role? Assess risk-- consequence times probability of occurrence. And you do it as a function of alternative policy choices. And then it's also our job-- confidence in the assessment of risks.
The distribution of risks and benefits, and then a traceable accounts of aggregations. We've talked about all those already. What about the role of decision makers? Their job is to negotiate the acceptability of risks and policies that alter risks, make choices, and then also suggest the assessment process.
We don't always choose to study what they think is important. We often tend to look at what's interesting, or what we can get data on, rather than on what they need. So that's where the merger of scientists and the decision-making process works well.
And whether we're going to sit there and worry a lot about giving a bad forecast, or whether we're going to worry a lot about we didn't hedge by ignoring the forecast because of uncertainty, that, in a way, is the decision-maker's choice. We should not be withholding information because of that, though. We should just be putting the confidence that we have in it out there. How much time?
AUDIENCE: You have 12-plus or minus.
SCHNEIDER: Oh, good, I'll get there. Thank you. You want to spin that wheel and see what I get now? So here's the competing paradigms between the science and policy communities. So it's common in policy, right. The incorrect forecast taken to be true is Type I-- decision to ignore is Type II. We've already said that. There's just a quick summary.
And what this argues-- and don't bother to read it, because I just said it-- is that risk managers often prefer to hedge against potentially damaging event. And those of us giving the forecast don't like to give bad advice. So that's something, which if both sides understand it, and we can do a better job than in doing it.
Of course, Type II error version, that's Henry, right? Stop them at the outset and don't wait until the results are presented to us, and the troubles have developed. So he's clearly-- and I am, personally, as well. My value systems are in Type II error version up to a point. I don't know what I'd do in the LA case.
I really want to have first decimal point odds before I'd announce it. But those are not easy choices. But it's the nature of the business that we're in when we're starting to bring science together in the policy world.
Okay, now, what about dangerous anthropogenic interference? I do this. I'll just show you a few minutes of the personal work that I do. We also have assessed-- oh, sorry about that-- came out lousy. Mike Mastrandrea, my colleague and I, published a paper on probabilistic assessment.
We always put the quotes around dangerous-- not just because we're quoting the UN Framework Convention, but because I don't know what it means. It's a how-do-you-define-it paper. This is in PNAS.
And this is our mantra every single time I give a talk-- "owing to the many model-dependent assumptions inherent in the use of such highly simplified models, we emphasize that our quantitative results using this simple model and not intended to be taken literally.
We do suggest the probabilistic framework and methods to be taken seriously. They produce relative trends and general conclusions that represent risk-management approach." So that's the frame. Now, if you're going to give anybody something quantitative, you have to make assumptions, and you're going to run a model, which is going to give you numerical answers.
And so what our philosophy is, let's run the model many times under different beliefs and see how that changes the outcome. So now, we're going to talk about climate policy as risk management. We've already seen those other slides. And that means assessing risk is a function of policy choices.
So what are the policy choices? Well, we did run the fan of uncertainties that IPCC had, but not in the particular study I'm going to show you. We went and used O'Neill and Oppenheimer's scenarios. And the reason I use them is I think my own personal belief is an overshoot is more likely than anything else.
I wish I had these in color. I just got it. Oh, I'll just walk you through it. So this is ready to forcing watts per square meter associated with various emissions and scenarios. 2,000, 2,100, 2,250. 500 parts per million-- CO2 equivalent-- a little bit below doubling over preindustrial.
So scenario one is SC-- slow change, gradually sidles up to the equilibrium, stabilization. Here's scenario two. That's that one. That is a more rapid change, but it still stabilizes. So in this case, we start out using the coal and oil. But we invest a lot in alternatives that we invent our way out of the problem after a big orgy.
This is an even bigger orgy. This is more than doubling CO2 equivalent. So there is the slowly changing world, the rapidly changing world. And then there are these two. This one is an overshoot of the 500 that comes down to 500.
So what we do in this period is bigger orgy invent your way out of it. And then in here, even bigger, the same thing. So these are overshoot scenarios, which found personal view is more probable than the slow changes. I don't know what you think, Mort, John.
You're going to make me hang out here and do it. Mike, if you could convince me you had deity status, and you put me on some very painful electric device and asked me to sketch what I think the scenario is going to look like, I'm going to probably sketch one in between that one and that one.
I think we're going to overshoot and come down. I do think we're going to get smart and stop this, but not right away. But okay-- doesn't matter what I think. Our job is not what I think. It's what the models tell you under a range of assumptions in the literature.
So this is an interesting one, because this is a tremendous challenge to science. So DAI-- dangerous anthropogenic interference. What's the threshold below, which we're okay, and above, which we're not? I don't believe in such a thing.
It's in the framework convention, because already, we've passed a threshold where some species will probably driven to extinction. I'm going to guess the polar bear ecosystem is already functionally extinct and so forth. But it's not a lot of systems. And the degree of magnitude is not so much.
Above 4 degrees-- I think we're going to have world-wide systems in every sector and every location very seriously affected and to a deep degree. So, therefore, where you set it is back to that value judgment about things. So if you take what the EU did, they pick 2 degrees above preindustrial.
2 degrees above preindustrial means 0.1.4 above 2,000 at the last IPCC report, because it'd already warmed up 6/10 and 1.25 above now. That's pretty probable. So the likelihood of exceeding it is high. So you have the mean exceed its amplitude. And then you have this-- what we call DY.
It's like air conditioning. You want to figure out what's your air conditioning load. What do you have to have in the power plants? You add up degree days. So you look at the degree years. What's the total area under the curve? And time really matters, because if it happened sooner, it's harder to adapt. If it happens later, it's easier. So that's basically the thresholds that we're looking at.
Okay, sorry for the complexity. But here we have the CDF an exceed in CDF. I'll stop after the next one. We've gone far enough. This is the probability of exceeding. So here's the EU threshold, right. It's 1.4, because they set two degrees above preindustrial, have to take the 6/10 away, because this is above 2,000.
So there is their threshold. So now, you're going to ask, what's the probability of exceeding that for the slow change scenario in 2200 or for the overshoot in 2200? That's the yellow. Those are these two. Okay, where did that come from? That came from taking the three functions I showed you, the probability functions of climate sensitivity and stupidly averaging with equal weights.
Now, there's a lot of reasons not to do that. But we're only here to demonstrate framework. So we did it rather than pick one of them. It doesn't matter if the numbers change. The object is to see what it means. So there is about a 45% chance that that threshold of the EU will be exceeded above the 2,000 level for even the slow-changing one.
There's a 55% chance in the overshoot in 2,200. But remember, it exceeded and then came down, supposing what mattered was the top temperature. I don't know. Maybe that's going to trigger some kind of ice melt you can't stop or some species extinction. This is what I said, what a challenge, the scientific community.
We not only do we have to figure out how to do climate transients driven by transient emissions scenarios, we got to figure out how to do impacts transients, because that's probably the scenario we're going to get. And if you take a look at the peak temperature in the 500 overshoot, you have a 77% chance of exceeding.
So we can end with this one. It's worse than I just said, because now, here we are. Seven years later, there's been an increase in temperature. It's now, they say, 0.75, not 0.6. So the new EU dangerous number is 1.25. And if I move this thing back, now, we're looking at a 55% chance on the slope for 2007.
Now, all these numbers, again, don't take them literally. But the way that I think we need to frame the problem is as a risk management problem. And the key to that is to try to get people to be a bit more literate about what that means. And I'll end gratuitously, Ron, because I stole this from him when he and I were at Ways and Means Committee giving testimony.
And you made the mistake of having it on a jump drive that I quickly took. All right. So there's the first one, right. There's your famous one from your unmitigated CO2 tripling. Have you all seen this before? All right, I'll give you a little shameless pandering. I think this is very cool.
All right, so now, let's invest whatever it takes-- trillions of dollars. Who knows-- there's all kinds of debates about that. And now you get a new wheel. So there's the wheel with only CO2 doubling. Oh, I don't know how you did this. This is really, really cool. So when I show this to my students, they like it a lot.
I said, so, okay, kids, which world do you want to be on? That one, that one, that one. I said, but it might cost trillions of dollars. And then they ask a very good question, well, what, does that mean? I don't know what trillions of dollars means. And I said, well, it means a large fraction of current GDP and attrivial fraction of future GDP.
In fact, I've done work on this. It's on my website-- Christian Azar and I. And what we basically showed was that if you really projected growth rate in the economy at 2% per year, and you lost 2% of the economy 100 years from now, that's like $7 trillion. Just in one year, it sounds unimaginably high-- 50% of the current US economy.
But at 2% per year, that's made up in two years. So what you end up with is being 500% per capita richer in 2102 with a $350 a ton carbon tax that keeps you to 450 PPM, versus, being 500% richer in 2,100 with whatever you get on the roulette wheel. And for me, that's cheap insurance, even though it's an absolute a lot of money.
So the question, then, is, how's the political world going to react? And when you get people literate enough to understand these ideas so that they realize that these really don't require that sophisticated an understanding. We're talking about risk management with a life support system.
And that's how, at least, I try to frame it when I talk to those groups. And it's very tough to frame it this way when you have 20-second sound bite on the evening news. It's a little easier in Congress. And graphics like that, Ron, on your group, thank you. This helps a lot. Thank you, all.
PRINN: So we are ready for questions-- so just put your hand up, and I think the acoustics--
PRINN: --are pretty good here.
SCHNEIDER: And I did it under the time you told me to stop.
PRINN: Yes, yeah. Well, within that aeroba that I did.
SCHNEIDER: Oh, all right.
PRINN: Yes, at the back.
AUDIENCE: Thank you. I was just in New Orleans where I was out on legging with all the night's award and hustling to launch the national day of action was overwhelming a couple of weekends ago. And, although, I really welcomed the spirit with which you said it, and I probably should do that knowing and intended just slightly.
I'm not convinced that one post. If you've been in New Orleans, it doesn't look like it isn't cost of the land is notorious. And so I [INAUDIBLE] suggest that aspect of this important battle, if you will. And secondly, I happened to hear Bill Nordhaus was up at Harvard a week or so ago. And he's done some recent work where he seems to be-- he's an economist at Yale--
SCHNEIDER: Yeah, I know him very well.
AUDIENCE: And I wonder if you could just briefly summarize what you think he's saying currently, and your understanding-- your assessment of his current analysis.
SCHNEIDER: Let me start in the back with Bill. I don't know what his current one is, because I'll be at the Energy Modeling Forum this summer, and he'll show it. Do you know, John or Mort or somebody what Bill's done lately? I know he's been trying to incorporate nonlinear effects and making his damaged functions have much longer tails. But I don't know--
AUDIENCE: He has a new version. It's now brand new. I reviewed his paper. I think that make it a cast, actually, either that or a [INAUDIBLE] very low CEUs-- [INAUDIBLE] changes. And so it's relatively optimistic assessment of--
SCHNEIDER: I, obviously, remember-- in this business, as I said, don't ever take model-dependent results very literally till you unpack the assumptions. And, in fact, I know Bill's dice model very well, because the C, it's right at dynamic integrated climate economy model. The C is Schneider and Thompson 81.
And the reason I recommended to Bill to use that simple two-box globally-average model, because it was just as bad as the globally-average economic model that he was using. And Bill did not build it to forecast the world. He built it to do sensitivity analysis for which I thought that was responsible.
Now, how some people have used it is another story. So we'll argue that out. My largest debate with Bill, and sometimes John and the whole community, is I don't use market interest rates when I'm discounting sustainability-- not on cost benefit analysis, because I don't think my child is a more valuable life than my grandchild.
I use it when we're doing cost-effectiveness. That is if you've picked a target for reduction based upon a political negotiation about weighing across all the multiple metrics-- remember, I said this things traded in markets and ecosystems and species and inequity, which you're just talking about in New Orleans.
How you aggregate those into any single thing depends upon the weight you put on the various metrics. That's not science. That's values. So my own view is you should reaggragate multiple times with different values and different weight functions and then have a political negotiation.
But once you've picked in the political world how much you want to stay under or 550, 450-- whatever it is-- then you, of course, want to have the economist with discount rates with every good tool they have to try to get us there most cost-effectively. That's my own personal view, which is that picking targets is not amenable to cost-benefit analysis.
But finding out how to do it once you pick the targets is, in fact, requires for efficient resource use good economic methods. We could argue that one, John or others. But that's my own prejudice. As far as sinking the one end of the boat is concerned, there is no doubt. And Henry said that. Remember, he said the poor will be hurt more.
But New Orleans was going to happen with or without climate change. I have been to no less than 10 meetings in my life before New Orleans with people showing slides-- I could show you. I don't have it on this chart-- of it mostly below sea level, saying any category three-plus hurricane is going to make that city history.
And what did they do? They chose to cut taxes to top brackets rather than invest $10 billion on fixing the levees. That was a political decision, not a geophysics decision. And what warming does, right, Kerry, is it increases the likelihood of some increased intensity storms sometime over a city's lifetime.
And, therefore, we don't know whether the 1/2 degree warmer in the Gulf was worth what-- 3 inches or 2 feet of the storm surge. We'll never know in any one storm. So we can't say we did it. On the other hand, what you do is you shrink the time between these catastrophic events by warming it up.
So right now, we're 6, 7/10 of a degree warming for that one. You want to talk about sinking both ends of the boat-- when you're talking about 3 to 6 degrees warming, there's no literature on very many benefits after 3 degrees.
And also, in a world of that kind of stress-- and this is what a number of military advisors and strategic planners are saying-- is that kind of a world is not a world that lowers the probability of conflict. And, therefore-- and it's not that climate change is going to create conflict.
You have to go to an area where human behavior and historical animosity and all kinds of factors make it vulnerable. Now, you add on top of it an outside force that increases stress. We don't need that on top of what we've already got. So that's what I would argue.
So I think Henry was using it, metaphorically, meaning the larger changes down the stream, and that we wouldn't be exempt from world problems associated with their getting hit harder than us. And I think that's probably fair. I don't think he was trying to imply that the scale of our economy that would be damaged would be the same as the scale of the Honduran. It was in Honduras in Hurricane Mitch.
So we lose 1% for a year. And they lose 40%. What was the number? It was really bad. So that didn't sink the ship yet, because it's still just isolated. It's when it goes beyond isolated that that becomes a problem.
PRINN: Yeah, at the back.
AUDIENCE: Two-- where okay interactions between permission? How do you feel that the [INAUDIBLE] professionals leave uncertainty between policymakers of developed countries and developing countries?
SCHNEIDER: That's a very good question. I don't know if everybody heard it. How do I see the perceptions of policymakers in developed and developing countries? First of all, you could have said, how do I see the perceptions of policymakers in the developed country called the United States?
And I can tell you the perceptions in California differ radically from the perceptions in the West Wing, which now different radically. And I think Ron would agree after our testimony. And I'm now done three since then from the Congress. So within us, there is very, very different perceptions. And a lot of that comes out of, a, risk aversion and, b, general ideology.
Do you believe protecting the commons is your job? And do you believe protecting entrepreneurial rights is your job? And that's deep ideology. And that often drives people's perceptions of the seriousness of the problem is whose ox gets gored? The victims of the climate change or the victims of the policy?
So that is not even constant across developed countries. Look at Europe versus the US or California versus inside Washington. And even as I said, inside Washington, the other part of Washington. So now, let's go back to the developing divide, and that's bigger.
And it's bigger, because we argue a ton of carbon emitted in Beijing does exactly the same thing to the ecology and the sea level as a ton emitted in Boston. Okay, everybody agrees. Therefore, [? Bird Hegel ?] said on the day that Henry testified to the White House when the rest of us did that. Therefore, they don't want a policy if there isn't a level playing field.
It's not fair that our industries should be taxed. And we'll have leakage from China. And they went in all those arguments. And they forgot that we had a 100-year head start, that we had the Victorian Industrial Revolution with sweatshops, coal burning, internal combustion engines. And we got rich doing it.
And then as we got richer, we started to cut emissions and started to worry about this problem. And they said, well, they-- India, China-- when we catch up to you, we'll take a target. So, of course, we said, back, yeah. But you got four times the population. We can't multiply that times the emissions that we have, because then we'll end up CO2 Quintupling or more. And that's not sustainable.
So now we have the blame fingers going out. I've seen this at cops all the time. You colonialist power, a monopolistic capital, hoarding your resources, keeping your technology, and making us buy it, and then not putting your consumptive patterns up for negotiation.
And then, of course, on the other side you hear, you inefficient place without proper market incentives to stimulate people and with corrupt governments and too many people. And, of course, who's right? Both of them. So how do you solve the problem? Trust-- ha. You talk to each other. You say, okay, you can't repeat the Victorian Industrial Revolution with your population at the rate that we did it.
But you don't have to, because we now have invented high tech. We have more of the patents and the scales than you do. I'll tell you what-- we'll fund our companies, give them tax breaks, or other money. They can go over there with a public private partnership with you. You each co-own the patents, go invent something, get rich doing it.
You need to have a leapfrogging over the Victorian Industrial Revolution, because right now, what you're going to get is blame. And it's slowly changing. But what we need instead is cooperative solutions. And I think now, we are closer to that than I have seen in my career on this. And I started working on this stuff in 1970.
There was a small window in 1988. But that got closed by this information campaign. And we didn't have as strong a scientific community buy-in as we do now. So I actually think we're going to get somewhere.
And this plain, ordinary incentives to companies to really try to be the ones that get that turnover sooner, because the ones who come up with the cheaper batteries to store things, and the ones that come up with the hot tower solar at near coal prices, even if it's a little above, if we have a shadow price, that will even it up. They're going to be very rich. And they should be.
So I'm hopeful that we'll get around it that way. But the perception in the developing world that if I'm going to-- let me frame it nastily. Their view is that they're willing to hold at least, in rhetoric, the sustainability agenda of the planet, hostage their view of equity.
And we're willing to hold it hostage to our view of consumption. Neither of those are okay. And I'm hopeful when we have some leadership that we'll be able to make some good deals.
AUDIENCE: I'm just wondering what [INAUDIBLE] nuclear power effects that are part of your solution.
SCHNEIDER: You're asking me about nuclear power when this is one of the temples of good thinking about that. I'll answer it by going way back to the Bush 1 administration. When the president made the statement in the campaign and afterward that it would be no new taxes-- and then the environmental groups were saying, no new nukes.
So I got asked once by many more than once by a reporter. And the reason the answer is still relevant, because it's what I still believe, is I said, I try hard not to be an ideologue. And I try to never say no new nukes or no new taxes. What I'll say is no inefficient inequitable taxes-- no dangerous overpriced oversubsidized energy systems.
Now, do I think that's nuclear? Then yeah. Does that have to be nuclear in the future? Not necessarily. But I'm not going to get into the debate of the details. What I would argue is steel with nonmeltdownable cycles with waste disposal. And I don't know whether it's ceramics or something that's acceptable.
I don't know that we want to subsidize liability limitations. It's a subsidy without also subsidizing solar wind and plug-in hybrids. So my own view is if we're going to invent our way out of the problem and build an economy, we probably need to subsidize everything. And I wouldn't leave nuclear out.
But one area where I worry is in countries that do not already have nuclear weapons. I worry about whether or not anything that we sell them in terms of radioactive products don't need to be taken back, that you have to have the cost of the process include repossession.
I don't mean you have to take it back where it was. But you have to be responsible for that. And maybe that has to be part of a deal. And it should be part of the cost. I think the Europeans, now, are requiring computer companies to repossess the computers at the end so they don't end up in all the landfills.
And so I think doing some version of whole system accounting and not just for nuclear but for everything. And then finding a way to transition from now till then is the way to go. So I'm going to repeat-- I'm not going to be ideological about it. I don't personally think the costs are going to encourage it very much in this country till people figure out a way to bring down the costs. I doubt it's going to be a serious option.
But that doesn't mean it couldn't be in some places or even here if that happens. Others here must know much more about it than I though. I encourage my environmental friends not to be ideological.
PRINN: Okay, this looks like a--
SCHNEIDER: Let's go out and have a reception.
PRINN: to stop. There is now a reception that's up on the fourth floor. In MIT terms, 34-401 A and B. And that's on the fourth floor of this building, of course. I'd like to thank Steve Schneider for a very stimulating lecture.