Three Degrees of Conflagration: Dr. Andrew Weaver on Modeling Global Warming

By James Turner
October 22, 2008 | Comments: 8

You may also download this file. Running time: 00:34:19

Subscribe to this podcast series via iTunes. Or, visit the O'Reilly Media area at iTunes to find other podcasts from O'Reilly.

James Turner: This is James Turner for O'Reilly Media. I'm speaking today with Dr. Andrew Weaver, Canada Research Chair in Climate Modeling and Analysis in the School of Earth and Ocean Sciences at the University of Victoria. Dr. Weaver recently served as the Lead Author of the Inter-Governmental Panel on Climate Change and is currently the Chief Editor of the Journal of Climate. Good day; and thank you for talking to us.

Andrew Weaver: Thank you for having me on your show.

JT: To begin I'm curious what leads someone to a career as a Climate Researcher; you were doing your Graduate work in the mid-'80s when global warming was just beginning to appear on the radar--certainly not the--excuse the expression--hot topic that it is today. Did you have some inkling that this was going to be one of the consuming issues of the next century?

AW: Absolutely not at the time; you know my PhD is actually in Mathematics but I always have enjoyed trying to do relevant math, that is math that I feel is applicable to what happens on a daily basis. So my thesis was on Atmosphere and Mathematics Applied to Atmospheric and Oceanic Physics--Physics problems and it wasn't until I went did a Post-Doc at University of Washington in Seattle that I was steered into the climate problem by a guy there called Ed Sarachik, quite a visionary at the time. And so it was back in 1989 that I got interested in the problem.

JT: Talking a little bit about the current state of the art and climate modeling, it strikes me--is climate modeling a bit like the Hollywood CGI blockbuster in the sense that you need people who are really good at developing models and then you also need people who are really good at implementing models, or do people come to the table who have like both skill sets?

AW: No; you're absolutely right. A climate model basically is like a laboratory. It's a combination of many, many hundreds of hours of works by people all over the world, so that what you have there is you have people who go out and study the physics of some very particular process. For example they might study how mixing occurs in the ocean, turbulence in the ocean; they might study cloud processes. Then you get others who try to represent those processes on a bigger scale, so what does it mean for that physics on a scale of say the size of a town. Others look at how different aspects of the system interact; how does the ocean interact with the atmosphere, we're building fundamental understanding of nature and then in the climate model what you do is you start to put it all together. I' m a kind of an end-person; that is I'm putting stuff together that other people have done the experiments in the field and then making those projections both back into the past as well as into the future. But you can't--you certainly can't do anything as a Planet Scientist.

JT: Right; but is there even in there a division between people who are developing the equations for example and the people who are actually writing the code?

AW: Historically there has been because there was a generation of scientists who basically got going when the computers weren't around and so a lot of the early analysis in many fields of physics was theoretical--analytical. That is it was equations without being able to truly solve them; but with the advent of the computer now we're able to solve equations that you can solve with pencil and paper. So now the theoretician is equally versed with numerical computational solutions of these equations as they are with analytic solutions. So I think that distinction has in recent years moved away from being there.

JT: Is there anything like an Open Source philosophy in climate modeling or is kind of every team off in its own corner with its own model?

AW: Well that's a great question. I mean our group here, I mean I have a--we have something known as the UVic Earth System Climate Model; that is Open Source. Anybody can have it and do whatever they want, within reason of course. And many groups are like that; it goes back to long, rich history of that stems from the NOAA Geo-Physical Fluid Dynamic Laboratory in Princeton which have historically always made their ocean model code available. NCAR--National Center for Atmospheric Research in Boulder have always made the climate system model available, within the community--we always share everything, but there are groups around the world in various government centers that are unable to release their code freely to people. But most people do; so the code is Open Source.

JT: In the good old days things like computational fluid dynamics was like gobs and gobs of Fortran code. What's the gearing under a modern climate model made of and do the normal rules of software engineering apply to this type of work?

Living, Reinvented: The Technology of Abundance and Constraints

Learn more about this topic at O'Reilly Emerging Technology Conference — Happening March 9-12, 2009 in San Jose, CA. Building a Bright Green Future is one of the keynotes from this upcoming conference. Learn more.

AW: [Laughs] Basically that's a great question because basically what--historically a lot of these climate models have originated from that gobs and gobs of Fortran code that was done in computational fluid dynamics except gobs and gobs more of Fortran is added on top. If you go around the world, most climate models are based on Fortran; that's an old language , the later versions of Fortran and they have pre- processors. That is, you write the actual code itself in Fortran but you'll have a pre-processor which is maybe C or some other language which will pick the bits of code and put the model together based on what you're trying to look at, the question you're trying to look at. For example, maybe you don't really need to look at sea ice processes if you're focusing on specifically on what's going over the Sahara. So you might just in the process pick bits of code that are relevant to the problem in question. So they are hundreds of thousands of lines long now; these things are huge and they are Fortran and a lot of this modern software engineering is kind of happening, but not in climate modeling. Although now we've had to take advantage of the parallel processing machines to really get our speed up. So the parallelizing and vectorizing of new code are becoming more and more the norm.

JT: That kind of leads to another question I had which is that climate modeling is notorious for requiring massive super-computing resources. Is the existence of the new distributing computing effort such as the BOINC from Stanford which is being used for drugs, research, and for SETI@HOME and things like that changing the game about who has the resources to do large-scale modeling?

AW: Well I don't--often it's a question of priorities of the government d'jour whether or not there are the resources to look at some area of science. I mean in Canada where I am, the Federal government over previous years has spent a lot in insuring that the computers are state-of-the-art--IBM systems right now. In--the same in NCAR; some have used Crays or NEC Super Computing system so I would say that the existence of computer power is not a limitation. The limitation is the desire of those who hold the purse strings to actually allocate the resources of a particular branch of science over another branch. So it depends basically on the issue d'jour actually.

JT: Is the problem right now more compute-bound or model-bound?

AW: Interesting; I think in the US it's probably a little bit of both. It used to be in the US compute-bound until very recently, until a very massive procurement happened for NCAR in the US where now it's less compute-bound and more the combination of the two. And that's true over most of the parts of the world that now it's going hand-in-hand. As you understand the physics more and more and you want to include better and better representation of the physics you need more and more computing power to be able to do so, so it's working in tandem. So I would say it's not bounded by either one by itself but by the combination of the two.

JT: How do you validate the correctness of climate models in the sense that short-term weather predictions can be validated against next year's hurricanes and blizzards? You can see if they track the way you think they will but a climate model really can't be validated that way because by the time you have enough data to validate it--it would already be decades into the target period.

AW: Yeah; I mean that's a great question. So again, the validation process in the climate model is very complicated. First of all, each of the individual components of that model are historically validated; that is the ocean component will be validated in looking at ocean examples; the atmospheric component could be validated in as much as you know in numerical weather prediction procedures, very--it's not much differences between the atmospheric component of climate models and the numerical weather prediction models--just much coarser resolution. But when you put it altogether, how do you validate that? Well the way you do it is a number of ways. One is you look at the statistics of the variability and you compare the statistics of the variability in the model to the statistics of the variability in nature. We can't look at the weather in the climate model but we can look at how it represents the El Nino Southern Oscillation, how the global signature of that variability looks and how that looks to years when we have El Nino. We can also then look at the kind of mean climatology and the seasonal cycle; how does the sea ice grow and disappear on any given year in the summer versus the winter? That's looking at the validation of the season cycle, which is the very component of these models. We can look at validation of precipitation field, climatological fields which our climatological data network. We can look historically, take our models and run them back to periods prior to civilization, last glacial maximum and others, and see whether the large scale features that are inferred by the proxy paleo record are captured in the climate model; things like a reduction of the strength of the sinking in the North Atlantic and the last glacial maximum; things like an expanse of ice sheets in the last glacial maximum; things like pollen records which suggest various temperatures in the last glacial maximum. The same thing in terms of the mid-Holicene when there was a slight warm period; you'd want to look at that too. And then you can also examine the transient evolution of the climate system from 1850 or from pre-industrial time to today and compare the transient evolution of what we have seen to that which the models say we should have seen. That is hind-casting. So at a combination of these you come up with an understanding of how good or bad or average your particular climate model is.

JT: I'd like to move on now to the work you did with the Inter-Governmental Panel on Climate Change. That document must have been a delicate one to write; it's one of those places where Science and Politics and Geo-Politics become inseparable. What are the realities of trying to produce a consensus document in that kind of environment?

AW: It's actually a fascinating procedure and process to be part of. I mean it's truly the most comprehensive thoroughly reviewed rigorous scientific assessment process I've ever heard of--and certainly being part of. I mean this happens over the many, many years; I mean it's three--about three or even pushing four years from beginning to end of the whole process. In fact we're starting now for the next one that will happen in a few years. The first meeting is going to happen in February coming up. So what happens then is a bunch of Scientists covering various areas of the climate system get together and they're tasked with writing a chapter. They have to go out and survey all the literature, come back, put it together, argue amongst themselves as to priorities or lack thereof, contact colleagues in the field and get their input and put together a very, very skeletal zero sort of draft, which is then sent out to the wider membership of those writing the IPCC Report. They get the feedback from that and then put together the first draft. That first draft is then sent out to experts around the world and literally thousands and thousands of comments come in and you have to respond to all of these comments, and you revise your manuscript in the group to deal with the comments. It goes out for review a second time and more comments come back and then you revise it again. Within a group there's probably maybe ten authors of a chapter. Now to get consensus there is actually pretty easy ironically--I mean sure there's arguments but consensus-- we don't argue whether gravity exists. We might argue a little bit about one paper, the assumptions underlying the analysis conducted in the paper, scientific manuscript that we're assessing--were actually valid in the context they were and so it's not actually that difficult. Scientists typically are able to reach consensus on these topics; sometimes it's delicate. And there is no political interference at that stage. The only political issues come at the final plenary when, because this is a document that's subject to UN regulations, the summary for policy makers has to be approved word-for-word unanimously by every Member State and that is where special interests start to rear their ugly heads.

JT: And that's when you get the beyond a 95-percent certainty, 90-percent certainty--?

AW: Well no; that's where you will get the decisions as to the certainty, those are based on chapters so those are based on discussions within chapters and the feedback we get from the international community through the peer review system. At the actual final plenary you will have governments who are unable to change the science. The regulations will not allow them to change the scientific content but they can question the certainty question or question inclusion or deletion of certain bits, so you might find that some special interests would not like this to be said, so they'll raise that objection By and large it ends up getting approved much in the same form as it was originally submitted but there's just some negotiations that happen in the final plenary.

JT: Many people consider the IPCC paper to be a wakeup call to old governments that the time for denial is over. Do you think it was a sea change or do you think it's going to be a blip on the way?

AW: Well, what I find interesting about the issue of global warming is that unlike many other environmental issues out there, but very similar to the ozone depletion issue of 30 years ago, the people speaking the loudest, who you will typically see speaking out about it are scientists working in the area. It's not the kind of typical environmental movement who are leading the charge, whilst certainly they're supportive, but you've got scientists standing up and saying we need to deal with this, so we've been saying that for a long time. I don't know why still to this day why the IPCC Report in 2007 had so much impact. Perhaps it was because it followed shortly after Al Gore's Inconvenient Truth; perhaps it was because Hurricane Katrina hit the US; perhaps it was because of other international weather-related events or climate-related events. Nobody quite knows but certainly there was a public awareness and willingness to actually take this and recognize oh yes, we need to start to deal with this.

JT: A common criticism of efforts to curb carbon emissions in the US is that with Asia's carbon output exploding, significant reductions in the West will become insignificant. Apart from the two wrongs don't make a right argument, what is the real incentive for the industrialized nations to cut carbon output?

AW: Well this is a great question because I'm in Canada and that's basically the single most important reason that a region like Canada doesn't want to actually do anything. Well first of all, number one, a lot of the growth a lot of the emissions in Asia for example are associated with goods we consume in America and Canada, so it's not quite fair to say that all their emissions are internal. They're not; there's a lot of external emissions associated with our consumption here and the production of the good and the fine. So number two is though if you look, the climate actually cares about not so much the emissions in any given year but the cumulative emissions of carbon dioxide; that is the carbon cycle as such that what matters is the total amount of human emitted carbon dioxide that's put into the atmosphere. If you actually compare since pre-industrial times, 80-percent of the amount of emissions is coming from 20-percent of the world's population. That is 80-percent of the emissions is coming from essentially the big developed world. Now China is certainly coming onboard but I would view that--rather than viewing that as a negative as a huge opportunity. What is happening here is China is improving their standard of living; every single in person in China is a consumer of energy and every single person in China will want to be a consumer of energy that is not producing emissions, so we who have the economic and technological wherewithal to develop those solutions need to get a move on. So I view it as, rather than taking this as a negative, viewing this as what a business opportunity this will be when we have a marketplace. Can you imagine if we in North America had designed the electric cars like the hybrids but 100-percent electric and we're now marketing them in China? Could you imagine that people in China are buying cars and India they're buying cars? What an opportunity.
Instead what's happening is our car manufacturers are suffering because they made a strategic decision to go and build big cars numbers of decades ago whereas in Japan they're building these hybrids that people are buying left, right, and center and they're making gobs of money. So it--you have to recognize that the world is going to move towards reducing emissions, so we can either be leaders and use that as an opportunity, a catalyst for change to steer our economy in a way to take advantage of that or we can do more of the same. But what is more of the same giving us? More of the same is giving us, look around; that's your local bank. The credit crunch is happening and so on and so forth and telling us now we need to think about doing things differently. And I would argue that would mean mirroring the economy and the environment together to look at using the environment as a means--or the catalyst and spurn the economy in new directions that are less fossil-fuel intensive.

JT: When the average person hears about a three-degree rise in temperature over the next century it may not sound that bad; especially if you're paying for fuel oil in the Northeast right now. What does three-degrees translate to on the ground? Is it an extinction level event or is it just something that's going to make--well just is not the right term--is it something that's going to make life miserable for us and unlivable for many?

AW: Three-degree warming if you're in some parts of the world like, well you might think well okay; let me just take a look at my winter and it's minus five--three-degrees, minus two. It doesn't work that way; okay? Three-degrees, first of all, is amplified; it's a global average--it's amplified where the feedback from the climate systems are stronger. So that means the warming is actually amplified in high latitudes. But it's not only warming; it's as you move the whole system towards warmer temperatures you're changing the statistics of weather about the mean climate as well. What does that mean? For example, the likelihood of say five days of 39-degrees of weather in Chicago is-- whatever it is today. If you warm the whole climate system up by three-degrees well now you're talking about a likelihood of 42-degrees and at those extremes that's a big difference. When you start to move up that, you move up those extremes more and more. The same with precipitation; the US has got a bit of a problem right now because if you look at where it's getting its water from, a lot of it's coming in the Southern US from high-plains aquifer which is actually not being replenished at the rate water is being extracted. So you have in the projections an increased drying of the subtropical latitude, we include the Southern US. So this means that water becomes more and more available [in some areas] but with populations growing and water going down you start to see some problems there. So it's not only about warming in that context; it's about warming as its affect to water.
In terms of species, three-degree warming means widespread, and I don't mean 10-percent. I'm talking about 25-percent committing to extinction of 25-percent--more than 25-percent of all species on the planet. I mean this is huge. These are big changes so it also starts to get--with three-degrees, what does that mean? It means we've passed the point of no return for Greenland and likely West Antarctica ice sheet; six meters from Greenland, six from West Antarctica or so or five from West Antarctica; we're committing ourselves to many meters of sea level rise. What happened overnight will take centuries, but that's a real problem if you live in Holland. I mean, what is 11-meters of sea level rise going to do to you in Holland? Well it means you don't have a country anymore. So then is that good for me sitting comfortably at three-degrees above warming in my house in New York? Well no; I now got millions of Dutch people who are saying hey; what are you going to do with me? It's all around the world like this. So to me the biggest problem is environmental refugees will be a very great problem and again this sows the seeds for discontent because in many parts of the world they do not have the ability to adapt economic abilities. They're also not the ones who caused the problem, so they're going to be looking to us and saying well heck; you caused the problem. What are you going to do about it? So I think this is a huge issue that I think we need to deal with because if you could frame it in terms of national security, you could frame it in terms of the internal security through water availability, you can frame it through just mortality and a species or even human comforts with extreme weather events. So all of the above are reasons why we need to act now.

JT: The more we study global climate, the more carbon tipping points we seem to be encountering--things like the biomass trapped in the Arctic Tundra. Just today the World Wildlife Federation released a statement that they think we may be passed the point of no return for polar sea ice. Do you think these tipping points or ones yet unknown may cause an unexpectedly rapid acceleration of global warming and are you trying to incorporate them into new models as you go along?

AW: I think you'd find that first of all most people in the community would agree that Arctic sea ice in the summer is gone. We can't stop that because it's going to happen anyway; it's going to go ice-free because of the fact that we've got so much warming in store because of emissions we've already done that. But in terms of feedback this is a great unknown. What we do know is we know for example that as frozen soil, permafrost thaws you expose organic material to oxygen or if it's water on top--no oxygen. In either case it decays and that decay will add another form of greenhouse gases. We know the sign of that feedback is positive; it does accelerate the warming but we don't know the magnitude and we don't know the stability of some of these northern regions. So this is a huge area of unknown right now, so it's something that the scientific community is trying to grapple with to try to get a handle on the magnitude. We know the sign, but what the magnitude is; so this is also why we have more uncertainty as to the upper bound of warming this century as opposed to the lower bound, which we're much more confident in. You know, again, I am an optimist; I think we'll recognize long before we have to pass these kinds of methane or carbon thresholds that we've solved this problem--like we did and rose to the challenge for chlorofluorocarbons back in the early 1990s with the Montreal Protocol.

JT: So finally I'd like to turn a little bit to your recent work doing climate projections for Discovery's Project Earth. First of all how did you come to be involved in that project?

AW: Well it's interesting; I know exactly what happened--is the Discovery Channel had decided they were going to look at some of these geo-engineering projects. And so after they had done this, they contacted a couple of people and I think one of them was they contacted people at NOAA in fact, NOAA GFDL in Princeton and asked them if they could run some experiments for them. But NOAA, the GFDL Model is an outstanding climate model, one of the best if not the best in the world but it can't run these experiments overnight because it takes an awful lot of computing power and also there's national interests at stake here; they have to be careful. But they recommended then that they contact us because we have a climate model which has a little simpler atmosphere, which is designed to look at these kinds of what-if experiments; it doesn't resolve individual weather processes like the NOAA one does, but it looks at the statistics of the phenomenon. So then they just contacted us and said hey what do you want? And we were deeply suspicious to begin with, deeply suspicious on the intent, but after some research--I mean this is the same company that did Walking with Dinosaurs. We felt that we can do this as long as our caveats are expressed as to what we believe are the ethical questions of actually dealing with the geo-engineering problem.

JT: How much of the work you did with the show involved developing new models and how much was just plugging different parameters into models you already had?

AW: It was 100-percent us taking emissions scenarios or--or Aledo changes or engineering fixes that were prescribed to us based on the people in the field and actually using our existing climate model that is being validated and evaluated. Everything we do is based on this; and then just running it with different input. It was not building new models; it was all about running what we had there.

JT: The series seems to provide a more in-depth analysis of benefits and risks compared to typical science content on television but at the end of the day it's so--it still seemed to me to have a kind of Amazing Race / Mythbusters flavor to it in that they had short timeframes in terms of projects, they had kind of an emphasis on the drama of the experiments; they did a lot of one-shot trials. Do you think you can do good science under those types of conditions? And do you think that the data that they collected was in any way valid?

AW: Well I think I was actually pretty impressed--I don't know what their budget was--at some of the things they did. I mean they didn't obviously have a goal of writing scientific papers in it but I was in talking to them--I understand that they built a carbon scrubber with David Keith in Calgary here that actually worked, which is a good thing. That's a piece of technology that I think you'll find a lot of people think is a good one where you actually take carbon out of the atmosphere and store it underground. Then they started to mess around with clouds and they actually had these big cloud experiments, but a lot of these things, I think it's actually a fascinating show or series of shows inasmuch as it puts out the whole questions out there because people have to recognizes that our--we have one planet. It's getting warmer. You know you might--either it is or it isn't but the fact is--it is; we know that. The scientific community has known that for a long time. We are going to have troubles unless we do one of two things; one is cut our emissions which I think is the right approach, or two is we start to play around with Mother Nature which is the geo-engineering report. You see the kind of gee-wow things that they're doing; well think about that on a massive scale and frankly it scares the pa-toobies out of me because that I can imagine putting these reflectors out in space and them being stuck to reflect sunlight and in the show we found that if you put too many and you take us back to kind of mini Ice Age conditions. I don't think that's the right approach; it's a dangerous approach; it's a hubris approach. You know we are somehow masters of our planet. There's--I don't think we can control everything like we want to and so I think the one--the best approach is to reduce emissions. But I think it at least allows the discussion to take place and I think it's important that--that discussion take place because really there are two options. One is--reduce emissions or two is this geo-engineering and frankly I don't like a lot of these geo-engineering ideas.

JT: That leads into the question I was about to ask, which is of the solutions you got involved in which one do you consider the most promising and which one scares the be-jeebies out of you?

AW: Well the most promising without a question was the carbon scrubbers one and I actually think that that's a good idea, because for example, the carbon scrubbers, what they did there is they developed a mechanism to actually take carbon dioxide out of the atmosphere, take it out and store it and capture it. You might argue that that's basically the inverse of what we've been doing as humans--as humans we've been putting it in and we can take it right back out. So to me it makes a lot of sense there because it's basically--it's not messing around with Mother Nature directly; it's kind of taking out the problem so--and I think that does have a lot of hope. And but even imagine for example that--that being done in parallel with an airline industry which I have a lot of difficulty thinking of off-the-shelf solutions for jet fuel but you might imagine that you could capture something like that and with jet fuel and you might get somewhere. The scary one is putting these shiny mirrors in space; I mean oh my goodness. [Laughs] I think that's messing around with the amount of sunlight we have coming in from space is not something I want to be anything part of, so I don't think that was right. Again you can imagine it being stuck on all these umbrellas, stuck on reflect mode and suddenly we're reflecting solar radiation for--let's not go there.

JT: Yeah; it strikes me as a hard thing to turn off quickly.

AW: Yeah; yeah exactly. You--well precisely, whereas the scrubber is a good one and planting trees; there's nothing wrong with planting trees. It doesn't have--as you saw in the series, it didn't have a lot of effect but it is a means of mitigation. It can't account for all the emissions we've done but there-- that's always a good thing. Plant some trees; it looks nice. You're helping nature; you're building biodiversity so that's a good thing.

JT: Last question I've got for you, the Carbon Establishment as it were has--seems to have altered its approach over the last few years moving from there's no proof the earth is warming stance to it's warming but there's nothing we can do about it. As the series itself points out, many of the schemes could be used as an excuse to slow down--not to slow down carbon output because we're just going to fix it technologically. Is there a little bit of a damned if you do, damned if you don't aspect to this kind of research because doing this research could deflect attention from reducing carbon production?

AW: You know I don't buy those arguments--there's nothing we can do about it. I simply do not buy that and the example I use from Canada and you probably have similar examples in the US but here's an example. In Canada we--any electric transmitted car is classified as a low-speed vehicle, an LSV vehicle. Transport Canada sets regulations that does not allow these LSVs to go more than 40 kilometers an hour. The reason why is that they haven't passed the same safety tests. Okay; well then you're not allowed to drive on any city streets in Canada a car that doesn't make 50 kilometers an hour. What we have there is a Catch 22. We have regulation insuring that these things don't drive more than 40 kilometers an hour because then it beats in with regulation that doesn't let them on the streets unless they drive 50 kilometers an hour. We have so much protectionism out there of existing kinds of technologies that I think we need to be dealing with this as a side and saying you know what; this isn't a free market. This is a regulated market, one that's regulated in favor of the status quo. So big oil money won't want the status quo changed because big oil money is there to produce profits for its shareholders and its shareholders want big oil to be used. So but we have to move beyond that and we can do that by creating a level playing field for other companies and also as consumers by looking at what we buy and asking the question what is the carbon output of this product and guiding these operations to our own consumption and behavior.

JT: I actually want to follow-up on that very quickly. If you think about big oil for example, certainly one of the poster-child(s) for it during most of its life has been T. Boone Pickens. Do you think it's significant that you're starting to see them move at least publicly to look more at alternative energy sources?

AW: I think so; people who are running big oil companies and big oil corporations they're in the energy business. They're there to make money; they're not there to --if they see a wind of change they might on the one hand make it look like they don't want the change to be regulated by government, but you can bet these people are looking out for alternates. The wind power in Texas for example; where is the leader in North America? It's Texas in wind power; Calgary in Canada and Alberta in Canada is taking the lead here. So as an optimist I think we are moving in that direction and I think that the sooner and the faster the better. I mean if nothing else, we don't have an unlimited supply of oil and a lot of that supply is in from parts of the world that are not the most stable these days so why would we want to rely on oil from these parts of the world? Why wouldn't we embrace these renewables? And I actually put nuclear power as--not a renewable but certainly as an interim energy source that I think we should be looking at very seriously because the emissions of carbon are zero and what we do is we have a storage problem. But that's a local problem that we have to contain in our backyard and we're not relying on someone else to deal with our problem.

JT: I've been speaking today with Dr. Andrew Weaver, who is the Canada Research Chair in Climate Modeling and Analysis at the School of Earth and Ocean Sciences in the University of Victoria. I'd like to thank you for speaking to us and I look forward to the next paper out of the IPCC.

AW: Thank you very much again for having me.

You might also be interested in:


"A climate model basically is like a laboratory"

Maybe, but the earth most definitely is not!

When climate models are able to model things like clouds accurately and not just as parameters, we cam MAYBE, start to take them seriously, but the simple fact is climate models are unable to do this as we do not understand enough about clouds.

Cuckoo, you seem to be saying that because climate models don't include all the details like clouds we can't even "start to take them seriously." But there's a long section in this interview which addresses how the accuracy of the model can be determined. This section about prediction was not very detailed, but you seem to be not thinking through the implications. If a model predicts both past and current weather patterns, then whether we understand everything about clouds is irrelevant.

You seem to be arguing from a theoretical stance ("We don't know enough about clouds, therefore these models can't be accurate") rather than looking at the evidence. If the model can make accurate predictions, that's an extremely reliable indication that knowing everything about clouds doesn't matter one teeny little bit.


Hansen, the IPCC etc all say that climate models are unreliable and recognise that climate models are unable to accurately predict future climate. Climate models are tweaked to show that they can accurately predict the past climate and therefore it is assumed they can accurately predict the future. This is simply not true. The earth is not a lab, the atmosphere mixes in such complex ways that there is little hope that computers will be able to predict future climate

Clouds are reduced to a parameters, when we don't know nearly enough to even reduce them to parameters and can't reduce the scaling sufficiently to see how clouds impact the atmosphere.

The sooner the fallacy that computers are able to accurately predict future climate is put to bed, the better.

Jim G, you say "If a model predicts both past and current weather patterns, then whether we understand everything about clouds is irrelevant."
Agreed, but have the models been able to predict current weather patterns? Temperatures have been flat for nearly 10 years now, did the models predict that?


I agree with you the IPCC is an excellent source of information about the accuracy of climate models. The Intergovernmental Panel on Climate Change has the backing of both the World Meteorological Organization (WMO) and the United Nations Environment Programme.

However, they do not agree with your low opinion if climate models. Perhaps you are not familiar with their more recent research? For example, see the report “Climate Change 2007.” Chapter 8 is entitled “Climate Models and Their Evaluation.” ( )

This chapter states on the first page “Climate models are based on well-established physical principles and have been demonstrated to reproduce observed features of recent climate . . . and past climate changes.” Overall they assess climate models as greatly improved and accurate. They even mention “Substantial progress has been made in understanding the inter-model differences in equilibrium climate sensitivity. Cloud feedbacks have been confirmed as a primary source of these differences.”

Now, about Hansen. I presume you mean James Hansen, of the NASA Institute for Space Studies. I think Hansen has some interesting points, and I am not enough of a scientist to follow the detailed criticisms he makes of global modeling. But I do think it very relevant that all of Hansen’s criticisms revolve around the current climate models UNDER-estimating global warming. You don’t actually state whether you think the current climate models over-estimate or under-estimate the danger. But if you are citing Hansen as a critic, do you agree the problem is that the models are understating the crisis?

I ask because sometimes on other web sites people cite Hansen as an authority, and then try to claim his specific concerns with modeling accuracy can somehow be interpreted to mean that climate modeling is generally bad and therefore there’s nothing to worry about. Hansen’s arguments are completely focused on the situation being worse than the IPCC thinks it is.


I am hesitant to reply. The particular fact you mention is often used by people who are “cherry picking” facts to argue against the scientific consensus on global warming. I don’t know what the models predicted about the last 10 years, although the IPCC paper I cited earlier might address this.

More important, I think, is that the global average temperature is clearly and unarguably going up, and that is consistent with the current climate models. Furthermore, the rate of increase has been getting worse. For a great summary graphic of temperature averaged across the last century see . Yes, at the very end of the data you see the rate flattens out a little. But there’s no question as one looks across the decades as to the overall increase.


Chapter 8 also goes on to say:

Nevertheless, models still show significant errors. Although these are generally greater at smaller scales, important largescale problems also remain. For example, deficiencies remain in the simulation of tropical precipitation, the El Niño-Southern Oscillation and the Madden-Julian Oscillation (an observed variation in tropical winds and rainfall with a time scale of 30 to 90 days). The ultimate source of most such errors is that many important small-scale processes cannot be represented explicitly in models, and so must be included in approximate form as they interact with larger-scale features. This is partly due to limitations in computing power, but also results from limitations in scientific understanding or in the availability of detailed observations of some physical processes. Significant uncertainties, in particular, are associated with the representation of clouds, and in the resulting cloud responses to climate change.

James Hansen may believe the climate models are under estimating the potential rise in temps, but what he is also saying is the climate models are simply not up to the task, which is what I believe. The earth is not a lab that can be modelled on a computer, whether or not it's a land, ocean or combined model. The processes are simply too complex as admitted by the IPCC

One thing I will say, Jim, is it is very refreshing to speak somebody on this subject who doesn't resort to the name calling defence, so prevalent in these discussions.


Just spotted this on the BBC website

A few quotes for you

"These are some of the largest cloud systems in the world and we know that they must play a very significant role in climate change, yet we know that climate models do not represent them very well," he explained.
"We hope to finally hit some of the uncertainties in current climate models on the head."

Hi Doc saw some of your amazing scenarios on Discovery and would like to know if you could run a model on say one then two billion home owners painting their roofs white to reflect back the suns rays ?
How many roofs would have to be painted to say equal the the landmas of Greenland should it lose its ice cover ?
Governments could give tax rebates to people painting their roofs white as an incentive.
Charles Segal

News Topics

Recommended for You

Got a Question?