#194 29 min 29 sec Degrees of uncertainty: Modeling the climate impact of greenhouse gases
Dr Malte Meinshausen is Senior Researcher at the Potsdam Institiute for Climate Impact Research, Germany, and Honorary Senior Research Fellow at the School of Earth Sciences, University of Melbourne. He holds a Ph. D. in "Climate Science & Policy", a Diploma in "Environmental Sciences" from the Swiss Federal Institute of Technology, and an M.Sc. in "Environmental Change and Management" from the University of Oxford, UK. Before joining the Potsdam Institute for Climate Impact Research (PIK) in 2006, he was a Post-Doc at the National Center for Atmospheric Research in Boulder, Colorado. He has been contributing author to various chapters in the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC AR4). Until May 2011, he was leading the PRIMAP ("Potsdam Real-Time Integrated Model for probabilistic Assessment of emission Path") research group at PIK before relocating to Melbourne. Since 2005, he is a scientific advisor to the German Environmental Ministry related to international climate change negotiations under the UNFCCC.
Host: Dr Shane Huntington
Producers: Eric van Bemmel, Kelvin Param
Audio Engineer: Gavin Nebauer
Episode Research: Dr Dyani Lewis
Voiceover: Nerissa Hannink
Series Creators: Kelvin Param & Eric van Bemmel
I’m Shane Huntington. Thanks for joining us. In its long history, our planet has undergone many shifts in climate. Some of these have occurred over a protracted period such as the Ice Ages, whilst others have been rapid, causing widespread extinctions of animal and plant life. The human induced shifting climate we are now experiencing has many unique factors when compared to previous changes. Of particular interest is the impact, over the next 100 years, of mitigation strategies, that is measures we can take to reduce greenhouse gas emissions, as well as to speed up their removal from the atmosphere.
Today on Up Close we speak to a scientist whose climate models specifically deal with the impact of these mitigations efforts. Dr Malte Meinshausen is a senior researcher at the Potsdam Institute for Climate Impact Research, an honorary visiting researcher at the School of Earth Sciences at the University of Melbourne. He is also a scientific advisor to the German government and has been a member of the German delegation to the United Nations Framework Convention on Climate Change, the UNFCCC, since 2005. Welcome to Up Close, Malte.
Thank you very much for having me.
We often hear about limiting global warming to two degrees above pre-industrial levels. How did we come to this number and what does it mean for us?
Well it's kind of a trade-off between what we thought is still possible to avoid and what was impossible to tolerate in terms of climate impact. So it's a trade-off. Global warming at two degree is certainly not a safe level for many ecosystems. It means probably for many people at the coastline it means they have to shift further inland, et cetera. So it's not a safe level. But it's kind of a compromise. And of course the compromise is partly value judgments. So what science can provide us,is the insight into all these thresholds that we are going to cross, as global warming is getting higher and higher. But finally a political decision of saying what is too much? So governments came to the agreement in Copenhagen that two degrees is too much and they confirmed that a couple of times thereafter.
Now some governments and primarily those from the small island states are understandably concerned that two degrees is too much for them, because it does mean that some of these islands are going to be drowning in the rising seas. So they proposed a target of one and a half degree, which we are probably not likely to meet right away. But we certainly can work towards coming down to one and a half degree again. We're at the moment about 0.8 degrees above pre-industrial levels.
Two degrees sounds like a very nice convenient round number. How much uncertainty is there in this particular figure we've picked? How bad would it be if it was 2.1 versus 1.9, for example?
That's hard to tell. We are dealing with large risks. So for example the risk could be that, if we sustain two degrees warming over 200 years et cetera, the Greenland Ice Shield will start melting. It won't melt immediately. It won't melt in 10 years. It won't melt in 100 years. But over the long-term, it would mean seven metres of sea level rise. And we don't exactly know where these thresholds are. We just know that, with enhanced global warming, we are getting more and more of these impacts what we have never seen before, as long as humankind is on the planet. So it's hard to say whether it's 2.1. It's obviously there are more impacts going to be with every additional warming that we are going to have. So again, it's a value judgment science can provide that somewhat fuzzy picture about the impacts that we are going to expect at the different warming levels., wWhether it's 1.9 or two degrees,. Iin the end we have some natural viability anyway on top of the warming. So I think the distinction is rather do we aim for two degrees? Do we aim for two and a half? Do we aim for one and a half degrees? That should be kind of the framework that we are thinking about.
Now you're a climate modeller. Can you give us an idea of how that is distinguished from, for example, a long-term weather forecaster?
That's something completely different. Because in weather, we are trying to really predict what is going to happen in the next days, maybe weeks. But it's rather on the daily scale. As a climate modeller, we are not concerned whether there's going to be sunshine or rain in the next days or whether there's going to be sunshine or rain on 1 January, 2060. What we are concerned about is what does the mean climate and how is that changing? The mean climate is defined as a time period of 30 years or 20 to 30 years. So then are we going to see dramatic shifts there in terms of winter precipitation, snow cover, the droughts in the summer, et cetera? This is what the timeframes that we are dealing with.
So although some of the larger models have the same cause, the weather forecasters use these earth system dynamic models and the climate modellers use them and just run them on a coarser scale but much, much longer, there are some similarities but still the question is a completely different one. The one is about the weather and the other one is about the climate.
Now when modelling the climate of an entire planet, and unfortunately earth is a very complex planet as planets go, what types of things do you need to take into account to get a realistic picture of that 20 to 30 year scale you're talking about?
That's a good question. We are all the time working on it, trying to figure out what we need to take into account there. Particularly one problem is that we do an experiment with a system earth that we have never done before and it's hard to run multiple experiments with the system earth. So we have one experiment. We see the observations over the last 150 years, how the earth climate changed due to the human influences. There are a range of factors that we have to account. First of all, it is our emissions and primarily the CO2 emissions. We know that CO2 is the primary culprit for what we are doing to the climate.
But then there's a range of others. G greenhouse gases, like methane or nitrous oxide. Methane for example has a shorter lifetime. If we emit it today, it will mostly be gone after 20 years. But CO2 is going to affect the earth climate for hundreds and thousands of years. Then on the even shorter lifetime there are aerosols, for example, sulphate aerosols, nitrate aerosols and organic carbon et cetera. The shorter the lifetime, the more regional the impact is going to be.
So we have to know about our emissions. We have to know about how they are going to change the atmospheric abundances, the concentrations of these substances. Then we have to know about the feedbacks in the system, which is for example the carbon cycle. If you put CO2 in the atmosphere and the CO2 concentration rises, then for example plants grow a bit better. They have more CO2 to do photosynthesis with. So they grow a bit better. We have changing precipitation patterns. We have warming. That is for example increasing soil respiration. So we have that carbon cycle feedbacks, the dynamics, et cetera, the CO2 fertilisation that has a lot to do about the link between emissions and then the ultimate concentration levels that we are going to see.
Then finally we want to know about the feedbacks. How do these concentrations actually affect the warming in the end? And thatIt is frequently coined climate sensitivity. So if we double CO2 concentrations, then the question is how much warmer is the earth climate average going to be? That number, between two and four and a half degrees and with the best estimate of three, that number hasn't changed dramatically over the last 10 years.
There's one paradox, because, if we work harder to understand the climate system, we know more about the feedbacks. But that at the same time can mean that we are not actually decreasing uncertainties in the future, because we know about all the little things, pieces and strings that hold the whole system together. Then each little piece and each little feedback, there is some uncertainty. So paradoxically, as we know more about the climate, the uncertainties about what is the future warming going to be can actually increase, not necessarily decrease. But over the longer timeframe of course they are going to decrease.
So to summarise, what we have to know, we have to know about our emissions. We have to know about all these feedback processes that then determine the concentrations. Then we have to know about the link between concentrations and temperatures. This is normally called the climate sensitivity. Of course then, for the local climate impacts, we have to know a whole lot about how does the global warming affect local precipitation patterns, et cetera, et cetera, in order to break it down actually to the scale where impacts matter.
Are we learning anything from a couple of our neighbour planets, being Venus and Mars, where there are obviously very significant climate scenarios that have gone on in the past and certainly in the present in the case of Venus that may give us an insight into our potential future for earth or our past?
[Laughs]. Well, if we are looking at any of these neighbouring planets and see any similarity in what our potential future could be, then we are in deep trouble. Because, as you know, they are not the most inhabitable planets. There is, every now and then, sort of the fear or the slight possibility that we could have runaway greenhouse effects. Then our earth would turn into something more similar to Venus. This is obviously a concern, but it is a hardly quantifiable concern whether there are really one-way climate affects.
The question is, and there the community might be a little bit more divided, in how much we treat that as a mere speculation that we should not further investigate or how much we should treat that as a concern with mainly only a per mille chance, but still, since the affects would be so dramatic, maybe we should take them seriously, like we would take a per mille chance of an airplane going down, we would take very seriously and would investigate further.
So, within the climate science community, I would say there are some people who more investigate the core issues about climate sensitivity. Then there are some topics on the margin where we don't actually know a lot about and there are some theories that it could go dramatically more and turn our earth into Venus. But I would say the chances are very, very small.
I'm Shane Huntington. My guest today is climate scientist, Dr Malte Meinshausen. We're talking about climate change mitigation and modelling, here on Up Close, coming to you from the University of Melbourne Australia.
Now one of the things I would like to understand with regards to the modelling is where the two degree figure fits into the modelling that you do. Can you explain that?
Yesp. The modelling that I have done is trying to synthesise the knowledge that is out there is multiple different climate models, so in carbon cycle models and radiative forcing scheme codes and large climate models that look at the dynamics of the earth. So trying to synthesise that understanding and its uncertainties and then answering the question, how much emission reductions do we have to do in order to have a good chance of staying below two degrees?
There, for example, we did a paper in 2009 combining carbon cycle and uncertainties and climate system response uncertainties for the first time in a larger modelling framework. Then we came to the result that, as humanity, we have a global emission budget between 2000 and 2050 of 1000 gigatonnes of carbon dioxide. So 1000 gigatonnes of carbon dioxide is our emission budget. If we exceed that emission budget then our chances of keeping global warming to be lower than two degrees will be not below 25 per cent. It might be 50/50. It might be worse. But if we want to have a good chance or a chance of at least 75 per cent to stay below two degrees, then we have to keep within that emission budget over the timeframe.
Now 1000 gigatonnes CO2 sounds like a whole lot. But actually in the last 10 years we burned over a third of it. If we continue at the current rate, and actually we're increasing the rate as we burn fossil CO2, we are going to exhaust that emission budget before 2030. Then kind of the people in 2030, the decision-makers, are left with a question do they stop immediately emissions? Or do they just accept that we are going to increase warming to beyond two degrees? So what comes out of this research, and this is not only our research, there are many different research teams around that, in order to stay below two degrees, we need to peak global emissions before 2020. This is in this decade and all the signs in China, in Australia, in the industrialisedt countries, are actually not looking in the right direction.
In the industrialised countries, we more or less have a plateau of emissions. Actually we had a bit higher emissions in the 1980s, et cetera. But we don't have a fast enough decrease of emissions. In the developing countries of course we have a huge boom in some of their economies. So we have an increasing fossil fuel use. They legitimately say, well, we have our development needs and we have to increase further our energy services. Now the question is whether - – and that leads to a whole different topic - – whether these energy services need to be met by fossil fuels. But if they are, as we did in the industrialised countries, then it looks rather grim for staying below two degrees.
This sort of risk mitigation modelling, I guess you could call it, brings into the discussion the issue of political will. Do you model for that? Can you model for that? When you put those sorts of factors in, how are we looking?
I would say the modelling is kind of in three sections. The first section that I primarily deal with is just the climate science. TIn there, we answer a simple question about how much emission reductions are necessary in order to stay below a climate target that policy-makers chose. The second bulk of the modelling would be finding out the economic and technical feasibility of those emission pathways that actually would keep us below two degrees. That has been done in huge model inter-comparison exercises, again involving a huge number of groups. There, the result is relatively clear. Yes, it's economically possible. It's technically possible. We have the technical means to start that path towards two degrees in the next two decades. We need more technologies thereafter. But it comes at a moderate cost. So the technical and economic feasibility is given.
Then, as you mentioned, the third thing is the political feasibility. I'm not sure whether there are any models yet around to capture that. I think scientists are well-advised not to try to model this, because otherwise you quickly get into a self-reinforcing cycle. If climate scientists, as citizens, believe, well, this is never going to happen, these dramatic reductions. Then they are saying well this pathway isn't feasible. So policy-makers say well scientists told us this pathway isn't feasible. In the end, we are not going to get there.
Humankind did large achievements in the past, if really the political will and the public was behind it and there were dramatic shifts possible in quite short timeframes. So I think there is some hope. Of course there is a lot of reason as well for pessimism. But as scientists we are well -advised not to trying to model that political will, nor make any inferences about it.
Now one of the key concepts scientifically that we deal with in all of these models is radiative forcing. Can you give us an idea of what is meant by this term and whether we can do anything to change that element of the model?
Why does the burning of fossil CO2s actually do a warming of the globe? The reason is because the CO2 acts as a radiative forcer. It kind of prevents the long-wave radiation to leave back into space and that the primary affect effect is there in the upper troposphere, so in the upper layers of the atmosphere. So the standard unit, how we measure radiative forcing, is what Watt per square metre? If we think of a little bicycle light-bulb having one watt Watt per square metre, this is about the amount of energy that we put on this planet. It sounds terribly small. But we put it on every square metre of this planet. In fact, it's not only one wattWatt, it's 1.6 watts Watts per square metre. Then that kind of accumulates over time and that leads to that warming that we are concerned about. So it's one watt Watt per square metre all across the planet, actually 1.6, that we are currently at. We are looking at, by the end of the century, if we do not mitigate emissions, we are at maybe eight times of that, so maybe six watts Watts per square metre or eight and a half watts Watts per square metre, but again summarised over the whole surface of the planet,. Iit's an incredible amount of energy that is leading to the global warming.
I'm Shane Huntington and my guest today is climate scientist, Dr Malte Meinshausen. We're talking about climate modelling and climate change mitigation here on Up Close, coming to you from the university of Melbourne Australia.
When we put your models into play and how they roll out over the next century and what they tell us about the climate, what assumptions do you have to make to run these models and what does that give us in terms of uncertainty in the outcomes?
The first assumption that we have to do in every modelling study where we want to project future climate is what are the emissions going to be? There, we have to assume scenarios. We have to assume either a high emission scenario or a low emission scenario. This is kind of okay. We play with it. At the end, the results will be conditional on the emission scenario that we assumed. There are other assumptions of course in every detail of the modelling and we try to make them as neutral as possible. By neutral, I mean we try to constrain the possible range that these parameters can have by using historical observations. The simplest example is we observed in the instrumental record how the earth system warmed over the last 150 years. So what we can do is put the CO2 emissions, put the methane emissions, put the nitrous oxide and the aerosol emissions into our models, try to model the past century and then we find out, by this historical constraining about which model parameter seems sensible, which seem not.
But this is separate from kind of the really big models. You don't constrain the really big models with historical observations. In the really big models, you rather put in physical mechanisms. You use lead lab results. You use observations of very small fine-grain processes. You don't constrain the really big models by the overall global warming observations. So there, of course, the modellers have to make judgments about how important it is for example that cloud entrainment process. Or can I somehow parameterise the sub-grid scale processes that I can't put into a model which has a resolution that is too coarse for some of the cloud processes?
The reason why we can't put it in is simply because we don't have the computational and resources. These models are something of the most resource hungry computational exercises that one can do. Frequently the super computers are used to run these climate models. But therey are obviously judgments that have to be made. But since 20 groups make these judgments, largely independently, we have a good feeling for which are the important parameters, which are the not so important parameters and then every modelling group of course tries to do its best. Then, by looking at the range of modelling results, we get a feeling for the overall uncertainty because every modelling group will decide slightly differently on these assumptions that one can take.
Do we currently have a solid enough understanding of the carbon cycle of the planet for these models? Or are there gaps that we haven't quite filled in yet?
Yes, we have a good solid understanding. But there are of course gaps and stuff where we want to know about, where we want to include more processes, et cetera. So for example, over nitrogen fertilisation. I mentioned at the beginning that plants grow because of additional CO2 in the atmosphere. But plants can as well grow because of additional nitrogen that is put on their feet. So understanding about how much of that nitrogen fertilisation actually drove plant growth over the last decades will help us to come to better projections for how is the carbon cycle reacting in the future. There are still gaps in regionally specific plant types et cetera. I mentioned nitrogen fertilisation in how the rainfall patterns change in different parts.
There are some uncertainties in some regions about whether there's more or less rainfall. That obviously is going to have a huge impact on the vegetation that is living there, for example, the Amazon regions. One or two models out of 20 project that the Amazon Basin will be largely turned into a Savannah over the next 100 years. This obviously would have a large impact on the CO2 concentrations in the atmosphere because there's a whole lot of carbon stored in the Amazon rainforest. There we have to investigate further what are the robust responses. This is currently just a matter of uncertainty. But it's not an uncertainty that would be squashing all the certainties that we have about the system, because we know it's going to get warmer. We just don't know precisely by how much.
What are the models telling us at the moment about the capacity of our oceans to continue to absorb CO2?
Currently, the oceans roughly absorb maybe a third of the CO2 emissions that we put in the atmosphere. The land, the terrestrial biospheres, so the plants, absorb, and other parts, so that roughly half of the anthropogenic emissions are currently sucked up by both land and oceans. The current misperception is that, well, if half of the CO2 that we put in the atmosphere is sucked up, why don't we just reduce our emissions by half and then all is fine, because the nature sucks it up. But the problem with that is it's a dynamic response. If we decrease our emissions as well, these things will decrease. It's a bit like communicating pipes in which you have water.
So we have three large pools of the carbon reservoirs which is ocean, land terrestrial biosphere and the atmosphere. If we put more in the atmosphere of course more goes into the other carbon pools. But if we want to reduce CO2 concentrations, and for example we are currently at around 390 ppm in the atmosphere and a very prominent goal is, over the long-term, to reduce that again to 350, in order to make that reduction we have to take our CO2 from the atmosphere. But then CO2 will come as well out of the oceans. So the oceans in the long-term don't necessarily have always to act as a sink. If we want to decrease CO2 concentrations, they are going to act as a source, because a lot of carbon is already in there.
There are uncertainties about how much the ocean, carbon is disappearing in the oceans. We just have to further investigate how quickly the oceans are kind of stopping as a carbon sink. But one thing is clear, they are not just whopping mopping up the carbon and the carbon disappears there. They are just, as I said, the communicating pipes. So if we really want to stop the climate from warming further, there's no way around basically zero carbon emissions.
Malte, just to finish up, do the economic arguments towards all these mitigation strategies work? Are we likely to see a shift in the sort of economies of the world towards a scenario where we'll be in a sustainable situation in the next sort of 20 to 50 years?
That's the $100 million question. There's some science that - look, for example, at Europe. There, climate mitigation is largely thought of as an innovation and technology driving agenda, because, for example in Germany, the Germans want to have the tools ready to deliver to the world, which are needed in a zero carbon economy. So huge investments are going in to developing these tools and technologies and et cetera. So it's kind of considered as a win strategy, both for jobs and for the climate. In other parts of the world, it's not yet considered a win strategy. But undoubtedly ahead of the curve is China. They're installing huge amounts of wind-power. They're installing huge amounts of solar PV. They have all the technology there for electrical vehicles that maybe drive us all around in the future.
So I think, if countries really see it as this is our chance to innovate in our economy, to have manufacturing jobs, to have high-paid jobs actually in order to have that clean energy revolution, then I think a shift can quickly happen. But I think this is not something that we can predict with any certainty. The course of the international community might go the one way or the other. In the end, you wonder whether, as humans, we always need kind of forceful wake-up and reminders by climate change that we actually feel before we start action. All the economic analysis suggests that, once we embark on the course towards a zero carbon economy, it's not a painful course. It actually has multiple side benefits of cleaner air, of energy security, et cetera, et cetera. So it's a win-win-win situation. But of course there are large interests in continuing the path that we are on, which is using fossil fuels to provide us with the energy services we ask for. Whether there's kind of a flip in the system and the large economies of this world consider it as a win-win strategy for them, only in the next decade can tell.
Dr Malte Meinshausen, senior researcher at the Potsdam Institute for Climate Impact Research, and honorary visiting researcher at the School of Earth Sciences at the University of Melbourne. Thank you for being our guest today on Up Close and talking about modelling, the impact of our mitigation efforts on climate change.
Thank you very much.
Relevant links, a full transcript and more info on this episode can be found at our website, at upclose.unimelb.edu.au. Up Close is a production of the University of Melbourne Australia. This episode was recorded on 5 April, 2012. Our producers for this episode were Kelvin Param and Eric van Bemmel. Audio engineering by Gavin Nebauer. Background research by Dyani Lewis. Up Close is created by Eric van Bemmel and Kelvin Param. I'm Shane Huntington, until next time, goodbye.
show transcript | print transcript | download pdf
© The University of Melbourne, 2012. All Rights Reserved.