First, what does the spread of simulations look like? The following figure plots the global mean temperature anomaly for 55 individual realizations of the 20th Century and their continuation for the 21st Century following the SRES A1B scenario. For our purposes this scenario is close enough to the actual forcings over recent years for it to be a valid approximation to the simulations up to the present and probable future. The equal weighted ensemble mean is plotted on top. This isn't quite what IPCC plots (since they average over single model ensembles before averaging across models) but in this case the difference is minor.

It should be clear from the above the plot that the long term trend (the global warming signal) is robust, but it is equally obvious that the short term behaviour of any individual realisation is not. This is the impact of the uncorrelated stochastic variability (weather!) in the models that is associated with interannual and interdecadal modes in the models - these can be associated with tropical Pacific variability or fluctuations in the ocean circulation for instance. Different models have different magnitudes of this variability that spans what can be inferred from the observations and in a more sophisticated analysis you would want to adjust for that. For this post however, it suffices to just use them 'as is'.

We can characterise the variability very easily by looking at the range of regressions (linear least squares) over various time segments and plotting the distribution. This figure shows the results for the period 2000 to 2007 and for 1995 to 2014 (inclusive) along with a Gaussian fit to the distributions. These two periods were chosen since they correspond with some previous analyses. The mean trend (and mode) in both cases is around 0.2ºC/decade (as has been widely discussed) and there is no significant difference between the trends over the two periods. There is of course a big difference in the standard deviation - which depends strongly on the length of the segment.

Over the short 8 year period, the regressions range from -0.23ºC/dec to 0.61ºC/dec. Note that this is over a period with no volcanoes, and so the variation is predominantly internal (some models have solar cycle variability included which will make a small difference). The model with the largest trend has a range of -0.21 to 0.61ºC/dec in 4 different realisations, confirming the role of internal variability. 9 simulations out of 55 have negative trends over the period.

Over the longer period, the distribution becomes tighter, and the range is reduced to -0.04 to 0.42ºC/dec. Note that even for a 20 year period, there is one realisation that has a negative trend. For that model, the 5 different realisations give a range of trends of -0.04 to 0.19ºC/dec.

Therefore:

• Claims that GCMs project monotonic rises in temperature with increasing greenhouse gases are not valid. Natural variability does not disappear because there is a long term trend. The ensemble mean is monotonically increasing in the absence of large volcanoes, but this is the forced component of climate change, not a single realisation or anything that could happen in the real world.
• Claims that a negative observed trend over the last 8 years would be inconsistent with the models cannot be supported. Similar claims that the IPCC projection of about 0.2ºC/dec over the next few decades would be falsified with such an observation are equally bogus.
• Over a twenty year period, you would be on stronger ground in arguing that a negative trend would be outside the 95% confidence limits of the expected trend (the one model run in the above ensemble suggests that would only happen ~2% of the time).

A related question that comes up is how often we should expect a global mean temperature record to be broken. This too is a function of the natural variability (the smaller it is, the sooner you expect a new record). We can examine the individual model runs to look at the distribution. There is one wrinkle here though which relates to the uncertainty in the observations. For instance, while the GISTEMP series has 2005 being slightly warmer than 1998, that is not the case in the HadCRU data. So what we are really interested in is the waiting time to the next unambiguous record i.e. a record that is at least 0.1ºC warmer than the previous one (so that it would be clear in all observational datasets. That is obviously going to take a longer time.

This figure shows the cumulative distribution of waiting times for new records in the models starting from 1990 and going to 2030. The curves should be read as the percentage of new records that you would see if you waited X years. The two curves are for a new record of any size (black) and for an unambiguous record (> 0.1ºC above the previous, red). The main result is that 95% of the time, a new record will be seen within 8 years, but that for an unambiguous record, you need to wait for 18 years to have a similar confidence. As I mentioned above, this result is dependent on the magnitude of natural variability which varies over the different models. Thus the real world expectation would not be exactly what is seen here, but this is probably reasonably indicative.

We can also look at how the Keenlyside et al results compare to the natural variability in the standard (un-initiallised) simulations. In their experiments, the decadal mean of the period 2001-2010 and 2006-2015 are cooler than 1995-2004 (using the closest approximation to their results with only annual data). In the IPCC runs, this only happens in one simulation, and then only for the first decadal mean, not the second. This implies that there may be more going on than just the tapping into the internal variability in their model. We can specifically look at the same model in the un-initiallised runs. There, the differences between first decadal means spans the range 0.09 to 0.19ºC - significantly above zero. For the second period, the range is 0.16 to 0.32 ºC. One could speculate that there is actually a cooling that is implicit to their initialisation process itself. It would be instructive to try some similar 'perfect model' experiments (where you try and replicate another model run rather than the real world) to investigate this further though.

Finally, I would just like to emphasize that for many of these examples, claims have circulated about the spectrum of the IPCC model responses without anyone actually looking at what those responses are. Given that the archive of these models exists and is publicly available, there is no longer any excuse for this. Therefore, if you want to make a claim about the IPCC model results, download them first!

Much thanks to Sonya Miller for producing these means from the IPCC archive.

Share This

Global cooling appears to be the “flavour of the month”. First, a rather misguided media discussion erupted on whether global warming had stopped, based on the observed temperatures of the past 8 years or so (see our post). Now, an entirely new discussion is capturing the imagination, based on a group of scientists from Germany predicting a pause in global warming last week in the journal Nature (Keenlyside et al. 2008).
Specifically, they make two forecasts for global temperature, as discussed in the last paragraphs of their paper and shown in their Figure 4 (see below). The first forecast concerns the time interval 2000-2010, while the second concerns the interval 2005-2015 (*). For these two 10-year averages, the authors make the following prediction:

“… the initialised prediction indicates a slight cooling relative to 1994-2004 conditions”

Their graph shows this: temperatures in the two forecast intervals (green points shown at 2005 and 2010) are almost the same and are both lower than observed in 1994-2004 (the end of the red line in their graph).

Figure 4 from Keenlyside et al '08

The authors also make regional predictions, but naturally it was this global prediction that captivated most newspaper stories around the world (e.g. BBC News, Reuters, Bloomberg and so on), because of its seeming contradiction with global warming. The authors emphasise this aspect in their own media release, which was titled: Will Global Warming Take a Short Break?

That this cooling would just be a temporary blip and would change nothing about global warming goes without saying and has been amply discussed elsewhere (e.g. here). But another question has been rarely discussed: will this forecast turn out to be correct?

We think not – and we are prepared to bet serious money on this. We have double-checked with the authors: they say they really mean this as a serious forecast, not just as a methodological experiment. If the authors of the paper really believe that their forecast has a greater than 50% chance of being correct, then they should accept our offer of a bet; it should be easy money for them. If they do not accept our bet, then we must question how much faith they really have in their own forecast.

The bet we propose is very simple and concerns the specific global prediction in their Nature article. If the average temperature 2000-2010 (their first forecast) really turns out to be lower or equal to the average temperature 1994-2004 (*), we will pay them € 2500. If it turns out to be warmer, they pay us € 2500. This bet will be decided by the end of 2010. We offer the same for their second forecast: If 2005-2015 (*) turns out to be colder or equal compared to 1994-2004 (*), we will pay them € 2500 – if it turns out to be warmer, they pay us the same. The basis for the temperature comparison will be the HadCRUT3 global mean surface temperature data set used by the authors in their paper.

To be fair, the bet needs an escape clause in case a big volcano erupts or a big meteorite hits the Earth and causes cooling below the 1994-2004 level. In this eventuality, the forecast of Keenlyside et al. could not be verified any more, and the bet is off.

The bet would also need a neutral arbiter – we propose, for example, the director of the Hadley Centre, home of the data used by Keenlyside et al., or a committee of neutral colleagues. This neutral arbiter would also decide whether a volcano or meteorite impact event is large enough as to make the bet obsolete.

We will discuss the scientific reasons for our assessment here another time – first we want to hear from Keenlyside et al. whether they accept our bet. Our friendly challenge is out – we hope they will accept it in good sportsmanship.

(*) We adopt here the definition of the 10-year intervals as in their paper, which is from 1 November of the first year to 31 October of the last year. I.e.: 2000-2010 means means 1 November 2000 until 31 October 2010.

(This article is published simultaneously in German on the KlimaLounge weblog.)

_______________________
Update: Andy Revkin has weighed in at "dot earth".

Update 5/11/08: so has Anna Barnett at Nature's 'climate feedback' blog

Share This

The uncertainties in the impacts of rising greenhouse gases on multiple systems are significant: the potential impact on ENSO or the overturning circulation in the North Atlantic, probable feedbacks on atmospheric composition (CO₂, CH₄, N₂O, aerosols), the predictability of decadal climate change, global climate sensitivity itself, and perhaps most importantly, what will happen to ice sheets and regional rainfall in a warming climate.

The reason why paleo-climate information may be key in these cases is because all of these climate components have changed in the past. If we can understand why and how those changes occurred then, that might inform our projections of changes in the future. Unfortunately, the simplest use of the record - just going back to a point that had similar conditions to what we expect for the future - doesn't work very well because there are no good analogs for the perturbations we are making. The world has never before seen such a rapid rise in greenhouse gases with the present-day configuration of the continents and with large amounts of polar ice. So more sophisticated approaches must be developed and this meeting was devoted to examining them.

The first point that can be made is a simple one. If something happened in the past, that means it's possible! Thus evidence for past climate changes in ENSO, ice sheets and the carbon cycle (for instance) demonstrate quite clearly that these systems are indeed sensitive to external changes. Therefore, assuming that they can't change in the future would be foolish. This is basic, but not really useful in a practical sense.

All future projections rely on models of some sort. Dominant in the climate issue are the large scale ocean-atmosphere GCMs that were discussed extensively in the latest IPCC report, but other kinds of simpler or more specialised or more conceptual models can also be used. The reason those other models are still useful is that the GCMs are not complete. That is, they do not contain all the possible interactions that we know from the paleo record and modern observations can occur. This is a second point - interactions seen in the record, say between carbon dioxide levels or dust amounts and Milankovitch forcing imply that there are mechanisms that connect them. Those mechanisms may be only imperfectly known, but the paleo-record does highlight the need to quantify these mechanisms for models to be more complete.

The third point, and possibly the most important, is that the paleo-record is useful for model evaluation. All episodes in climate history (in principle) should allow us to quantify how good the models are and how appropriate the our hypotheses for climate change in the past. It's vital to note the connection though - models embody much data and assumptions about how climate works, but for their climate to change you need a hypothesis - like a change in the Earth's orbit, or volcanic activity, or solar changes etc. Comparing model simulations to observational data is then a test of the two factors together. Even if the hypothesis is that a change is due to intrinsic variability, a simulation of a model to look for the magnitude of intrinsic changes (possibly due to multiple steady states or similar) is still a test both of the model and the hypothesis. If the test fails, it shows that one or other elements (or both) must be lacking or that the data may be incomplete or mis-interpreted. If it passes, then we a have a self-consistent explanation of the observed change that may, however, not be unique (but it's a good start!).

But what is the relevance of these tests? What can a successful model of the impacts of a change in the North Atlantic overturning circulation or a shift in the Earth's orbit really do for future projections? This is where most of the attention is being directed. The key unknown is whether the skill of a model on a paleo-climate question is correlated to the magnitude of change in a scenario. If there is no correlation - i.e. the projections of the models that do well on the paleo-climate test span the same range as the models that did badly, then nothing much has been gained. If however, one could show that the models that did best, for instance at mid-Holocene rainfall changes, systematically gave a different projection, for instance, of greater changes in the Indian Monsoon under increasing GHGs, then we would have reason to weight the different model projections to come up with a revised assessment. Similarly, if an ice sheet model can't match the rapid melt seen during the deglaciation, then its credibility in projecting future melt rates would/should be lessened.

Unfortunately apart from a few coordinated experiments for the last glacial period and the mid-Holocene (i.e. PMIP) with models that don't necessarily overlap with those in the AR4 archive, this database of model results and tests just doesn't exist. Of course, individual models have looked at many various paleo-climate events ranging from the Little Ice Age to the Cretaceous, but this serves mainly as an advance scouting party to determine the lay of the land rather than a full road map. Thus we are faced with two problems - we do not yet know which paleo-climate events are likely to be most useful (though everyone has their ideas), and we do not have the databases that allow you to match the paleo simulations with the future projections.

In looking at the paleo record for useful model tests, there are two classes of problems: what happened at a specific time, or what the response is to a specific forcing or event. The first requires a full description of the different forcings at one time, the second a collection of data over many time periods associated with one forcing. An example of the first approach would be the last glacial maximum where the changes in orbit, greenhouse gases, dust, ice sheets and vegetation (at least) all need to be included. The second class is typified by looking for the response to volcanoes by lumping together all the years after big eruptions. Similar approaches could be developed in the first class for the mid-Pliocene, the 8.2 kyr event, the Eemian (last inter-glacial), early Holocene, the deglaciation, the early Eocene, the PETM, the Little Ice Age etc. and for the second class, orbital forcing, solar forcing, Dansgaard-Oeschger events, Heinrich events etc.

But there is still one element lacking. For most of these cases, our knowledge of changes at these times is fragmentary, spread over dozens to hundreds of papers and subject to multiple interpretations. In short, it's a mess. The missing element is the work required to pull all of that together and produce a synthesis that can be easily compared to the models. That this synthesis is only rarely done underlines the difficulties involved. To be sure there are good examples - CLIMAP (and its recent update, MARGO) for the LGM ocean temperatures, the vegetation and precipitation databases for the mid-Holocene at PMIP, the spatially resolved temperature patterns over the last few hundred years from multiple proxies, etc. Each of these have been used very successfully in model-data comparisons and have been hugely influential inside and outside the paleo-community.

It may seem odd that this kind of study is not undertaken more often, but there are reasons. Most fundamentally it is because the tools and techniques required for doing good synthesis work are not the same as those for making measurements or for developing models. It could in fact be described as a new kind of science (though in essence it is not new at all) requiring, perhaps, a new kind of scientist. One who is at ease in dealing with the disparate sources of paleo-data and aware of the problems, and yet conscious of what is needed (and why) by modellers. Or additionally modellers who understand what the proxy data depends on and who can build that into the models themselves making for more direct model-data comparisons.

Should the paleo-community therefore increase the emphasis on synthesis and allocate more funds and positions accordingly? This is often a contentious issue since whenever people discuss the need for work to be done to integrate existing information, some will question whether the primacy of new data gathering is being threatened. This meeting was no exception. However, I am convinced that this debate isn't the zero sum game implied by the argument. On the contrary, synthesising the information from a highly technical field and making it useful for others outside is a fundamental part of increasing respect for the field as a whole and actually increases the size of the pot available in the long term. Yet the lack of appropriately skilled people who can gain the respect of the data gatherers and deliver the 'value added' products to the modellers remains a serious obstacle.

Despite the problems and the undoubted challenges in bringing paleo-data/model comparisons up to a new level, it was heartening to see these issues tackled head on. The desire to turn throwaway lines in grant applications into real science was actually quite inspiring - so much so that I should probably stop writing blog posts and get on with it.

The above condensed version of the meeting is heavily influenced by conversations and talks there, particularly with Peter Huybers, Paul Valdes, Eric Wolff and Sandy Harrison among others.

Share This

Ed Lorenz had a reputation of being shy and quiet, and this was indeed the impression he gave on first meeting. Indeed raypierre was interviewed by Ed at MIT in 1979 for his first faculty job — and remembers having to ask most of the questions as well as answer them. But he also remembers a lot of timely support from Ed that helped smooth over the somewhat rocky transition from basic turbulence theory to atmospheric science. The longer you were around Ed, the more you came to appreciate his warmth and sense of humor. He was an avid hiker, and many in the community (our own Mike Mann included) have recollections of time on the trail with him around the hills of Boulder and elsewhere.

Lorenz launched the modern era of the study of chaotic systems, which has profound implications both within and beyond atmospheric science. We'll say more about that in a bit, but the monumental work on chaos should not leave Lorenz's other contributions to atmospheric science completely in its shadow. For example, in a 1956 MIT technical report, Ed introduced the notion of "empirical orthogonal functions" to atmospheric science, and this technique now plays a central role in diagnostic studies of the atmosphere-ocean system. He also pioneered the study of angular momentum transport in the atmosphere, and of atmospheric energetics. Among other things, he introduced the important notion of "available potential energy," which quantifies the fact that not all of the potential energy can be tapped by allowable rearrangements of the atmosphere. Later, he pioneered the concept of resonant triad instability of atmospheric waves, an idea that has repercussions for the sources of atmospheric low frequency variability. As if that weren't enough Ed also introduced the concept of the "slow manifold" — a special subset of solutions to a nonlinear system which evolve more slowly than most solutions. The atmospheric equations support a lot of very quickly changing solutions, like sound waves and gravity waves, but on the whole what we think of as "weather" or "climate" involves more ponderous motions evolving on time scales of days to years. Ed's work on this subject launched the study of how such slowly evolving solutions can exist, and how to initialize a numerical model so as to minimize the generation of the fast transients. This is now part and parcel of the whole apparatus of data assimilation and numerical weather forecasting.

Ed was not a user of general circulation models. His essential approach was to crystallize profound phenomena into very small sets of equations for how a handful of variables change with time. He left behind him a dozen or so such models, each of which would repay many lifetimes of study. He was indeed a master of "seeing the world in a grain of sand." You can read about some of these models in the talk raypierre gave at the 1987 Lorenz 'retirement' symposium — not that this slowed him down!

Now let's take a closer look at that butterfly effect. Despite the fact that there are no butterflies or tornadoes in climate models, Lorenz's discoveries and their implications played a central role in climate modelling efforts and in the most recent IPCC report.

The notion of the butterfly effect itself was drawn from a simple but astute observation of the way the solutions of certain nonlinear equations behave when they are solved using a computer. Start with a greatly simplified representation of thermal convection, first formulated by Barry Saltzman using a technique called "low order modelling." If you run a simulation using these equations and then try and replicate it using starting values that only differ in the last decimal place, you will find that the simulations quickly diverge from one another - and by quickly, it means that the differences grow exponentially fast. Lorenz found this phenomenon by accident, but quickly recognised the profound implications. If the real weather system displayed the same behaviour, it meant that since however well one knew the initial conditions of the atmosphere, there would always be some uncertainty, that uncertainty would be quickly magnified, rendering weather forecasts useless after a few exponential doubling times. The practical implication is that - even if you had a perfect model - for every halving of the error in the initial conditions you only get one extra time period of useful forecast. Given this time period is only a few hours in many cases, the practicality of true weather forecasts for periods longer than two weeks or so, is vanishingly small.

The mathematically inclined reader who takes a look at Ed's early papers on what is now called the "Lorenz Attractor" will be astonished at the depth and modernity of his ideas about chaos. This line of work was no mere remark on a numerical exercise. Lorenz actually teased out the geometry of chaos — the many-leaved structure of the attractor — realizing that it was no simple geometric entity like a sphere or a folded sheet of paper. It was indeed "strange" in a sense which he made geometrically precise. This is why the work had such lasting impact on the area of pure mathematics known as dynamical systems theory. He went beyond that to develop or apply many fundamental concepts in chaotic systems, quantitatively formulating various measures of predictability and connecting the Lyapunov exponent — a certain precise mathematical characterization of chaos — with the structure of strange attractors. But that's for the mathematicians. What makes Lorenz's work interesting to the entity on the Clapham omnibus is the notion of sensitive dependence on initial conditions. Some have even seen in this deterministic chaos the resolution to the problem of free will!

The idea that small causes can have large and disproportionate effects is not at all new of course. The poem: "For the want of a nail, the battle was lost" (medieval in origin) encapsulates that well, and popular culture is full of such examples, "It's a wonderful life" (1946) and Ray Bradbury's "A Sound of Thunder" (1952) for instance. Curiously, Bradbury's story also involves a butterfly, and since it predates Lorenz's coining of the phrase by a decade or so, people have speculated that there was a connection to Lorenz's choice of metaphor (he started off with a seagull in his 1963 Trans. N.Y. Ac. Sci. paper). But that doesn't appear to be the case (see here for a history). It's worth adding that all of Lorenz's papers were exceptional in their clarity and are well worth tracking down as an example of science writing at its best.

However, the difference between the long-standing popular conception and Lorenz's work is that he demonstrated this effect in a completely deterministic system with no random component. That is, even in a perfect model situation, useful predictability can be strongly limited. Strictly speaking, Poincaré first described this effect in the classic three-body problem in the early 1900s, but it was only with the onset of electronic computers, as used by Lorenz, that this became more widely recognised. To throw in another popular culture reference, Tom Stoppard's Arcadia has a character, Septimus, who drives himself mad trying to calculate chaotic solutions to the logistic map by hand.

So what does this have to do with the IPCC?

Even though the model used by Lorenz was very simple (just three variables and three equations), the same sensitivity to initial conditions is seen in all weather and climate models and is a ubiquitous phenomenon in many complex non-linear flows. It is therefore usually assumed that the real atmosphere also has this property. However, as Lorenz himself acknowledged in 1972, this is not directly provable (and indeed, at least one meteorologist doesn't think it does even though most everyone else does). Its existence in climate models is nonetheless easily demonstratable.

But how can climate be predictable if weather is chaotic? The trick lies in the statistics. In those same models that demonstrate the extreme sensitivity to initial conditions, it turns out that the long term means and other moments are stable. This is equivalent to the 'butterfly' pattern seen in the figure above being statistically independent of how you started the calculation. The lobes and their relative position don't change if you run the model long enough. Climate change then is equivalent seeing how the structure changes, while not being too concerned about the specific trajectory you are on.

Another way of saying it is that for the climate problem, the weather (or the individual trajectory) is the noise. If you are trying to find the common signal that is a signature of a particular forcing then averaging over a number of simulations with different weather works rather well. (There is a long standing quote in science - "one person's noise is another person's signal" which is certainly apropos here. Climate modellers don't average over ensemble members because they think that weather isn't important, they do it because it gives robust estimates of the signal they are usually looking for.)

The ensemble approach, and indeed the multi-model ensemble approach, used in IPCC then derives directly from Lorenz's insights into his serendipitous numerical problem.

Share This

When registering at the conference, we got a packet called 'Planet Earth; Directions for Use'. As far as I know, this is a new feature apparently offered by the EGU. The box says 'EGU cares…' and it contains 4 sheets: Biosphere, Hydrosphere, Litho- and Pedosphere, and the Atmosphere. The Biosphere sheet is concerned about the biodiversity, the hydropshere discusses water shortage and loss of marshland issues, the litho- & pedosphere mentions the fact that fossil fuels are finite and soil erosion, and the atmosphere discusses AGW.

Of course this is a gimmick, and perhaps it is even aimed at the wrong target group. These issues are more or less taken as given by the majority of the EGU community by now, it seems. It's more pressing, however, that the rest of the world population understand the problems.

Actually, it was refreshing arriving at the harbor of sanity in the EGU meeting, after the the insane climate-change debate circus in Norway at the moment – lead by a number of academics who start to look more and more like crack pots, and a right-wing populist political party taking after Inhofe.

What were the highlights? It's impossible to cover everything, and I only sampled some talks which are most relevant to my own work. But one important talk was about setting the global climate models' initial state (ocean) to describe the current climate. The intention was to capture subsequent slow natural variations (decadal variation) associated with the thermohaline circulation (THC, not to be confused with other meanings according to a recent article in Eos: VOLUME 89 NUMBER 11 11 March 2008).

Apparently, if the global climate model is initialized with the current state, then the global mean temperature may not rise much over the next decade or so, and then suddenly bounce up and converge with the current scenarios. But some critiques argue that forcing models to have a prescribed state, will trow them off balance, and that the model will try to recover its balance for the next few model years.

Another presentation discussed the possibility for slow climatic variations to be predicted 10 years in advance (potential decadal predictability), and concluded that there is a potential for over the North Atlantic regions – associated with the THC. But an increasing AGW may destroy this possibility, as the predictability reduces when the world gets warmer.

There are some interesting and newish data coming of age: radio occultation. This involves measuring the bending of GPS signal through the atmospheric limb, as measured between different satellites. The atmospheric temperature and humidity affect radio signals refractive index. But there is only short data series (~10 years), but so far the temperature trends are consistent with the models for similar intervals. But these are independent to the satellite MSU data, and do not suffer in the same way from differences between satellite instruments, etc.

My personal favourite this time was a talk on 'recurrence based transition analysis'. The presentation was subtly slick and so nicely executed that it can only be done on a Mac. The talk was very clear, no excess number of words, and to the point.

There were many good talks, but some common mistakes. Well, at least I'm a bit slow when having attended a few dozen presentations, so presenters speaking fast or too crowded Powerpoint slides risk losing me. There is supposed to be a golden rule called 'Seven by seven': no more than seven bullet points, consisting of no more than 7 words! And one should speak slowly and repeat the important points.

So what did people talk about? What was 'The buzz-word'? There was no obvious paradigm shift, and I didn't catch one single theme that was the vogue of the day, but there were some issues that kept popping up: decadal predictability, cryosphere and the polar regions, model ensembles and probabilities, regional modelling and extremes.

What I find striking with such monster conferences is the sheer scale of diversity in terms of geo-subjects that people study. There were rows upon rows of posters in several large rooms, in addition to the talks.

There seems to be a great secret of Powerpointerism: the programmers at Microsoft designed a right mouse click option to show a presentation straight away without showing the subsequent page. There is a systematic neglect of this functionality, so that the Microsoft guys must have implemented this one in vain.

The best quote that i heard on this conference was: 'A trend is a trend is a trend …'. In other words, there is no definite definition of a trend, at least not to statisticians who like to use more complex lagged correlation models. Something to bear in mind for those who fit linear lines to data points - in order to study trends - and then use the goodness of fit to say whether the trend is 'significant' or not.

Another bad habit is showing latitudinal profiles of zonal mean values as if the points at high latitudes are equal to those near the equator. What they really compare are oranges and apples, as the low latitude zonal means involve higher degrees of freedom than at high latitudes. I have explained this in more technical details in a GRL article from 2005, but such graphs can be found even the latest IPCC report. Though it may be a minor point, it makes models look worse than they actually are, as part of the spread towards the poles can be attributed increasing statistical fluctuations when the number of degrees becomes less. Thus, the results would stand stronger taking this aspect into account.

I was pleased to hear that some colleagues in the German weather service sometimes use RealClimate for inspiration to their monthly seminars. What was more unexpected, however, was being met with a slide showing 'Natural Trendy?' on a session that I had been invited to give a talk.

Furthermore, it turned out that LiCohn and Koutsoyiannis, one of them the author of the very paper that I had criticized, sat down next to me. We nevertheless had a very civilized and friendly chat, deciding to disagree on the matter of natural trends.

But Dr. Koutsoyiannis commended us for being respectful in our reply to his comments. I think this is a very important issue – we have to be respectful, sincere, and show courtesy in our criticism, even when we argue why we think that a paper has flaws. This brings us back to the discussion on blogs and journals.

I think that we have built up a reputation only because we deliver relevant quality analysis. We are very much aware that we some day may be mistaken, so it's important to be humble and check our drafts amongst ourselves. But when a question was asked about the importance of blogs like RealClimate in the session, the answer was that they were good entertainment.

Vienna is a pleasant city with many pretty sights. The only annoying thing is that one often has to breathe in local pollution from the next table when dining in restaurants. Austria is one of the few western European countries that has not introduced a smoking ban in restaurants it seems.

Share This

The net loss in volume and hence sea level contribution of the Greenland Ice Sheet (GIS) has doubled in recent years from 90 to 220 cubic kilometers/year has been noted recently (Rignot and Kanagaratnam, 2007). The main cause of this increase is the acceleration of several large outlet glaciers. There has also been an alarming increase in the number of photographs of meltwater draining into a moulin somewhere on the GIS, often near Swiss Camp (35 km inland from the calving front). The story goes—warmer temperatures, more surface melting, more meltwater draining through moulins to glacier base, lubricating glacier bed, reducing friction, increasing velocity, and finally raising sea level. Examining this issue two years RealClimate suggested this was likely the correct story. A number of recent results suggest that we need to take another look at this story.

The Acceleration:

Jakobshavn Glacier, West Greenland, retreated 30 km from 1850-1964, followed by a stationary front for 35 years. Jakobshavn has the highest mass flux of any glacier draining an estimated 6% of the GIS. The glacier terminus region also had a consistent velocity of 19 meters/day (maximum of 26 m in glacier center), from season to season and year to year, the glacier seemed to be in balance, as I noted in a 1989paper. This is the fastest glacier in the world, no steroids needed. After 1997 it began to accelerate and thin rapidly, reaching an average velocity of 34 m/day in the terminus region. The glacier thinned at a rate of up to 15 m/year and retreated 5 km in six years. Jakobshavn has since slowed to near its pre-1997 speed, the terminus retreat is still occurring, but likewise is.

Helheim Glacier, East Greenland had a stable terminus from the 1970’s-2000. In 2001-2005 the glacier retreated 7 km and accelerated from 20 m/day to 33 m/day, while thinning up to 130 meters in the terminus region. Kangerdlugssuaq Glacier, East Greenland had a stable terminus history from 1960-2002. The glacier velocity was 13 m/day in the 1990’s. In 2004-2005 it accelerated to 36 m/day and thinned by up to 100 m in the lower reach of the glacier. Helheim and Kangerdlugssuaq combined drain 8 % of GIS. Hence, they are more than canaries in the coal mine. In 2006, the velocity of Helheim and Kangerdlugssuaq decreased to near the 2000 level, the terminus of Helheim advanced a bit (Howat et al., 2007).

The first mechanism for explaining the change in velocity is the "Zwally effect", which relies on meltwater reaching the glacier base and reducing the friction through a higher basal water pressure. A moulin is the conduit for the additional meltwater to reach the glacier base. This idea, proposed by Jay Zwally, was observed to be the cause of a brief seasonal acceleration of up to 20 % on the Jakobshavns Glacier in 1998 and 1999 at Swiss Camp (Zwally et al., 2002). The acceleration lasted two-three months and was less than 10% in 1996 and 1997 for example. They offered a conclusion that the “coupling between surface melting and ice-sheet flow provides a mechanism for rapid, large-scale, dynamic responses of ice sheets to climate warming”. The acceleration of the three glaciers had not occurred at the time of this study and they were not concluding or implying that the meltwater increase was the cause of the aforementioned acceleration. However, many others have made this assertion and are investigating (Stearns and Hamilton, 2007). Examination of recent rapid supra-glacial lake drainage documented short term velocity changes due to such events, but they had little significance to the annual flow of the large glaciers outlet glaciers (Das et.al, 2008).

The second mechanism is a "Jakobshavn effect", coined by Terry Hughes, (1986), where a force small imbalance of forces caused by some perturbation can cause a substantial non-linear response. In this case an imbalance of forces at the calving front propagates up-glacier. Thinning causes the glacier to be more buoyant, even becoming afloat at the calving front, and is responsive to tidal changes. The reduced friction due to greater buoyancy allows for an increase in velocity. This is akin to letting off the emergency brake a bit. The reduced resistive force at the calving front is then propagated up glacier via longitudinal extension in what R. Thomas calls a backforce reduction (Thomas, 2003 and 2004). For ice streaming sections of large outlet glaciers (in Antarctica as well) there is always water at the base of the glacier that helps lubricate the flow. This water is, however, generally from basal processes, not surface melting.

If the Zwally effect is the key than since meltwater is a seasonal input, velocity would have a seasonal signal. If the Jakobshavn effect is the key the velocity will propagate up-glacier, the terminus velocity will be impacted by tides, and there will be no seasonal cycle.

On Jakobshavn the acceleration began at the calving front and spread up-glacier 20 km in 1997 and up to 55 km inland by 2003 (Joughin et al., 2004). On Helheim the thinning and velocity propagated up-glacier from the calving front. Each of the glaciers fronts did respond to tidal variations indicating they had become afloat, detached from their bed (Hamilton et al, 2006). This had been the case at Jakobshavn for the last 50 years, but not for Helheim or Kangerdlussuaq. In each case the major outlet glaciers accelerated by at least 50%, much larger than the impact noted due to summer meltwater increase. On Jakobshavn the acceleration was not restricted to the summer, persisting through the winter when surface meltwater is absent.

As a result of the above Luckman et al. ( 2006) concluded:

“The most plausible sequence of events is that the thinning eventually reached a threshold, ungrounded the glacier tongues and subsequently allowed acceleration, retreat and further thinning. It is reasonable to believe that the 1998 Jakobshavn speed-up, also following a long period of stability, was triggered by the same processes of thinning but occurred earlier and after a shorter period of thinning because the tongue was already afloat.”

Examination of the acceleration of other glaciers such as the Petermann Glacier indicate a much smaller acceleration than that observed on three glaciers we have focused, and indeed it is in the summer and of a magnitude that the Zwally effect could explain (Rignot, 2005). Other large outlet glaciers such as the Rinks and Daugaard-Jensen have been stable since 1960 (Stearns et al, 2005). Many other lesser outlet glaciers have accelerated substantially.

That each of the three glaciers has a reduced velocity in 2006 and 2007 despite some exceptional melt conditions in 2007 further suggests that meltwater is not the dominant driver of the acceleration of the main outlet glaciers. Temporarily, there appears to be a force imbalance at the glacier fronts. This will reduce the annual contribution to rising sea level from glacier dynamic changes. The bad news is that the degree of acceleration that can occur via the Jakobshavn effect is greater in these cases than that from the Zwally effect. The Zwally effect is nonetheless real and also implies a direct sea level impact of greater melt.

The Jakobshavn is of particular importance as it has a bed below sea level for at least 80 km inland from the terminus. In this reach there are no significant pinning points, or abrupt changes in slope or width (Clarke and Echelmeyer, 1996) that would help stabilize the glacier during retreat. It is the only outlet glacier of GIS to lack these, and can then (via backforce reductions) tap into the heart of GIS. We know that surface melting is a slow process for raising sea level. but as Greenland’s major outlet glaciers have recently shown, rapid acceleration can quickly deliver large volume of ice to the ocean. The pace of change is not glacial.

Clarke, T.S. & Echelmeyer, K. 1996: Seismic-reflection evidence for a deep subglacial trough beneath Jakobshavns Isbræ, West Greenland. Journal of Glaciology 42(141), 219–232.

Thomas RH (2004), Force-perturbation analysis of recent thinning and acceleration of Jakobshavn Isbrae, Greenland, Journal of Glaciology 50 (168): 57-66.

Share This

This week, the story has taken a curious new twist. In a letter published in Nature Geoscience, Pielke presents such a comparison for a longer period, 1990-2007 (see Figure). Lesson 1 learned - 17 years is sufficient. In fact, the very first figure of last year's IPCC report presents almost the same comparison (see second Figure).

Pielke's comparison of temperature scenarios of the four IPCC reports with data

There is a crucial difference, though, and this brings us to Lesson 2. The IPCC has always published ranges of future scenarios, rather than a single one, to cover uncertainties both in future climate forcing and in climate response. This is reflected in the IPCC graph below, and likewise in the earlier comparison by Rahmstorf et al. 2007 in Science.

IPCC Figure 1.1 - comparison of temperature scenarios of three IPCC reports with data

Any meaningful validation of a model with data must account for this stated uncertainty. If a theoretical model predicts that the acceleration of gravity in a given location should be 9.84 +- 0.05 m/s2, then the observed value of g = 9.81 m/s2 would support this model. However, a model predicting g = 9.84+-0.01 would be falsified by the observation. The difference is all in the stated uncertainty. A model predicting g = 9.84, without any stated uncertainty, could neither be supported nor falsified by the observation, and the comparison would not be meaningful.

Pielke compares single scenarios of IPCC, without mentioning the uncertainty range. He describes the scenarios he selected as IPCC's "best estimate for the realised emissions scenario". However, even given a particular emission scenario, IPCC has always allowed for a wide uncertainty range. Likewise for sea level (not shown here), Pielke just shows a single line for each scenario, as if there wasn't a large uncertainty in sea level projections. Over the short time scales considered, the model uncertainty is larger than the uncertainty coming from the choice of emission scenario; for sea level it completely dominates the uncertainty (see e.g. the graphs in our Science paper). A comparison just with the "best estimate" without uncertainty range is not useful for "forecast verification", the stated goal of Pielke's letter. This is Lesson 2.

In addition, it is unclear what Pielke means by "realised emissions scenario" for the first IPCC report, which included only greenhouse gases and not aerosols in the forcing. Is such a "greenhouse gas only" scenario one that has been "realised" in the real world, and thus can be compared to data? A scenario only illustrates the climatic effect of the specified forcing - this is why it is called a scenario, not a forecast. To be sure, the first IPCC report did talk about "prediction" - in many respects the first report was not nearly as sophisticated as the more recent ones, including in its terminology. But this is no excuse for Pielke, almost twenty years down the track, to talk about "forecast" and "prediction" when he is referring to scenarios. A scenario tells us something like: "emitting this much CO2 would cause that much warming by 2050″. If in the 2040s the Earth gets hit by a meteorite shower and dramatically cools, or if humanity has installed mirrors in space to prevent the warming, then the above scenario was not wrong (the calculations may have been perfectly accurate). It has merely become obsolete, and it cannot be verified or falsified by observed data, because the observed data have become dominated by other effects not included in the scenario. In the same way, a "greenhouse gas only" scenario cannot be verified by observed data, because the real climate system has evolved under both greenhouse gas and aerosol forcing.

Pielke concludes: "Once published, projections should not be forgotten but should be rigorously compared with evolving observations." We fully agree with that, and IPCC last year presented a more convincing (though not perfect) comparison than Pielke.

To sum up the three main points of this post:

1. IPCC already showed a very similar comparison as Pielke does, but including uncertainty ranges.

2. If a model-data comparison is done, it has to account for the uncertainty ranges - both in the data (that was Lesson 1 re noisy data) and in the model (that's Lesson 2).

3. One should not mix up a scenario with a forecast - I cannot easily compare a scenario for the effects of greenhouse gases alone with observed data, because I cannot easily isolate the effect of the greenhouse gases in these data, given that other forcings are also at play in the real world.

Share This

We need to start with some definitions. Sensitivity is defined as the global mean surface temperature anomaly response to a doubling of CO₂ with other boundary conditions staying the same. However, depending on what the boundary conditions include, you can get very different numbers. The standard definition (sometimes called the Charney sensitivity), assumes the land surface, ice sheets and atmospheric composition (chemistry and aerosols) stay the same. Hansen's long term sensitivity (which might be better described as the Earth System sensitivity) allows all of these to vary and feed back on the temperature response. Indeed, one can imagine a whole range of different sensitivities that could be clearly defined by successively including additional feedbacks. The reason why the Earth System sensitivity might be more appropriate is because that determines the eventual consequences of any particular CO₂ stabilization scenario.

Traditionally, the decision to include or exclude a feedback from consideration has been based on the relevant timescales and complexity. The faster a feedback is, the more usual it is to include. Thus, changes in clouds (~hours) or in water vapour (~10 days) are undoubtedly fast and get included as feedbacks in all definitions of the sensitivity. But changes in vegetation (decades to centuries) or in ice sheets (decades(?) to centuries to millennia) are slower and are usually left out. But there are other fast feedbacks that don't get included in the standard definition for complexity reasons - such as the change in ozone or aerosols (dust and sulphates for instance) which are also affected by patterns of rainfall, water vapour, temperature, soli moisture, transport and clouds (etc.).

Not coincidentally, the Charney sensitivity corresponds exactly to the sensitivity one gets with a standard atmospheric GCM with a simple mixed-layer ocean, while the Earth System sensitivity would correspond to the response in a (as yet non-existent) model that included interactive components for the cryosphere, biosphere, ocean, atmospheric chemistry and aerosols. Intermediate sensitivities could however be assessed using the Earth System models that we do have.

In principal, many of these sensitivities can be deduced from paleo-climate records. What is required is a good enough estimate of the global temperature change and measures of the various forcings. However, there are a few twists in the tale. Firstly, getting 'good enough' estimates for global temperatures changes is hard - this has been done well for the last century or so, reasonably for a few centuries earlier, and potentially well enough for the really big changes associated with the glacial-interglacial cycle. While sufficient accuracy in the last few centuries is a couple of tenths of a degree, this is unobtainable for the last glacial maximum or the Pliocene (3 million years ago). However, since the signal is much larger in the earlier periods (many degrees), the signal to noise ratio is similar.

Secondly, although many forcings can be derived from paleo-records (long-lived greenhouse gases from bubbles in the ice cores most notably), many cannot. The distribution of sulphate aerosols even today is somewhat uncertain, and at the last glacial maximum, almost completely unconstrained. This is due in large part to the heterogenity of their distribution and there are similar problems for dust and vegetation. In some sense, it is the availability of suitable forcing records that suggests what kind of sensitivity one can define from the record. A more subtle point is that the 'efficacy' of different forcings might vary, especially ones that have very different regional signatures, making it more difficult to add up different terms that might be important at any one time.

Lastly, and by no means leastly, Earth System sensitivity is not stable over geologic time. How much it might vary is very difficult to tell, but for instance, it is clear that from the Pliocene to the Quaternary (the last ~2,5 million years of ice age cycles), the climate has become more sensitive to orbital forcing. It is therefore conceivable (but not proven) that any sensitivity derived from paleo-climate will not (in the end) apply to the future.

We've often gone over the Charney sensitivity constraint for the Last Glacial Maximum. There is information about the greenhouse gases (CO₂, CH₄ and N₂O), reconstructions of the ice sheets and vegetation change, and estimates of the dust forcing. A recent estimate of the magnitude of these forcings is around 8 +/- 2 W/m2 (Schneider von Deimling et al, 2006). This implicitly includes other aerosol changes or atmospheric chemistry changes in with the sensitivity (or equivalently, assumes that their changes are negligible). So given a temperature change of about 5 to 6ºC, this gives a Charney sensitivity of around 3ºC (ranging from 1.5 to 6 if you do the uncertainty sums).

Hansen suggests that the dust changes should be considered a fast feedback as well (as could the CH₄ changes?) and that certainly makes sense if vegetation changes are included on the feedback side of the equation. Since all of these LGM forcings are the same sign (i.e. they are all positive feedbacks for the long term temperature change), that implies that the Earth System sensitivity must be larger than the Charney sensitivity on these timescales (and for this current geologic period). So far so good.

Hansen's first estimate of the Earth System sensitivity is based on an assumption that GHG changes over the long term control the amount of ice. That gives a scaling of 6ºC for a doubling of CO₂. This is however problematic for two reasons; first most of the power of this relationship is derived from when there were large N. American and European ice sheets. It is quite conceivable that, now that we are left with only Greenland and Antarctica, the sensitivity of the temperature to the ice sheets is less. Secondly, it subsumes the very special nature of orbital forcing - extreme regional and seasonal impacts but very little impact on the global mean radiation. Hansen's estimate assumes that an overall cooling of the same magnitude of the LGM would produce the same extent of ice sheets that was seen then. It may be the case, but it is not a priori obvious that it must be. Hansen rightly acknowledges these issues, and suggests a second constraint based on longer term changes.

Unfortunately, prior to the ice core record, our knowledge of CO₂ changes is much poorer. Thus while it seems likely that CO₂ decreased from the Eocene (~50 million years ago) to the Quaternary through variations related to tectonics, the exact magnitude is uncertain. For reasonable values based on the various estimates, Hansen estimates a ~10 W/m2 forcing change over the Cenozoic from this alone (including a temperature-related CH₄ change). The calculation in the paper is however a little more subtle. Hansen posits that the long term trend in the deep ocean temperature in the early Cenozoic period (before there was substantial ice) was purely due to CO₂ (using the Charney sensitivity). He then plays around with the value of the CO₂ concentration at the initiation of the Antarctic ice sheets (around 34 million years ago) to get the best fit with the CO₂ reconstructions over the whole period. What he ends up with is a critical value of ~425 ppm for initiation of glaciation. To be sure, this is fraught with uncertainties - in the temperature records, the CO₂ reconstructions and the reasonable (but unproven) assumption concerning the dominance of CO₂. However, bottom line is that you really don't need a big change in CO₂ to end up with a big change in ice sheet extent, and that hence the Earth System sensitivity is high.

So what does this mean for the future? In the short term, not much. Even if this is all correct, these effects are for eventual changes - that might take centuries or millennia to realise. However, even with the (substantial) uncertainties in the calculations and underlying assumptions, the conclusion that the Earth System sensitivity is greater than the Charney sensitivity is probably robust. And that is a concern for any policy based on a stabilization scenario significantly above where we are now.

Share This

Myles' piece stresses that criticism of papers in the peer-reviewed literature needs to be in the peer-reviewed literature and suggests that informal criticism (such as on a blog) might undermine that.

We actually agree that there is a real tension between a quick and dirty pointing out of obvious problems in a published paper (such as the Douglass et al paper last December) and doing the much more substantial work and extra analysis that would merit a peer-reviewed response. The approaches are not however necessarily opposed (for instance, our response to the Schwartz paper last year, which has also lead to a submitted comment). But given everyone's limited time (and the journals' limited space), there are fewer official rebuttals submitted and published than there are actual complaints. Furthermore, it is exceedingly rare to write a formal comment on an particularly exceptional paper, with the results that complaints are more common in the peer reviewed literature than applause. In fact, there is much to applaud in modern science, and we like to think that RC plays a positive role in highlighting some of the more important and exciting results that appear.

Myles' piece, while ending up on a worthwhile point of discussion, illustrates it (in my opinion) with a rather misplaced example that involves RC - a post and follow-up on the Stainforth et al (2005) paper and the media coverage it got. The original post dealt in part with how the new climateprediction.net model runs affected our existing expectation for what climate sensitivity is and whether they justified a revision of any projections into the future. The second post came in the aftermath of a rather poor piece of journalism on BBC Radio 4 that implied (completely unjustifiably) that the CPDN team were deliberately misleading the public about the importance of their work. We discussed then (as we have in many other cases) whether some of the responsibility for overheated or inaccurate press actually belongs to the press release itself and whether we (as a community) could do better at providing more context in such cases. The reason why this isn't really germane to Myles' point is that we didn't criticise the paper itself at all. We thought then (and think now) that the CPDN effort is extremely worthwhile and that lessons from it will be informing model simulations some time into the future. Our criticisms (such as they were) were mainly associated instead with the perception of the paper in parts of the media and wider community - something that is not at all appropriate for a peer-reviewed comment.

This isn't the place to rehash the climate sensitivity issue (I promise a new post on that shortly), so that will be deemed off-topic. However, we'd be very interested in any comments on the fundamental issue raised - how do (or should) science blogs and traditional peer-review intersect and whether Myles' perception that they are in conflict is widely shared.

Share This

[This is one of an occasional series on the science of mitigation/adaptation/geo-engineering that we hope to continue. Since this isn't our core expertise, we'd especially appreciate balanced contributions from other scientists.]

One of the central challenges of controlling anthropogenic climate change is developing technologies that deal with emissions from small, dispersed sources such as automobiles and residential houses. Capturing these emissions is more difficult as they are too small to support infrastructure, such as pipelines, and may be mobile, as with cars. For these reasons, proposed solutions, such as switching to using hydrogen or electricity as a fuel, rely on the carbon-free generation of electricity or hydrogen. That implies that the fuel must be made either by renewable generation (wind, solar, geothermal etc.), nuclear or by facilities that capture the carbon dioxide and store it (CCS).

There is however an alternative that gets some occasional attention: Air Capture (for instance, here or here). The idea would be to let people emit the carbon dioxide at the source but then capture it directly from the atmosphere at a separate facility.

The removal of carbon dioxide directly from the atmosphere is a natural phenomenon that occurs in the surface ocean or during photosynthesis. Ocean absorption is a result of both the higher concentration of CO₂ in the atmosphere and the alkaline nature of seawater (Note that this absorption that is leading to the “other” CO₂ problem, ocean acidification - which may prove detrimental to coral reefs and other organisms that use carbonate). Land-based air capture is an effort to enhance this mechanism at an industrial scale so that CO₂ can be removed from the atmosphere under controlled conditions. Given that it is performed under controlled conditions, we can use more alkaline solutions to improve the rate of capture without adversely affecting the biosphere.

Industrial air capture is based on the absorption of CO₂ using alkali earth metals such as sodium or potassium. The process is a variant of the Kraft Process used in most pulp and paper mills and as such, benefits from a long industrial history. The CO₂ is absorbed into solution, transferred to lime via a process called causticization and released in a kiln. With some modifications to the existing processes, mainly an oxygen-fired kiln, the end result is a concentrated stream of CO₂ ready for storage or use in fuels. An alternative to this thermo-chemical process is an electrical one in which an electrical voltage is applied across the carbonate solution to release the CO₂. While simpler, the electrical process consumes more energy as it splits water at the same time. It also depends on electricity and so unless the electricity is renewable or nuclear, will result in the storage of more CO₂ than the chemical process.

If the technology is well established and, aside from the oxygen combustion of lime, dates back over 50 years, what stops it from being used and what might change in the future?

The main barrier is the efficiency of the energy requirements during the reducing process. Air capture requires energy to move the air, manufacture the absorbing solutions and solids as well as to produce the oxygen, fuel and make up chemicals. All of these items will result in additional CO₂ emissions, which reduce the efficiency and therefore the benefits. The second important consideration, and maybe the dominant one, is cost. Air capture has to be more economical than the proposed alternatives (hydrogen, electricity, renewables, greater efficiency etc.). It should be stated clearly that air capture is not a viable alternative to capture at large, point source emitters such as power plants since it will always be more efficient to capture and store carbon dioxide from more concentrated streams. So while there are any non-CCS fossil fuel plants, Air Capture is a non-starter.

But recent suugestions have re-thought air capture as a thermal process. The early incarnations of air capture used electricity as the energy source and therefore depended on carbon-free sources. A thermal Air Capture system uses heat that can be generated on-site, reducing the inefficiencies associated with producing electricity, but of course it still needs a source of (carbon-free) heat. Notably, this implies that air capture could reduce greenhouse gas emissions independently of developments in the power generation or transportation sector. Preliminary experimentation has shown that causticization can occur at ambient temperatures and that conventional vacuum filtration is sufficient to avoid large evaporation penalties. These developments reduce the total energy required for the process by about half compared to the conventional method and thereby reduce the amount of CO₂ that would need to be sent to storage.

However, the cost of air capture is still basically unknown. Estimates have varied wildly and real numbers will only come from pilot projects over the next few years. In some sense, that puts this technology on par with the hydrogen economy with expansion potentially starting sometime after 2015. For now there are far easier (efficiency) and cheaper (power plants) ways of reducing emissions of CO₂ and so air capture is not a replacement for other efforts to reduce emissions. But in the long run, all carbon sources will require mitigation - including the transportation sector - and at that time air capture could be the most cost effective option for some sources. It is not any kind of panacea though.

Share This