According to Freeman Dyson, the current climate models are woefully inadequate because they do not account properly for the effect of plants over time:
I was a little curious about whether the changes in plant life due to increased CO2 in the air or increased temperature could act as a negative feedback and stabilize CO2 level. So I did a seach on Algae Blooms, thinking that they’re probably the fastest growing biomasses, and came across pictures of massive phytoplankton blooms - one of them was 57,000 square kilometers, which my back of the envelope calculations say could have had as much carbon in it as about 2% of the entire U.S. carbon consumption for a year.
I also stumbled across a number of articles saying that phytoplankton levels in the ocean are increasing, and that global warming may be to blame.
So, how accurately are these predicted, and do current climate models consider them?
The interaction between the biosphere and atmospheric gases is a subject of ongoing research. You basically need on-ground field information combined with satellite data, and the two related to each other accurately. I know that ocean absorption of carbon was already something mentioned as a major factor in the models when I was taking university courses about this. Perhaps jshore or somebody can tell you how well the current models incorporate this information, but I can tell you that it’s an important part of current projects in Earth Observation.
ESA’s GlobColour project combines data from several satellites, suitably adjusted in calibration for interusability, to give continuing data on global ocean colour to allow study, forecasting, and input to climate models. The service includes an unbroken time-series going back over a decade, and will continue under the joint ESA-European Commission program GMES (Global Monitoring for Environment and Security), as part of the GMES Marine Core Service. ESA’s main satellite to gather this kind of information is Envisat – ESA and the EC are preparing a more advanced, dedicated ocean-monitoring satellite as we speak, called Sentinel-3. The aim is not only a source of historical data as an input for climate and ocean models, but to even support ‘operational oceanography’ – short term forecasting like we already do for meteorology. Envisat gives global coverage every three days, but the combined data from the various satellites gives reliable daily global coverage.
Anyway, all of this is to say that if it’s not already well accounted in the models, then they’re working on it, and they got the tools now, and it’s being taken very seriously, backed as a main element of ESA’s Earth Observation program, and a public service by the EU. As I recall, the GlobColour data have been checked and validated by surface measurements, so they’re ready for use.
More information about GlobColour is available at the links above, and in an article from the ESA Bulletin available in PDF form here.
The climate models do include carbon cycle modeling, which is indeed important given that only about half the CO2 that we are emitting is remaining in the atmosphere. The largest fraction of the half being absorbed is being absorbed by the oceans but some is being absorbed by the land biosphere. On the other hand, there are predictions that some of these sinks may begin to saturate in the future.
Note that Dyson’s statement that “this fact, that the exchange of carbon between atmosphere and vegetation is rapid, is of fundamental importance to the long-range future of global warming, as will become clear in what follows” is a reference to his discussion later in the article about producing genetically engineered trees that will eat carbon and store it in the ground. (He does not try to make the claim that the fact that these exchanges occur rapidly means that the other half of human emissions will naturally disappear rapidly, which is good because such a claim would not be correct.)Here is the take on Freeman’s claim from David Archer at RealClimate:
However, obviously it is hard to predict future technology so I think there is also a more basic argument to apply to these sort of technological solutions and that is that such solutions are unlikely to be developed and brought to market when there is no cost to emitting CO2 into the atmosphere. So, even if such technology does get developed, that simply seems to be an argument that, once there is a cost to CO2 emissions, the market will develop low-cost ways to reduce or sequester those emissions. And, my answer to that would be: Great! All the more reason to put a pricetag on carbon today. I’m pretty ambivalent about what technological solutions are developed to reduce the buildup of CO2 in the atmosphere (as long as they don’t have too strongly undesirable side effects). But, the economic incentives to develop these aren’t there as long as the costs of CO2 remain externalized.
jshore, if so much carbon is being absorbed by the ocean (as biomass, dissolved gas, etc.), does this affect the ocean acidity? Are changes in the dissolved concentrations of CO[sub]2[/sub] making the ocean more acidic, with potential consequences for biology? I’m not sure of the magnitudes of the effects.
To say “the climate models do include carbon cycle modeling” is a misleading simplification. For example, in the Santer study, of the 19 models:
79% did not include mineral dust
74% did not include sulfate aerosol indirect effects
74% did not include sea salt
63% did not include organic carbon
63% did not include land use change
53% did not include black carbon
53% did not include volcanic aerosols.
42% did not include solar irradiance
37% did not include tropospheric and stratospheric ozone
As you can see, the models vary greatly in what forcings they include, and only some climate models include the carbon cycle. For example, the biosphere comes in peripherally (as “land use change”) in only about a third of the models. Only about half of the models include black carbon. I am not aware of any models that contain changes in plant number/species/extent (other than human-caused changes in land use/land cover).
In addition, a large part of the biosphere (in terms of climate) is oceanic plankton. I do not know of any climate model that includes changes in plankton …
It is not possible to make any blanket statement about what the models include, as each one is different, and includes different forcings.
In my opinion, plants matter a lot, and are poorly represented or not represented at all in most models.
A “back of the envelope” business-as-usual (BAU) calculation looks like this:
Carbon in the soil ~ 1600 Gtonnes
Carbon in the plants ~610 Gtonnes
If we extend the 1980-2000 rise in fossil fuel emissions linearly to 2100, we’ll be emitting about 15 Gtonnes of carbon per year in 2100.
If we extend the 1990-2000 rise in fossil fuel emissions linearly to 2100, we’ll be emitting about 13 Gtonnes of carbon per year in 2100.
OK, let’s use 20 Gtonnes per year in 2100 to represent the highest end of the “Business as Usual” scenario, that gives us plenty of room to spare. We’re currently emitting about 7 Gtonnes/yr of carbon, so the average over the remainder of the century will be around (20+7)/2, or 13.5 Gtonnes/yr. There’s 92 years left in the century, that gives emissions of about 1250 Gtonnes over the century.
However, David Archer conflates the “release [of] carbon to the air” with the amount added to “the land carbon reservoir”, which are very different things. Of all the emissions, some amount will be absorbed by natural processes in the land and ocean. The “Bern Model” used by the IPCC estimates that if the emissions rise linearly to about 14Gt/yr by the year 2100, the CO2 level will rise to ~550 ppmv. Since each ppmv represents ~2.13 tonnes of C, and the current level is about ~380 ppmv, this is a net change of about 360 Gtonnes of C that means that Dyson’s trees would have to soak up.
Since the total carbon on land is about 2200 Gtonnes, that’s only about 15% of the total carbon stored on land that the trees would have to deal with.
My estimate is greatly above some of the IPCC scenarios, as would be expected. The B1 scenario sees 189 Gtonnes of carbon emissions between now and 2100. The average of all of the IPCC scenarios is ~1,350 GTonnes emitted from now to the end of the century, slightly above my estimate of 1,250 Gtonnes. Again, much of this will be sequestered by natural processes, leaving about ~400 Gtonnes to be sequestered by Dyson’s magical trees. Again this is far from the total carbon content of the land, only about 20%.
Note that I make no statement about the validity Dyson’s claim for his mystery trees, which I find quite speculative … I just don’t like bad numbers. Dyson’s claim may well be wrong … but David Archer’s counter-claim does nothing to show that.
Of the models that include biomass, do any of them assume an increase in the biomass due to increased temperature and CO2 concentrations?
Plankton blooms seem to be increasing in total size. A single bloom that I looked at contained about 2% of the carbon emitted from the U.S. in a year. That seems like a very large number. If increased CO2 concentrations result in an increase in blooms, then that’s a moderating effect. Also, it seems that plankton blooms release chemicals which seed clouds leading to increased cloud coverage.
It just seemed to me that these interactions are very complex (just modeling how much of a plankton bloom gets sequestered seems very difficult), and yet this is a major part of the CO2 cycle. What am I missing? How can we get projections of global temperatures out 100 years and claim that we can be accurate within a couple of degrees, when we’re still learning major things about huge forces at play?
intention: Having checked some of your numbers but not all of them, I see some issues with what you have calculated…but do agree that it seems like Archer’s estimate may be high. The problems that I have with your estimate is that you are counting the sequestering of carbon that is occurring on land due to natural processes as taking some of that carbon away. However, what Archer clearly speaks of is the fact that the total land sink would have to double…I.e., he is not just talking about Dyson’s magic trees but about what would have to happen to the land sink in total. I also believe that he may be making his statement from the pre-industrial baseline…i.e., what will happen to the total land sink of carbon relative to what it was before we started emitting significant amounts of CO2. Those two things together seem to me that they would go a fair bot of the way toward fixing this discrepancy you find…although probably not all the way. It does seem that Archer may not have been counting the amount that will be sequestered by the ocean rather than by the land sink but I’m not sure.
Sam, I am not really up on the carbon cycle modeling, so I would have to read up more to answer this (which I don’t have the chance to do right now). The appropriate chapter of the IPCC report to look at, by the way, appears to be Chapter 7, especially Section 7.3.
Well, we do know that the fraction of the CO2 that we have been emitting that has remained in the atmosphere as opposed to being absorbed has remained fairly constant over the time period (at least 50 years…I think more) that we have good data on this. So, I don’t think the uncertainties are as large as you think. If anything, such processes that are available to absorb excess CO2 for a while may tend to saturate.
Also, when you say “accurate within a couple of degrees”, we are talking about changes that are somewhat, but not a lot, larger than that…So, in fact the error bars on the temperature change is pretty large in percentage terms, which reflects the various uncertainties: the uncertainty in how much we will emit (which is clearly also a choice that is in our hands), the uncertainty in how much the sinks will absorb, and the uncertainty in the climate sensitivity from the increase in greenhouse gas concentrations. [And, needless to say, the errors that exist go in both directions…i.e., in addition to the possibility that there is something that will cause the amount of CO2 in the atmosphere to stay on the low side of what we’d expect based on our emissions, there are also things (such as the release of greenhouse gases from melting permafrost areas) that could make things worse than we expect.] It really is a grand experiment that we are performing!
jshore, thanks for the clarification. As near as I could tell, Archer was allegedly calculating the amount that Dyson’s magical trees would have to sequester. This does not include the amount that would otherwise be sequestered by natural processes.
Because otherwise, if he is talking about how much the land would have to sequester in total, his statement has nothing to do with Dyson’s trees. It’s just what the land would sequester, his trees don’t appear in that calculation.
Even if we include the total land sequestration, as you point out, he’s still overstating the case. About two-thirds of all excess carbon is sequestered by the ocean. So taking his figure (about 2,000 Gtonnes) and dividing it by three gives about 600 to 700 Gtonnes for the land to absorb.
It is worth bearing in mind that the climate system (soil, plants, ocean, and air) contains about 43,000 Gtonnes of carbon (excluding the carbon buried in fossil fuels). By burning the fuels, we add about 0.01% per year, or 1% per century, to the total.
I think the question that he is addressing is simply, if we have some way to sequester more in the land sink (plants and soils) by whatever process, magical trees or something else, how much does the land sink have to increase in order to store all the carbon we have put into the atmosphere since the beginning of the industrial revolution through 2100?
Ah…but here is where I realized (in the middle of the night) that you and I are likely making a mistake: Imagine that tomorrow we invent some magical way to instantaneously pull all of the CO2 we have added out of the air to take us back to 280ppm. What will then happen? The answer is that the waters of the oceans will find themselves supersaturated with CO2 relative to what is now in the air and so they will start outgassing CO2 to the air! In other words, if we really want to get back down to pre-industrial levels, we will not in the end have any of the excess stored in the oceans because the oceans are only storing extra CO2 now because of the higher partial pressure of CO2 in the atmosphere. (There is one issue not considered here…which is ocean overturning that sequesters the CO2 deep into the ocean, but this is apparently a slow process…on the order of a thousand years or more…and so I don’t think we can count on it sequestering much. I believe that most of the CO2 that has gone into the ocean due to our activities is still relatively close to the surface of the ocean and thus rapidly available to be liberated if the concentration in the atmosphere falls.)
Well, at current rates (7 Gtonnes per year), it is greater than 1.5% per century and increasing. However, I think it is important to realize that the vast majority of that is in the oceans (using your numbers, I get ~95%)…and presumably a lot of that is deep in the oceans. So, I don’t know how relevant that is, particularly on the timescales of interest to us.
Interesting, and true. However, I don’t think that’s what he’s talking about, because in that case his number would be too small. To return the world to the pre-industrial level would require that the trees soak up all of the excess CO2 that was produced since 1750. It’s not just the ocean that’s in equilibrium with the atmosphere, it’s the soil and the plants as well.
Most of it, as you point out, is in the deep oceans. The exchange rate between the deep oceans and the surface is the subject of much research these days, with no firm answers.
The difficulties are the same ones that bedevil all climate research – the object under study is huge, there’s processes that are very dispersed and hard to measure, and we don’t know everything that’s going on.
It is fairly easy to calculate the CO2 exchange rate between say your bathtub and the air. But the ocean has waves, and currents, and variations in saltiness, and parts of it are covered by fresh water from river or rain. In addition, life makes things very messy. Plankton take CO2 from the air into their bodies. When they die, their bodies sink … but how far do they get before they’re eaten, and if they are not eaten, what is the water depth?
Also, the mixing process are not just from wind, wave, and density. The “deep scattering layer” is the visible sign of one of the largest animal migrations on earth. Every night, when it gets dark, billions of tiny creatures (shrimps and copepods and diatoms and zooplankton and the like) swim up from the depths to feed. With the first light they go down deep again. Of course, the mixing effects of this are huge … and also very hard to calculate.
Finally, we are only starting to get a handle on things like the average speed of the major ocean currents (particularly the deep ones), the amount of salinity driven vertical mixing, and the like.
So in short, you are not the only one who does not “know how relevant that is, particularly on the timescales of interest to us.”
I came in in the middle and haven’t directly addressed this question.
As I showed above, a two thirds of the nineteen climate models used in the Santer study did not even include changes in plant life on the land. I can tell you right now that none of the major GCMs include the effect of plankton.
And as you point out, the amount of carbon in their tiny plant bodies is huge. However, their effects do not stop there.
Plankton use some of the sun’s energy to effect chemical transformations. This energy does not go into warming the ocean.
Phyto-plankton are at or near the surface, so they shade the deeper waters. The energy they absorb is not all used in chemical changes. Some is absorbed as heat, preferentially warming the surface waters. This concentrates the sun’s warmth in the shallow layer, to the benefit of the phyto-plankton.
The most unusual effect is that when plankton get hot, they can emit an aerosol that is a cloud nuclei precursor, increasing the cloudiness above them and shading the hot plankton. Amazing.
So … the answer to your question about whether climate models have a dynamic plankton component is … definitely not, as far as I know. I’ve never heard of it … might be one out there somewhere, but the major models, no.
Returning to Dyson, here’s what he said. He said that there is about an 8% peak to trough wiggle in the CO2 levels every year. If we could prevent the CO2 from returning from the trough to the peak each time (remove it from the system), in about 12 years we could cut the CO2 level in half. In his words:
So, his quote is actually quite different from Archer’s attack. He’s saying that we can, in effect, create a new carbon sink, diverting the carbon away from the atmosphere. Archer, on the other hand, says:
Note that Dyson has not said that he was planning to fill up the land carbon reservoir, that’s Archer’s fantasy. Dysons point is quite different, which is that once we have the atmospheric CO2 “in [biotechnology’s] grasp”, we will be able to pretty much do what we want with it. Make it into a pile of black carbon. Make it into proteins, or precursors for plastics. Who knows what uses we might find?
Without taking any stand on whether this is technologically feasible, or where this carbon might end up, how much carbon would this involve? Well, according to the Mauna Loa data, the current concentration is about 385 ppmv. 8% of that is about 31 ppmv, or about 66 Gtonnes per year. This can be compared with our current fossil fuel emissions of ~ 7 Gtonnes/year.
Thus, at the end of the day, Archer is attacking a straw man. Dyson made no claim that he would necessarily put the carbon in the “land carbon reservoir”, that’s only one of many options with bioengineered trees.
I also found Archers mockery of Dyson quite revealing. Archer says:
Dyson, of course, said nothing of diamonds. However, much stranger things have already come out of bioengineering – goats that produce human hormones, to pick just one example. Thus, while trees making diamonds is unlikely, there are a host of more useful things made out of carbon than just diamonds. Through his mockery and exaggeration, Archer would have us believe that bioengineering trees to produce useful items, chemicals, and raw materials out of carbon is as likely as bioengineering trees to produce diamonds … and if you believe that, you’re as foolish as Archer.
My best to all,
w.
PS - Well, I wandered over to RealClimate to get this information, so I thought I’d post to ask about the numbers … BZZZT!
My simple, polite, scientific question was censored … typical of the RC bullshit. Those cowards never post interesting questions, just people gushing about how wonderful the place is. Scientific site my ass, they’re wimps who never engage in real question and answers. They post up red herring bullshit about trees creating diamonds, and then refuse to let people comment, while pretending their doors are open. Pusillanimous pathetic pricks …
And, Archer’s point is simply that the size of that pile of that black carbon pile is going to be roughly the size of all of the carbon in all of the plants and soils in the entire land biosphere. Is this feasible? I don’t know nor does Archer. However, it certainly seems unlikely to come to fruition in the world that you desire where the atmosphere can be used as a free sewer for CO2.
By the way, I noticed that Figure 7.3 in the IPCC report has a more detailed carbon cycle figure than I have seen before and includes things like the exchange between the surface ocean and the deep ocean, both directly and via marine biota. It also separates out the anthropogenically-induced fluxes from the natural ones.
No, the size of the pile of black carbon (or plastic precursors, or whatever the trees might produce) was not Archer’s point at all. He ignored that it might not be black carbon at all, it might be precursors for plastics, or proteins, or medicines.
Archer blithely assumed it would be just carbon, that it was headed back into the land, and his point was whether the land could absorb it all. His claim is that Dyson’s trees will have to put all of the carbon back into the land carbon reservoir. He said:
That was not what Dyson had proposed. I pointed that out before. As you are making the same identical claim after I pointed out the difference, perhaps my writing was not clear. I hope this clarifies the difference between what Dyson said, and what Archer claimed.
Sounds very interesting. Is it in the TAR, or AR4? Do you have a citation?
All the best,
w.
PS - “Free sewer for CO2”??? Please, let’s leave the hype for the tabloids. Last I looked, there was CO2 in the atmosphere before there were humans … who was using the atmosphere as a “free sewer” back then? Nasssty hobbitses? Dinosaurs? What’s the next claim, that plants ruined the primordial atmosphere by using it as a “free sewer for O2”? …
Why would that matter? Black carbon (or diamonds or whatever) would seem to be the most compact form? Any form where the carbon is diluted with other stuff will just be less compact.
I don’t really see the difference. The carbon has to be stored somewhere. I suppose that perhaps Dyson might propose some method of taking the carbon that the trees magically take out of the atmosphere and burying it in the ocean or something…but then you still have the issue of doing this. I.e., it becomes a carbon sequestration thing (although perhaps the trees will put it in a form that is nicer for sequestering…I don’t know). In the end, I think Archer’s basic point still stands, which this is a large amount of carbon to store…and my point still stands, which is that the technology to do this is unlikely to be developed in a system where there is no financial incentives to do so.
AR4…Chapter 7. I meant this to refer back to my link in my previous post noting that chapter as being the useful one to read on the issue of the carbon cycle. Sorry that this was unclear.
A lot of things, such as mercury, occur in nature but we still don’t want our atmosphere (or waterways) used as free sewers for the portion of it that we emit. (In the case of mercury, in close analogy with CO2, a large portion of mercury emissions is from burning coal.)
I’m sorry my writing is not clear, and perhaps Dyson’s is not as well. What he said is that we can bioengineer trees to sequester the carbon in a useful form, such as say precursors for plastics. This raw material would be converted to plastic and used all over the world for a host of things. He is not talking about storing the carbon, he is talking about using the carbon.
Many thanks, I’ll take a look at that.
Mercury is a heavy metal, which is a poison no matter where it is, even in minute amounts. Are you claiming that CO2 in minute amounts is a poison as well? Because if not, your analogy is meaningless.
For pollutants, the ideal situation would be for their to be none in the air … I don’t think we want to do that with CO2 … because it isn’t a pollutant.
Storing excess carbon in the biomass is really just a temporary solution. Plants and animals die, and when they do their carbon is going to be consumed and reintroduced into the atmosphere. The fundamental problem is that we are taking carbon that is fixed, and changing it into a form that can be absorbed by the atmosphere. In order to solve that problem, we need to take the carbon out of the atmosphere, and somehow make it impossible to be reabsorbed by the atmosphere. Storing it in living things does not do that.
Somehow bioengineering trees to fix carbon into useful products as a method to solve global warming is somewhere between insane delusion and wild pipe dream. Even if we ignore the problem coming up with a plant that can do so, the sheer amount of carbon being emitted makes this scenario impossible. The amount of land, water, and labor required to cultivate a plant on that scale is staggering, and simply impractical to an extreme degree.