The Carbon Cycle

Pieter Tans
National Oceanic and Atmospheric Administration

I am a very cautious guy. There is talk about a change in sea level as a result of global warming and that people in low-lying countries might be in some danger of seeing their homes being flooded. Well, I grew up in The Netherlands, and I lived about one meter below sea level for a long time. But being a very cautious guy, I moved to Colorado, five and a half thousand feet above sea level.

I will be talking about CO2. Now, if you want to tackle the greenhouse gas problem, CO2 is the molecular species that we have to come to grips with. Not only is it the most important anthropogenic greenhouse gas, it is also very long-lived. I will try to make a case this morning that every kilogram of CO2 emitted counts, in the long run.

Furthermore, I will say nothing new here. Everything I am going to say is known, and has been known for decades by geochemists. For instance, people like Dave Keeling and Wally Broecker in the '70s already put forward the ideas I will present here.

Figure 1 is a schematic of part of the natural carbon cycle. The rapidly exchanging reservoirs are, to the left, plants and soils, the atmosphere in the middle, the oceans to the right, and on top, the fossil fuel reserves that we are dumping into the system.

The units I have used here are ten to the 15th mol or petamol, if it were oil or petroleum, it would correspond to 17 cubic kilometers of oil, or four cubic miles. The atmosphere holds 50 petamols. The amount of fossil fuels that we are burning is half a petamol per year, or in other words, two cubic miles of oil if all the fossil fuel were in the form of oil.

The oceans hold much more than the atmosphere. The oceans hold 3100, the atmosphere only 50, so that the latter is a relatively small reservoir. Plants and soils hold about three to four times the amount the atmosphere holds.

There are two big arrows. This is air-sea exchange between the oceans and the atmosphere, and photosynthesis and respiration between plants and soils and the atmosphere. There are about seven petamols of carbon exchange between the oceans and the atmosphere, and four petamols per year photosynthesized and given back to the atmosphere through respiration.

At this point, I have not yet mentioned the slowly exchanging reservoirs, so I'll add them. We then have a somewhat more complete, albeit still very schematic, picture of the carbon cycle. They are the rock reservoirs, the really large reservoirs of carbon, carbonate rock and kerogen. The oceans have 3100 petamols of carbon; carbonate rock holds five million petamols, more than a thousand times the oceans. However, the fluxes from the oceans to the carbonate rock and from carbonate rock back to the oceans and to the atmosphere are very small compared to these fast exchanges like air-sea exchange.

Take for instance, continental weathering. There is about .02 petamol per year of carbonate rock that is weathered. So these reservoirs are large, but they are almost immobile. They almost play no role in the fast- exchanging system of carbon reservoirs.

I have calculated the results of adding fossil fuels to the atmosphere in a simple carbon cycle model. Figure 2 shows the assumed fossil fuel combustion as a function of time that I have calculated for this model. Here on the X axis is the time scale, starting in 1800, and I go out to the year 2600. The assumption is that all the fossil fuel reserves, 400 petamols of carbon, that is considered economically recoverable will be burned. In both of these scenarios, all of these recovered fossil fuels will be consumed.

The upper curve, the solid curve, is the one in which the initial rate of increase of fossil fuel combustion is two percent per year, until we run into limitations. As it takes more and more effort to get it out of the ground, it will taper off again, according to a logistic curve. That is just a possible scenario of fossil fuel burning.

The dashed line represents the case in which we increase initially the amount of fossil fuel consumption by 1 percent per year. However, in both cases, the same total amount is burned.

What happens? We will only be looking at the atmosphere-ocean system first. What happens is a giant titration reaction (see Figure 3).

Carbon dioxide will combine with water to form carbonic acid. The carbonic acid in the oceans will react with carbonate ions, giving two bicarbonate ions. It will also react with borate ions, giving boric acid. This is the third reaction. These reactions are very well known. The thermodynamic constants have been worked out or measured to great precision, so this is something we really know very well.

Figure 4 is a titration diagram. On the bottom axis is the partial pressure of CO2 in the atmosphere in microatmospheres on a logarithmic scale. Figure 4 represents the well known chemical equilibrium between atmospheric CO2, dissolved CO2, carbonate and bicarbonate ions, boric acid and borate ions.

The solid vertical line shows you the chemical equilibrium concentrations when the CO2 level in the atmosphere was at the pre-industrial level, say in year 1850. The dashed line shows you the same for the CO2 level in the atmosphere now. As you can see, the upper curve is the carbonate ion concentration, the curve underneath that is the borate ion concentration. They are both being depleted by those two reactions that I showed you in the previous Figure. The figure shows how the thermodynamic equilibrium between the surface ocean and the atmosphere develops as CO2 in the atmosphere goes up. This is known. Of course, you will be more interested to see how this plays out in time as expressed in the atmosphere for the two fossil fuel combustion scenarios I showed in Figure 2. This is shown in Figure 5. As a reminder, the two fossil fuel scenarios, with the initial rise of 2 percent and 1 percent, are drawn (not to scale) at the bottom. The upper curve represents the partial pressure of CO2 in the atmosphere if we manage to combust all the fossil fuels that are recoverable according to the first scenario. The partial pressure will go up around the year 2200 to 2,000 ppm. This is only part of the carbon cycle, I remind you. This curve is based only on the response of the atmosphere and the oceans.

Now, if we are more conservative, and we have measures in place to slow down the rate of increase of fossil fuel burning, and we increase by only 1 percent a year, the resulting atmospheric CO2 level is shown on the dashed curve. One sees that it doesn't matter in the long run. In other words, if we manage to burn it all, every ton counts. If you follow the slower scenario, by the time it has all been burned, the CO2 level in the atmosphere around the year 2500, 2600, is the same as when you burn it all at once.

One more thing I should say. I calculated this for a long time into the future. You can see that, by the time all the fossil fuels have been burned, the CO2 in the atmosphere is still decreasing. Why is that? Because the oceans are still not completely in equilibrium with the atmosphere. The total amount of carbon in the atmosphere- ocean systems of course is constant under these assumptions. But the atmosphere is still losing a little bit to the oceans. That is because the turnover time of the oceans is slow. The exchange time of the deep oceans with the atmosphere in this model I took to be 1500 years. This is loosely based on carbon-14 estimates of ocean turnover.

What I just showed you is very simplistic, because there are more reservoirs involved. So let's add the reservoir of plants and soils, and see what happens. I have divided the reservoir of plants and soils into two. One I call plants, which have an average life of maybe forty years, to which I have added so-called intermediate soil carbon, which also may stay in the soil for anywhere between 20 and 100 years. What I call explicitly soils is very long-lived.

There is a component in soils that accumulates very slowly, but it is large. There is more carbon in slowly accumulating soils than there is in the atmosphere. So a typical turnover time for that component in soils, especially at high latitudes, might be a thousand years. Therefore, I split this reservoir up into these two components, one with an aggregate turnover time of 60 years, plants and shorter-lived soils, and long-lived soils with a turnover time of a thousand years.

Next, I have made some very rough assumptions. It was assumed that the rate of photosynthesis responds positively to increased CO2. This type of response has been measured in greenhouse experiments, lab experiments with fairly small plants in pots. It appears to be a very general response that plants will photosynthesize faster at higher CO2 levels, however, we are not so sure that this actually works in the real world, because there are lots of other limitations.

I took an expression for the enzyme kinetic response to higher CO2 levels from Allen, in Global Biogeochemical Cycles, Volume 1, page 1, and I put that into this simple model. Respiration, or decay, is assumed to be proportional to biomass. Figure 6 shows you a dramatic difference. If you allow the biosphere to absorb carbon, you get the middle curve. The lower curve shows the initial 2 percent rise scenario of fossil fuel burning (not to scale). The upper curve is the same as in Figure 5, with no change in plant and soil carbon. It looks as if the carbon taken up by plants and soils has not been emitted into the atmosphere- ocean system, so the carbon that is stored in plants and soils does make a difference. With our assumptions, the CO2 in the atmosphere will be lower.

How does this play out over the long run? I ran it out to the year 7,000 (Figure 7), still with this very simple model. The middle curve includes growth of the biosphere, the upper curve excludes it. Even in the long run, as long as CO2 in the atmosphere stays high, plants will photosynthesize faster than they do today. Under the simple assumptions that I made, they will have a higher amount of stored carbon in them. Both curves go down for the first few thousand years. This is due to deep ocean mixing. How likely is this? Practically all terrestrial ecologists do not believe that this will happen. On the contrary, most of them believe that there will be stresses on ecosystems due to ourselves -- deforestation is a very simple example of that -- but also due to climate change, if the climate changes significantly. There will be significant pressure on ecosystems to move to areas where they presently are not.

Many ecologists tend to think that plants and soils will be storing less carbon instead of more into the future. But for the simple assumptions that I have made, namely that the only response to higher CO2 in a changed climate is more rapid photosynthesis, Figure 8 shows how it plays out. The fast biota, on a time scale of 60, 70 years, will grow in response to higher atmospheric CO2, then decrease and finally level off. The slowly increasing soils will also absorb more carbon in the long run. The fact that the response levels off at the end is only related to the fact that CO2 in the atmosphere also almost levels off in this model.

You may have noticed that none of the carbon added to the system has left. It has only been re-distributed. Whether we intend it to or not, atmospheric CO2 will stabilize if we wait long enough.

I still have omitted a very important long term feature. That is, the role of the rock reservoirs. On a time scale of 7,000 years, they do play a role, because they impact on what I earlier called the titration reaction of CO2 in the oceans.

Figure 9 shows the reactions. As we add more CO2 to the oceans, the deep seas become more acidic. One of the reactions in the long run will be that some calcium carbonate on the ocean floor will dissolve. It will turn into calcium ions and carbonate ions.

It was these carbonate ions that the CO2 initially reacted with. This reaction is repeated here as the second reaction, and when you add the first and second reactions, you get the third. They show that calcium ions add to the oceans' capacity to hold CO2.

Note that this type of calcite weathering takes place both in the deep oceans and on the land. CO2 in the atmosphere, and especially in soils where its concentration is very high, will also react with calcium carbonate (calcite), giving two bicarbonate ions.

Finally, there is a geologic reaction that takes place deep under the earth that I won't go into.

Only the first of these processes I will consider in Figure 10, whereby calcite of the ocean floor is dissolved. What is the result of that? The upper curve again is the very first curve I showed, considering the ocean and the atmosphere only. If we add a rock reservoir to that, no biology here, just inorganic reactions, then the middle curve shows what calcium carbonate dissolution might do.

There are assumptions in here as well. But first of all, let me point that for the first thousand years, it doesn't make any difference whether this carbonate dissolves on the ocean floor or not. That is because it takes a long time for the acidity (CO2) that we are adding at the surface of the ocean to reach the deeper parts of the ocean, those parts of the ocean floor that have calcium carbonate in them.

Next, I made an assumption here which partially determines the rate of atmospheric CO2 decrease, namely that it takes a thousand years for the product of the calcium ion and carbonate ion concentrations to restore itself to a value where the calcium cycle in the oceans is again in balance. So that is an assumption. And it is probably too fast. Articles in the literature suggest that it could also be 3,000 years, but I was conservative and I said, let's maximize this effect and see what happens. So I made it relatively fast. In other words, it is quite possible that the rock cycle or the dissolution of calcite hardly lowers the atmosphere CO2 curve by the year 7000.

To take a look at the really long term effect, I assumed (erroneously) that the product of the calcium ion and carbonate ion concentrations will restore itself to a proper calcium balance, so that the calcium cycle is in balance, in a hundred years, which is extremely unlikely, and is certain not to happen. However, I assumed that just to see what might happen in the really long run.

Figure 11 shows that, in the very long run, the CO2 in the atmosphere doesn't go down anymore. The upper curve still represents the original case, with no changes in the biosphere and no calcite dissolution. In the middle curve, we start out at 280 parts per million in pre-industrial times, and in the year 7000, even with very rapid calcite dissolution, we still have 370 parts per million in the atmosphere after we have burned all 400 petamols of fossil fuel carbon before the year 2500. In other words, this is permanent.

Not entirely permanent, of course. There is also a silicate weathering cycle and burial of organic matter. This will eventually control atmospheric CO2, but it may take 100,000 to a million years.

I hope I have pointed out the importance of management of the carbon cycle, the importance of the biological reservoirs. The amount of carbon stored in biological reservoirs appears to us in this context as if it hasn't been emitted into the giant oceanic titration system.

The research that we do is focused on that. We want to find out right now what the role is of terrestrial systems in absorbing carbon.

Figure 12 shows our main tool. This is an observing network, where we have regular air samples from all of these places and we are looking for the fingerprint of carbon sources and sinks in the atmosphere. Not only do we measure the CO2 itself, we measure a host of other gasses, but especially in the case of CO2, we also measure the carbon-13 and carbon-12 isotopic ratios of CO2. That is important, because when plants take up carbon, they discriminate against the uptake of C-13, thereby enriching the isotopic ratio of the CO2 that stays behind in the atmosphere. When the CO2 is taken up by the oceans, there is no isotopic effect.

In other words, looking carefully at the isotopic ratios, we can distinguish whether plants are taking up the CO2, or the oceans are taking up the CO2.

I will show you something that we published recently. Figure 13 shows during '92 and '93 what we obtained from this fingerprint of CO2 and its isotopic ratio all over the earth in the atmosphere, a picture of sources and sinks of carbon as a function of latitude. So latitude is at the bottom, the South Pole to the left, the North Pole to the right. The solid curve portrays the sum of all natural sources and sinks of CO2, either due to the oceans or the terrestrial biosphere. The dashed line is the contribution to the total sources and sinks from the oceans. The dot- dash is the contribution of plants. Always, the dashed and the dot-dash lines will add up to the solid line.

What do we see? At least during '92 and '93, there is tremendous uptake of CO2 at mid-latitudes in the Northern Hemisphere by plants. The uptake is about half as large as the total combustion of fossil fuels. So this is fortunate, this is good news. People in the oil and coal industry might love it. But like I said, we don't know if this is going to last. Biologists are generally very skeptical that this will keep happening for decades. In fact, we know that in 1994, terrestrial uptake at the mid-latitudes in the Northern Hemisphere was much smaller than during '92 and '93. So we know that it varies a lot from year to year. It just so happened that when we got our isotopic analysis on line, there were two big years of terrestrial uptake.

Go To Detection of Climate Change