The past few weeks and years have seen a bushel of papers finding that the natural world, in particular perhaps the ocean, is getting fed up with absorbing our CO2. There are uncertainties and caveats associated with each study, but taken as a whole, they provide convincing evidence that the hypothesized carbon cycle positive feedback has begun.
Of the new carbon released to the atmosphere from fossil fuel combustion and deforestation, some remains in the atmosphere, while some is taken up into the land biosphere (in places other than those which are being cut) and into the ocean. The natural uptake has been taking up more than half of the carbon emission. If changing climate were to cause the natural world to slow down its carbon uptake, or even begin to release carbon, that would exacerbate the climate forcing from fossil fuels: a positive feedback.
The ocean has a tendency to take up more carbon as the CO2 concentration in the air rises, because of Henry’s Law, which states that in equilibrium, more in the air means more dissolved in the water. Stratification of the waters in the ocean, due to warming at the surface for example, tends to oppose CO2 invasion, by slowing the rate of replenishing surface waters by deep waters which haven’t taken up fossil fuel CO2 yet.
The Southern Ocean is an important avenue of carbon invasion into the ocean, because the deep ocean outcrops here. Le Quere et al.  diagnosed the uptake of CO2 into the Southern Ocean using atmospheric CO2 concentration data from a dozen or so sites in the Southern hemisphere. They find that the Southern Ocean has begun to release carbon since about 1990, in contrast to the model predictions that Southern Ocean carbon uptake should be increasing because of the Henry’s Law thing. We have to keep in mind that it is a tricky business to invert the atmospheric CO2 concentration to get sources and sinks. The history of this type of study tells us to wait for independent replication before taking this result to the bank.
Le Quere et al propose that the sluggish Southern Ocean CO2 uptake could be due to a windier Southern Ocean. Here the literature gets complicated. The deep ocean contains high concentrations of CO2, the product of organic carbon degradation (think exhaling fish). The effect of the winds is to open a ventilation channel between the atmosphere and the deep ocean. Stratification, especially some decades from now, would tend to shut down this ventilation channel. The ventilation channel could let the deep ocean carbon out, or it could let atmospheric carbon in, especially in a few decades as the CO2 concentration gets ever higher (Henry’s Law again). I guess it’s fair to say that models are not decisive in their assessment about which of these two factors should be dominating at present. The atmospheric inversion method, once it passes the test of independent replication, would trump model predictions of what ought to be happening, in my book.
A decrease in ocean uptake is more clearly documented in the North Atlantic by Schuster and Watson . They show surface ocean CO2 measurements from ships of opportunity from the period 1994-1995, and from 2002-2005. Their surface ocean chemistry data is expressed in terms of partial pressure of CO2 that would be in equilibrium with the water. If the pCO2 of the air is higher than the calculated pCO2 of the water for example, then CO2 will be dissolving into the water.
The pCO2 of the air rose by about 15 microatmospheres in that decade. The strongest Henry’s Law scenario would be for the ocean pCO2 to remain constant through that time, so that the air/sea difference would increase by the 15 microatmospheres of the atmospheric rise. Instead what happened is that the pCO2 of the water rose twice as fast as the atmosphere did, by about 30 microatmospheres. The air-sea difference in pCO2 collapsed to zero in the high latitudes, meaning no CO2 uptake at all in a place where the CO2 uptake might be expected to be strongest.
One factor that might be changing the pressure of CO2 coming from the sea surface might be the warming surface waters, because CO2 becomes less soluble as the temperature rises. But that ain’t it, as it turns out. The surface ocean is warming in their data, except for the two most tropical regions, but the amount of warming can only explain a small fraction of the CO2 pressure change. The culprit is not in hand exactly, but is described as some change in ocean circulation, caused maybe by stratification or by the North Atlantic Oscillation, bringing a different crop of water to the surface. At any event, the decrease in ocean uptake in the North Atlantic is convincing. It’s real, all right.
Canadell et al  claim to see the recent sluggishness of natural CO2 uptake in the rate of atmospheric CO2 rise relative to the total rate of CO2 release (from fossil fuels plus land use changes). They construct records of the atmospheric fraction of the total carbon release, and find that it has increased from 0.4 back in about 1960, to 0.45 today. Carbon cycle models (13 of them, from the SRES A2 scenario) also predict that the atmospheric fraction should increase, but not yet. For the time period from 1960 to 2000, the models predict that we would find the opposite of what is observed: a slight decrease in the atmospheric fraction, driven by increasing carbon uptake into the natural world. Positive feedbacks in the real-world carbon cycle seem to be kicking in faster than anticipated, Canadell et al conclude.
There is no real new information in the Canadell et al  analysis on whether the sinking sink is in the ocean or on land. They use an ocean model to do this bookkeeping, but we have just seen how hard it is to model or even understand some of the observed changes in ocean uptake. In addition to the changing ocean sink, drought and heat wave conditions may change the uptake of carbon on land. The infamously hot summer of 2003 in Europe for example cut the rate of photosynthesis by 50%, dumping as much carbon into the air as had been taken up by that same area for the four previous years [Ciais et al., 2005].
The warming at the end of the last ice age was prompted by changes in Earth’s orbit around the sun, but it was greatly amplified by the rising CO2 concentration in the atmosphere. The orbits pushed on ice sheets, which pushed on climate. The climate changes triggered a strong positive carbon cycle feedback which is, yes, still poorly understood.
Canadell, J.G., C.L. Quere, M.R. Raupach, C.B. Field, E.T. Buitehuis, P. Ciais, T.J. Conway, N.P. Gillett, R.A. Houghton, and G. Marland, Contributions to accelerating atmospheric CO2 growth from economic activity, carbon intensity, and efficiency of natural sinks, Proc. Natl. Acad. Sci. USA, doi 10.1073, 2007.
Ciais, P., M. Reichstein, N. Viovy, A. Granier, J. Ogee, V. Allard, M. Aubinet, N. Buchmann, C. Bernhofer, A. Carrara, F. Chevallier, N. De Noblet, A.D. Friend, P. Friedlingstein, T. Grunwald, B. Heinesch, P. Keronen, A. Knohl, G. Krinner, D. Loustau, G. Manca, G. Matteucci, F. Miglietta, J.M. Ourcival, D. Papale, K. Pilegaard, S. Rambal, G. Seufert, J.F. Soussana, M.J. Sanz, E.D. Schulze, T. Vesala, and R. Valentini, Europe-wide reduction in primary productivity caused by the heat and drought in 2003, Nature, 437 (7058), 529-533, 2005.
Le Quere, C., C. Rodenbeck, E.T. Buitenhuis, T.J. Conway, R. Langenfelds, A. Gomez, C. Labuschagne, M. Ramonet, T. Nakazawa, N. Metzl, N. Gillett, and M. Heimann, Saturation of the Southern Ocean CO2 sink due to recent climate change, Science, 316 (5832), 1735-1738, 2007.
Schuster, U., and A.J. Watson, A variable and decreasing sink for atmospheric CO2 in the North Atlantic, J. Geophysical Res., in press, 2007.