The possibility that human emissions of CO2 into the atmosphere might change the climate of the earth was first pointed out by Arrhenius in 1896. However, little attention was paid to this idea at the time, for the study of biogeochemical cycles in the atmosphere and ocean had only just begun. Questions of human physiology, on the other hand, captured far more attention. Amongst these was a puzzle concerning respiration in the human lung. The surface area of the lung was known, as was the approximate rate at which inert gases could transfer across it. It was also known that blood was alkaline, and so most dissolved inorganic carbon would be carried in the form of bicarbonate ions (HCO3-), due to the chemical equilibrium:
CO2 + H2O <==> HCO3- + H+ (Keq = approx 10-6 mol l-1)
Therefore, bicarbonate must be converted to CO2, before it can pass into the lung and be exhaled. The rate of this reaction had been measured, and found to be quite slow. In 1928 a calculation was made of the expected flux of CO2 across the surface of the lung based on this reaction rate, and the result was far too low, implying that within a few minutes the concentration of CO2 in the blood would rise to fatal levels. Clearly, some unknown mechanism must have evolved to deal with this problem. It was postulated that an enzyme "carbonic anhydrase" existed to convert bicarbonate rapidly to CO2. After five years this enzyme was isolated, although its structure and the mechanism of catalysis were not known until much later.
Now that the global carbon cycle is the focus of much research, we have come across a similar puzzle concerning the calculation of the flux of CO2 across the surface of the ocean. Although it is difficult to measure the CO2 air-sea exchange rate directly, we know the approximate rate at which inert trace gases cross the air-sea interface, as a function of windspeed and temperature. We can extrapolate these results to CO2, and calculate the global average exchange rate. However, this result comes out to be about 60% lower than that predicted from the global 14C budget. Recent field measurements of the exchange of inert gases at sea using the "dual tracer" technique have confirmed that the apparent discrepancy between the inert gas and 14C average exchange rates cannot be explained simply by physical processes such as diffusion and bubble transport.
Seawater is also alkaline, and therefore most dissolved inorganic carbon exists in the form of bicarbonate ions, but only CO2 can be exchanged with the atmosphere. If HCO3- could be converted to or from CO2 faster than the time that it takes for the CO2 to diffuse across the thin surface microlayer, which is the rate-limiting process for air-sea gas transfer, then the exchange rate for CO2 could be considerably enhanced compared to that of inert gases. Unfortunately, due to the slow rate of reaction, it is possible to calculate that such an enhancement would not be significant in average sea conditions.
Unless, perhaps, the reaction was catalysed by the same enzyme, carbonic anhydrase. Marine microalgae need to take up CO2 for photosynthesis. Although there is plenty of bicarbonate in seawater, only CO2, being uncharged, can cross the cell membrane. In the last decade it has been found that many species of marine algae produce carbonic anhydrase at the cell surface, to provide a rapid conversion of bicarbonate to CO2 particularly when the CO2 concentration in the water is unusually low. If some of this enzyme escaped to the sea surface microlayer, it might also catalyse air-sea gas exchange.
Although the amount of enzyme required is greater than would be reasonably expected in bulk seawater, many surface active molecules and also some phytoplankton are known to accumulate in the microlayer, the "skin of the sea". It is very difficult to make reliable measurements of microlayer chemistry directly, but perhaps by measuring the CO2 air-water exchange rate itself during an algal bloom in a laboratory tank, we might be able to observe a catalysis effect.
The results of such experiments are described in chapter 8. For some, but not all of the species of marine microalgae investigated during blooms in the tank, there was a distinct catalysis of the air-water CO2 exchange rate, compared to that measured without algae in similar conditions. The algal blooms sometimes drew water pCO2 down to very low levels, and catalysis was greatest in these conditions, as predicted from a model of physiological demand for carbonic anhydrase.
The CO2 exchange rate in the "control" experiments without algae also increased at lower water pCO2. A slight increase is to expected, because CO2 also reacts directly with hydroxyl (OH-) ions, and these are more abundant at lower pCO2. However, the measured enhancement due to uncatalysed chemical reaction was considerably greater than predicted by a reaction-diffusion model, and also seemed to be influenced by the air pCO2. To investigate this further, many measurements were made over a range of stirring speeds, temperatures and pCO2, and these results are presented in chapter 7.
We must also bear in mind that the net global flux of CO2 from the air to the sea is the small difference of much larger regional and seasonal fluxes. Generally, CO2 enters the ocean where the water is cold or during intense algal blooms, and leaves it in warmer regions, or where water upwells from below. If catalysis of the exchange rate is significant, it is only likely to affect regions of CO2 influx, due to the physiological demand for the enzyme at low pCO2, and this bias might greatly magnify the impact of enzyme catalysis on the net global air-sea CO2 flux.
To investigate this, the chemical enhancement was calculated for every square degree of the ocean for every month, using satellite windspeed, and temperature data and modelled pCO2 data. The enzyme was distributed using both satellite chlorophyll data and the physiological model. The effect of using an increased rate for the reaction between CO2 and OH-, to match the results from the laboratory experiments, was also investigated. The results are reported in chapter 9. Comparison of fluxes calculated using many different scenarios illustrates the subtle effects of intercorrelation between the kinetic and thermodynamic parameters controlling air-sea gas exchange.
When the enzyme is distributed using the physiological model, its effect on the net global CO2 flux can indeed be greater than its effect on the global average exchange rate. Unfortunately, however, this magnifying effect will not help us to explain the discrepancy between the 14C and inert tracer exchange rates, because the flux of 14C, is effectively one-way only, into the sea, and is therefore little affected by water pCO2. Conversely, the 14C flux does not constrain the possible effect on the 12C flux, and should not be considered as an "empirical" measurement which takes into account all these "minor" kinetic processes. To accurately determine the net global air-ocean flux, we will need to measure all the contributing factors separately for each region and season, and also to consider intercorrelations at a local scale.
To achieve this, detailed studies of each process are essential, but the global problem must always be borne in mind. This thesis, therefore, covers a wide range of scales of space and time, from the transfer of protons around the carbonic anhydrase molecule, to the global circulation of carbon in the oceans. It also attempts to bring together two previously separate areas of research which share many puzzles in common: CO2 transfer across the skin of the sea, and CO2 transfer across biological membranes. Air-sea CO2 exchange will be introduced in more detail in chapter one, whereas the physiology of carbonic anhydrase in marine algae will be covered in chapter two.
If you jumped into this page from elsewhere, you may find it more convenient to go to the "frames" version of this thesis, alternatively you can use the links below to jump straight to a particular chapter:
Title Page Abstract Contents Figures Overview Chapter1 Chapter2 Chapter3 Chapter4 Chapter5 Chapter6 Chapter7 Chapter8 Chapter9 Chapter10 Appendix References Acknowledgements Links