The exchange of CO2 between ocean and atmosphere is controlled by the air-sea difference in CO2 partial pressure at the surface and by the efficiency of the transfer processes. The efficiency of the transfer processes is mainly determined by the resistance to the transfer in the ocean. Most investigations describe the transfer efficiency in terms of transfer velocity (k). Traditional estimates of air-sea exchange use relatively simple empirical wind-speed-dependent expressions. There are, however, several other physical processes contributing to air-sea transfer, including microwave breaking, spray, bubbles as well as buoyancy at the air-sea interface. The present study shows the important of oceanic turbulence on the efficiency of the air sea transfer. In the low to intermediate wind speed regime (3 to 10 m/s) a deep ocean mixed layer and cooling at the water surface induced an increased transfer velocity (Rutgersson et al., 2009).

Figure 1 shows transfer velocity determined from direct flux measurements with the eddy-covariance method taken in the Baltic Sea in Northern Europe (Rutgersson et al., 2008). The transfer velocity from measurements agrees relatively well with a traditional wind-speed dependent formulation (Wanninkhof, 1992) for some of the data (grey lines and dashed black line in Figure 1). For the data with strong convection in the water and a deep mixed layer the transfer velocity from the measurements are significantly higher than expected from the wind speed dependent function (Wanninkhof, 1992; black solid line in Figure 1). Using the convective velocity scale of the water (analogous to atmospheric convective scaling), the enhancement of the transfer velocity is clearly dependent on the water-side convective velocity.

The transfer velocity is determined by the diffusivity in the water. The efficiency of the transfer can also be expressed in terms of the resistance to transfer (r = 1/k). Resistances can be added, and this is done differently if the resistances act in serial or parallel. For the case with limited mixed-layer depth (Period II in Figure 1), the transfer is entirely controlled by the very large resistance in the water, and the resistance in the atmosphere can be neglected. On the other hand, for the case with less resistance in the water (and a higher transfer velocity in Period I), atmospheric processes can be expected to become increasingly important.

Figure 1:Averages of normalised transfer velocity for wind speed intervals (3, 4, 5, 6 ±0.5 m/s). Data are separated into Periods I and II as well as into stable and unstable atmospheric stratification. Period I represent large mixed layer depth and Period II, small mixed layer depth. Error bars represent ± the standard error, defined as one standard deviation divided by the square root of the number of data for each interval and do not include the measurement error.

References

Rutgersson, A., M. Norman, B. Schneider, H. Pettersson, and E. Sahlée, The annual cycle of carbon dioxide and parameters influencing the air–sea carbon exchange in the Baltic Proper, J. Mar. Syst., 74, 381-394., 2008.

Rutgersson A. and Smedman, A. Enhancement of CO2 transfer velocity due to water-side convection, J. Marine Syst., 80, 125-134,. 2010

Wanninkhof, R., Relationship between wind speed and gas exchange over the ocean, J. Geophys. Res., 7373—7382, 1992.