Gas exchange across the air-sea interface can be calculated using concentrations in the ocean and atmosphere and the transfer velocity (k) across the interface (Liss and Merlivat, 1986). Typically, k has been scaled with wind speed (U10); a recent parameterisation of this was developed by Nightingale et al. (2000):
k = 0.222(U10)2 + 0.333U10
However, a modern measurement technique, eddy correlation, has been directly applied to dimethylsulphide (DMS). This produces an alternative parameterisation of k with wind speed (Blomquist et al., 2006):
kDMS = (U10)1.3
Blomquist et al. (2006) attribute this variation to the differential effect of gas solubility within bubbles. This highlights the important point that parameterising transfer velocities to wind speed alone is clearly simplistic and dependent upon the gas considered and the environmental conditions specific to when each transfer velocity measurement was made. Amongst other things, transfer velocities can be affected by the impact of bubbles (e.g. Asher et al., 1996; Woolf, 1993), the sea-surface microlayer (e.g. Upstill-Goddard et al., 2003) and the impact of raindrops (e.g. Ho et al., 2000). In a recent assessment, Woolf (2005) recommends that satellite-retrieved transfer velocities can be estimated more accurately from a combination of wind speed and significant wave height than by wind speed alone.
Implementation Group 2, Physical processes controlling air-sea exchange, also examines Dry/wet deposition.