The exchange of carbon dioxide and other trace gases across the air-sea interface plays an important role in global and regional biogeochemical cycles. The gas transfer velocity (k) is thought to be controlled by near-surface turbulence at low to moderate wind speeds and by bubble-mediated processes at higher wind speeds. At low to moderate wind speeds, small-scale waves including microbreaking disrupt the diffusive boundary layer, contribute to mixing at the surface, and enhance exchange. Likewise, at higher wind speeds, large-scale wave breaking, or whitecapping, generates mixing and additionally enhances gas transfer via bubble-mediated exchange. The parameterization for k based on the direct covariance fluxes is shown to have a cubic dependence on wind speed. This result supports the hypothesis that, if bubble mediated exchange is important, the transfer velocity should increase proportionally with whitecap coverage, since whitecap coverage been shown to increase with at least a cubic dependence on wind speed. However, the very large uncertainties under high wind speed conditions limit the universality of this result and the role of breaking waves and bubble modulated transfer.

Here, we present results of the combination of turbulence, deep-ocean wave statistics, whitecapping, and CO2 gas exchange measured during the Southern Ocean Gas Exchange Experiment (SO GasEx) with sustained conditions between 10-20 m s-1. Directional ocean wave spectra, significant wave height, peak wave period, and peak wave direction were obtained with a Wave and Surface Current Monitoring System (WaMoSŪ II). WaMoSŪ II also has the capability to resolve two-dimensional maps of surface elevation snapshots with the significant advantage of continuous availability of wave data in rough seas. In addition, significant wave height was measured using a laser altimeter as well as a nadir-looking microwave system. Oceanic turbulent kinetic energy dissipation rates were measured using a pulse-coherent Doppler sonar mounted at 2-m depth from a drifting surface buoy. Lastly, wave-breaking statistics and whitecapping coverage are reported using two high-resolution digital cameras from the flying bridge. We present results of process studies that investigate the various models for gas transfer that incorporate turbulence and wave-breaking statistics with the goal of developing a focused parameterization.