Observations from a suite of platforms deployed in the coastal ocean are being combined with numerical models and simulations to investigate the processes that couple the atmosphere and ocean.

Abstract. Laboratory results showing that the air-water gas transfer velocity k is correlated with mean square wave slope have been cited as evidence that a wave-related mechanism regulates k at low to moderate wind speeds [JShne et al., 1987;Bock et al., 1999]. Csanady [1990] has modeled the effect of microscale wave breaking on air-water gas transfer with the result that k is proportional to the fractional surface area covered by surface renewal generated during the breaking process. In this report we investigate the role of microscale wave breaking in gas transfer by determining the correlation between k and AB, the fractional area coverage of microscale breaking waves. Simultaneous, colocated infrared (IR) and wave slope imagery is used to verify that AB detected using IR techniques corresponds to the fraction of surface area covered by surface renewal in the wakes of microscale breaking waves. Using measurements of k and AB made at the University of Washington wind-wave tank at wind speeds from 4.6 to 10.7 m s -1, we show that k is linearly correlated with AB, regardless of the presence of surfactants. This result is consistent with Csanady's [1990] model and implies that microscale wave breaking is likely a fundamental physical mechanism contributing to gas transfer.

[1] The role of microscale wave breaking in controlling the air-water transfer of heat and gas is investigated in a laboratory wind-wave tank. The local heat transfer velocity, k H , is measured using an active infrared technique and the tank-averaged gas transfer velocity, k G , is measured using conservative mass balances. Simultaneous, colocated infrared and wave slope imagery show that wave-related areas of thermal boundary layer disruption and renewal are the turbulent wakes of microscale breaking waves, or microbreakers. The fractional area coverage of microbreakers, A B , is found to be 0.1-0.4 in the wind speed range 4.2-9.3 m s À1 for cleaned and surfactant-influenced surfaces, and k H and k G are correlated with A B . The correlation of k H with A B is independent of fetch and the presence of surfactants, while that for k G with A B depends on surfactants. Additionally, A B is correlated with the mean square wave slope, hS 2 i, which has shown promise as a correlate for k G in previous studies. The ratio of k H measured inside and outside the microbreaker wakes is 3.4, demonstrating that at these wind speeds, up to 75% of the transfer is the direct result of microbreaking. These results provide quantitative evidence that microbreaking is the dominant mechanism contributing to air-water heat and gas transfer at low to moderate wind speeds.

No abstract