Air-Sea Gas Exchange in High Winds
Christopher Zappa, Lamont Associate Research Professor, Lamont-Doherty Earth Observatory, Columbia University
Poor understanding of the complex physical controls of air-sea exchanges under high winds, in particular with respect to uptake and release of greenhouse gases, remains one of the major uncertainties in biogeochemical models and climate predictions. Adequate characterization of gas transfer across the air-sea interface is not only essential to quantify local and global sinks and sources of CO2 but also to budget many other trace gases that influence Earth’s radiation.
At high wind speed, breaking waves become a key factor to take into account when estimating gas fluxes. Breaking results in additional upper ocean turbulence and generation of bubble clouds. Efforts have been made towards including the effect of bubble mediate transfer to reduce the uncertainties around gas transfer velocity K estimates at high wind speed. These parameterizations model the transfer velocity due to breaking as a function of fractional whitecap coverage (W) or windsea Reynolds numbers.
Limitations in these gas transfer models arise from the large scatter in W parameterizations that is not yet fully understood. Recent work has focused on linking W variability to wave field statistics in addition to wind speed. Other limitations arise from the lack of observations at high wind speed. While gas transfer coefficients and W have regularly been obtained under winds of up to ~15 m s-1, few gas transfer measurements at wind speeds of 15 to 30 m s-1 exist, and almost none with coincident wave physics observations.
The High Wind Gas exchange Study (HiWinGS) offers a diverse data set that presents the unique opportunity to gain new insights on the poorly understood aspects of air-sea interaction under high winds. The HiWinGS cruise took place in the North Atlantic during October and November 2013. Wind speeds exceeded 15 m s-1 25% of the time amounting to a total of 189 hours of wind speeds above 15 m s-1 of which 48 hours wind speeds greater than 20 m s-1. On October 25th, wind speeds exceeded 25 m s-1 with gusts of 35 m s-1 during the St Jude storm. Continuous measurements of turbulent fluxes of heat, momentum and gas were taken from the bow of the R/V Knorr. Visible imagery was acquired from the port and starboard side of the flying bridge during daylight hours at 20Hz and directional wave spectra were obtained when on station from a wave rider buoy. Additional wave field statistics were computed from a laser altimeter as well as from a Wavewatch III hindcast.
Taking advantage of the range of physical forcing and variable wave conditions sampled during HiWinGS we investigate how W and K vary with sea state, contrasting pure windseas to swell dominated periods. We distinguish between wind seas and swell based on a separation algorithm applied to directional wave spectra as described in [Hanson and Philips, 2001]. For mixed sea, system alignment is considered when interpreting results.
The role of bubble-mediated transfer depends on gas solubility. The four gases (CO2, DMS, acetone, and methanol) sampled during HiWinGS ranged from being mostly waterside controlled to almost entirely airside controlled. While bubble-mediated transfer appears to be small for moderately soluble gases like DMS, the importance of wave breaking turbulence transport has yet to be determined for all gases regardless of their solubility. This will be done by correlating measured gas transfer velocities to estimates of active whitecap fraction (WA) and turbulent kinetic energy dissipation rate (ε). WA and ε are estimated from moments of the breaking crest length distribution derived from the imagery, focusing on young seas, when it is likely that large-scale breaking waves (i.e., whitecapping) will dominate the TKE dissipation rate.