Bubble‐Mediated Gas Transfer and Gas Transfer Suppression of DMS and CO₂

Abstract

Direct dimethyl sulfide (DMS) flux measurements using eddy covariance have shown a suppression of gas transfer at medium to high wind speed. However, not all eddy covariance measurements show evidence of this suppression. Processes, such as wave-wind interaction and surfactants, have been postulated to cause this suppression. We measured DMS and carbon dioxide eddy covariance fluxes during the Asian summer monsoon in the western tropical Indian Ocean (July and August 2014). Both fluxes and their respective gas transfer velocities show signs of a gas transfer suppression above 10 m/s. Using a wind-wave interaction, we describe a flow separation process that could be responsible for a suppression of gas transfer. As a result we provide a Reynolds number-based parameterization, which states the likelihood of a gas transfer suppression for this cruise and previously published gas transfer data. Additionally, we compute the difference in the gas transfer velocities of DMS and CO 2 to estimate the bubble-mediated gas transfer using a hybrid model with three whitecap parameterizations. Plain Language SummaryInvestigating the air-gas transfer of dimethyl sulfide (DMS) and CO 2 , we estimate the influence of bubble-mediated gas transfer.Furthermore, we explore the phenomena of gas transfer suppression. The gas transfer between atmosphere and ocean should increase with increasing wind speed. At certain wind speed the amount of gas transferred flattens. We provide a wind-wave interaction as possible explanation of this phenomenon.

The influence of transformed Reynolds number limitation on gas transfer parameterizations and global DMS and CO<sub>2</sub> fluxes

“…We apply the correction to two data sets (Knorr11 (Bell et al, 2017) and SO234-2/235 (Zavarsky et al, 2018)) of DMS gas transfer velocities. Both data sets experienced gas transfer limitation at high wind speed.…”

“…Below Re tr ≤ 6.7 · 10 5 a flow separation between the sea surface and the wind flowing above the wave limits gas trans-25 fer (Zavarsky et al, 2018). As a result, common wind speed parameterizations of k are not applicable (Eq.…”

“…We use a linear parameterization, ZA18 (Zavarsky et al, 2018), obtained during a cruise in the Indian Ocean (SO234-2/235). This linear fit does not contain data points which are influenced by gas transfer limitation.…”

“…In this study n=− 1 2 is used. In contrast to commonly accepted gas transfer velocity parameterizations, parameterizations based on direct flux measurements 15 by eddy covariance systems have shown a decrease or flattening of k with increasing wind speed at medium to high wind speed (Bell et al, 2013(Bell et al, , 2015Yang et al, 2016;Blomquist et al, 2017;Zavarsky et al, 2018). …”

“…Here we use the transformed Reynolds number Re tr (Zavarsky et al, 2018) to identify instances of gas transfer limitations.…”