Abstract
Abstract. The ocean-atmosphere flux of a gas can be calculated from its measured or estimated concentration gradient across the air-sea interface and the transfer velocity (a term representing the conductivity of the layers either side of the interface with respect to the gas of interest). Traditionally the transfer velocity has been estimated from empirical relationships with wind speed, and then scaled by the Schmidt number of the gas being transferred. Complex, physically based models of transfer velocity (based on more physical forcings than wind speed alone), such as the NOAA COARE algorithm, have more recently been applied to well-studied gases such as carbon dioxide and DMS (although many studies still use the simpler approach for these gases), but there is a lack of validation of such schemes for other, more poorly studied gases. The aim of this paper is to provide a flexible numerical scheme which will allow the estimation of transfer velocity for any gas as a function of wind speed, temperature and salinity, given data on the solubility and liquid molar volume of the particular gas. New and existing parameterizations (including a novel empirical parameterization of the salinity-dependence of Henry's law solubility) are brought together into a scheme implemented as a modular, extensible program in the R computing environment which is available in the supplementary online material accompanying this paper; along with input files containing solubility and structural data for ~90 gases of general interest, enabling the calculation of their total transfer velocities and component parameters. Comparison of the scheme presented here with alternative schemes and methods for calculating air-sea flux parameters shows good agreement in general. It is intended that the various components of this numerical scheme should be applied only in the absence of experimental data providing robust values for parameters for a particular gas of interest.
Highlights
The rate of exchange of a trace gas between the atmosphere and ocean is often calculated from observed or inferred concentrations of the gas of interest, from point measurements in individual field studies or averaged or modelled data used in regional or global budgets of trace gas fluxes
The supplementary material contains a full implementation of the scheme, along with input data for 90 trace gases of possible interest to end users, coded in the open-source R software environment, which is freely available for use in other studies
As applied to the effect of dissolved salts on non-electrolyte solubility, there are two competing processes which can be conceptually related to volume at boiling point (Vb) and (Kow): (i) the salting-out effect of the increasing free energy cost of forming “cavities” in the water matrix to accommodate the non-electrolyte molecules with increasing ionic strength, and (ii) the salting-in effect of the shift in the hydration/dissociation of non-electrolyte molecules in favour of increasing solubility as a result of the increasing polarity of the water-ion matrix with increasing ionic strength (Masterton and Lee, 1970; Masterton, 1975; Zhou and Mopper, 1990; Ni and Yalkowsky, 2003)
Summary
The rate of exchange of a trace gas between the atmosphere and ocean (or other water surface) is often calculated from observed or inferred concentrations of the gas of interest, from point measurements in individual field studies or averaged or modelled data used in regional or global budgets of trace gas fluxes. Approaches more detailed than empirical relationships between wind speed and transfer velocity are often used e.g. NOAA COARE (Fairall et al, 1996, 2003) or MESSY AIR-SEA (Pozzer et al, 2006) Such generally physically-based schemes require extensive information on the physical properties of the gas and/or more detailed physical forcing data than wind speed alone and are, more appropriate for well-studied gases such as CO2, O2 or DMS (dimethyl sulfide), where concentration uncertainty is better constrained and physical characteristics of the gases have been determined experimentally, or where a detailed study of physical parameters is commonly undertaken, e.g. in eddy-covariance or other “micro-meteorological” flux studies. The supplementary material contains a full implementation of the scheme, along with input data for 90 trace gases of possible interest to end users, coded in the open-source R software environment, which is freely available for use in other studies
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
