Parameterising CO2 air-sea gas transfer with wave breaking energy dissipation rate, sea state, and wind speed

Air-sea gas exchange has up-scale ramifications for global climate and ocean biogeochemistry that are of paramount relevance. Gas transfer velocity (k) measurements or appropriate parameterizations for them are required to quantify the fluxes and budgets of the important trace gases (e.g., CO₂, DMS, and CH₄). Where gas flux and concentration gradients are not explicitly measured, k is subdivided into diffusive and bubble-mediated components – each parameterized. Although diffusive transfer velocity, k_s , has been well-described by power-law relationships involving the Schmidt number Sc, large variability exists in parameterizations for bubble-mediated gas transfer velocity, k_b. Since k_b is driven primarily by entrainment of gases through wave breaking, the uncertainty is acutely problematic at high winds where gas flux measurements are scarce. To address the paucity of such data, the High Wind Gas Exchange Study (HiWinGS) directly calculated gas transfer velocity of CO2 (k_CO2) from flux and concentration gradient measurements taken in the Labrador Sea from October 9 – November 13, 2013, where 10-meter neutral wind speeds ranged between 1.8 – 25.2 m s^-1. We use these data to validate a novel gas transfer velocity parameterization constructed using output from a wave hindcast obtained with the spectral wave model (ecWAM) forced with the European Centre for Medium-Range Weather Forecasts (ECMWF) 5th Generation Reanalysis (ERA5). Our parameterisation combines a diffusive term based on wind speed and Sc, and a bubble-mediated term based on gas solubility, wave age, and wave breaking energy dissipation rate to capture gas transfer velocity. We compare our results to common wind-speed-only parameterisations and more recent sea-state based relationships.