Global Synthesis of Air-Sea CO2 Transfer Velocity Estimates From Ship-Based Eddy Covariance Measurements

Abstract

The air-sea gas transfer velocity (K660) is typically assessed as a function of the 10-m neutral wind speed (U10n), but there remains substantial uncertainty in this relationship. Here K660 of CO2 derived with the eddy covariance (EC) technique from eight datasets (11 research cruises) are reevaluated with consistent consideration of solubility and Schmidt number and inclusion of the ocean cool skin effect. K660 shows an approximately linear dependence with the friction velocity (u*) in moderate winds, with an overall relative standard deviation (relative standard error) of about 20% (7%). The largest relative uncertainty in K660 occurs at low wind speeds, while the largest absolute uncertainty in K660 occurs at high wind speeds. There is an apparent regional variation in the steepness of the K660-u* relationships: North Atlantic ≥ Southern Ocean > other regions (Arctic, Tropics). Accounting for sea state helps to collapse some of this regional variability in K660 using the wave Reynolds number in very large seas and the mean squared slope of the waves in small to moderate seas. The grand average of EC-derived K660(−1.47 + 76.67u*+ 20.48u*2 or 0.36 + 1.203U10n+ 0.167U10n2) is similar at moderate to high winds to widely used dual tracer-based K660 parametrization, but consistently exceeds the dual tracer estimate in low winds, possibly in part due to the chemical enhancement in air-sea CO2 exchange. Combining the grand average of EC-derived K660 with the global distribution of wind speed yields a global average transfer velocity that is comparable with the global radiocarbon (14C) disequilibrium, but is ~20% higher than what is implied by dual tracer parametrizations. This analysis suggests that CO2 fluxes computed using a U10n2 dependence with zero intercept (e.g., dual tracer) are likely underestimated at relatively low wind speeds.

Air-Sea Trace Gas Fluxes: Direct and Indirect Measurements

“…The data clearly imply values of CE on the order 4-6 cm hr -1 at the lowest wind speeds but even at U 10 = 5 ms -1 it is doubtful the difference between observations of Dk and the estimates of the bubble component are significant. This result is similar to (Yang et al, 2022) who computed a grand-averaged k c660 from eight field programs (including the ones used here) and estimate CE by differencing their average with dual tracer estimates of k c660 (Ho et al, 2006). Because the dual tracer studies are done with non-reactive gases, the argument is that they do not include CO 2 CE.…”

“…Because the dual tracer studies are done with non-reactive gases, the argument is that they do not include CO 2 CE. Yang et al (2022) found CE values of about 4 cm hr -1 for U 10 < 10m s -1 . These results suggest t r on the order of 3 s (a r = 0.33) should be used in (38) to estimate CE if the difference between CO 2 and dual tracer is due to chemical enhancement only.…”

Finale: impact of the ORCHESTRA/ENCORE programmes on Southern Ocean heat and carbon understanding

“…This O/E air–sea flux observations and analysis have also provided: the first Southern Ocean evidence for the suppression of CO 2 gas exchange velocity due to the presence of surface surfactants [ 24 ], an important finding particularly for Earth system models; information on the role of ice melt in air–sea CO 2 flux estimates [ 11 , 25 , 26 ]; the first global synthesis of air–sea CO 2 gas exchange velocity measurements derived from eddy covariance, demonstrating that earlier and widely used geochemical tracer-based CO 2 exchange estimates were likely underestimated at low wind speeds [ 27 ]; and the discovery that SST variability in the Southern Ocean demonstrates scale-invariant (fractal) properties, making it plausible that SST can be statistically extrapolated from coarser to finer spatial scales [ 28 ].…”

Response of Surface Ocean pCO2 to Tropical Cyclones in Two Contrasting Basins of the Northern Indian Ocean

“…$$$\delta {F}_{{CO}_{2}}=\sqrt{{(0.2)}^{2}+{(0.003)}^{2}+{{(0.057)}^{2}+(0.0006)}^{2}}\approx 0.2$$$ In the process of calculating CO 2 air‐sea flux, many algorithms have been put forth which relate the gas transfer velocities to a variety of factors regulating the transfer of gases across the air‐sea boundary (Garbe et al., 2014; Gutierrez‐Loza et al., 2022; Leighton et al., 2018; Ribas‐Ribas et al., 2018; Yang et al., 2021). Empirical formulations which relate transfer velocities to wind speeds have reported quadratic (Wanninkhof 2014; Prytherch & Yelland 2021) or higher dependencies and even other functionalities (Deiki & Melville 2018; Frankignoulle, 1988; Gu et al., 2021b; Li et al., 2021; Monahan & Spillane 1984; Wanninkhof & McGillis, 1999; Wanninkhof et al., 2009; Yang et al., 2022). Wanninkhof (1992) showed that a quadratic dependence exists between gas transfer velocities and wind speeds within the range of global wind speeds of 0–30 m/s determined at intervals of 0.5 m/s.…”