| 23 Dec 2025
Global Gridded Air-sea Oxygen Flux Inferred from a Machine Learning-based Dissolved Oxygen Product
Abstract. We present estimates of the monthly open ocean air-sea O2 flux from 2004 to 2024 on a 1°×1° grid spanning 64.5°S to 79.5°N. The flux is computed using 4-dimensional ocean dissolved oxygen (DO) fields derived from Argo float and shipboard observations via machine learning algorithms (GOBAI-O2, Sharp et al., 2023). Flux uncertainties are quantified by generating a large ensemble of flux estimates, first propagating errors from the dissolved oxygen product, then computing fluxes across all combinations of three gas exchange parameterizations that account for bubble-mediated flux and three wind products. We apply DO adjustments to align our resolved global annual mean fluxes with those derived from scaling global ocean heat uptake and regional annual mean fluxes with those derived from ocean inversions. Our results show larger seasonal flux variations at high latitudes than at low latitudes, with clear differences between major ocean basins. Both adjusted and unadjusted annual mean flux estimates exhibit strong ocean O2 sinks at high latitudes, weak sources in the low-to-mid latitude subtropics, and weak sinks near the Equator. The annual mean adjustment significantly enhances ocean O2 uptake in the northern high latitudes and tropical regions while reducing outgassing in the northern subtropics. We evaluate our flux seasonal cycles and annual mean values by comparing forward transport simulations with atmospheric O2 observations from global airborne surveys and surface sampling stations. The simulations reproduce the observed mean seasonal cycles well, but some differences remain in the annual mean latitudinal gradients. We analyze fractional variance contributions from DO, wind, and gas exchange scheme uncertainties and their interactions at regional and hemispheric scales for both climatological monthly and annual mean fluxes. This dataset marks a major improvement over existing air-sea O2 flux products, as it includes the resolution of interannual variability in the flux seasonal cycle, the use of advanced machine learning-based DO fields that better represent complex spatial and temporal patterns, and robust uncertainty quantification across various scales.
As in many previous studies of atmospheric oxygen, this manuscript overlooks the fundamental and controlling influence of water vapor on oxygen availability (Kowalski et al., 2025). The repeated assumption of a constant atmospheric oxygen mole fraction of 0.2094 is inappropriate for many terrestrial surface environments and introduces a systematic bias into the air–sea flux calculations in equations (2) through (4). By neglecting humidity, the authors implicitly assume that the amount of oxygen in air is invariant, whereas it is directly constrained by the presence of water vapor.
The effect is readily illustrated by contrasting typical polar and tropical surface air compositions. In very dry polar air, the water vapor mole fraction may be as low as ~0.0004, leaving dry air to comprise ~99.96% of the total. Under such conditions, representative mole fractions for nitrogen, oxygen, and argon are approximately 0.7808, 0.2095, and 0.0093, respectively. In contrast, tropical surface air commonly exhibits water vapor mole fractions exceeding 0.03, reducing dry air to ≤97% of the total. The corresponding mole fractions of nitrogen, oxygen, and argon are then reduced to less than 0.7577, 0.2033, and 0.0090, respectively—representing a decrease of more than 6000 ppm for oxygen. Assuming a constant oxygen mole fraction therefore leads to a substantial overestimation of the oxygen available for dissolution in tropical surface waters.
This issue is more clearly framed in terms of oxygen’s true thermodynamic control: according to Henry’s law, oxygen solubility depends on its partial pressure. By Dalton’s law, total atmospheric pressure is the sum of the partial pressures of all constituents. Elevated water vapor partial pressure in tropical air does not generally coincide with elevated surface pressure; in fact, mean sea-level pressure near the equator is typically below the global average of 1013.25 mb. Consequently, increased humidity necessarily depresses the partial pressure of dry air and, proportionally, that of oxygen. Ignoring this effect systematically skews estimates of oxygen exchange at the air–sea interface.
The manuscript overlooks this humidity dependence in several key locations, including (i) the second paragraph of the Introduction, (ii) Section 2.3 as a whole, and (iii) Section 2.7, which notes that water vapor is indeed measured but its effect on oxygen availability is not taken into account. Given that humidity is already observed, the omission is both unnecessary and consequential.
The authors should therefore recompute oxygen mole fractions and partial pressures in a manner consistent with atmospheric humidity. Doing so is essential for physically correct estimates of air–sea oxygen fluxes.
Reference
Kowalski, A. S., Janssens, I. A., and Pérez-Priego, O., 2025, Water vapour dynamics as a key determinant of atmospheric composition and transport mechanisms, Biogeosciences, 22, 8005–8012, https://doi.org/10.5194/bg-22-8005-2025