Underestimation of multi-decadal global O 2 loss due to an optimal interpolation method

Abstract. The global ocean's oxygen content has declined significantly over the past several decades and is expected to continue decreasing under global warming, with far-reaching impacts on marine ecosystems and biogeochemical cycling. Determining the oxygen trend, its spatial pattern, and uncertainties from observations is fundamental to our understanding of the changing ocean environment. This study uses a suite of CMIP6 Earth system models to evaluate the biases and uncertainties in oxygen distribution and trends due to sampling sparseness. Model outputs are sub-sampled according to the spatial and temporal distribution of the historical shipboard measurements, and the data gaps are filled by a simple optimal interpolation method using Gaussian covariance with a constant e-folding length scale. Sub-sampled results are compared to full model output, revealing the biases in global and basin-wise oxygen content trends. The simple optimal interpolation underestimates the modeled global deoxygenation trends, capturing approximately two-thirds of the full model trends. The North Atlantic and subpolar North Pacific are relatively well sampled, and the simple optimal interpolation is capable of reconstructing more than 80 % of the oxygen trend in the non-eddying CMIP models. In contrast, pronounced biases are found in the equatorial oceans and the Southern Ocean, where the sampling density is relatively low. The application of the simple optimal interpolation method to the historical dataset estimated the global oxygen loss to be 1.5 % over the past 50 years. However, the ratio of the global oxygen trend between the sub-sampled and full model output has increased the estimated loss rate in the range of 1.7 % to 3.1 % over the past 50 years, which partially overlaps with previous studies. The approach taken in this study can provide a framework for the intercomparison of different statistical gap-filling methods to estimate oxygen content trends and their uncertainties due to sampling sparseness.



Figure S3 .
Figure S3.Same as Figure S2 but for the trend reconstructed from optimal interpolation of sub-sampled model output.

Figure S4 .
Figure S4.Time series of Subpolar North Atlantic (SPNA) upper ocean (0-1,000m) column O2 inventory from World Ocean Database 2018 (black solid) and CMIP6 models (color) from (left) full model output and (right) subsampled and optimally interpolated model output.

Figure S5 .
Figure S5.Time series of Subtropical North Atlantic (STNA) upper ocean (0-1,000m) column O2 inventory from World Ocean Database 2018 (black solid) and CMIP6 models (color) from (left) full model output and (right) subsampled and optimally interpolated model output.

Figure S6 .
Figure S6.Time series of Equatorial Atlantic (EQAT) upper ocean (0-1,000m) column O2 inventory from World Ocean Database 2018 (black solid) and CMIP6 models (color) from (left) full model output and (right) subsampled and optimally interpolated model output.60

Figure S8 .
Figure S8.Time series of Mediterranean (MED) upper ocean (0-1,000m) column O2 inventory from World Ocean Database 2018 (black solid) and CMIP6 models (color) from (left) full model output and (right) subsampled and optimally interpolated model output.

Figure S9 .
Figure S9.Time series of Subpolar North Pacific (SPNP) upper ocean (0-1,000m) column O2 inventory from World Ocean Database 2018 (black solid) and CMIP6 models (color) from (left) full model output and (right) subsampled and optimally interpolated model output.

Figure S12 .
Figure S12.Time series of Subtropical South Pacific (STSP) upper ocean (0-1,000m) column O2 inventory from World Ocean Database 2018 (black solid) and CMIP6 models (color) from (left) full model output and (right) subsampled and optimally interpolated model output.

Figure S13 .
Figure S13.Time series of Equatorial Indian (EQID) upper ocean (0-1,000m) column O2 inventory from World Ocean Database 2018 (black solid) and CMIP6 models (color) from (left) full model output and (right) subsampled and optimally interpolated model output.

Figure S14 .
Figure S14.Time series of Subtropical South Indian (STSI) upper ocean (0-1,000m) column O2 inventory from World Ocean Database 2018 (black solid) and CMIP6 models (color) from (left) full model output and (right) subsampled and optimally interpolated model output.

Figure S15 .
Figure S15.Time series of Southern Ocean (SO) upper ocean (0-1,000m) column O2 inventory from World Ocean Database 2018 (black solid) and CMIP6 models (color) from (left) full model output and (right) subsampled and optimally interpolated model output.

Figure S16 .
Figure S16.Time series of Arctic Ocean (AO) upper ocean (0-1,000m) column O2 inventory from World Ocean Database 2018 (black solid) and CMIP6 models (color) from (left) full model output and (right) subsampled and optimally interpolated 125 model output.