A seamless ensemble-based reconstruction of surface ocean pCO<sub>2</sub> and air–sea CO<sub>2</sub> fluxes over the global coastal and open oceans

Abstract

Abstract. We have estimated the air–sea CO2 fluxes (fgCO2) over the global ocean from the open sea to the continental shelves. Fluxes and associated uncertainty were computed from an ensemble-based reconstruction of CO2 sea surface partial pressure (pCO2) maps trained with observations from the Surface Ocean CO2 Atlas v2020 database. The ensemble mean (which is the best estimate provided by the approach) fits independent data well and a broad agreement between the spatial distribution of model-data differences and the ensemble standard deviations (which are our model uncertainty estimate) is seen. The space-time varying uncertainty fields identify oceanic regions where improvements in data reconstruction and extensions of the observational network are needed. Poor reconstructions of pCO2 are primarily found over the coasts and/or in regions with sparse observations, while fgCO2 estimates with largest uncertainty are observed over the open Southern Ocean (44° S southward), the subpolar regions, the Indian gyre, and upwelling systems. Our estimate of the global net sink for the period 1985–2019 is 1.643 ± 0.125 PgC yr−1 including 0.150 ± 0.010 PgC yr−1 for the coastal net sink. Results suggest that the open ocean Subtropical Pacific (between 18° N–49° N) has the strongest CO2 sink (0.485 ± 0.014 PgC yr−1) among the basins of the world, followed by the open ocean sub-basins in the Southern hemisphere. The coastal Subpolar Atlantic (between 49° N–76° N) is the most significant coastal net sink, amounting to one third of the total coastal uptake; the northern Pacific continental shelves (north of 18° N) are the next contributors. The Equatorial Pacific (between 18° S–18° N) is the predominant source emitting 0.523 ± 0.016 PgC yr−1 of CO2 back to the atmosphere. Based on the mean flux density per unit area, the most intense CO2 drawdown is, however, observed over the Arctic (76° N poleward) followed by the Subpolar Atlantic and Subtropical Pacific for both open ocean and coastal sectors. The mean efflux density over the Equatorial Pacific remains the highest, but similar densities can also be found along other strong upwelling systems in the equatorial band.

The impact of the South-East Madagascar bloom on the oceanic CO<sub>2</sub> sink

“…In a broader context, Landschützer et al (2016) suggested that the carbon uptake tended to increase slightly in 1998-2011 in the Subtropical Indian Ocean (their figure 3). We will see that such a change in the CO 2 fluxes in this region is also revealed in the CMEMS-LSCE-FFNN model (Chau et al, 2021). Note that if at that location 27°S/55°E…”

“…In order to complement the results based on regional in-situ data and evaluate the CO 2 sink anomalies in this region back to 1996, we also used results from a neural network model that reconstructs monthly fCO 2 fields and air-sea CO 2 fluxes. The fCO 2 fields were obtained from an ensemble-based feed-forward neural network model (named CMEMS-LSCE-FFNNN) described in Chau et al (2021). To take into account the period in austral summer 2020 when the SEMB was particularly strong, we used the latest temporal extension of the model which relies on the most recent version of the SOCAT data-base (SOCAT-v2021, Bakker et al, 2021).…”

“…To take into account the period in austral summer 2020 when the SEMB was particularly strong, we used the latest temporal extension of the model which relies on the most recent version of the SOCAT data-base (SOCAT-v2021, Bakker et al, 2021). For a full description of the model, access to the data and a statistical evaluation of fCO 2 reconstructions please refer to Chau et al (2021).…”

On the potential of a neural network-based approach for estimating XCO2 from OCO-2 measurements

“…This would allow an easy and rapid quality indicator for the selection of observations that may be used for XCO2 estimates, either using 400 a physics-based algorithm or a NN approach. variables are varied adequately (e.g., Chau et al 2021), or analytical estimates (Aires et al, 2004) should be explored.…”

A monthly surface pCO<sub>2</sub> product for the California Current Large Marine Ecosystem

“…The data products provided by Laruelle et al (2017) and Landschützer et al (2020b) -hereafter L17 and L20, respectively Most data-based estimates of oceanic CO2 uptake have considered only the open ocean (e.g., Rödenbeck et al, 2013;Landschützer et al, 2014;Iida et al, 2015;Denvil-Sommer et al, 2019;Gregor et al, 2019;Watson et al, 2020); however, coastal ocean CO2 uptake is estimated to be about 10% of the open-ocean figure (Laruelle et al, 2010;2014;Bourgeois et al, 2016;Roobaert et al, 2019;Chau et al, 2021) and may be changing at a different rate relative to open-ocean CO2 uptake (Laruelle et al, 2018). Therefore, augmenting global open-ocean pCO2(sw) data products to include the coastal ocean is quite valuable (Fay et al, 2021).…”