Alison R. Gray¹, Jaime Palter²

1. University of Washington
2. University of Rhode Island

Estimates of contemporary global air-sea carbon dioxide (CO₂) flux (Takahashi et al. 2009; Landschützer et al. 2014) suggest that subtropical western boundary currents (WBCs) and their zonal extensions are key regions of oceanic carbon uptake (Figure 1a). These narrow, intensified currents, which transport water poleward along the western edge of every ocean basin before separating from the continental shelf and turning eastward, are associated with maxima in mean velocity, eddy kinetic energy, air-sea heat flux, and nutrient transport, as well as deep mixed layers on their equatorward flanks. The prominence of these regions in the global air-sea flux of CO₂ leads to a basic question: What is the role of the WBCs and their zonal extensions in the global carbon cycle? The answer, which involves the physics and biogeochemistry of the ocean-atmosphere system across a wide range of temporal and spatial scales, reveals the complex nature of these systems and their importance within the climate system, while also highlighting a number of the gaps in our current understanding.

Figure 1: (a) Climatological mean annual sea-air CO₂ flux referenced to the year 2000 adapted from Takahashi et al. 2009. Blue (red) areas are ocean sink (source) regions for atmospheric CO₂. (b) Surface eddy kinetic energy calculated from the 2011-2013 daily AVISO sea surface height product. The white circle in a and b indicates the location of the Kuroshio Extension Observatory mooring. WBC systems are labeled as follows: Kuroshio Extension (KE), Gulf Stream (GS), Agulhas Return Current (ARC), East Australian Current (EAC), and Brazil-Malvinas Confluence (BMC). Courtesy A. Fassbender and S. Bishop.

Air-sea carbon fluxes are primarily controlled by the difference between the partial pressure of CO₂ in the atmosphere, which has been steadily increasing since the 1870s due to the burning of fossil fuels, and the partial pressure of CO₂ in seawater (pCO₂), which can be influenced by many factors (Sarmiento and Gruber 2006). The contemporary oceanic carbon uptake in the subtropical WBCs and their extensions comes about from both thermally and biologically driven decreases in pCO₂. The first of these is tied to the role that WBCs play in the large-scale circulation of the ocean. These currents supply the poleward return flow for the large-scale wind-driven circulation that, through Sverdrup dynamics, generates upper ocean equatorward flow across the vast subtropical gyres (Vallis 2006). As a result, the vigorous (on the order of 1 m s^-1) flows in the WBCs rapidly transport warm water from the tropics to mid-latitudes. At mid-latitudes in winter, the atmospheric storm track advects frigid air across the continents and eastward over these warm currents, leading to the largest air-sea heat fluxes in the global ocean. This physical phenomenon has a direct impact on air-sea carbon flux through the solubility effect, whereby ocean cooling reduces pCO₂ and thus drives the system towards more uptake by the ocean. The second mechanism for lowering surface ocean pCO₂ and producing a flux into the ocean is biological production of organic matter in the euphotic zone, which also occurs at high rates in and around WBCs and their extension regions. This biological productivity is supported by deep convective mixing on the equatorward fringes of the WBCs, which brings nutrients up from the seasonal thermocline, as well as the cross-WBC transport of nutrients. Together these two factors ¾ solubility effects and primary productivity ¾ act to decrease pCO₂ in WBCs and their extensions, creating hotspots of ocean carbon uptake.

The impact of air-sea CO₂ fluxes on the global climate system hinges on the fate of carbon after it enters the surface ocean, and in this regard, the WBC regions also play a crucial role.  The significant heat loss to the atmosphere and strong winds that combine to create substantial solubility-driven carbon uptake also lead to some of the deepest wintertime mixed layers in the global ocean on the equatorward side of the WBCs. These thick mixed layers are subsequently capped by lighter waters during springtime restratification and then subducted into the ocean interior to form the subtropical mode waters (STMWs) found in each basin. The carbon contained in these mode waters is thus isolated from the atmosphere. How, when, and where STMWs subsequently re-enter the ocean mixed layer governs the influence of WBC carbon uptake on atmospheric carbon concentrations and helps determine the future evolution of the ocean carbon sink (Gruber et al. 2002). Alternatively, carbon taken up in the surface ocean can enter the interior ocean via sinking organic matter that is then remineralized. Depending on the depths to which biological particles sink, this carbon can be trapped below the surface for potentially much longer timescales.

In addition to the role that STMWs play in the transfer of carbon from the surface ocean to the thermocline, the process of mode water formation has several other effects on the oceanic carbon cycle. The deepening of the mixed layer entrains subsurface waters that are higher in dissolved inorganic carbon and nutrients. The resulting increase in mixed layer carbon acts to reduce the pCO₂ gradient across the air-sea interface, dampening ocean carbon uptake. The increase in nutrients, however, can also stimulate phytoplankton growth, leading to more biologically driven CO₂ uptake. The balance between these processes, which can vary enormously in space and time, will regulate the total carbon uptake and storage in STMWs.  If inorganic carbon and nutrients are entrained into the mixed layer at the same ratio as is removed in sinking organic particles, then primary production will not have a net impact on oceanic CO₂ uptake. However, the physical processes that govern both mixed layer dynamics and the upper ocean circulation, as well as the biological processes that determine productivity and export, are frequently decoupled in both space and time. Accordingly, variability in biologically driven carbon uptake can exist on timescales of seasons to decades or longer.

All of the processes described thus far are believed to have been operating since well before anthropogenic perturbations to the atmospheric CO₂ concentration. Reconstructions of atmospheric CO₂ from ice cores suggest that, averaged over decades, Holocene concentrations were near steady state (Ciais et al. 2013). Therefore, the carbon taken up by the ocean in WBC regions was balanced by outgassing elsewhere, in the climatological mean. The carbon that moves through the climate system through this pre-industrial carbon cycle, referred to as natural carbon, is often conceptually separated from the anthropogenic carbon added to the atmosphere by fossil fuel burning, of which approximately 30% has been absorbed by the ocean (Le Quéré et al. 2016). Modeling and observational studies point to WBCs as being important regions for uptake of anthropogenic carbon, and the STMWs formed on the equatorward flanks of the WBCs account for a significant portion of the anthropogenic carbon storage (Sabine et al. 2004; Iudicone et al., 2016). Climate model-based projections indicate that WBCs and their extensions will continue to be important sinks of carbon as the climate warms (McKinley et al. 2017). However, from observations it is quite difficult to disentangle the natural carbon cycle from the anthropogenic perturbation, and thus natural variability may significantly affect the rates of carbon uptake and storage that we observe.

Together, solubility and biological effects as well as anthropogenic carbon uptake act in concert to create the hotspots of ocean carbon uptake in the WBCs that we observe today. As we become progressively better able to observe and model the climate system, with longer observational records and greater resolution in both space and time, the more variability we find in these highly dynamic regions of the global ocean. WBCs are notable for significant energy at the mesoscale level, i.e., motions at the scales of approximately one month and 100 km at the WBC latitudes, as evidenced by the mean variability in the altimetric sea surface height (Figure 1b). Intense eddying occurs here because WBCs are regions of strong fronts and thus steeply tilted isopycnals. The significant potential energy inherent in this condition leads to the growth of baroclinic instabilities and the substantial mesoscale variability observed in the WBCs.

Mesoscale features can impact the oceanic carbon cycle in a number of different ways. Mesoscale motions strongly affect mixed layer depths, with implications for mode water formation, subduction, and ventilation; the net effect is to increase the stratification of the upper ocean. Many studies have shown that mesoscale eddies, in releasing the potential energy associated with tilted isopycnals, pump heat out of the deep ocean and towards the surface (Gregory 2000; Gnanadesikan et al. 2005; Palter et al. 2014). Given that natural carbon increases with depth in the ocean due to the remineralization of sinking organic matter, this eddy-driven vertical exchange is expected to reduce the strength of the biological pump. On the other hand, anthropogenic carbon concentrations are highest at the ocean’s surface and decrease downward, so that the global-average effect of mesoscale eddies would be expected to decrease the ocean uptake of anthropogenic CO₂, relative to an ocean without such eddies.  To infer the current generation of climate models’ ability to represent the net effect of eddies on carbon through common parameterizations, we take cues from an analysis of ocean heat uptake in models of varying resolution (Griffies et al. 2015): Parameterizations were successful in their qualitative representation of the upward transport of heat, albeit at a reduced efficiency relative to the simulations that resolved a rich spectrum of mesoscale eddies. This conclusion suggests that, in the model analyzed, parameterized eddies might underestimate the upward pumping of natural carbon from the ocean interior and overestimate the unbalanced downward transport of anthropogenic carbon by the time-mean circulation.

As a result of the significant horizontal and vertical shear associated with the high levels of mesoscale activity found in the WBCs, these regions are also hotspots of variability at the submesoscales, or motions at scales of approximately 10 km and one day. Such motions, which can be generated through a number of different mechanisms, can be associated with substantial instantaneous vertical velocities and are thought to be critical in the restratification of the mixed layer. Submesoscale motions can lead to strong physical export of particles (Omand et al. 2014), and their effects are only partially parameterized in current generation of climate models.  While a number of ambitious field campaigns have been conducted in recent years in order to better understand these types of motions (e.g., Shcherbina et al. 2015), there are still many open questions regarding the impact of small-scale instabilities on the uptake and storage of carbon through both biophysical coupling and effects on mode water subduction and ventilation.

Diverse physical and biological processes, at a range of temporal and spatial scales, clearly contribute to making the WBCs hotspots of oceanic carbon uptake. While the large-scale mean state of the system is relatively well-understood, a number of challenges remain that currently limit our understanding of the role that WBCs and their extensions play in the global carbon cycle. The spatial and temporal variability in the natural carbon cycle of these systems is not well-characterized, due to the difficulties inherent in both observing and modeling these extremely turbulent regions. Given that the WBC uptake of total carbon (the sum of the natural and anthropogenic components) is determined by the balance of large fluxes both into and out of the ocean, variability in the processes that govern these fluxes can have important effects on the ocean carbon cycle. Although our knowledge of mesoscale and submesoscale physics has increased immensely over the past few decades, how these motions impact biogeochemical cycling at small scales and how these effects feed back on the larger-scale carbon uptake and storage are still open questions. In addition, the Southern Hemisphere WBCs are generally much less studied than their counterparts in the North Atlantic and North Pacific basins, despite the fact that these systems connect to and interact with the Southern Ocean where model estimates indicate approximately 50% of the ocean uptake of anthropogenic carbon has occurred (Frölicher et al. 2015). Indeed, one of the biggest bottlenecks in both quantifying and understanding the role of WBCs in the global carbon cycle remains the chronic scarcity of observations in the Southern Hemisphere.

As we work toward addressing these issues, it remains critically important to develop a mechanistic understanding of the myriad processes that are involved in carbon uptake in the WBCs and their extensions. In this way, we will be able to translate this knowledge into better predictions of future changes in the role of WBCs in the global carbon cycle.

References

Ciais, P., and Coauthors, 2013: Carbon and Other Biogeochemical Cycles. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, T. F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex, and P. M. Midgley, Eds., Cambridge University Press, Cambridge, United Kingdom, and New York, USA, doi:10.1017/CBO9781107415324.015.

Frölicher, T. L., J. L. Sarmiento, D. J. Paynter, J. P. Dunne, J. P. Krasting, and M. Winton, 2015: Dominance of the Southern Ocean in anthropogenic carbon and heat uptake in CMIP5 models. J. Climate, 28, 862–886, doi:10.1175/JCLI-D-14-00117.1

Gnanadesikan, A., R. D. Slater, P. S. Swathi, and G. K. Vallis, 2005: The energetics of ocean heat transport, J. Climate, 18, 2604–2616, doi:10.1175/jcli3436.1.

Gregory, J. M., 2000: Vertical heat transports in the ocean and their effect on time-dependent climate change. Climate Dyn., 16, 501-515, doi:10.1007/s003820000059.

Griffies, S. M. and Coauthors, 2015: Impacts on ocean heat from transient mesoscale eddies in a hierarchy of climate models. J. Climate, 28, 952–977, doi:10.1175/JCLI-D-14-00353.1.

Gruber, N., C. D. Keeling, and N. R. Bates, 2002: Interannual variability in the North Atlantic ocean carbon sink. Science, 298, 2374–2378, doi:10.1126/science.1077077.

Iudicone, D., and Coauthors, 2016: The formation of the ocean’s anthropogenic carbon reservoir. Scientific Reports, 6, 35473. http://doi.org/10.1038/srep35473

Landschützer, P., N. Gruber, D. C. E. Bakker, and U. Schuster, 2014: Recent variability of the global ocean carbon sink. Global Biogeochemical Cycles, 28, 927–949. http://doi.org/10.1002/2014GB004853

Le Quéré, C., and Coauthors, 2016: Global Carbon Budget 2016. Earth Syst. Sci. Data., 8, 605-649. DOI:10.5194/essd-8-605-2016.

McKinley, G. A., A. R. Fay, N. S. Lovenduski, and D. J. Pilcher, 2017: Natural variability and anthropogenic trends in the ocean carbon sink. Annu. Rev. Mar. Sci., 9, 125–50. http://doi.org/10.1146/annurev-marine-010816-060529

Omand, M. M., E. A. D’Asaro, C. M. Lee, M. J. Perry, N. Briggs, I. Cetini, and A. Mahadevan, 2015: Eddy-driven subduction exports particulate organic carbon from the spring bloom. Science, 348, 222–225. http://doi.org/10.1126/science.1260062

Palter, J. B., S. M. Griffies, B. L. Samuels, E. D. Galbraith, A. Gnanadesikan, and A. Klocker, 2014: The deep ocean buoyancy budget and its temporal variability. J. Climate, 27, 551–573, doi:10.1175/JCLI-D-13-00016.1.

Sabine, C. L., and Coauthors, 2004: The oceanic sink for anthropogenic CO₂. Science, 305, 367–371. http://doi.org/10.1126/science.1097403

Sarmiento, J. L., and N. Gruber, 2006: Ocean Biogeochemical Dynamics. Princeton University Press, New Jersey, USA, ISBN:9781400849079.

Shcherbina, A. Y., and Coauthors, 2015: The LatMix summer campaign: Submesoscale stirring in the upper ocean. Bulletin of the American Meteorological Society, 96, 1257–1279. http://doi.org/10.1175/BAMS-D-14-00015.1

Takahashi, T., and Coauthors, 2009: Climatological mean and decadal change in surface ocean pCO₂, and net sea–air CO₂ flux over the global oceans. Deep-Sea Res. Part II: Top. Stud. Oceanogr., 56, 554–577, doi:10.1016/j.dsr2.2008.12.009.

Vallis, G. K., 2006: Atmospheric and Oceanic Fluid Dynamics: Fundamentals and Large-Scale Circulation.  Cambridge University Press, Cambridge, UK, ISBN: 9780521849692.

Keith B. Rodgers¹, Ping Zhai¹, Daniele Iudicone², Olivier Aumont³, Brendan Carter⁴, Andrea J. Fassbender⁵, Stephen M. Griffies⁶, Yves Plancherel⁷, Laure Resplandy⁸, Richard D. Slater¹, Katsuya Toyama⁹

1. Princeton University
2. Stazione Anton Dohrn, Italy
3. Sorbonne Universités, LOCEAN/IPSL, France
4. University of Washington
5. Monterey Bay Aquarium Research Institute
6. NOAA Geophysical Fluid Dynamics Laboratory
7. University of Oxford, UK
8. Princeton University
9. Japan Meteorological Agency, Japan

A long-standing question regarding the ocean carbon cycle is whether western boundary currents (WBCs) and their extension regions provide an important pathway for anthropogenic carbon (C_ant) uptake, thereby contributing to the known importance of these regions in the climate system. Successive versions of the Lamont Doherty Earth Observatory air-sea carbon dioxide (CO₂) flux climatologies (Takahashi et al. 2002, 2009) indicate that, at the very least, there is a broad correspondence between the maxima in CO₂ fluxes (uptake) and surface heat fluxes (release of heat to the atmosphere over the western subtropical gyres and WBCs). Motivated to understand the mechanistic controls on the ocean carbon cycle in these regions, a number of modeling and observationally based studies have drawn on multiple platforms to constrain fluxes both at the surface and in the interior of this region (Fassbender et al. 2017, and references therein; Nakano et al. 2011).

In particular, Nakano et al. (2015) and Iudicone et al. (2016) have emphasized the value of invoking a density-based framework for understanding the relationship between heat and carbon fluxes in WBCs and their extension regions building on the earlier studies of Iudicone et al. (2008, 2011). Within this framework, WBCs are best understood in the context of the shallow subtropical cell overturning structures (McCreary and Lu 1994; Lu and McCreary 1995) that connect the equatorial upwelling regions with their poleward subtropical water mass formation regions. The role of the subtropical cells in the climate system is to export excess heat absorbed by the coupled system in the equatorial regions to the subtropics, where heat is released along WBCs to the atmosphere. The heat released in WBC regions results in the densification of surface waters and the filling of the subtropical mode water reservoirs. In summary, this overall heat exchange process manifests itself in the ocean as a poleward divergence of warm surface waters in the upper branch of the subtropical cells and a subsurface convergence of cooler thermocline waters. The modeling studies of Nakano et al. (2011, 2015) and Iudicone et al. (2016) argue that the upper branches of the subtropical cells that feed WBCs accumulate C_ant over broad scales and that this accumulation is a first-order process in setting the C_ant inventory of the subtropical and subpolar mode waters reservoirs. In particular, Nakano et al. (2011) demonstrated that the earlier study of Rodgers et al. (2008), which had argued instead for convection in mode water formation regions to determine C_ant uptake, was not supported by the large-scale Lagrangian diagnostics. Although Iudicone et al. (2016) was able to identify an integrated net transfer of C_ant to higher density class waters associated with WBCs, given their global focus, they did not consider the specific pathways over which such transfers can occur.

Before proceeding to a mechanistic evaluation of the pathways and mechanisms regulating the transmission of C_ant to higher density water masses in WBCs in a global model, it is instructive to first consider a simple budget of C_ant over the Southern Hemisphere given by a global ocean carbon cycle model. The model considered is a global non-eddying (nominally 1°) configuration of the Geophysical Fluid Dynamics Laboratory’s (GFDL’s) MOM5-BLING model (Griffies 2012; Galbraith et al. 2011), forced at the surface with CORE-II normal-year forcing (Large and Yeager 2009). The model was spun up for 1000 years, and from 1860-1995 two runs were performed with identical climatological circulation states. For one of the runs, the surface boundary condition for CO₂ gas exchange followed observational reconstructions, and for the second run, pre-industrial CO₂ was maintained in the atmosphere. Following the definition of C_ant of Zhai et al. (2017), who used the same modeling configuration, C_ant in the model is the difference between the carbon variables for these two runs. Given our focus on waters for which s₀  ≤  27.1, we use potential density for our analysis, following the method presented by Iudicone et al. (2016).

Figure 1. For the Southern Hemisphere, the simulated density-binned inventory of anthropogenic carbon (Cant) in 1995 (solid black), the cumulative air-sea fluxes of Cant over 1861-1995 (dashed black), and the density-binned inventories of Cant from the GLODAPv1 data product of Sabine et al. (2004) (solid blue). For each case, the units are 1015 moles of Cant per Δσ=0.1 kg m-3 density interval.

Figure 1 shows the observationally derived density-binned C_ant inventory from the GLODAPv1 product (Sabine et al., 2004) with the density-binned C_ant inventories in 1995 and the cumulative air-sea C_ant fluxes over 1860-1995 simulated by the model. The cumulative fluxes were density-binned by month and then summed separately for each individual density class over the full period, 1860-1995. The base of the directly ventilated thermocline in the model at 30°S, calculated following the method of Sallée et al. (2013), has been identified to be at s₀ = 26.4 in our model configuration. Comparing the density-binned inventories of C_ant from GLODAPv1 (Sabine et al. 2004) with the model state in 1995 reveals that the model captures the first-order structure of the total C_ant inventories over mode and intermediate waters. This diagnostic view reveals that C_ant uptake by gas exchange tends to quantitatively explain storage patterns in waters lighter than the base of the thermocline, as these lighter waters are less likely to be transferred into the interior via subduction. In contrast, in the ocean interior for waters denser than the base of the thermocline, storage tends to exceed uptake (see Iudicone et al. 2016, for a discussion and analytical approximation of these distributions). This is consistent with the idea that WBCs and their extension regions may be serving as “gateways” for the net transfer of C_ant from thermocline to sub-thermocline waters.

Figure 2. Maps of fluxes of Cant across the σ0=26.4 horizon for the year 1995, derived using the water mass transformation diagnostics presented in Equations 1 and 2 of Zhai et al. (2017): (a) the buoyancy-driven component, (b) the diffusive transformation term, (c) the contribution from tracer diffusion, and (d) the total diapycnal fluxes. The units are gC m-2 yr-1 and positive values indicate a net transfer from lighter to denser water masses.

In order to identify the specific mechanisms whereby WBCs and their extension regions sustain exchanges between thermocline water masses and subpolar water masses across the base of the thermocline, we consider, for the year 1995, a decomposition into the three dominant drivers in Figure 2. We begin with the contribution from buoyancy exchange with the atmosphere (Figure 2a). The sign convention is such that net buoyancy loss to the atmosphere, resulting in densification of water parcels that contain C_ant, results in a positive flux. Thus over the WBCs and their extension regions, the diagnostic reveals a structural and significant annual mean flux of C_ant across s₀ = 26.4 from subtropical to subpolar water masses (positive), with a smaller flux in the opposing sense from subpolar to subtropical water masses (negative). The diffusive transformation contribution (Figure 2b) reveals a smaller net flux of C_ant from subpolar water masses into the thermocline across s₀ = 26.4. Thus, this term opposes in its sign the buoyancy-driven component over the WBC and extension regions. A smaller amplitude flux derives from the tracer diffusion contribution (Figure 2c). The total diapycnal flux is shown in Figure 2d, which includes additional terms such as cabbeling – i.e. when two separate water parcels mix to form a third that sinks. Taken together, the results emphasize an interplay of mechanistic drivers over the WBC regions that sustain diapycnal fluxes across the thermocline base and the central importance of heat loss to the atmosphere among the drivers of diapycnal exchanges.

Figure 3. Overturning schematics for the Southern Ocean (three-dimensional domain Y<30°S) for (a) mass fluxes, and (b) for Cant fluxes over the year 1995 in the MOM5-BLING simulation. The coarse graining into thermocline water (TW), subantarctic mode water (SAMW), and Antarctic intermediate water (AAIW) has been accomplished using the algorithmic approach of Sallée et al. (2013). The net freshwater forcing (precipitation minus evaporation, or pme) is shown at the sea surface for the mass fluxes. The mass units are Sverdrups (109 kg s-1) and for Cant are PgC yr-1.

The net cycling of C_ant through the ocean’s overturning structures in 1995 can be better appreciated by considering the net transfers between three coarse-grained layers over the Southern Hemisphere:  subtropical thermocline waters (TW) (s₀ < 26.4), subantarctic mode water (SAMW), (26.4 < s₀ < 27.1), and a deeper layer that will be referred to as Antarctic intermediate water (AAIW) (27.1<s₀). Although the deeper layer also aggregates water masses denser than AAIW, our interest is in quantifying fluxes across the SAMW/AAIW interface. For mass (Figure 3a), it can be seen that the principal formation source of SAMW is from AAIW (70%), with only 30% emanating from TW. This stands in stark contrast to the formation sources for C_ant in SAMW (Figure 3b), where the TW formation source dominates over the AAIW source within the overturning circulation. In fact, the net of TW-to-SAMW formation sources is of nearly the same amplitude as the net gas exchange uptake of C_ant by the SAMW layer over 1995, suggesting highly efficient transfer of C_ant to the ocean interior. We wish to emphasize the strong degree of amplification in the TW formation source of C_ant relative to mass for SAMW. While the Revelle factor (Revelle and Suess 1957) is expected to contribute to this amplification, a detailed attribution study of the discrepancies between mass and C_ant is yet to be realized.

The model results presented here underscore an important role for WBCs and their extension regions in the ejection of C_ant from the thermocline into denser waters. Building on the density framework for understanding C_ant pathways first developed and presented by Iudicone et al. (2011; 2016), our analyses reveal important net diapycnal transfers of C_ant to the ocean interior, consistent with the uptake pathways emphasized in Nakano et al. (2011). Furthermore, these analyses substantiate direct attribution of heat fluxes in WBCs and their extension regions as first-order contributors of C_ant storage in sub-thermocline waters associated with the shallow subtropical cell overturning structures.

Ejection of C_ant from the thermocline in WBCs and their extension regions has implications for the climate system for two reasons. First, the Revelle factor of low-latitude and thermocline waters is known to be less than that of circumpolar waters (Sabine et al. 2004), meaning that despite higher temperatures, low-latitude waters have an enhanced capacity to absorb C_ant from the atmosphere than high-latitude waters. Thus filling a large subpolar reservoir such as SAMW with a subtropical formation source may lead to more efficient storage with respect to a circumpolar formation source. Second, denser subpolar interior water masses are expected to have longer interior renewal or re-emergence timescales for their C_ant than subtropical waters (Toyama et al. 2017), and the longer the delay before re-emergence, the weaker the Revelle factor impact will be on regulating future C_ant uptake by the ocean. Given the potential significance of the Revelle factor in regulating carbon-climate feedbacks, it will be important to determine whether this entry pathway for C_ant might change under future perturbations to the physical state of the ocean.

Viewed in light of the study of Toyama et al. (2017), the net transfer of C_ant from thermocline to subpolar water masses across s₀=26.4 should be associated with re-emergence of C_ant from the thermocline into the ocean’s mixed layer over the Southern Ocean (Toyama et al. 2017). We think it is important to combine the Lagrangian diagnostics for re-emergence applied in that study and the water mass transformation diagnostics applied here within a consistent modeling framework. It is also important to test the sensitivity of the formation sources for the important subtropical and subpolar mode water reservoirs to resolution for eddy-permitting or eddy-resolving model configurations. Of equal importance is the development of new observational constraints on the surface and near-surface formation sources of mode waters, through the development and application of quasi-conservative tracers of water mass transformations. One promising quasi-conservative tracer for this purpose is oceanic radiocarbon, which for the Southern Hemisphere has distinct subtropical and circumpolar surface ocean signatures.

References

Fassbender, A. J., C. L. Sabine, M. F. Cronin, and A. J. Sutton, 2017: Mixed-layer carbon cycling at the Kuroshio Extension Observatory. Global Biogeochem. Cycles, 31, doi:10.1002/2016GB005547.

Galbraith, E. D., and Coauthors, 2011: The impact of climate variability on the distribution of radiocarbon in CM2Mc, a new Earth System Model. J. Climate, 24, 4230-4254, doi:10.1175/2011JCLI3919.1.

Griffies, S. M., 2012: Elements of the Modular Ocean Model (MOM5) (2012 release), GFDL Ocean Group Technical Report No. 7, NOAA/Geophysical Fluid Dynamics Laboratory, 618pp.

Iudicone, D., G. Madec, and T. J. McDougall, 2008: Water-mass transformations in a neutral density framework and the key role of light penetration. J. Phys. Oceanogr., 38, 1357-1376, doi:10.1175/2007JPO3464.1.

Iudicone, D., K. B. Rodgers, I. Stendardo, O. Aumont, G. Madec, L. Bopp, O Mangoni, and M. Ribera d’Alcala, 2011: Water masses as a unifying framework for understanding the Southern Ocean carbon cycle. Biogeosci., 8, 1031-1052,doi:10.5194/bg-8-1031-2011.

Iudicone, D., K. B. Rodgers, Y. Plancherel, O. Aumont, T. Ito, R. M. Key, G. Madec, and M. Ishii, 2016: The formation of the ocean’s anthropogenic carbon reservoir. Sci. Rep., 6, 35473; doi:10.1038/srep35473.

Large, W. G., J. C. McWilliams, and S. C. Doney, 1994: Ocean vertical mixing:  A review and a model with a nonlocal boundary layer parameterization. Rev. Geophys., 32, 363-403, doi:10.1029/94RG01872.

Large, W. G., and S. Yeager, 2009: The global climatology of an interannually varying air-sea flux data set. Climate Dyn.., 33, 341-364, doi:10.1007/s00382-008-0441-3.

Lu, P., and J. P. McCreary, 1995: Influence of the ITCZ on the flow of thermocline water from the subtropical to the equatorial Pacific Ocean. J. Phy. Oceanogr., 25, 3076-3088, doi:10.1175/1520-0485(1995)025<3076:IOTIOT>2.0.CO;2.

McCreary, J. P., and P. Lu, 1994: Interaction between the subtropical and the equatorial ocean circulations: The subtropical cell. J. Phys. Oceanogr., 24, 466-497.

Nakano, H., H. Tsujino, M. Yasuda, T. Hirabara, T. Motoi, M. Ishii, and G. Yamanaka, 2011: Uptake mechanisms of anthropogenic CO₂ in the Kuroshio Extension region in an ocean general circulation model. J. Oceanogr., 67, 765-783, doi:10.1007/s10872-011-0075-7.

Nakano, H., M. Ishi, K. B. Rodgers, H. Tsujino, and G. Yamanaka, 2015: Anthropogenic CO₂ uptake, transport, storage, and dynamical controls in the ocean imposed by the meridional overturning circulation. Global Biogeochem. Cycles, 29, doi:10.1002/2015GB005128.

Revelle, R., and H. E. Suess, 1957: Carbon dioxide exchange between atmosphere and ocean and the question of an increase of atmospheric CO₂ during the past decades. Tellus, 9, 18-27, doi:10.1111j.2153-3490.1957.tb01849.x.

Rodgers, K .B., J. L. Sarmiento, O. Aumont, C. Crevoisier, C. de Boyer Montégut, and N. Metzl, 2008: A wintertime uptake window for anthropogenic CO₂ in the North Pacific. Global Biogeochem. Cycles, 22, doi:10.1029/2006GB002920.

Sabine, C., and Coauthors, 2004: The oceanic sink for anthropogenic CO₂. Science, 305, 367-371, doi:10.1126/science.1097403.

Sallée, J.-B., E. Schuckburgh, N. Bruneau, A. J. S. Meijers, T. J. Bracegirdle, Z. Wang, and T. Roy, 2013: Assessment of Southern Ocean water mass circulation and characteristics in CMIP5 models:  Historical bias and forcing response. J. Geophys. Res. Oceans, 118, 1830-1844, doi:10.1002/jgrc.20135.

Takahashi, T., S. C. Sutherland, C. Sweeney, A. Poisson, N. Metzl, B. Tilbrook, N. R. Bates, R. Wanninkhof, R. A. Feely, and C. L. Sabine, 2002: Global sea-air CO₂ flux based on climatological surface ocean pCO₂ and seasonal biological and temperature effects. Deep-Sea Res., Part II, 49, 1601-1622, doi:10.1016/S0967-0645(02)0003-6.

Takahiashi, T., and Coauthors, 2009: Climatological mean and decadal change in surface ocean pCO₂ and net sea-air CO₂ flux over the global oceans. Deep-Sea Res., Part II, 56, 554-577, doi:10.1016/j.dsr2.2008.12.009.

Toyama, K., K. B. Rodgers, B. Blanke, D. Iudicone, M. Ishii, O. Aumont, and J. L. Sarmiento, 2017: Large re-emergence of anthropogenic carbon into the ocean’s surface mixed layer sustained by the ocean’s overturning circulation. J. Climate, 30, 8615-8631, doi:10.1175/JCLI-D-16-0725.1.

Zhai, P., K. B. Rodgers, S. M. Griffies, R. D. Slater, D. Iudicone, J. L. Sarmiento, and L. Resplandy, 2017: Mechanistic drivers of re-emergence of anthropogenic carbon in the Equatorial Pacific, Geophys. Res. Lett., 44, doi:10.1002/2017GL073758.