News & Views | Published:

Climate science

Ocean circulation drove increase in CO2 uptake

Nature volume 542, pages 169170 (09 February 2017) | Download Citation

The ocean's uptake of carbon dioxide increased during the 2000s. Models reveal that this was driven primarily by weak circulation in the upper ocean, solving a mystery of ocean science. See Letter p.215

The ocean has absorbed approximately 40% of the carbon dioxide emitted to the atmosphere by the burning of fossil fuels and cement production during the industrial era1,2. This uptake is variable: the amount of CO2 absorbed declined in the 1990s, but increased sharply during the 2000s (refs 3,4). However, the processes controlling this reinvigoration of ocean carbon uptake remain a puzzle. On page 215, DeVries et al.5 report that most of the increase that occurred in the 2000s can be explained by changes in ocean overturning circulation — the sinking of cold, dense surface waters to the deep ocean, and the compensatory rising of deep waters. This suggests that biological processes and temperature-driven changes in solubility have relatively minor roles.

Ocean carbon uptake is proportional to a quantity known as ΔpCO2, which is a measure of the difference between the amount of CO2 in the atmosphere and that in the ocean surface6. If there were no changes in ocean temperature, biology or circulation, one would expect ocean carbon uptake to increase in proportion to atmospheric CO2 levels. Ocean models and estimates based on atmospheric and ocean-surface data suggest that ocean CO2 uptake declined during the 1990s, relative to what might have been expected from atmospheric CO2 concentrations, and that this decline occurred primarily in the Southern Ocean3,4,7. This reduction was linked7,8 to intensification of the westerly winds over the Southern Ocean, which brought carbon-rich deep waters to the surface and so decreased ΔpCO2.

Observations of CO2 levels in the atmosphere and ocean surface suggest that this trend reversed in the early 2000s (refs 4,9). The Southern Ocean again played a crucial part in this increase in ocean carbon uptake3,10, but uptake also increased at northern mid-latitudes during this period4. The post-2000 increase in ocean carbon uptake cannot be explained by a simple reversal of the mechanism that caused the slowdown during the 1990s, because westerly winds continued to strengthen during the 2000s. An analysis of temperature-driven effects on the solubility of CO2 suggested that these have a substantial role in the uptake of CO2 in some regions3,4, but cannot explain it entirely. Other processes are needed to fully explain the observed changes3,4. DeVries and colleagues' paper is the first to robustly quantify the role of circulation change in the recent decadal shift in CO2 uptake, providing the missing piece of this puzzle.

DeVries et al. used a new ocean inverse-modelling approach to quantify changes in circulation. Inverse models use observations of the state of a system to estimate the combination of parameters most likely to explain the observed conditions. The authors estimated parameters that drive ocean circulation using observations of ocean temperature, salinity and the concentrations of trace chemicals such as chlorofluorocarbons and radiocarbon, which are strongly affected by ocean circulation but not by ocean biology. This approach revealed decadal variability in upper ocean circulation of up to 50%. Changes between the 1980s and 1990s occurred primarily in the Southern Hemisphere, whereas changes between the 1990s and 2000s involved many regions of the ocean.

The authors then coupled their ocean inverse model to an ocean carbon-cycle model to quantify how these changes in circulation would have affected ocean carbon uptake. The results suggest that upper ocean circulation slowed during the 2000s, reducing the upwelling of carbon-rich deep waters and increasing ΔpCO2, and therefore also increasing ocean CO2 uptake (Fig. 1). To estimate the total increase in uptake, the authors used data from the Surface Ocean pCO2 Mapping (SOCOM) project9, which reconstructs the ocean CO2 sink around the world. By selecting the subset of SOCOM models that best match observations of ΔpCO2, the authors conclude that the total increase in uptake between 2001 and 2011 was approximately 0.8 petagrams of carbon (PgC; 1 Pg is 1015 g) per year per decade. About half of this change can be explained by the increase in atmospheric CO2 from human activities during this period, and the researchers estimate that the remaining trend of 0.4 PgC yr−1 decade−1 is due to ocean circulation alone.

Figure 1: Factors affecting the ocean carbon sink.
Figure 1

The amount of carbon dioxide absorbed by the ocean during the 2000s increased compared with that absorbed during the 1990s. DeVries et al.5 report that the large-scale circulation of the ocean (the overturning circulation) was weaker during the 2000s than during the 1990s, especially in the upper ocean, and therefore brought less CO2-rich water from the ocean depths to the surface. This effect, taken together with increases in anthropogenic CO2 emissions, increased the CO2 gradient between the atmosphere and the ocean, driving CO2 uptake by the ocean. Temperature-driven solubility effects were about tenfold smaller than circulation-driven changes. Because most of the observed change in ocean CO2 uptake during the 2000s can be explained by changes in circulation and atmospheric CO2, it is likely that ocean biology had only a modest role.

These findings suggest that biological processes are unlikely to have had a major role in the post-2000 increase in ocean carbon uptake. This does not imply that ocean biology has remained constant in response to circulation changes, only that the changes have probably had a relatively modest net impact on carbon uptake. For example, a reduction in upper ocean circulation might lead to a reduced nutrient supply at the ocean surface, lowering biological CO2 uptake, whereas temperature-driven changes in the depth of the mixed layer (the depth to which the ocean surface is well mixed by winds) might affect light availability, potentially increasing biological CO2 uptake. The net effect of such changes on the ocean carbon sink might have been limited during the past decade, but the changes might have affected the structure of ocean biological communities. Furthermore, the ability to represent these biological processes accurately in models might be important for predicting other ocean–climate feedbacks11 and future carbon uptake.

The current generation of models — used to estimate global carbon uptake during recent decades and to simulate Earth's response to climate change — do not fully capture the observed decadal-scale variability4. This represents a considerable gap in our ability to assess contemporary carbon budgets and predict future change. Inverse methods such as that used by the authors are currently among the best tools with which to infer how parameters must have varied to match observations, but provide limited insight as to why. So, although DeVries and colleagues' work is a major advance in our understanding of these trends in ocean carbon uptake, it remains unclear for how long the increased carbon uptake observed during the 2000s will persist. More detailed mechanistic studies are needed to tell us what the future holds.



  1. 1.

    & in Climate Change 2013: The Physical Science Basis. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (eds Stocker, T. F. et al.) 486 (Cambridge Univ. Press, 2013).

  2. 2.

    et al. Biogeosciences 10, 2169–2191 (2013).

  3. 3.

    et al. Science 349, 1221–1224 (2015).

  4. 4.

    , & Glob. Biogeochem. Cycles 30, 1396–1417 (2016).

  5. 5.

    , & Nature 542, 215–218 (2017).

  6. 6.

    J. Geophys. Res. 97, 7373–7382 (1992).

  7. 7.

    et al. Science 316, 1735–1738 (2007).

  8. 8.

    , & Glob. Biogeochem. Cycles 22, GB3016 (2008).

  9. 9.

    et al. Biogeosciences 12, 7251–7278 (2015).

  10. 10.

    et al. Geophys. Res. Lett. 42, 7623–7630 (2015).

  11. 11.

    , , , & Geophys. Res. Lett. 38, L07704 (2011).

Download references

Author information


  1. Sara E. Mikaloff Fletcher is at the National Institute of Water and Atmospheric Research (NIWA), Greta Point, Wellington 6021, New Zealand.

    • Sara E. Mikaloff Fletcher


  1. Search for Sara E. Mikaloff Fletcher in:

Corresponding author

Correspondence to Sara E. Mikaloff Fletcher.

About this article

Publication history




By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing