Decoding the ocean’s carbon cycle from dissolved oxygen variations

The ocean plays a vital role in shaping Earth’s climate by absorbing carbon dioxide (CO₂) from the atmosphere and storing it for long time period (Fig. 1, left). One of the key mechanisms behind this process is the biological carbon pump—a system driven by several biological processes in the upper ocean (Fig. 1, right). In this process, dissolved CO₂ in seawater is converted into organic matter through photosynthesis by phytoplankton. Some of this organic material sinks into deeper layers of the ocean as carcass and feces, effectively transporting carbon from the surface to the depths. However, directly measuring all the pathways of carbon transport in the biological carbon pump is extremely challenging. As a result, scientists still do not fully understand where and how much CO₂ is actually taken up and stored in the ocean through this process.

Fig. 1. Schematic diagram of the global carbon cycle (left) and the biological carbon pump (right). In the left panel, numbers indicate the amount of carbon stored in each reservoir (such as the atmosphere and the ocean, in units of billion tons of carbon), while numbers on the arrows represent carbon fluxes (billion tons of carbon per year). Black or white numbers denote preindustrial values, and red numbers indicate changes resulting from human activities.

To tackle this problem, our study used a different approach. Instead of directly observing the carbon flux, we analyzed the dissolved oxygen (DO) concentration in seawater, for which much more data are available (Fig. 2). During photosynthesis in the upper ocean, phytoplankton produce oxygen in nearly the same proportion as the amount of CO₂ they consume. Therefore, if we can estimate the net annual biological oxygen production from DO data, we can infer how much organic carbon was produced during the same period.

Fig. 2. Overview of dissolved oxygen (DO) observations in the ocean. DO data have historically been obtained primarily from ship-based observations (upper left), and more recently from autonomous profiling floats (upper center and right). (Bottom) Annual number of observations used in this study (bars) and the proportion of ocean areas (black line) with sufficient DO data coverage to describe the annual cycle.

Dissolved oxygen in the surface ocean, however, is influenced not only by biological processes but also by physical processes such as air–sea gas exchange, ocean currents (horizontal and vertical advection), and mixing or diffusion (Fig. 3, center). In this study, we utilized over 400,000 DO observational profiles (Fig. 2) together with the latest knowledge and datasets on physical ocean processes. By quantitatively accounting for all these physical contributions, we were able to isolate the biological component of oxygen variation. This allowed us to estimate, for the first time, the net annual oxygen production by biological activity across the entire global ocean (Fig. 3).

Fig. 3. Conceptual diagram of the dissolved oxygen (DO) budget in the upper ocean (center) and its annual components (left and right). The DO concentration in the surface ocean varies due to physical processes—such as air–sea gas exchange, advection, and diffusion—and biological processes, including photosynthesis and respiration. Red (blue) regions indicate areas where oxygen is supplied to (removed from) the upper ocean on an annual average basis by each process.

When converted into carbon units, the global ocean carbon uptake via the biological carbon pump was found to be 7.4 ± 2.1 billion tons of carbon per year, significantly lower than previous estimates of about 13 billion tons of carbon per year. Moreover, the global distribution we obtained for the first time revealed that the biological carbon pump plays a particularly important role in carbon uptake in high-latitude and tropical regions (Fig. 4).

Fig. 4. Latitudinal distribution of the biological carbon pump and air–sea carbon dioxide exchange. In both hemispheres, at high latitudes (poleward of 40°) and in the tropics (20°S–20°N), more carbon is transported to the ocean interior than is supplied through the sea surface. This highlights the crucial role of the biological carbon pump in the ocean’s carbon uptake.

Looking ahead, by further refining this method and combining it with new ocean observations planned around Japan under the Habitable Japan program, we aim to deepen our understanding of material and carbon cycles in the surrounding seas.

For more details:
Yamaguchi, R., S.  Kouketsu, N. Kosugi, and M. Ishii (2024): Global upper ocean dissolved oxygen budget for constraining the biological carbon pump. Communications Earth & Environment5, 732, doi:10.1038/s43247-024-01886-7.

(Ryohei Yamaguchi@A01-1, ECHOES. November 2025)