U.S. Dept. of Commerce / NOAA / OAR / PMEL / Publications

The oceanic sink for carbon dioxide

Christopher L. Sabine and Richard A. Feely

Pacific Marine Environmental Laboratory, NOAA, Seattle, Washington

Chapter 3 In Greenhouse Gas Sinks, D. Reay, N. Hewitt, J. Grace, and K. Smith (eds.), CABI Publishing, Oxfordshire, UK, 31–49 (2007).
Copyright CAB International 2007. Further electronic distribution is not allowed.

3.3 Modern Carbon Fluxes

Since the pre-industrial period, atmospheric CO2 concentrations have increased from 280 ppm to nearly 380 ppm. This increase in CO2 drives the sea water to absorb CO2 from the atmosphere so that surface sea water is pushed to achieve thermodynamic equilibrium with the atmospheric partial pressure. Figure 3.2 shows a summary of the additional fluxes in the modern ocean resulting from human activity and rising atmospheric CO2. The role of the ocean in the global carbon cycle has changed from being a net source of CO2 to the atmosphere to a net sink for CO2 of ~2 Pg C/year (Sabine et al., 2004a).


Fig. 3.2. Schematic representation of the ocean carbon cycle with pre-industrial fluxes and reservoir sizes (upright) and average values for the 1980s and 1990s (italic). Fluxes (arrows) are in Pg C/year and reservoir sizes (numbers in square brackets) are in Pg C. (Modified from Sabine et al., 2004a.)

Today, the average pCO2 of the atmosphere is ~7 ppm higher than the global ocean pCO2. This small air–sea difference, when spread across the entire surface of the ocean, is sufficient to account for the oceanic uptake of anthropogenic CO2. The pCO2 values in mixed-layer waters, which exchange CO2 directly with the atmosphere, are affected primarily by changes in temperature, dissolved inorganic carbon (DIC) and total alkalinity (TAlk). While the water temperature is regulated by physical processes, including solar energy input, sea–air heat exchanges and mixed-layer thickness, DIC is primarily controlled by the physical processes of air–sea exchange and upwelling of subsurface waters as well as the biological processes of photosynthesis and respiration. Biological production removes carbon from surface waters to form organic material. As organisms die and sink to the ocean interior, they decompose, releasing the carbon once again to the water. This process contributes to higher pCO2 and DIC concentrations in deep ocean waters relative to the surface waters. As pCO2 increases when the water is warmed and decreases as a result of biological uptake, the oceanic uptake and release of CO2 is governed by a balance between the changes in sea water temperature, net biological utilization of CO2 and circulation processes in the upper ocean (Zeebe and Wolf-Gladrow, 2001).

Taro Takahashi of Lamont–Doherty Earth Observatory and his collaborators have amassed a database of more than 1.7 million surface ocean pCO2 measurements, spanning more than 30 years, and derived a pCO2 climatology for the global ocean (Takahashi et al., 2002). These data have been used to determine global and regional sea–air CO2 fluxes with an average annual global open-oceanic uptake of 1.5 ± 0.4 Pg C/year for a nominal year of 1995 (Takahashi et al., 2002; revised by T. Takahashi, New York, 2005, personal communication). This flux estimate represents the total net flux in 1995. The total anthropogenic flux would be the difference between the 1995 net sea–air flux and the pre-industrial net sea–air flux (i.e., −1.5 − 0.6 = −2.1 Pg C/year)—a flux consistent with earlier estimates based on models.

3.3.1 Effect of surface sea water pCO2 variations on CO2 fluxes

Figure 3.3 shows the distribution of total net sea–air CO2 fluxes. The darker shades indicate oceanic areas where there is a net source of CO2 to the atmosphere and the lighter shades indicate regions where there is a net sink of CO2. The equatorial Pacific is a strong source of CO2 to the atmosphere throughout the year as a result of upwelling that brings deep, high CO2 waters to the surface in the central and eastern regions. This upwelling, and thus the CO2 flux to the atmosphere, is heavily modulated by the El Niño–southern oscillation (ENSO) cycle. During strong El Niño years the equatorial Pacific CO2 source can drop to zero. During La Niña the CO2 source to the atmosphere is enhanced. High CO2 outgassing fluxes are also observed in the tropical Atlantic and Indian oceans throughout the year. The Arabian Sea becomes a significant source of CO2 to the atmosphere in the late summer and early fall months as the south-east monsoon generates intense upwelling off the Arabian peninsula.


Fig. 3.3. Global climatology of the annual net sea–air CO2 flux (mol CO2/m2/year) based on interpolation of sea–air pCO2 differences as in 1995. (From Takahashi et al., 2002.)

Strong convective mixing also brings up high CO2 values in the north-western sub-Arctic Pacific and Bering Sea during the northern winter. However, just outside this region there is a seasonal oscillation in CO2 flux. The geochemical response of the ocean to changing temperatures is to decrease the pCO2 by 4.23%/°C of sea water cooling (Takahashi et al., 1993). In some regions, decreasing temperatures in the winter can lower the ocean’s pCO2 values sufficiently to counteract the elevated CO2 brought to the surface from stronger winter time mixing (e.g., temperate North Pacific and North Atlantic oceans). The fluxes out of the ocean from elevated temperatures during summer are limited by stratification, resulting in a small net annual flux into the ocean. Similar seasonal changes are observed in the southern temperate oceans, but are out of phase by half a year.

Intense regions of CO2 uptake are seen in the high-latitude northern ocean in summer and in the high-latitude South Atlantic and southern Indian oceans in austral summer. This uptake is associated with high biological utilization of CO2 in thin mixed layers. As the seasons progress, vertical mixing of deep waters eliminates this uptake of CO2. These observations indicate that the CO2 flux in high-latitude oceans is governed primarily by deep convection in winter and biological uptake during the spring and summer months, whereas in the temperate and subtropical oceans, the flux is governed primarily by water temperature. Outside the equatorial belt, the ∆pCO2 (sea water pCO2—atmospheric pCO2) is highest during winter in subpolar and polar waters, whereas it is highest during summer in the temperate regions. Thus, the seasonal variation of ∆pCO2 and, consequently, the shift between net uptake and release of CO2 in subpolar and polar regions are about 6 months out of phase with that in the temperate regions.

3.3.2 Effect of gas transfer velocity on CO2 fluxes

The ∆pCO2 maps are combined with solubility (s) in sea water and the kinetic forcing function, the gas transfer velocity (k), to produce the flux equation:

F = ks∆pCO2                                     (3.1)

where the gas transfer velocity, k, is controlled by near-surface turbulence in the liquid boundary layer. Laboratory studies in wind–wave tanks have shown that k is a strong but non-unique function of wind speed (Wanninkhof et al., 2002). Results from various wind–wave tank investigations and field studies indicate that factors such as fetch, wave direction, atmospheric boundary layer stability and bubble entrainment influence the rate of gas transfer. Moreover, surfactants can inhibit gas exchange through their damping effect on waves. The commonly used gas transfer parameterizations have been based solely on wind speed, in large part because k is strongly dependent on wind, global and regional wind speed data are readily available and effects other than wind speed have not been well quantified (Wanninkhof et al., 2002). Table 3.1 shows the regional variations of the climatological sea–air exchange fluxes.


Table 3.1.

Using an alternative gas exchange formulization, however, can suggest a different distribution of fluxes. For example, Wanninkhof and McGillis (1999) have suggested a cubic relationship to wind speed instead of the quadratic relationship of Wanninkhof (1992). The cubic relationship gives an uptake that is 45% larger than the quadratic relationship (Table 3.2). This primarily results from a larger CO2 uptake in the high-latitude sink regions because of the stronger impact of the higher winds on the gas exchange (Feely et al., 2001). More studies of gas exchange processes at high wind speed regimes are required before determining whether the quadratic, cubic or some other newly developed relationship is appropriate for high wind speeds.


Table 3.2.

3.3.3 CO2 flux variability

Over the last 40 years, the growth rate of CO2 in the atmosphere has experienced interannual variations as large as ±2 Pg C/year (Francey et al., 1995; Keeling et al., 1996). There is an ongoing controversy on the relative contributions of this variability from atmosphere–land and atmosphere–ocean exchanges (Fig. 3.4). Time series measurements of atmospheric CO2, 13C and O2/N2 sources have suggested that ocean flux variations must be in the order of 1–2 Pg C/year (Francey et al., 1995; Keeling et al., 1996; Rayner et al., 1999; Battle et al., 2000; Bousquet et al., 2000). However, ocean modelling and revised inverse models (Winguth et al., 1994; Le Quéré et al., 2000, 2003; Obata and Kitamura, 2003; McKinley et al., 2004; Peylin et al., 2005; Wetzel et al., 2005) as well as empirical approaches (Lee et al., 1998; Park et al., 2006) have suggested a much smaller ocean variability of ~0.3–0.5 Pg C/year.


Fig. 3.4. Comparison of (a) atmospheric mean annual growth rate, (b) land CO2 flux anomalies and (c) ocean CO2 flux anomalies between 1980 and 1995 (in Gt C/year). The grey zone denotes the range of the inversion models, and the dark line denotes the mean. The coloured lines show the ocean models. (After Peylin et al., 2005.)

Of the few direct time series measurements made over large ocean regions so far, only the equatorial Pacific Ocean (Feely et al., 1997, 1999) and the Greenland Sea (Skjelvan et al., 1999) have shown large year-to-year variations in sea–air CO2 flux. However, there are not many data-sets with which to evaluate such flux variability directly. The variability observed in the equatorial Pacific and North Atlantic oceans is not sufficient to account for all of the variability estimates, but other regions including the Southern Ocean and subtropical regions have not been studied sufficiently to determine their contributions to oceanic variability. Recent ocean model results have suggested that after the equatorial Pacific, the Southern Ocean and the northern extra-tropical regions are also important regions showing significant interannual variability in sea–air CO2 flux (Peylin et al., 2005; Wetzel et al., 2005). Resolving this controversy and imposing stricter constraints on carbon cycle models will require more detailed observations of the magnitude and causes of variability in the sea–air CO2 flux and other carbon-related species in the ocean, as well as continued atmospheric measurements of temporal and spatial distributions of CO2, 13C and O2/N2.

Return to previous section or go to next section

PMEL Outstanding Papers

PMEL Publications Search

PMEL Homepage