Please note: You are viewing the unstyled version of this website. Either your browser does not support CSS (cascading style sheets) or it has been disabled. Skip navigation.

Variations in Oceanic CO2 Concentration, Transport and Divergence in the Atlantic

  Email    Print  PDF  Change text to small (default) Change text to medium Change text to large

1 March 2004 – 31 August 2006

Dr. Alison Macdonald
Woods Hole Oceanographic Institution, Woods Hole, MA 02543

Program Manager: Dr. Kathy Tedesco, NOAA/OAR/OGP/GCC

Related NOAA Strategic Plan Goal:
Goal 2. Understand climate variability and change to enhance society’s ability to plan and respond.

Ocean circulation redistributes carbon zonally, meridionally and vertically, moving waters through, i.e. towards and away from, formation sites, bringing under-saturated waters to the surface and bearing natural and anthropogenic carbon to depth. It effects the location and magnitude of storage and can provide feedbacks on the atmospheric CO2 gradient. The Atlantic's central role in the global thermohaline circulation suggests that this basin should be an important laboratory for understanding the ocean carbon cycle and possible temporal variations in that cycle. Much of what is known about the oceanic carbon cycle has been sought through numerical models and indirectly through atmospheric measurements. With the now relative abundance of CO2 data provided by the World Ocean Circulation Experiment (WOCE), the Joint Global Ocean Flux Study (JGOFS), the National Oceanographic and Atmospheric Administration's Ocean Atmosphere Exchange Study (OACES) and other programs, direct
estimates of oceanic CO2 flux and storage can be made. The project is designed to create a detailed Atlantic, WOCE-era data synthesis focusing on the circulation and associated oceanic transport and divergence of total inorganic carbon (TIC) and anthropogenic carbon (Cant) derived from a box inverse model, and considering temporal differences in carbon & related property concentrations where repeat transects are available.

Project Goals:
(1) focusing on the Atlantic, where many of the newly available one-time and repeat WOCE/OACES/JGOFS hydrographic transects which included carbon measurements were performed (Figure 1), to synthesize these high-resolution, zonal and meridional data into an inverse model capable of providing an estimate of the three dimensional circulation of the full Atlantic Ocean which could then be used to investigate the uptake, transport and storage of total inorganic carbon (TIC), anthropogenic carbon (Cant) and pre-industrial or natural (Cpre),

(2) to further our understanding of possible temporal changes in the sources and sinks of carbon throughout as much of the Atlantic as had been covered by repeat transects which included carbon parameters.

Using the available hydrographic data set, we are working on the development of two different inverse box models. The first is a simple 2 box version of the North Atlantic; the second is the full Atlantic model. The following basic description is valid for all the models we are formulating. Each model is defined by a set of simultaneous linear equations which are solved using a Gauss-Markov type tapered weighted least squares (Wunsch, 1996; Ganachaud, 1999; Macdonald et al., 1998). Keeping in  mind that not all equations include all terms, the general equation which is formed from observed variables describes conservation of a property within a density layer or set of layers and has the form:

where in this equation, for clarity, the unknown values are in boldface type; i, j and k are indices to the station pairs, CTD pressure surfaces in the layer(s) and in the associated Ekman layer, respectively; s represents the indices to each segment (i.e. box side or portion of a box side); 􀀀 is in situ density; vR, vE, and vG are the unknown reference, known geostrophic and initial Ekman velocities, respectively; Ci,j,k and CU/L are the average property concentration or concentration anomaly at pressure surface for a pair and on a density interface, respectively; A & a are the horizontal and vertical layer areas; E are Ekman correction factors defined for each segment; w, kz are dianeutral velocity and diffusivity terms; U, L indicate upper and lower bounding density interfaces; J are layer source/sink terms and V is the volume of the layer.

Most of the available measured properties (Table 1) from each hydrographic line (obtained from WHPO, now CCHDO, or from CDIAC/GLODAP (Carbon Dioxide Information Analysis Center/Global Ocean Data Analysis Project (Key et al., 2004, Lee et al., 2003) have been included in the quality control procedures and retained for use within the inversion or for analysis of absolute property transport. In general, the constraints which are applied to the data include conservation of top to bottom mass and silica, layer volume, layer salt and heat anomaly (Wijffels 1993; Ganachaud 1999, 2000) and some form of conservation of oxygen/phosphate. Currently, the latter is applied as a Redfield coupling, i.e., Jox s substituted by Jph assuming Redfield ratios (Redfield, 1963) with P:O2 = 1:170 (Anderson and Sarmiento, 1994). To complete the model we supply along with the initial estimates of the unknowns, initial estimates of the (co)variance in the solution and the constraints.

6. The North Atlantic Model
Our initial model was simple, i.e., it contained no meridional lines, and therefore, was not subject to crossover issues. Focused on the North Atlantic, its two boxes are defined by data sets taken over a time period spanning less than one year between 1997 and 1998 across 24°N, 47°N, and nominally 57°N, between Newfoundland and Greenland and between the Hebrides and Greenland. Each of these lines is a repeat of a previous section occupied in the early 1990s. A comparison (as done for 24oN) of the early 1990’s to late 1990’s fields of TIC, Cant and other carbon related parameters is being done as part of this project. We use this model testing purposes. Its results helping us understand the effect of the meridional data used in the full model.

For the North Atlantic model using a basic set of conservation constraints (as described above), and initial reference surfaces as defined by Macdonald et al. (2003) at 24oN and from averaged float velocities across most of the two northern latitudes, a TIC flux across 24oN is suggested which is not significantly different from the previous estimate using the 1998 24oN data alone (Figure 2). As expected and more importantly, however, the uncertainties have been reduced by half from those found using a single line model at 24°N (Macdonald et al., 2003). At 57°N, our estimate of net TIC transport and uncertainty is surprisingly similar to previous findings given that different data sets and different analysis techniques were employed. At 47°N, the model produced the interesting result of a very strong southward transport of TIC (2.1±0.7 PgCyr-1) implying that strong uptake of carbon is occurring further north than originally expected, and that the outgassing within the subtropics which may be (depending on which 24°N value you choose to believe) nearly as strong. This model also suggests a relatively weak northward transport of Cant across 47°N.

These results are affected by the strong variability at the western end of the 47oN transect. We are currently working on basing our reference levels and transport field for this region on the mooring data results of Schott et al. (2004) and the repeat transect work of Lumpkin and Speer (2003) and Lumpkin et al., 2006, however,  none of the different referencing schemes we have attempted thus fair have had a significant effect the results large southward transport of TIC across 47°N. What we have found is that the inclusion of constraints based on the conservation of phosphate and oxygen (1:170, Anderson and Sarmiento, 1994), significantly reduces the 47°N TIC transport value. This result is shown in the next section.

7. The Full Atlantic Model
When combined, the variety of TIC transport estimates in Figure 2 suggest there may be substantial uptake of CO2 to the north of 24°N and outgassing to the south (Roson et al., 2002; Alvarez et al., 2003). However, except for the Holfort’s (1998) values in the South Atlantic and the values provided by the North Atlantic model described above, all the estimates come from independent studies and include no guarantee of consistent physics and assumptions. To  provide a physical consistent vision of the Atlantic circulation using the available data we have created a full Atlantic model (Figure 1). The one time coverage is decent (Table 1), however, not all the stations include TIC or Cant estimates. Since in the initial runs we are not including direct constraints on these properties, it is  not necessary that they exist everywhere. We chose only to examine results where data is  available rather than mapping data to cover gaps. Our reasoning is that as we are looking for
sources and sinks, and we need to be careful not to create or remove what we are looking for through mapping errors.

The meridional lines running across the equator will not be used. In the equatorial regions, estimation of synoptic geostrophic velocities is problematic. We initially looked to handle this issue in the same manner as was done for the large Pacific inversion of Toole and Robbins through the use of direct velocity of measurements within the 2°S−2°N latitude band. In the Pacific inversion these are time averaged underway ADCP absolute velocity profiles which are mapped to the mid-point positions between station pairs and used as the initial guess in place of the geostrophic estimates. However, in the Atlantic, coverage will not allow for the same technique to be used (G. Johnson pers. comm.).

The full Atlantic model is currently made up of about “small” 40 boxes (Figure 1), 10 large boxes (defined by the zonal transects, Figure 3) and a super box which constrains the entire basin. We will choose to combine some of the smaller boxes if it is found that aliasing of spatial and temporal variations causes the model to produce unrealistic flow patterns. Within this system the “large” boxes (Figure 3) used in conjunction with the “small” boxes (Figure 1) act to hold the constraints together and keep the noise in the small boxes from overwhelming the full solution. Comparing the carbon data along the zonal lines (Figure 4), one can see the high  TIC values in the upper waters of the tropics where upwelling of deep older waters occurs, low TIC values in the deep north due to the water formation and higher values as depth to south where abyssal waters are older. In the anthropogenic carbon (Cant) profiles one can again see the effect of ventilation in the high deep values in the north and the high to low intermediate water values in both hemispheres as one moves from the poles to the equator.

The full Atlantic model combines the WOCE-era data in a physically consistent fashion and presents a rather different picture (Figure 5) of present day TIC oceanic transport and divergence. In particular, because it includes conservation P:O2, the transport of TIC across the 24°N transect in vastly reduced, removing completely   the sense that there is strong uptake in the subpolar North Atlantic. The model also includes constraints requiring conservation of TIC made possible because the model includes air-sea flux terms. Combining the transport estimates with the air-sea flux estimates (Figure 6) provides more a complete picture of the meridional circulation of TIC. Most notably, there is a strong southward transport of TIC from the northern hemisphere across the equator supplied in part by a substantial uptake of carbon from the atmosphere between 47°N and 24°N. There is also strong uptake in the southern ocean south of 30°S. As might be expected, outgassing occurs in the boxes surrounding the equator. The interior source/sink terms are small compared to the air-sea exchange terms and some effort still needs to be made to understand their magnitude and to investigate possible vertical structure in their patterns.

Anthropogenic carbon (Cant) is not available along all the lines and so presents a less complete picture. However, what is abundantly clear in Figure 7 is 1) the pattern of a general northward flow of Cant throughout the Atlantic Basin, 2) some disagreement as to the magnitude of the Cant transport, particularly in the Northern Basin, but 3) the sense that the Cant transport  estimates are still small compared to the TIC estimates indicating that without temporal changesign circulation (which are possible) Cant is not yet going to reverse the sense of the Atlantic carbon transport compared to its pre-industrial state.

This project which has been funded 50% through NOAA and 50% through NSF is in its last year, but the analysis is not yet complete. However, we have recently obtained NSF funding to continue this work over the next three years and expect to publish the results from the full Atlantic model in the coming year.

Hansell. D. A., H. W. Ducklow , A. M. Macdonald and M. O'Neil Baringer, 2004, Metabolic poise in the North Atlantic Ocean diagnosed from organic matter transports, Limnology and Oceanography, 49, 1084−1094.

Macdonald A. M., M. O'Neil Baringer, R. Wanninkhof, K. Lee and D. W. R. Wallace, 2003, A 1998−1992 comparison of inorganic carbon and its transport across 24.5°N in the Atlantic. Deep Sea Research II, 50, 3041−3064.

Macdonald, A. M., North Atlantic CO2 transport and divergence, Invited talk, EGU Meeting, Nice, France, 2004.

Also presented at the Cooperative Institute for Climate and Ocean Research (CICOR) Executive Board Meeting, WHOI, May, 2004. Extended version given at the Wednesday PO lunch seminar in October 2004.

Macdonald, A. M., Oceanic CO2 Transport, Divergence and Air/Sea Exchange in the North Atlantic, 2nd Annual CARINA general meeting and open science conference, 2003. Also presented at the NSF NACP PI Meeting, May 2003.

Macdonald, A. M., M. O'Neil Baringer, D. W. R. Wallace and R. Wanninkhof, Carbon Transport at 24.5°N in the Atlantic, Presented by R. Wanninkhof at the CIMAS external review, 2003.

Macdonald, A. M., CO2 Transport and Divergence in the North Atlantic, Seminar at AOML in Miami, March 2005.

Macdonald, A. M., Ocean Carbon Transport, talk presented at the CARBOOCEAN data synthesis workshop in Laugarvatn, Iceland June 2006

Alvarez, M., A.F. Rios, F.F. Pérez, H.L. Bryden, and G. Roson,, 2003, Transport and budgets of total inorganic carbon in the subpolar and temperate North Atlantic, Global Biogeochemical Cycles, 17, 10.1029/2002GB001881.

Anderson, L.A., and J.L. Sarmiento, 1994, Redfield ratios of remineralization determined by nutrient data analysis, Global Biogeochemical Cycles, 8, 65−80.

Ganachaud, A.S., 1999, Large-scale oceanic circulation and fluxes of freshwater, heat, nutrient and oxygen, Ph.D. thesis, Massachusetts Institute of Technology/Woods Hole Oceanographic Institution Joint Program in Oceanography, Cambridge, Mass., 266 pp.

Ganachaud, A. and C. Wunsch, 2000, Oceanic meridional overturning circulation, mixing, bottom water formation rates and heat transport, Nature, 408, 453-456.

Holfort, J., K.M. Johnson, B. Schneider, G. Siedler, and D.W.R.Wallace, 1998, The meridional transport of dissolved inorganic carbon in the South Atlantic Ocean, Global Biogeochemical Cycles, 12(3), 479−499.

Key, R.M., A. Kozyr, C.L. Sabine, K. Lee, R. Wanninkhof, J.L. Bullister, R.A. Feely, F.J. Millero, C. Mordy and T.-H. Peng, 2004, A global ocean carbon climatology: Results from Global Data Anlaysis Project (GLODAP), Global Biogeochemical Cycles, 18, DOI:10.1029/2004GB002247.

Lee, K., S.-D. Choi, G.-H. Park, R. Wanninkhof, T.-H. Peng, R.M. Key, C.L. Sabine, R.A. Feely, J.L. Bullister, F.J. Millero, and A. Kozyr, 2003, An updated anthropogenic CO2 inventory in the Atlantic Ocean, Global Biogeochemical Cycles, 17(4),1116, DOI:10.1029 /2003GB002067.

Lumpkin, R., and K. Speer, 2003, Large-scale vertical and horizontal circulation in the North Atlantic Ocean, Journal of Physical Oceanography, 33, 1902−1920.

Lumpkin, R., K. Speer, and K.P. Koltermann, 2006, Transport across 48oN in the North Atlantic Ocean, Submitted.

Lundberg, L., and P.M. Haugan, 1996, A Nordic Seas-Arctic Ocean carbon budget from volume flows and inorganic carbon data, Global Biogeochemical Cycles, 10, 493−510.

Macdonald, A.M., M.O. Baringer, R. Wanninkhof, K. Lee, and D.W.R. Wallace, 2003, A 1998−1992 comparison of inorganic carbon and its transport across 24.5°N in the Atlantic, Deep-Sea Research II, 50, 3041−3064.

Mikaloff Fletcher, S.E. , N. Gruber, A.R. Jacobson, S.C. Doney, S.Dutkiewicz, M. Gerber, M. Follows, F. Joos, K. Lindsay, D. Menemenlis, A. Mouchet, S.A. Miller and J.L. Sarmiento, 2006, Inverse Estimates of Anthropogenic CO2 Uptake, Transport and Storage by the Ocean, Global Biogeochem. Cycles, 20, doi:10.1029/2005GB002530.

Redfield, A.C., B.H. Ketchum, and F.A. Richards, 1963, The influence of organisms on the comparison of sea-water, In: Hill, M. N. (Ed.), The Sea. Vol. 2, Wiley-Interscience, New York, pp. 26−77.

Rosón, G., A.F. Rios, A. Lavín, H.L. Bryden and F.F Pérez, 2002, Carbon distribution, fluxes and budgets in the subtropical North Atlantic, Journal of Geophysical Research, 108(C5), DOI:10.1029/1999JC000047.

Schott, F.A., R. Zantopp, L. Stramma, M. Dengler, J. Fischer and M. Wibaux, 2004, Circulation and deep-water export at the western exit of the Subpolar North Atlantic, Journal of Physical Oceanography, 34, 817−843.

Wijffels, S.E., 1993, Exchanges between hemispheres and gyres: a direct approach to the mean circulation of the equatorial Pacific, Ph.D. thesis, Massachusetts Institute of Technology/Woods Hole Oceanographic Institution Joint Program in Oceanography, WHOI-93-42, 267 pp.

Wunsch, C., 1996. The Ocean Circulation Inverse Problem. Cambridge University Press 442 pp.

I have had little specific recent interaction with NOAA on this project although I do stay in contact with R. Wannikhof and M. Baringer at AOML. I also recently discussed these results with R. Feely at (PMEL)

Summary of Education and Outreach Activity
Over the past year I was in contact with teachers from three different Massachusetts middle schools over the past year, but was unable to make dates to actually visit the classrooms. Nevertheless, I have stayed involved with COSEE New England. I have attended two Telling Your Story (TYS) workshops (one at the Univ. of CT, the other at Univ. of Mass. Boston). At the former I participated as a panel member and at the latter I also gave an example TYS presentation to a group of Boston middle school students. I have also agreed to join the COSEE New England team.

Last updated: August 19, 2008

whoi logo

Copyright ©2007 Woods Hole Oceanographic Institution, All Rights Reserved, Privacy Policy.
Problems or questions about the site, please contact