GEOPHYSICAL RESEARCH LETTERS, VOL. 32, L01602, doi:10.1029/2004GL021346, 2005
The importance of continental margins in the global carbon cycle
Frank E. Muller-Karger,1 Ramon Varela,2 Robert Thunell,3 Remy Luerssen,1
Chuanmin Hu,1 and John J. Walsh1
Received 26 August 2004; revised 19 October 2004; accepted 1 December 2004; published 6 January 2005.
[ 1 ] Approximately half of the world’s net annual
photosynthesis occurs in the oceans (48 Pg C y 1).
Areas bordering continents (bottom <2000 m) support 10–
15% of this production. We used satellite data to compute
annual global net primary production (1998 – 2001), and
derived the global particulate organic carbon (POC) flux
settling below the permanent thermocline and to the seafloor
using an empirical model of POC remineralization.
Approximately 0.68 Pg C y 1 sink below the thermocline
on continental margins, compared to 1.01 Pg C y 1 in the
deep ocean. Over 0.62 Pg C y 1 settles to the seafloor on
margins, compared to 0.31 Pg C y 1 to deep ocean
sediments. At least 0.06 Pg C y 1 may be buried in
sediments on margins. Therefore, margins may be
responsible for >40% of the carbon sequestration in the
ocean. These regions must be accounted for in realistic
models of the global carbon cycle and its linkages to climate
change. Citation: Muller-Karger, F. E., R. Varela, R. Thunell,
R. Luerssen, C. Hu, and J. J. Walsh (2005), The importance of
continental margins in the global carbon cycle, Geophys. Res. Lett.,
32, L01602, doi:10.1029/2004GL021346.
1. Background
[2] There is still a great deal of discussion about whether
primary production in coastal zones is exported or respired
[Gattuso et al., 1998]. Nearshore areas are generally considered to be heterotrophic, where CO2 sequestration is
thought to be small because of respiration and denitrification. It has been estimated to be of the order of 0.2 Pg C y 1
[Smith and Hollibaugh, 1993; Walsh, 1991]. Given the
small surface area of continental margins relative to the
area of the global ocean, and the complexity of oceanographic processes on these margins, they are frequently
viewed as ‘‘coastal’’ environments. Furthermore, the difficulty in increasing the spatial resolution of global numerical
models has led to a tendency to ignore primary production
along margins altogether. Boundary conditions for global
ocean carbon cycle models are typically set at the shelf
break, with no exchange between the deep ocean and
continental shelf waters assumed. Past efforts to assess
global ocean primary production using models or historical
data also aggregate analyses into 1° 1° resolution and are
1
College of Marine Science, University of South Florida, St. Petersburg,
Florida, USA.
2
Estación de Investigaciones Marinas de Margarita, Fundación La Salle
de Ciencias Naturales, Isla de Margarita, Venezuela.
3
Department of Geological Sciences, University of South Carolina,
Columbia, South Carolina, USA.
Copyright 2005 by the American Geophysical Union.
0094-8276/05/2004GL021346$05.00
restricted to oceanic depths greater than 200 m [see, e.g.,
Gregg et al., 2003].
[3] Therefore, two relevant questions are: How good are
global carbon assessments if continental margins are
omitted? And, how much particulate organic carbon
(POC) reaches ocean depths where it may be considered
‘‘sequestered’’?
[4] New production [Dugdale and Goering, 1967] by
phytoplankton, i.e., that amount of CO2 fixed during photosynthesis that is available for export from the euphotic zone,
is estimated to be about 2.9 Pg C y 1 in the deep ocean
[Walsh, 1991; Liu et al., 2000] (>2000 m water depth, a total
of 31 107 km2). Over the smaller area of the world’s
continental shelves and slopes (<2000 m depth, 4.8 107 km2), new production is estimated at 3.7 Pg C y 1
[Walsh, 1991; Liu et al., 2000; Jahnke et al., 1999]. These
estimates are based on carefully made but sparse measurements that do not account for variation in time. They suggest
that margins play an important role in the global carbon
budget, but the question remains as to whether this carbon is
sequestered or not. There are relatively few observations of
the vertical flux of settling particles in the deep ocean
[Francois et al., 2002; Lutz et al., 2002; Jahnke, 1996;
Deuser et al., 1990; Conte et al., 2001] and on margins
[Jahnke et al., 1999; Honjo, 1982; Dymond and Lyle, 1994;
Thunell et al., 1996, 2000; Muller-Karger et al., 2004]. These
studies all observe an extremely rapid decrease in the flux of
organic particles with depth in the upper water column
[Francois et al., 2002; Lutz et al., 2002], which leaves only
a very small fraction to reach the average depths of the ocean
(about 4,000 m) [Jahnke, 1996]. In this study we use these
historical studies as a basis to examine the importance of
carbon sequestration via the biological pump [Volk and Liu,
1988] on continental margins relative to the deep ocean.
2. Methods
[5] We used satellite data to estimate global POC fluxes
to the sea floor (depths >50 m). Sea surface temperature
(SST, °C), surface chlorophyll-a concentration (chl, mg
m 3), and Photosynthetically Active Radiation (PAR, mol
photons m 2 day 1) were derived from global satellite
observations (1998 – 2001). Monthly averages of SST
derived from the Advanced Very High Resolution Radiometer (AVHRR) were obtained from NASA’s Jet Propulsion
Laboratory Physical Oceanography Distributed Active
Archive Center [Vazquez et al., 1998]. Chlorophyll and
PAR were derived from the Sea-viewing Wide Field-ofview Sensor (SeaWiFS) Level-3 products, obtained from
NASA’s Goddard Space Flight Center (GSFC DAAC;
Version-4 chl and PAR [McClain et al., 1998; R. Frouin
et al. (2001); algorithm to estimate PAR from SeaWiFS
data available at http://daac.gsfc.nasa.gov/data/dataset/
L01602
1 of 4
MULLER-KARGER ET AL.: CONTINENTAL MARGINS AND CARBON CYCLE
L01602
Table 1. Annual Net Primary Production (NPP) and Particulate
Organic Carbon (POC) Flux in the World’s Oceans (Waters >50 m)a
Net PP [Pgb]
Flux [Pg]
C buried [Pg]
Global
(Flux to
Bottom)
Deep Water
(Flux to
Bottom)
Deep Water
(Flux to
800 m)
Margins
(Flux to
Bottom)
Margins
(Flux to
500 m)
47.91
0.93
0.15
38.92
0.31
0.09
1.01
8.99
0.62
0.06
0.68
a
Shown are POC flux to 800 m in the deep ocean and to 500 m on margins
(waters with bottom depth 50 – 2000 m), and to the bottom. Burial rates
estimated at 30% of the bottom flux in the deep ocean and 10% over margins.
b
1 Pg = 1015 g.
SEAWIFS/01_Data_Products/03a_PAR/index.html]).
These satellite data were mapped to 9 9 km2 per pixel at
the equator. To minimize errors in chlorophyll estimates
due to bottom reflection or other optical problems in
coastal waters, assessments were derived only for waters
deeper than 50 m. Net primary production (NPP) is an
estimate of the carbon fixed by photosynthesis minus plant
respiration in the euphotic zone. We computed monthly
mean global ocean NPP using the Vertically Generalized
Production Model (VGPM) [Behrenfeld and Falkowski,
1997]. The model estimates depth-integrated NPP
(gC m 2 day 1) based on surface chlorophyll (mg m 3),
surface PAR, and SST. The accuracy of the VGPM model
has been tested [Campbell et al., 2002], and it has been
used to study global primary production over short
[Behrenfeld et al., 2001] and decadal scales [Gregg et
al., 2003].
[6] Sinking POC flux was computed using an exponential decay model [Pace et al., 1987] (flux(Z) = 3.523*
NPP*Z 0.734). This model relates NPP to the depthdependent POC flux determined as part of the VERTEX
program in the northeastern Pacific [Martin et al., 1987]. A
number of open ocean studies have demonstrated that the
export ratio (the proportion of POC flux to integrated
euphotic zone net primary production) decreases exponentially with depth, due to decomposition and consumption of
the POC. Several empirical models have been derived from
these studies [Francois et al., 2002; Lutz et al., 2002; Pace
et al., 1987; Suess, 1980; Betzer et al., 1984]. The Pace et
al. [1987] model yielded the best fit to primary production
and sediment trap data from the Cariaco Basin [Thunell et
al., 2000; Muller-Karger et al., 2004].
[7] On shelves, only a relatively small fraction of the
POC moves laterally as it sinks [Etcheber et al., 1996; Peña
et al., 1999]. The statistical funnel [Deuser et al., 1990;
Siegel and Deuser, 1997] on margins is very shallow
compared to that of the deep ocean. For example, particles
settling at 100 m d 1 would intercept the shelf (<200 m) in
less than 2 days. Assuming horizontal velocities of about
20 cm s 1 throughout the water column, these particles
would travel laterally no more than about 20 km in a day, or
about two pixels in our images.
[8] Lateral transport of resuspended material at middepths has been observed in the mid-Atlantic bight off the
U.S. east coast, but only 1 – 6% of shelf production is
transported laterally offshore [Falkowski et al., 1994]. Also,
there are some areas where surface upwelling or river
plumes extend beyond the edge of the continental margin
(e.g., Peru and Gulf of Guinea upwelling, Amazon and
L01602
Orinoco River plumes). This leads to slightly higher fluxes
estimated in the deep ocean.
[9] We do not consider dissolved organic carbon (DOC)
and its role in the global carbon cycle, since its removal
from oxidation sites of surface water depends mainly upon
sinking water, not sinking POC [Walsh et al., 1992].
[10] To derive a first-order estimate of the POC flux to
the bottom, we constrained flux using a global digital
bathymetry database [Smith and Sandwell, 1997]. Continental margins were defined as regions where the bottom
lies between 50 and 2000 m, and the deep ocean where the
bottom is >2000 m.
[11] We assumed that carbon sequestration via the biological pump occurs when POC sinks below the permanent
thermocline, where the age of water is between 250 and
>500 years [Stuiver et al., 1983]. We defined the bottom of
the permanent thermocline as 800 m in the deep ocean and
500 m on continental margins [Ostlund et al., 1987].
[12] To estimate the amount of carbon buried in sediments, we assumed an average of 30% POC burial efficiency
in the deep sea [Jahnke, 1996; Dymond and Lyle, 1994] and
an arbitrary 10% on margins (R. A. Jahnke, personal
communication, 2004).
3. Results and Discussion
[13] Global ocean annual net primary production averaged 47.91 Pg C (1998– 2001; interannual variation <±2%;
Table 1 and Figure 1). We estimated an average of 8.99 Pg
C y 1 of NPP over continental margins (interannual variability <±2%). Previous estimates [Behrenfeld et al., 2001;
Gregg et al., 2003; Morel and Antoine, 2002] for the deep
ocean derived with the VGPM using the Coastal Zone Color
Scanner (CZCS), SeaWiFS, and AVHRR satellite data are
similar to those obtained in this study (38.92 Pg C y 1 NPP)
within about 20%. The differences are well within variation
associated with disparity in methods used to process the
SeaWiFS data.
[14] In the deep ocean, the total amount of POC that
reached 800 m was 1.01 Pg C y 1 (±2% interannual
variation). Over continental margins, approximately 0.68 Pg
C y 1 fell below 500 m (±1% interannual variation). Therefore, the oceanic biological pump sequesters 60% of the
carbon in the deep ocean and 40% on continental margins.
Clearly, carbon regenerated below 500 m on margins could
also move laterally into the deep ocean along isopycnals.
[15] Although the Pace et al. [1987] relationship may not
be applicable globally [Francois et al., 2002; Lutz et al.,
2002] and factors such as ballasting influence the efficiency
of organic carbon flux to the sea floor [Francois et al.,
2002; Armstrong et al., 2002], our results do provide a firstorder approximation of carbon sequestration in continental
margins relative to the deep ocean. There is evidence to
suggest that carbon flux along continental margins may
exceed that predicted from empirical models derived from
sediment trap data [Jahnke et al., 1990]. Also, sediment
traps may undersample POC flux in the mesopelagic zone
by as much as 40% [Francois et al., 2002; Lutz et al.,
2002]. Thus, our estimate of continental margin carbon flux
must be considered as a minimum value.
[16] The four-year average global POC flux that escaped
oxidation during descent to the bottom was 0.93 Pg C y 1
2 of 4
L01602
MULLER-KARGER ET AL.: CONTINENTAL MARGINS AND CARBON CYCLE
L01602
Figure 1. Average annual particulate organic carbon flux [gC m 2 y 1] deposited on the ocean bottom between 1998 and
2001, with continental margins outlined in white at the 2000 m depth contour.
(<±1% interannual variation). About 0.31 Pg C y 1 (±3%
interannual variation) were deposited on the sea floor in the
deep ocean, and about 0.62 Pg C y 1 (±2% interannual
variation) on margins (Table 1). Figure 1 shows that POC
flux to the seafloor over much of the deep sea is very small
(1.5 gC m 2 y 1). The largest flux (>3 gC m 2 y 1)
occurs under major divergences (equatorial upwelling zones
and upwelling areas along eastern ocean margins), under
the South Atlantic and Southern Indian Ocean Currents,
beneath the north Pacific convergence, and on the continental slopes and shelves. Seamounts and substantial portions of the mid-ocean ridges also show high bottom POC
fluxes, especially south of Iceland (>4 gC m 2 y 1); these
topographic highs received over twice the amount of POC
derived from surface production as the surrounding sea
floor. The POC flux to the bottom of continental margins
(>5 g C m 2 y 1) exceeded anything estimated for the deep
ocean. Thus, margins are characterized by both high production and high deposition of POC.
[17] It has been estimated that 0.13– 0.16 Pg C y 1 are
buried in the oceans, with 80– 85% of this occurring along
continental shelves and deltas [Berner, 1992; Hedges and
Keil, 1995]. Our calculated global organic carbon burial rate
of 0.15 Pg C (Table 1) matches those estimates. Our results
suggest that about 0.06 Pg C y 1 may be buried in margin
sediments deeper than 50 m (i.e., 40% of the organic carbon
stored annually in global marine sediments), and 0.09 Pg C
y 1 are buried in the deep sea. However, there is likely to be
significant variation in burial rates.
[18] A recent study [Gregg et al., 2003] suggested that
deep ocean NPP decreased 2.8 Pg C per decade between the
late 1980’s and the early 2000’s. This would lead to a
decrease of perhaps 0.007 Pg C y 1 in the carbon sequestered to >800 m in the deep open ocean. In contrast, just the
year-to-year variation in POC reaching the sea floor over
margins (±2%) is larger. Even a crude estimate of burial on
shelves is also 0.06 Pg C y 1.
[19] Our high-resolution results demonstrate that margins
play an important role in the global carbon cycle. Along
margins, natural and anthropogenic perturbations may have
an amplified effect on nutrient supply rates because of
vigorous exchanges of energy and matter between land,
atmosphere, and the ocean [Smith and Hollibaugh, 1993;
Walsh, 1991]. Ignoring the role of margins in the ocean’s
carbon cycle is a serious shortcoming of many carbon cycle
models that needs to be addressed.
[20] Acknowledgments. This work was supported by the National
Science Foundation, the National Aeronautics and Space Administration,
and the Consejo Nacional de Investigaciones Cientificas y Tecnologicas
(CONICIT, Venezuela). We thank the US SeaWiFS Project (NASA GSFC
Code 970.2), the Distributed Active Archive Center (NASA GSFC Code
902), and OrbImage for the production and distribution of the satellite data.
We also thank Ms. Haiying Zhang and Mr. Zhiqiang Chen for their help in
making the calculations described here. We are indebted to Dr. R. Jahnke
and to an anonymous reviewer for their many valuable comments on this
manuscript.
References
Armstrong, R. A., C. Lee, J. I. Hedges, S. Honjo, and S. G. Wakeham
(2002), A new, mechanistic model for organic carbon fluxes in the ocean
based on the quantitative association of POC with ballast minerals, Deep
Sea Res., Part II, 49, 219 – 236.
Behrenfeld, M. J., and P. G. Falkowski (1997), Photosynthetic rates derived
from satellite-based chlorophyll concentration, Limnol. Oceanogr., 42,
1 – 20.
Behrenfeld, M. J., et al. (2001), Biospheric primary production during an
ENSO transition, Science, 291, 2594 – 2597.
Berner, R. A. (1992), Comments on the role of marine sediment burial as a
repository for anthropogenic CO2, Global Biogeochem. Cycles, 6, 1 – 2.
Betzer, P., W. Showers, E. Laws, C. Winn, G. DiTullio, and P. Kroopnick
(1984), Primary production and particle fluxes on a transect of the equator at 153°W in the Pacific Ocean, Deep Sea Res., 31, 1 – 11.
Campbell, J., et al. (2002), Comparison of algorithms for estimating ocean
primary production from surface chlorophyll, temperature, and irradiance,
Global Biogeochem. Cycles, 16(3), 1035, doi:10.1029/2001GB001444.
Conte, M. H., N. Ralph, and E. H. Ross (2001), Seasonal and interannual
variability in deep ocean particle fluxes at the Oceanic Flux Program
(OFP)/Bermuda Atlantic Time-Series (BATS) site in the western Sargasso
Sea near Bermuda, Deep Sea Res., Part II, 48, 1471 – 1505.
Deuser, W. G., et al. (1990), Surface-ocean color and deep-ocean carbon
flux: How close a connection?, Deep Sea Res., 37, 1331 – 1343.
Dugdale, R. C., and J. J. Goering (1967), Uptake of new and regenerated
forms of nitrogen in primary productivity, Limnol. Oceanogr., 12, 685 –
695.
3 of 4
L01602
MULLER-KARGER ET AL.: CONTINENTAL MARGINS AND CARBON CYCLE
Dymond, J., and M. Lyle (1994), Particle fluxes in the ocean and implications for sources and preservation of ocean sediments, in Material Fluxes
on the Surface of the Earth, chap. 9, pp. 125 – 142, Natl. Res. Council,
Washington, D. C.
Etcheber, H., et al. (1996), Organic carbon fluxes and sediment biogeochemistry on the French Mediterranean and Atlantic margins, in Particle
Flux in the Ocean, SCOPE 57, edited by V. Ittekkot et al., chap. 12,
pp. 223 – 241, John Wiley, Hoboken, N. J.
Falkowski, P., P. Biscaye, and C. Sancetta (1994), The lateral flux of biogenic particles from the eastern North American continental margin to the
North Atlantic Ocean, Deep Sea Res., Part II, 41, 583 – 601.
Francois, R., S. Honjo, R. Krishfield, and S. Manganini (2002), Factors
controlling the flux of organic carbon to the bathypelagic zone of
the ocean, Global Biogeochem. Cycles, 16(4), 1087, doi:10.1029/
2001GB001722.
Gattuso, J.-P., M. Frankignoulle, and R. Wollast (1998), Carbon and carbonate metabolism in coastal aquatic ecosystems, Annu. Rev. Ecol. Syst., 29,
405 – 434.
Gregg, W. W., M. E. Conkright, P. Ginoux, J. E. O’Reilly, and N. W. Casey
(2003), Ocean primary production and climate: Global decadal changes,
Geophys. Res. Lett., 30(15), 1809, doi:10.1029/2003GL016889.
Hedges, J. I., and R. G. Keil (1995), Sedimentary organic matter preservation: An assessment and speculative synthesis, Mar. Chem., 49, 81 – 115.
Honjo, S. (1982), Seasonality and interaction of biogenic and lithogenic
particulate flux at the Panama Basin, Science, 218, 883 – 884.
Jahnke, R. A. (1996), The global ocean flux of particulate organic carbon:
Areal distribution and magnitude, Global Biogeochem. Cycles, 10, 71 –
88.
Jahnke, R. A., C. E. Reimers, and D. Craven (1990), Intensification of
recycling of organic matter at the sea floor near ocean margins, Nature,
348, 50 – 54.
Jahnke, R. A., J. R. Nelson, R. L. Marinelli, and J. E. Eckman (1999),
Benthic flux of biogenic elements on the southeastern U.S. continental
shelf: Influence of pore water advective transport and benthic microalgae,
Cont. Shelf Res., 20, 109 – 127.
Liu, K.-K., K. Iseki, and S.-Y. Chao (2000), Continental margin carbon
fluxes, in The Changing Ocean Carbon Cycle, Int. Geos. Bios.
Programme Book Ser., vol. 5, edited by R. B. Hanson, H. Ducklow,
and J. G. Field, chap. 7, pp. 187 – 239, Cambridge Univ. Press, New
York.
Lutz, M., R. Dunbar, and K. Caldeira (2002), Regional variability in the
vertical flux of particulate organic carbon in the ocean interior, Global
Biogeochem. Cycles, 16(3), 1037, doi:10.1029/2000GB001383.
Martin, J. H., G. A. Knauer, D. M. Karl, and W. Broenkow (1987), VERTEX:
Carbon cycling in the northeast Pacific, Deep Sea Res., 34, 267 – 285.
McClain, C. R., L. M. Cleave, G. C. Feldman, W. W. Gregg, S. B. Hooker,
and N. Kuring (1998), Science quality SeaWiFS data for global biosphere
research, Sea Technol., 39, 10 – 16.
Morel, A., and D. Antoine (2002), Small critters—Big effects, Science, 296,
1980 – 1982.
L01602
Muller-Karger, F. E., R. Varela, R. Thunell, E. Tappa, Y. Astor, H. Zhang,
and C. Hu (2004), Processes of coastal upwelling and carbon flux in the
Cariaco Basin, Deep Sea Res., Part II, in press.
Ostlund, H. G., H. Craig, W. S. Broecker, and D. Spencer (1987), GEOSECS
Atlantic, Pacific, and Indian Ocean Expeditions, vol. 7, Natl. Sci. Found.,
Washington, D. C.
Pace, M., G. Knauer, D. Karl, and J. Martin (1987), Primary production,
new production and vertical flux in the eastern Pacific Ocean, Nature,
325, 803 – 804.
Peña, M. A., K. L. Denman, S. E. Calvert, R. E. Thomson, and J. R. Forbes
(1999), The seasonal cycle in sinking particle fluxes off Vancouver Island,
British Columbia, Deep Sea Res., Part II, 46, 2969 – 2992.
Siegel, D. A., and W. G. Deuser (1997), Trajectories of sinking particles in
the Sargasso Sea: Modeling of statistical funnels above deep-ocean sediment traps, Deep Sea Res., Part I, 44, 1519 – 1541.
Smith, S. V., and J. T. Hollibaugh (1993), Coastal metabolism and the
oceanic organic carbon balance, Rev. Geophys., 31, 75 – 89.
Smith, W. H. F., and D. T. Sandwell (1997), Global sea floor topography
from satellite altimetry ship depth soundings, Science, 277, 5334.
Stuiver, M., P. D. Quay, and H. G. Ostlund (1983), Abyssal 14C and the age
of the world oceans, Science, 219, 849 – 851.
Suess, E. (1980), Particulate organic carbon flux in the ocean-surface productivity and oxygen utilization, Nature, 288, 260 – 263.
Thunell, R., et al. (1996), Plankton response to physical forcing in the Gulf
of California, J. Plankton Res., 18, 2017 – 2026.
Thunell, R., et al. (2000), Organic carbon flux in an anoxic water column:
Sediment trap results from the Cariaco Basin, Limnol. Oceanogr., 45,
300 – 308.
Vazquez, J., K. Perry, and K. Kilpatrick (1998), NOAA/NASA AVHRR
oceans pathfinder sea surface temperature data set, user’s reference
manual, version 4.0, Phys. Oceanogr. Distrib. Active Arch. Cent., Jet
Propul. Lab., Pasadena, Calif.
Volk, T., and Z. Liu (1988), Controls of CO2 sources and sinks in the Earth
scale surface ocean: Temperature and nutrients, Global Biogeochem.
Cycles, 2, 73 – 89.
Walsh, J. J. (1991), Importance of continental margins in the marine biogeochemical cycling of carbon and nitrogen, Nature, 350, 53 – 55.
Walsh, J. J., K. L. Carder, and F. M. Muller-Karger (1992), Meridional
fluxes of dissolved organic matter in the North Atlantic Ocean, J. Geophys. Res., 97, 15,625 – 15,637.
C. Hu, R. Luerssen, F. E. Muller-Karger, and J. J. Walsh, University of
South Florida, College of Marine Science, St. Petersburg, FL 33701, USA.
([email protected])
R. Thunell, Department of Geological Sciences, University of South
Carolina, Columbia, SC 29208, USA.
R. Varela, Estación de Investigaciones Marinas de Margarita, Fundación
La Salle de Ciencias Naturales, Apartado 144 Porlamar, Isla de Margarita,
Venezuela.
4 of 4