OCEAN CARBON SYSTEM, MODELING OF
S. C. Doney and D. M. Glover, Woods Hole
Oceanographic Institution, Woods Hole, MA, USA
& 2009 Elsevier Ltd. All rights reserved.
Introduction
Chemical species such as radiocarbon, chlorofluorocarbons, and tritium–
3
He are important tools
for ocean carbon cycle research because they can be
used to trace circulation pathways, estimate timescales, and determine absolute rates. Such species,
often termed chemical tracers, typically have rather
simple water-column geochemistry (e.g., conservative or exponential radioactive decay), and reasonably well-known time histories in the atmosphere or
surface ocean. Large-scale ocean gradients of nutrients, oxygen, and dissolved inorganic carbon reflect a
combination of circulation, mixing, and the production, transport, and oxidation (or remineralization) of organic matter. Tracers provide additional,
often independent, information useful in separating these biogeochemical and physical processes.
Biogeochemical and tracer observations are often
framed in terms of ocean circulation models, ranging
from simple, idealized models to full three-dimensional (3-D) simulations. Model advection and diffusion rates are typically calibrated or evaluated
against transient tracer data. Idealized models are
straightforward to construct and computationally
inexpensive and are thus conducive to hypothesis
testing and extensive exploration of parameter space.
More complete and sophisticated dynamics can be
incorporated into three-dimensional models, which
are also more amenable for direct comparisons with
field data. Models of both classes are used commonly
to examine specific biogeochemical process, quantify
the uptake of anthropogenic carbon, and study the
carbon cycle responses to climate change. All models
have potential drawbacks, however, and part of the
art of numerical modeling is deciding on the appropriate model(s) for the particular question at hand.
Carbon plays a unique role in the Earth’s environment, bridging the physical and biogeochemical
systems. Carbon dioxide (CO 2 ), a minor constituent
in the atmosphere, is a so-called greenhouse gas that
helps to modulate the planet’s climate and temperature. Given sunlight and nutrients, plants and some
microorganisms convert CO 2 via photosynthesis into
organic carbon, serving as the building blocks and
energy source for most of the world’s biota. The
concentration of CO 2 in the atmosphere is affected
by the net balance of photosynthesis and the reverse
reaction respiration on land and in the ocean.
Changes in ocean circulation and temperature can
also change CO 2 levels because carbon dioxide is
quite soluble in sea water. In fact, the total amount of
dissolved inorganic carbon (DIC) in the ocean is
about 50 times larger than the atmospheric inventory. The air–sea exchange of carbon is governed by
the gas transfer velocity and the surface water partial
pressure of CO 2 (pCO 2 ), which increases with
warmer temperatures, higher DIC, and lower alkalinity levels. The natural carbon cycle has undergone
large fluctuations in the past, the most striking during glacial periods when atmospheric CO 2 levels
were about 30% lower than preindustrial values.
The ocean must have been involved in such a large
redistribution of carbon, but the exact mechanism is
still not agreed upon.
Human activities, including fossil-fuel burning and
land-use practices such as deforestation and biomass
burning, are altering the natural carbon cycle. Currently about 7.5 Pg C yr
À 1 (1 Pg ¼ 10
15
g) are emitted
into the atmosphere, and direct measurements show
that the atmospheric CO 2 concentration is indeed
growing rapidly with time. Elevated atmospheric
CO 2 levels are projected to heat the Earth’s surface,
and the evidence for climate warming is mounting.
Only about 40% of the released anthropogenic carbon remains in the atmosphere, the remainder is
taken up in about equal portions (or 2 Pg C yr
À 1
) by
land and ocean sinks (Figure 1). The future magnitude of these sinks is not well known, however, and is
one of the major uncertainties in climate simulations.
Solving this problem is complicated because
human impacts appear as relatively small perturbations on a large natural background. In the ocean,
the reservoir of organic carbon locked up as living
organisms, mostly plankton, is only about 3 Pg C.
The marine biota in the sunlit surface ocean are quite
productive though, producing roughly 50 Pg of new
organic carbon per year. Most of this material is recycled near the ocean surface by zooplankton grazing or microbial consumption. A small fraction,
something like 10–20% on average, is exported to
the deep ocean as sinking particles or as dissolved
organic matter moving with the ocean circulation.
Bacteria and other organisms in the deep ocean feed
on this source of organic matter from above, releasing DIC and associated nutrients back into the
513
S. C. Doney and D. M. Glover, Woods Hole
Oceanographic Institution, Woods Hole, MA, USA
& 2009 Elsevier Ltd. All rights reserved.
Introduction
Chemical species such as radiocarbon, chlorofluorocarbons, and tritium–
3
He are important tools
for ocean carbon cycle research because they can be
used to trace circulation pathways, estimate timescales, and determine absolute rates. Such species,
often termed chemical tracers, typically have rather
simple water-column geochemistry (e.g., conservative or exponential radioactive decay), and reasonably well-known time histories in the atmosphere or
surface ocean. Large-scale ocean gradients of nutrients, oxygen, and dissolved inorganic carbon reflect a
combination of circulation, mixing, and the production, transport, and oxidation (or remineralization) of organic matter. Tracers provide additional,
often independent, information useful in separating these biogeochemical and physical processes.
Biogeochemical and tracer observations are often
framed in terms of ocean circulation models, ranging
from simple, idealized models to full three-dimensional (3-D) simulations. Model advection and diffusion rates are typically calibrated or evaluated
against transient tracer data. Idealized models are
straightforward to construct and computationally
inexpensive and are thus conducive to hypothesis
testing and extensive exploration of parameter space.
More complete and sophisticated dynamics can be
incorporated into three-dimensional models, which
are also more amenable for direct comparisons with
field data. Models of both classes are used commonly
to examine specific biogeochemical process, quantify
the uptake of anthropogenic carbon, and study the
carbon cycle responses to climate change. All models
have potential drawbacks, however, and part of the
art of numerical modeling is deciding on the appropriate model(s) for the particular question at hand.
Carbon plays a unique role in the Earth’s environment, bridging the physical and biogeochemical
systems. Carbon dioxide (CO 2 ), a minor constituent
in the atmosphere, is a so-called greenhouse gas that
helps to modulate the planet’s climate and temperature. Given sunlight and nutrients, plants and some
microorganisms convert CO 2 via photosynthesis into
organic carbon, serving as the building blocks and
energy source for most of the world’s biota. The
concentration of CO 2 in the atmosphere is affected
by the net balance of photosynthesis and the reverse
reaction respiration on land and in the ocean.
Changes in ocean circulation and temperature can
also change CO 2 levels because carbon dioxide is
quite soluble in sea water. In fact, the total amount of
dissolved inorganic carbon (DIC) in the ocean is
about 50 times larger than the atmospheric inventory. The air–sea exchange of carbon is governed by
the gas transfer velocity and the surface water partial
pressure of CO 2 (pCO 2 ), which increases with
warmer temperatures, higher DIC, and lower alkalinity levels. The natural carbon cycle has undergone
large fluctuations in the past, the most striking during glacial periods when atmospheric CO 2 levels
were about 30% lower than preindustrial values.
The ocean must have been involved in such a large
redistribution of carbon, but the exact mechanism is
still not agreed upon.
Human activities, including fossil-fuel burning and
land-use practices such as deforestation and biomass
burning, are altering the natural carbon cycle. Currently about 7.5 Pg C yr
À 1 (1 Pg ¼ 10
15
g) are emitted
into the atmosphere, and direct measurements show
that the atmospheric CO 2 concentration is indeed
growing rapidly with time. Elevated atmospheric
CO 2 levels are projected to heat the Earth’s surface,
and the evidence for climate warming is mounting.
Only about 40% of the released anthropogenic carbon remains in the atmosphere, the remainder is
taken up in about equal portions (or 2 Pg C yr
À 1
) by
land and ocean sinks (Figure 1). The future magnitude of these sinks is not well known, however, and is
one of the major uncertainties in climate simulations.
Solving this problem is complicated because
human impacts appear as relatively small perturbations on a large natural background. In the ocean,
the reservoir of organic carbon locked up as living
organisms, mostly plankton, is only about 3 Pg C.
The marine biota in the sunlit surface ocean are quite
productive though, producing roughly 50 Pg of new
organic carbon per year. Most of this material is recycled near the ocean surface by zooplankton grazing or microbial consumption. A small fraction,
something like 10–20% on average, is exported to
the deep ocean as sinking particles or as dissolved
organic matter moving with the ocean circulation.
Bacteria and other organisms in the deep ocean feed
on this source of organic matter from above, releasing DIC and associated nutrients back into the
513