Geoscience Reference
In-Depth Information
Small changes in this carbon pool, caused for exam-
ple by biological responses to ocean change, there-
fore have the potential to cause large changes in
atmospheric CO
2
concentration.
Based on the biological processes responsible for
carbon i xation, two biological carbon pumps can
be distinguished: (1) the organic carbon pump,
driven by photosynthetic CO
2
i xation, and (2) the
carbonate counter pump, generated by the forma-
tion of calcium carbonate (CaCO
3
) shell material by
calcifying plankton. While photosynthetic carbon
i xation lowers the CO
2
partial pressure in the
euphotic zone, causing a net l ux of CO
2
from the
atmosphere to the surface ocean, CaCO
3
precipita-
tion lowers
C
T
and total alkalinity in the surface
ocean, causing an increase in CO
2
partial pressure
(Chapter 2). Thus, the two biological carbon pumps
reinforce each other in terms of maintaining a verti-
cal
C
T
gradient whereas they counteract each other
with respect to their impact on air-sea CO
2
exchange
( Heinze
et al.
1991). The latter aspect has led to the
term 'counter pump'. With a global vertical l ux
of particulate organic carbon of approximately 10
Pg C yr
-1
compared with a CaCO
3
l ux of about 1 Pg
C yr
-1
the organic carbon pump clearly dominates
over the carbonate counter pump (Milliman 1993).
It is worth noting that dissolved organic carbon
produced in the surface layer and transported to
depth during deep-water formation, with an esti-
mated l ux of 2 Pg C yr
-1
( Hansell and Carlson 2002 ),
also contributes to the organic carbon pump.
equatorial upwelling regimes, where nutrient-rich
deep waters are transported into the mixed layer
promoting phytoplankton growth. Similarly, dis-
tinct biogeochemical provinces occur in coastal
upwelling regimes and along shallow continental
margins that are inl uenced by riverine input and/
or tidal forcing. The classical phytoplankton 'spring
bloom' is found in the North Atlantic, but not in
other subpolar waters (e.g. the subarctic north-east
Pacii c) where iron limitation prevents the complete
utilization of excess macronutrients (Martin and
Fitzwater 1988). In high-latitude polar systems, sea-
sonal light limitation restricts productivity over
much of the growing season, with large pulses of
productivity typically occurring over a single period
of the year (Arrigo
et al.
2008 ; Pabi
et al.
2008 ). Despite
some gross similarities between Arctic and Antarctic
marine ecosystems (e.g. the importance of seasonal
ice cover), large differences in physical circulation
and nutrient supply between these regions (both
macronutrients and trace metals) result in very dif-
ferent ecological and biogeochemical dynamics.
While a complete review of oceanic biomes is
well beyond the scope of this chapter (readers are
referred to the original work of Longhurst 1998 for
details), the broad concept is important for under-
standing the potential ecological effects of ocean
acidii cation. Productivity in the different ecological
domains is controlled by a unique combination of
physical, chemical, and biological factors (e.g. light,
macronutrients, trace metals, grazing, etc.), and
these factors will probably inl uence CO
2
-dependent
responses. For example, CO
2
-dependent growth of
phytoplankton is likely to differ in nutrient replete
versus nutrient-limited regions, and under condi-
tions of strong stratii cation versus deep vertical
mixing (see discussion below). Moreover, the
unique attributes of each ecological domain should
be considered when designing manipulative exper-
iments (see Section 6.4). A different approach should
be taken, for example, in designing manipulative
experiments in subtropical regions versus subpolar
waters. In the subtropics, experiments should aim
to mimic a tightly coupled, nutrient-limited envi-
ronment, since enrichment of macronutrients in
bottles would perturb the system into a new eco-
logical state. In contrast, steady-state approaches,
based on chemostat experiments (e.g. Sciandra
et al.
6.2.3 Biogeochemical provinces
Longhurst ( 1998 ) was the i rst to utilize satellite-
based observations of phytoplankton biomass
(based on inferred chlorophyll
a
distributions) to
characterize the broad-scale patterns of marine 'eco-
logical geography'. This work provided a concep-
tual and practical framework for classifying oceanic
regions into domains with spatially and temporally
coherent biological dynamics. The distribution of
these ecological domains is coupled with underly-
ing physical processes that drive chemical and bio-
logical gradients in the upper mixed layer. For
example, a clear distinction can be made between
the highly stratii ed subtropical gyre systems
(the 'trades biome'
sensu
Longhurst 1998 ) and the
