451
tested with respect to their capability in order to
adequately respond to dynamic forcing by organic
matter supply (Soetaert et al. 1996), to calculate
the global benthic oxygen demand (Christensen
2000) or the complete oxic/anoxic cycle in a global
gridded domain in order to gain a better understanding of control parameters and kinetic rate
constants (Archer et al. 2002). Benthic models
coupled to the external forcing of more or less
complex pelagic models exist in great variety. In
general, the right balance between complexity and
calculation time has to be found. Soetaert et al.
(2000) tested various approaches of physical,
benthic-pelagic models with respect to oxygen,
carbon, and nutrient cycling and concluded that
at least dynamic sediment models without depth
integration are required to produce useful results.
In particular, the definition of source and sink
terms or the definition of imposed fluxes at the
sediment/water interface is inappropriate.
However, a certain number of models with a
layered benthic module exist that have been
tested in order to explore various overarching
themes related to global change.
The steady-state model output of a coupled
model simulating ocean circulation, biogeochemically-controlled surface ocean productivity,
export of POC and CaCO 3 , and sediment
diagenesis is shown in Figure 12.20 (Archer et al.
1998). Without going into detail on the background of the study, this was chosen as an
arbitrary example in order to show some potentials
and limitations of these kind of approaches. The
general patterns of export production (a,b) and
sediment accumulation (c,d) follow very much the
known data distributions, so that we can conclude
a general agreement between observation and
model prediction. Overall, the graphs show that it
is possible to calculate parameter distributions of
different compartments representing a consistent
overall system. However, it is evident that the
spatial resolution of these model is very low and
regional characteristics can only be considered
and hence, investigated to a limited extent. Up to
now, the major advantage of these models is to
investigate long-term trends controlled by a
number of internal and external forcings.
As outlined in Chapter 6, phosphate is
strongly linked to ocean productivity and its
availability on geological time scales is largely
controlled by the oxidation state of sediments and
bottom water (Van Cappellen and Ingall 1994).
According to their model, global-scale events of
anoxia/dysoxia would lead to a more efficient
recycling of phosphorous and thus, to an
enhancement of ocean productivity and organic
carbon export. The increased POC accumulation,
in turn, would further cause the depletion of
oxygen in the bottom water and thus, further
enhance the recycling of phosphate at the
seafloor and lead into a positive feedback loop.
This, general mechanism could be proven by
Wallmann (2003) who invented a more sophisticated model to simulate the marine cycles of
POC, oxygen, nitrate, and phosphorus. Two
additional aspects deserve to be mentioned in this
regard: (1) Significant enhancement of productivity requires sufficient supply of nitrate which
needs to be provided by N 2 -fixation. (2) Productivity of the glacial ocean could have been enhanced by the falling sea level as the marine regression favors an increased POC export to the deep
ocean.
Heinze et al. (1999) developed a 10-layer sediment module coupled to a global-ocean circulation
model (GCM) specifically designed to study the
interactions of Si and C cycles and capable to
handle important processes like the oceanatmosphere exchange of CO 2 , biogenic particle
export production, as well as sediment accumulation and pore water geochemistry. The central
outcome of the study is that one critical and very
sensitive parameter in order to obtain reasonable
fits to various geochemical data in ocean, atmosphere and sediment is the export production of
POC and the corresponding rain ratio (Si/C CaCO3 /
C Corg ). This is basically the same problem, which
complicates the prediction of benthic mineralization processes by empirically derived export
functions (see above). Another very important
factor is the solubility of opal and thus, the
release of silicic acid from the sediments. A simple
test of halving or doubling the solubility in the
pore water leads to a significant increase or
decrease of atmospheric pCO 2 -levels (by either
enhancing or outcompeting CaCO 3 production).
This emphasizes the inherent coupling of Si and
C cycles and the capability of coupled models to
produce predictions, which would be possible to
deduce from isolated analyses of the specific
environmental compartments. The strong potential of drawing down atmospheric pCO 2 -levels by
significantly increasing Si-concentrations in the
ocean could be approved by subsequent studies
(Archer et al. 2000; Ridgwell 2002). Additionally, a
higher turnover rate of Si in the ocean (Heinze et
12.5
Applications
tested with respect to their capability in order to
adequately respond to dynamic forcing by organic
matter supply (Soetaert et al. 1996), to calculate
the global benthic oxygen demand (Christensen
2000) or the complete oxic/anoxic cycle in a global
gridded domain in order to gain a better understanding of control parameters and kinetic rate
constants (Archer et al. 2002). Benthic models
coupled to the external forcing of more or less
complex pelagic models exist in great variety. In
general, the right balance between complexity and
calculation time has to be found. Soetaert et al.
(2000) tested various approaches of physical,
benthic-pelagic models with respect to oxygen,
carbon, and nutrient cycling and concluded that
at least dynamic sediment models without depth
integration are required to produce useful results.
In particular, the definition of source and sink
terms or the definition of imposed fluxes at the
sediment/water interface is inappropriate.
However, a certain number of models with a
layered benthic module exist that have been
tested in order to explore various overarching
themes related to global change.
The steady-state model output of a coupled
model simulating ocean circulation, biogeochemically-controlled surface ocean productivity,
export of POC and CaCO 3 , and sediment
diagenesis is shown in Figure 12.20 (Archer et al.
1998). Without going into detail on the background of the study, this was chosen as an
arbitrary example in order to show some potentials
and limitations of these kind of approaches. The
general patterns of export production (a,b) and
sediment accumulation (c,d) follow very much the
known data distributions, so that we can conclude
a general agreement between observation and
model prediction. Overall, the graphs show that it
is possible to calculate parameter distributions of
different compartments representing a consistent
overall system. However, it is evident that the
spatial resolution of these model is very low and
regional characteristics can only be considered
and hence, investigated to a limited extent. Up to
now, the major advantage of these models is to
investigate long-term trends controlled by a
number of internal and external forcings.
As outlined in Chapter 6, phosphate is
strongly linked to ocean productivity and its
availability on geological time scales is largely
controlled by the oxidation state of sediments and
bottom water (Van Cappellen and Ingall 1994).
According to their model, global-scale events of
anoxia/dysoxia would lead to a more efficient
recycling of phosphorous and thus, to an
enhancement of ocean productivity and organic
carbon export. The increased POC accumulation,
in turn, would further cause the depletion of
oxygen in the bottom water and thus, further
enhance the recycling of phosphate at the
seafloor and lead into a positive feedback loop.
This, general mechanism could be proven by
Wallmann (2003) who invented a more sophisticated model to simulate the marine cycles of
POC, oxygen, nitrate, and phosphorus. Two
additional aspects deserve to be mentioned in this
regard: (1) Significant enhancement of productivity requires sufficient supply of nitrate which
needs to be provided by N 2 -fixation. (2) Productivity of the glacial ocean could have been enhanced by the falling sea level as the marine regression favors an increased POC export to the deep
ocean.
Heinze et al. (1999) developed a 10-layer sediment module coupled to a global-ocean circulation
model (GCM) specifically designed to study the
interactions of Si and C cycles and capable to
handle important processes like the oceanatmosphere exchange of CO 2 , biogenic particle
export production, as well as sediment accumulation and pore water geochemistry. The central
outcome of the study is that one critical and very
sensitive parameter in order to obtain reasonable
fits to various geochemical data in ocean, atmosphere and sediment is the export production of
POC and the corresponding rain ratio (Si/C CaCO3 /
C Corg ). This is basically the same problem, which
complicates the prediction of benthic mineralization processes by empirically derived export
functions (see above). Another very important
factor is the solubility of opal and thus, the
release of silicic acid from the sediments. A simple
test of halving or doubling the solubility in the
pore water leads to a significant increase or
decrease of atmospheric pCO 2 -levels (by either
enhancing or outcompeting CaCO 3 production).
This emphasizes the inherent coupling of Si and
C cycles and the capability of coupled models to
produce predictions, which would be possible to
deduce from isolated analyses of the specific
environmental compartments. The strong potential of drawing down atmospheric pCO 2 -levels by
significantly increasing Si-concentrations in the
ocean could be approved by subsequent studies
(Archer et al. 2000; Ridgwell 2002). Additionally, a
higher turnover rate of Si in the ocean (Heinze et
12.5
Applications