7.5. Model Results and Discussion
135
and Fe in and out of the surface and bottom waters are a function of the
concentration gradient of the nutrient and the velocity of the flux (measured in m day"). Because of deep-ocean remineralization of particulate
matter, the concentration of C, N and Fe is higher in the deep ocean than at
the surface . This results in a positive flux of nutrients into the surface water.
Changes in ocean circulation patterns occurring over geologic time scales
more than likely alter the magnitude of this flux (Boyle 1988).
7.5. Model Results and Discussion
To run the model, we initialize it with data for nutrient concentrations and
atmospheric CO 2 that reflect modern levels. We assume an initial CO 2 concentration of the atmosphere of 10.9
m", and [pC0 2
] of the atmosphere
of 345 uatm . We set Ct for surface and bottom water to 2.19 mM ( =
219mol/100 m- 3 ) and 2.31 mM (= 7854 mol/3400 m- 3 ) , respectively . Initial
[Fe] in surface and bottom waters are set at 0.26 pM ( = 2.6 108 mol/lOO
m- 3 ) and 1.2 pM ( = 4.08 10- 8 mol/3400 m- 3 ) , respectively. Initial N0 -
concentrations in surface and bottom waters are 12.0
3
( =1.2 mol/100
m- 3 ) and 30
=002 mol/3400 m- 3 ) .
The results of the first scenario are derived for an aeolian Fe flux of
2.048835 107 mol/tnom? per year and are shown as Graph 1 in Figures
7.5 and 7.6. In these figure are also plotted the results of model runs that assume, respectively, a two-fold (Graph 2), four-fold (Graph 3), eight-fold
(Graph 4) and sixteen-fold (Graph 5) increase in the aeolian flux of Fe.
Graph 1 in Figures 7.5 and 7.6 assumes that in HNLC regions primary
productivity , and the amount of CO 2 that can be removed from the atmosphere, are set by the combined Fe fluxes from upwell ing and aeolian deposition. An increase in the total flux of Fe into the surface waters (Graphs
2-5 in Figures 7.5 and 7.6) will change the complex chem istry of iron in
seawater. Phytoplankton are thought to only access the free ion of Fe for
growth (Hudson & Morel 1993; Morel, Rueter & Price 1991; Sunda 1994).
The free-ion concentration of Fe in solution is a function of the solubility,
degree of complexation by organic ligands, absorption/ scaveng ing and
photo-reduction/oxidation reactions in the surface and deep ocean (Iohnson, Gordon & Coale 1997). Therefore, a realistic model of iron/nitrate interactions should include a component that calculates the various pools of
iron , both bioavailable and refractory. On longer time scales considered in
this model , we assume that the free-ion activity of Fe is a linear, increasing
function of total Fe in the system, so that any increase of ambient Fe will
lead to an increase in production. Increasing the complexity of the model
with respe ct to Fe speciation does not change its qualitative behavior.
Because HNLC regions are far removed from terrestrial sources of Fe and
continental-shelf sediments, riverine Fe fluxes and advective Fe inputs to surface waters are extremely low. Therefore, the primary inputs of Fe required
135
and Fe in and out of the surface and bottom waters are a function of the
concentration gradient of the nutrient and the velocity of the flux (measured in m day"). Because of deep-ocean remineralization of particulate
matter, the concentration of C, N and Fe is higher in the deep ocean than at
the surface . This results in a positive flux of nutrients into the surface water.
Changes in ocean circulation patterns occurring over geologic time scales
more than likely alter the magnitude of this flux (Boyle 1988).
7.5. Model Results and Discussion
To run the model, we initialize it with data for nutrient concentrations and
atmospheric CO 2 that reflect modern levels. We assume an initial CO 2 concentration of the atmosphere of 10.9
m", and [pC0 2
] of the atmosphere
of 345 uatm . We set Ct for surface and bottom water to 2.19 mM ( =
219mol/100 m- 3 ) and 2.31 mM (= 7854 mol/3400 m- 3 ) , respectively . Initial
[Fe] in surface and bottom waters are set at 0.26 pM ( = 2.6 108 mol/lOO
m- 3 ) and 1.2 pM ( = 4.08 10- 8 mol/3400 m- 3 ) , respectively. Initial N0 -
concentrations in surface and bottom waters are 12.0
3
( =1.2 mol/100
m- 3 ) and 30
=002 mol/3400 m- 3 ) .
The results of the first scenario are derived for an aeolian Fe flux of
2.048835 107 mol/tnom? per year and are shown as Graph 1 in Figures
7.5 and 7.6. In these figure are also plotted the results of model runs that assume, respectively, a two-fold (Graph 2), four-fold (Graph 3), eight-fold
(Graph 4) and sixteen-fold (Graph 5) increase in the aeolian flux of Fe.
Graph 1 in Figures 7.5 and 7.6 assumes that in HNLC regions primary
productivity , and the amount of CO 2 that can be removed from the atmosphere, are set by the combined Fe fluxes from upwell ing and aeolian deposition. An increase in the total flux of Fe into the surface waters (Graphs
2-5 in Figures 7.5 and 7.6) will change the complex chem istry of iron in
seawater. Phytoplankton are thought to only access the free ion of Fe for
growth (Hudson & Morel 1993; Morel, Rueter & Price 1991; Sunda 1994).
The free-ion concentration of Fe in solution is a function of the solubility,
degree of complexation by organic ligands, absorption/ scaveng ing and
photo-reduction/oxidation reactions in the surface and deep ocean (Iohnson, Gordon & Coale 1997). Therefore, a realistic model of iron/nitrate interactions should include a component that calculates the various pools of
iron , both bioavailable and refractory. On longer time scales considered in
this model , we assume that the free-ion activity of Fe is a linear, increasing
function of total Fe in the system, so that any increase of ambient Fe will
lead to an increase in production. Increasing the complexity of the model
with respe ct to Fe speciation does not change its qualitative behavior.
Because HNLC regions are far removed from terrestrial sources of Fe and
continental-shelf sediments, riverine Fe fluxes and advective Fe inputs to surface waters are extremely low. Therefore, the primary inputs of Fe required