Nutrient Distribution and the Consequences of Differing Supply Ratios
73
two of the many essential plant nutrient molecules in the ocean: dissolved iron (Fe td
that passes a 04-m filter), and nitrogen as NO 3 . Each potentially limits phytoplankton
growth although at very different concentrations. A stoichiometric ratio for N:Fe cannot
be stated precisely, but it is thought to range from 0.06 to 20×10
4 for cells at optimal and
minimal growth rates respectively (Martin and Gordon, 1988). Where required, I shall
use the intermediate value of 15 × 10
4 proposed by Geider and LaRoche (1994).
Although nitrogen is abundant and readily accessible to plant cells in our present
era, the supply of iron is much more problematical, being an element whose abundant
reservoirs on Earth have been profoundly modified by biota. Oxygen, evolved by the
earliest photoautotrophic cyanobacteria, could accumulate in the atmosphere only after
the complete oxidation of free iron and its deposition as Fe 2 O 3 in the Banded Iron
Formations 2.5–3.0 billion years ago. The cellular chemistry of present-day microbiota
(and the cells of higher plants) retains the mark of their origin in an iron-rich environment; Fe is required in the elaboration of cytochromes and of the redox proteins involved
in photosynthesis, respiration, and nitrate reduction, for example.
A substantial proportion of dissolved iron now occurs in forms that are largely
unavailable to phytoplankton, principally together with strong Fe-complexing ligands
(Wells et al., 1994). Nevertheless, mechanisms exist to ensure that sufficient Fe can be
obtained by plant cells in today’s iron-poor environment, although cellular Fe:C ratios
may respond both to floristics and to dissolved iron concentration, unlike the relatively constant C:N:P ratio of Redfield (Fung et al., 2000). Iron-limitation for oceanic
phytoplankton typically occurs around 01 nM kg
−1 . Picoplanktonic Synechococcus are
not strongly Fe-limited, perhaps only because of their extremely small dimension, and
hence large surface area/volume ratio, even in equatorial high-N regimes. These cells also
produce Fe-binding siderophores that enable Fe acquisition even where dissolved iron
takes very low values indeed (Wells et al., 1994; Tortell et al., 1999). Some photosynthetic
cells (e.g., Ochromonas) are able to obtain Fe through ingestion of bacteria (Maranger
et al., 1998).
We should also perhaps recall here that Fe is recycled, like nitrogen, within the
microbial ecosystem of the mixed layer in the same manner, and with about the same
efficiency, as nitrogen (Bruland et al., 1991; Maranger et al., 1998). Indeed, for populations
of phytoplankton cells to be sustained in the mixed layer of an oligotrophic ocean, both
Fe and N (as well as other elements) must be recycled in situ. Indeed, some pre-Martin
observations of Fe excretion by mesozooplankton off India were used to support Cooper’s
1935 suggestion that regeneration of useful forms of oligonutrients in this way might be
critical for phytoplankton (Ramadhas and Venugopalam, 1977). Deeper than the mixed
layer, as shall be discussed later, Fe is regenerated by a mechanism that differs markedly
from that appropriate to the major nutrient molecules.
Iron presently enters the ocean from two sources: by wet or dry deposition of terrestrial dust everywhere at the sea surface, and by transport from shelf deposits or river
effluents into the adjacent water mass. The regional deposition of terrestrial dust at the
sea surface reflects aerosol trajectories from sources in arid lands, mostly from major
topographic depressions with dry alluvial soils. The mesoscale pattern within air masses,
and hence of dust deposition at the sea surface, is clearly different from the familiar
mesoscale, eddying circulation pattern in the ocean (see Color plate 2). The global pattern of dust deposition at the surface of the ocean is known only rather generally from
observations, so that many studies of the process have been based on the output from
models; these are perhaps regarded with too much confidence, for the uncertainties are
very great.
Nevertheless, simulation of sources and aerosol transport of Fe can now successfully
reproduce the seasonal pattern of accumulated haze observations at sea discussed by
Prospero (1981). The GOCART model represents the state of this art, but significantly
overestimates deposition in regions far from sources. The chemical composition of aerosol
73
two of the many essential plant nutrient molecules in the ocean: dissolved iron (Fe td
that passes a 04-m filter), and nitrogen as NO 3 . Each potentially limits phytoplankton
growth although at very different concentrations. A stoichiometric ratio for N:Fe cannot
be stated precisely, but it is thought to range from 0.06 to 20×10
4 for cells at optimal and
minimal growth rates respectively (Martin and Gordon, 1988). Where required, I shall
use the intermediate value of 15 × 10
4 proposed by Geider and LaRoche (1994).
Although nitrogen is abundant and readily accessible to plant cells in our present
era, the supply of iron is much more problematical, being an element whose abundant
reservoirs on Earth have been profoundly modified by biota. Oxygen, evolved by the
earliest photoautotrophic cyanobacteria, could accumulate in the atmosphere only after
the complete oxidation of free iron and its deposition as Fe 2 O 3 in the Banded Iron
Formations 2.5–3.0 billion years ago. The cellular chemistry of present-day microbiota
(and the cells of higher plants) retains the mark of their origin in an iron-rich environment; Fe is required in the elaboration of cytochromes and of the redox proteins involved
in photosynthesis, respiration, and nitrate reduction, for example.
A substantial proportion of dissolved iron now occurs in forms that are largely
unavailable to phytoplankton, principally together with strong Fe-complexing ligands
(Wells et al., 1994). Nevertheless, mechanisms exist to ensure that sufficient Fe can be
obtained by plant cells in today’s iron-poor environment, although cellular Fe:C ratios
may respond both to floristics and to dissolved iron concentration, unlike the relatively constant C:N:P ratio of Redfield (Fung et al., 2000). Iron-limitation for oceanic
phytoplankton typically occurs around 01 nM kg
−1 . Picoplanktonic Synechococcus are
not strongly Fe-limited, perhaps only because of their extremely small dimension, and
hence large surface area/volume ratio, even in equatorial high-N regimes. These cells also
produce Fe-binding siderophores that enable Fe acquisition even where dissolved iron
takes very low values indeed (Wells et al., 1994; Tortell et al., 1999). Some photosynthetic
cells (e.g., Ochromonas) are able to obtain Fe through ingestion of bacteria (Maranger
et al., 1998).
We should also perhaps recall here that Fe is recycled, like nitrogen, within the
microbial ecosystem of the mixed layer in the same manner, and with about the same
efficiency, as nitrogen (Bruland et al., 1991; Maranger et al., 1998). Indeed, for populations
of phytoplankton cells to be sustained in the mixed layer of an oligotrophic ocean, both
Fe and N (as well as other elements) must be recycled in situ. Indeed, some pre-Martin
observations of Fe excretion by mesozooplankton off India were used to support Cooper’s
1935 suggestion that regeneration of useful forms of oligonutrients in this way might be
critical for phytoplankton (Ramadhas and Venugopalam, 1977). Deeper than the mixed
layer, as shall be discussed later, Fe is regenerated by a mechanism that differs markedly
from that appropriate to the major nutrient molecules.
Iron presently enters the ocean from two sources: by wet or dry deposition of terrestrial dust everywhere at the sea surface, and by transport from shelf deposits or river
effluents into the adjacent water mass. The regional deposition of terrestrial dust at the
sea surface reflects aerosol trajectories from sources in arid lands, mostly from major
topographic depressions with dry alluvial soils. The mesoscale pattern within air masses,
and hence of dust deposition at the sea surface, is clearly different from the familiar
mesoscale, eddying circulation pattern in the ocean (see Color plate 2). The global pattern of dust deposition at the surface of the ocean is known only rather generally from
observations, so that many studies of the process have been based on the output from
models; these are perhaps regarded with too much confidence, for the uncertainties are
very great.
Nevertheless, simulation of sources and aerosol transport of Fe can now successfully
reproduce the seasonal pattern of accumulated haze observations at sea discussed by
Prospero (1981). The GOCART model represents the state of this art, but significantly
overestimates deposition in regions far from sources. The chemical composition of aerosol