72
Chapter 5: Nutrient Limitation: The Example of Iron
The clearest statement of belief in Martin’s suggestion that I have yet seen was contained
in a Web-published report of an open-ocean iron-fertilization experiment. It read:
as water is the primary factor determining how much life can be sustained in a particular
terrestrial environment so the supply of iron is the primary determinant of how much
life can be sustained in the ocean. Dust whipped off the continents by winds supplies iron to
the open ocean, and the productive regions receive more “iron rain” than the unproductive
regions.
From there, it is but a short step to the astonishing evocation of a relationship between
seasonality of chlorophyll biomass and of modeled dust deposition in the Southern Ocean
(Erickson et al., 2003).
Such statements are, I believe, mythical, although similar assumptions had already
led Banse and English (1993) to go so far as to suggest that the HNLC areas should
be regarded as one of only three characteristic natural domains of an ocean partitioned
according to the forcing and seasonality of phytoplankton growth. To confront the tension
between myth and reality will require special consideration of these areas in what follows.
In the years when this concept was being developed, the global chlorophyll fields
obtained by the CZCS sensors were not so familiar to most biological oceanographers as
such products are today. Now, we can appreciate that the HNLC areas are not distinguishable in global images of sea-surface chlorophyll as anomalies inexplicable by ocean
physics and by reference to Sverdrup. We can readily see that the distribution of surface
chlorophyll (and the pattern of primary production derived from it) in no way matches
the subaerial supply of Fe, but rather reflects those physical processes that induce (i) vertical motion of deeper water into the photic zone, and (ii) stability in the upper part of the
water column. This relationship, foreseen by Yentsch and Garside (1986), is valid down to
the mesoscale, and it is reasonable to believe that the observed correspondence between
phytoplankton growth and physical processes, at all scales, represents cause and effect.
These images also show us that the term HNLC itself is unhelpful because the
“LC” part of it is downright misleading: chlorophyll accumulates in these regions in
the same pattern as in other places having comparable physical conditions, but where
macronutrients are taken down to very low levels. It is certainly time to abandon
the term: where we observe persistent unused nitrate in a sufficiently lit, sufficiently
stratified euphotic zone we may assume that the vertical flux of this element exceeds
the requirements of phytoplankton, whose population growth is limited by some other
element. For this relationship, the dimensionless S proposed by Platt et al. (2003),
where S = (nitrate supply rate)/(nitrate equivalent of new production), is appropriate. In
so-called HNLC regions, S must take a value higher than unity, and I shall subsequently
refer to these simply as the high-S regions.
In the first edition of this book I suggested that these areas were, by virtue of their
nutrient regimes, “exceptional regions” that required to be, to some degree, treated
independently of the four major biomes of the seas discussed in the next chapter. This was
not a good idea, and now I much prefer to emphasize how the pattern of productivity—
both spatial and temporal—within each high-S area is forced primarily by regional
physics; these areas, in fact, function very similarly to comparable areas elsewhere in the
biomes within which they are located.
Nutrient Distribution and the Consequences
of Differing Supply Ratios
So, why should nitrate characteristically remain in excess in surface waters in some regions
even after the season of phytoplankton growth? Consider the relative distribution of just
Chapter 5: Nutrient Limitation: The Example of Iron
The clearest statement of belief in Martin’s suggestion that I have yet seen was contained
in a Web-published report of an open-ocean iron-fertilization experiment. It read:
as water is the primary factor determining how much life can be sustained in a particular
terrestrial environment so the supply of iron is the primary determinant of how much
life can be sustained in the ocean. Dust whipped off the continents by winds supplies iron to
the open ocean, and the productive regions receive more “iron rain” than the unproductive
regions.
From there, it is but a short step to the astonishing evocation of a relationship between
seasonality of chlorophyll biomass and of modeled dust deposition in the Southern Ocean
(Erickson et al., 2003).
Such statements are, I believe, mythical, although similar assumptions had already
led Banse and English (1993) to go so far as to suggest that the HNLC areas should
be regarded as one of only three characteristic natural domains of an ocean partitioned
according to the forcing and seasonality of phytoplankton growth. To confront the tension
between myth and reality will require special consideration of these areas in what follows.
In the years when this concept was being developed, the global chlorophyll fields
obtained by the CZCS sensors were not so familiar to most biological oceanographers as
such products are today. Now, we can appreciate that the HNLC areas are not distinguishable in global images of sea-surface chlorophyll as anomalies inexplicable by ocean
physics and by reference to Sverdrup. We can readily see that the distribution of surface
chlorophyll (and the pattern of primary production derived from it) in no way matches
the subaerial supply of Fe, but rather reflects those physical processes that induce (i) vertical motion of deeper water into the photic zone, and (ii) stability in the upper part of the
water column. This relationship, foreseen by Yentsch and Garside (1986), is valid down to
the mesoscale, and it is reasonable to believe that the observed correspondence between
phytoplankton growth and physical processes, at all scales, represents cause and effect.
These images also show us that the term HNLC itself is unhelpful because the
“LC” part of it is downright misleading: chlorophyll accumulates in these regions in
the same pattern as in other places having comparable physical conditions, but where
macronutrients are taken down to very low levels. It is certainly time to abandon
the term: where we observe persistent unused nitrate in a sufficiently lit, sufficiently
stratified euphotic zone we may assume that the vertical flux of this element exceeds
the requirements of phytoplankton, whose population growth is limited by some other
element. For this relationship, the dimensionless S proposed by Platt et al. (2003),
where S = (nitrate supply rate)/(nitrate equivalent of new production), is appropriate. In
so-called HNLC regions, S must take a value higher than unity, and I shall subsequently
refer to these simply as the high-S regions.
In the first edition of this book I suggested that these areas were, by virtue of their
nutrient regimes, “exceptional regions” that required to be, to some degree, treated
independently of the four major biomes of the seas discussed in the next chapter. This was
not a good idea, and now I much prefer to emphasize how the pattern of productivity—
both spatial and temporal—within each high-S area is forced primarily by regional
physics; these areas, in fact, function very similarly to comparable areas elsewhere in the
biomes within which they are located.
Nutrient Distribution and the Consequences
of Differing Supply Ratios
So, why should nitrate characteristically remain in excess in surface waters in some regions
even after the season of phytoplankton growth? Consider the relative distribution of just