Ecological Provinces in the Open Ocean
105
The coastal biome, as we have defined it in Chapter 6, shall therefore comprise not only
open and reasonably linear continental coastlines, but also marginal seas and archipelagos
where the islands are sufficiently large and numerous to modify the ocean circulation,
and important areas of shallow water. For instance, we shall include in this category
the Arctic, Indonesian, and Philippine archipelagos but not (for example) the oceanic
Marquesas or Cook Islands.
Our analysis will profit from careful comparisons between the three major oceans
and their marginal seas. We shall use the powerful comparative method of geographical
analysis and utilize the effects of differing dimension and boundary conditions between
different oceans and seas as natural experiments to understand regional oceanography
and ecology. Such apparently simple matters as the difference between how far Patagonia
and South Africa extend poleward have profound effects on the ecology of the tropical
Atlantic and Indian oceans.
Ecological Provinces in the Open Ocean
It was partly the availability of the global CZCS chlorophyll field that reactivated the search
for a satisfactory way of defining ecological provinces. Platt and Sathyendranath (1988)
suggested that calculations of primary production should be partitioned between biogeochemical provinces (BGCPs) within which photosynthetic parameters, and the form of
the chlorophyll profile, might be seasonably predictable. Their proposal rests on the fact
that regional differences exist in the photosynthetic response of phytoplankton to changes
in environmental conditions, probably associated with changes in the composition of the
phytoplankton. Following this proposal, but using different bio-optical criteria, Mueller
and Lang (1989) suggested how the northeast Pacific might be partitioned objectively
into provinces compatible with the concept of the BGCP.
At the same time, not every biological oceanographer thinks that ecological models
must be partitioned to achieve satisfactory global integration. Morel and Berthon (1989),
for instance, suggested that surface chlorophyll specifies subsurface chlorophyll distribution sufficiently well that a one-dimensional, vertically integrated model of primary
production may be scaled to the surface chlorophyll field and applied globally, as is
now done routinely by Behrenfeld and Falkowski. This is the approach now used in a
routine compilation of global maps of primary production using a vertically integrated
mode and is not so far from the concept of ecological continuity that is often assumed
when one-dimensional models of biological processes are incorporated into general circulation models. Ecological continuity requires either that the parameters of a biological
model remain constant everywhere or that they always respond in the same way to
changes in the environmental covariables such as temperature and nutrients. However,
biological responses to changes in environmental conditions are often species dependent
and, in nature, further complicated by species succession. Therefore, they may be highly
nonlinear.
In fact, Platt and Sathyendranath (1999) suggested that because the parameters descriptive of the phytoplankton profile respond to regional ocean physics they are “neither
globally uniform, nor invariant with season.” These, and physiological parameters (they
comment), are distributed in such a manner that the “functional structure of the ecosystem is determined everywhere by physical forcing, the variation in the physical forcing
from place to place, and by the associated biological response.” Consequently, “In the
ocean, these parameters are believed to be distributed in a manner that is not smoothly
continuous. Rather, they seem to have a piece-wise continuous distribution.”
Ideally, global fields of relevant data would carry information enabling us to set
objective boundaries between provinces or would demonstrate that a set of provinces
105
The coastal biome, as we have defined it in Chapter 6, shall therefore comprise not only
open and reasonably linear continental coastlines, but also marginal seas and archipelagos
where the islands are sufficiently large and numerous to modify the ocean circulation,
and important areas of shallow water. For instance, we shall include in this category
the Arctic, Indonesian, and Philippine archipelagos but not (for example) the oceanic
Marquesas or Cook Islands.
Our analysis will profit from careful comparisons between the three major oceans
and their marginal seas. We shall use the powerful comparative method of geographical
analysis and utilize the effects of differing dimension and boundary conditions between
different oceans and seas as natural experiments to understand regional oceanography
and ecology. Such apparently simple matters as the difference between how far Patagonia
and South Africa extend poleward have profound effects on the ecology of the tropical
Atlantic and Indian oceans.
Ecological Provinces in the Open Ocean
It was partly the availability of the global CZCS chlorophyll field that reactivated the search
for a satisfactory way of defining ecological provinces. Platt and Sathyendranath (1988)
suggested that calculations of primary production should be partitioned between biogeochemical provinces (BGCPs) within which photosynthetic parameters, and the form of
the chlorophyll profile, might be seasonably predictable. Their proposal rests on the fact
that regional differences exist in the photosynthetic response of phytoplankton to changes
in environmental conditions, probably associated with changes in the composition of the
phytoplankton. Following this proposal, but using different bio-optical criteria, Mueller
and Lang (1989) suggested how the northeast Pacific might be partitioned objectively
into provinces compatible with the concept of the BGCP.
At the same time, not every biological oceanographer thinks that ecological models
must be partitioned to achieve satisfactory global integration. Morel and Berthon (1989),
for instance, suggested that surface chlorophyll specifies subsurface chlorophyll distribution sufficiently well that a one-dimensional, vertically integrated model of primary
production may be scaled to the surface chlorophyll field and applied globally, as is
now done routinely by Behrenfeld and Falkowski. This is the approach now used in a
routine compilation of global maps of primary production using a vertically integrated
mode and is not so far from the concept of ecological continuity that is often assumed
when one-dimensional models of biological processes are incorporated into general circulation models. Ecological continuity requires either that the parameters of a biological
model remain constant everywhere or that they always respond in the same way to
changes in the environmental covariables such as temperature and nutrients. However,
biological responses to changes in environmental conditions are often species dependent
and, in nature, further complicated by species succession. Therefore, they may be highly
nonlinear.
In fact, Platt and Sathyendranath (1999) suggested that because the parameters descriptive of the phytoplankton profile respond to regional ocean physics they are “neither
globally uniform, nor invariant with season.” These, and physiological parameters (they
comment), are distributed in such a manner that the “functional structure of the ecosystem is determined everywhere by physical forcing, the variation in the physical forcing
from place to place, and by the associated biological response.” Consequently, “In the
ocean, these parameters are believed to be distributed in a manner that is not smoothly
continuous. Rather, they seem to have a piece-wise continuous distribution.”
Ideally, global fields of relevant data would carry information enabling us to set
objective boundaries between provinces or would demonstrate that a set of provinces