Primitive Equation Modelling ofPlankton Eeosystems
415
cies (Ducklow and Harris 1993). And while one group of plankton, diatoms, may
dominate the spring bloom overall, they stop reproduc ing when the supply of silieate is exhausted. Other species can then flourish until the nitrogen runs out. Later
in the year, the oligotrophic environment gives a temporary competitive advantage
to summer species. Notably dinoflagellates, which have evolved to grow in the seasonal thermocline, where they move up and down, getting nutrients below the
nutricline, and energy above it where the light is brighter. Similar specializations
exist in the zooplankton.
We showed earlier (18.9) how the changing abundances of competing species are
simulated in Virtual Plankton Ecosystems. We were content to perform numerical
experiments with idealized diatoms for that technical demonstration. The next step
is to simulate competition between real species. Within twenty years we shall have
computers that permit us to perform LE integrations with billions of particles. That
will allow us to create VPEs with a thousand populations, each represented by one
million particles (in a one-dimensional simulation). By the time marine biologists
have produced biologic al primitive equations for each of those species, we shall
already have in place the modelling technology to simulate demographic changes
in the plankton ecosystem at the level of species.
The value of such biodiverse simulations will depend on the stability, demographic noise and errors in the computation. Acquiring knowledge about the population ecology of biodiverse VPEs will take many years. Meanwhile, we shall
move ahead cautiously from the bridgehead secured with the WB food chain
model, and its variants described in 18.9. Our biodiversity strategy for the next
decade will be progressively to increase the number ofpopulations ofphytoplankton and zooplankton. In doing so, we shall have to as sume that each population is
representative ofhundreds or even thousands of species. The question to be considered here is how can we relate the populations in our models to aH the species they
are meant to represent? The answer lies in the concept of "functional groups"
(sometimes called "guilds", see May 1981).
18.16.1 Functional groups
AH the species in a functional group share the same biological primitive equations. Within the group, individual species and varieties are distinguished by the
values of the parameters found in those equations. Species in different functional
groups have different physiology and behaviour, and therefore different BPEs. This
is obviously true between trophic levels: only phytoplankton have equations for
photosynthesis, and only zooplankton have equations for foraging. But there are
also distinct functional groups within the same trophic level. For example, diatoms
use silicon, while other phytoplankton do not. And some phytoplankton can control
their motion through the water, others cannot.
Different functional groups have different equations, and therefore a different set
of parameters. It is helpful to describe the difference between functional groups by
their different locations in multi-dimensional parameter space. (lf this concept is
new to you, read the books by Kaufmann 1993, 1995.) The phytoplankton and
415
cies (Ducklow and Harris 1993). And while one group of plankton, diatoms, may
dominate the spring bloom overall, they stop reproduc ing when the supply of silieate is exhausted. Other species can then flourish until the nitrogen runs out. Later
in the year, the oligotrophic environment gives a temporary competitive advantage
to summer species. Notably dinoflagellates, which have evolved to grow in the seasonal thermocline, where they move up and down, getting nutrients below the
nutricline, and energy above it where the light is brighter. Similar specializations
exist in the zooplankton.
We showed earlier (18.9) how the changing abundances of competing species are
simulated in Virtual Plankton Ecosystems. We were content to perform numerical
experiments with idealized diatoms for that technical demonstration. The next step
is to simulate competition between real species. Within twenty years we shall have
computers that permit us to perform LE integrations with billions of particles. That
will allow us to create VPEs with a thousand populations, each represented by one
million particles (in a one-dimensional simulation). By the time marine biologists
have produced biologic al primitive equations for each of those species, we shall
already have in place the modelling technology to simulate demographic changes
in the plankton ecosystem at the level of species.
The value of such biodiverse simulations will depend on the stability, demographic noise and errors in the computation. Acquiring knowledge about the population ecology of biodiverse VPEs will take many years. Meanwhile, we shall
move ahead cautiously from the bridgehead secured with the WB food chain
model, and its variants described in 18.9. Our biodiversity strategy for the next
decade will be progressively to increase the number ofpopulations ofphytoplankton and zooplankton. In doing so, we shall have to as sume that each population is
representative ofhundreds or even thousands of species. The question to be considered here is how can we relate the populations in our models to aH the species they
are meant to represent? The answer lies in the concept of "functional groups"
(sometimes called "guilds", see May 1981).
18.16.1 Functional groups
AH the species in a functional group share the same biological primitive equations. Within the group, individual species and varieties are distinguished by the
values of the parameters found in those equations. Species in different functional
groups have different physiology and behaviour, and therefore different BPEs. This
is obviously true between trophic levels: only phytoplankton have equations for
photosynthesis, and only zooplankton have equations for foraging. But there are
also distinct functional groups within the same trophic level. For example, diatoms
use silicon, while other phytoplankton do not. And some phytoplankton can control
their motion through the water, others cannot.
Different functional groups have different equations, and therefore a different set
of parameters. It is helpful to describe the difference between functional groups by
their different locations in multi-dimensional parameter space. (lf this concept is
new to you, read the books by Kaufmann 1993, 1995.) The phytoplankton and