378 lohn Woods
18.1.2 Individual-Based Modelling
My lecture will introduce the modelling technology that is needed to exploit the
inherent strength of biological primitive equations; technology to create trustworthy simulations of the plankton ecosystem. Because BPEs describe the interaction
between a single micro-organism and its ambient environment, the modelling technology must be individually-based. The advantages of individual-based modelling
are well known. It has been widely used to simulate particular processes, especially
in zooplankton. This involves Biological Lagrangian Integration (BLI), in which
the life history of an individual organism is simulated by integrating the BPEs
along its trajectory through the water.
Here we are concemed with the more ambitious goal ofusing BLI to create comprehensive simulations ofthe plankton ecosystem (Fig. 18.1). To achieve that, we
must use the information contained in the life histories of individual plankters to
derive ac curate simulation of (1) the demographic history of each population, and
(2) the biofeedback to the physical and chemical environments. The modelling
technology developed to achieve that goal is called the Lagrangian Ensemble (LE)
method (Woods & Onken 1982). It links a dynamical sub-population of identical
plankters to each of a large number of particles behaving like individual plankters.
Techniques have been developed to monitor and control the errors in the demography and environment simulated by this method.
18.1.3 Virtual Plankton Ecosystems
Simulations ofthe plankton ecosystem based on the LE method are called Virtual
Plankton Ecosystems (VPE) to distinguish them from simpler (Eulerian) simulations that contain no infonnation about individual plankters. This distinction is
illustrated by the Azores VPE, derived from a simple food chain (NPZD) model.
The Azores VPE has a stable demographic attractor, which does not exhibit chaotic
fluctuations of the kind found in simulations by the Eulerian method. VPEs can be
used to parametrize Eulerian models (18.15). The computational cost of creating
VPEs was daunting in the last century, but the arrival of cheap parallel computers is
making it affordable in the 21 st century. You can explore the Azores VPE for yourself with a graphical analysis tool, called VEWdata, provided in a CD-ROM
(18.19).
The LE method was originally developed to simulate the plankton ecosystem
one-dimensionally in a vertical water column in the open ocean. It can be extended
to simulate the plankton ecosystem in shallow water (18.13) and in three-dimensions (18.14). The LE method is suitable for modelling a number ofphenomena of
practical interest in which plankton dynamics play an important role. These include
fisheries recruitment, pollution, epidemics of diseases that are transmitted to larger
animals, including humans, climate change by the greenhouse effect, and the
impact of enhanced UV radiation on life in the ocean. We now have mature technology that allows us to use biological primitive equation models to address practical problems.
18.1.2 Individual-Based Modelling
My lecture will introduce the modelling technology that is needed to exploit the
inherent strength of biological primitive equations; technology to create trustworthy simulations of the plankton ecosystem. Because BPEs describe the interaction
between a single micro-organism and its ambient environment, the modelling technology must be individually-based. The advantages of individual-based modelling
are well known. It has been widely used to simulate particular processes, especially
in zooplankton. This involves Biological Lagrangian Integration (BLI), in which
the life history of an individual organism is simulated by integrating the BPEs
along its trajectory through the water.
Here we are concemed with the more ambitious goal ofusing BLI to create comprehensive simulations ofthe plankton ecosystem (Fig. 18.1). To achieve that, we
must use the information contained in the life histories of individual plankters to
derive ac curate simulation of (1) the demographic history of each population, and
(2) the biofeedback to the physical and chemical environments. The modelling
technology developed to achieve that goal is called the Lagrangian Ensemble (LE)
method (Woods & Onken 1982). It links a dynamical sub-population of identical
plankters to each of a large number of particles behaving like individual plankters.
Techniques have been developed to monitor and control the errors in the demography and environment simulated by this method.
18.1.3 Virtual Plankton Ecosystems
Simulations ofthe plankton ecosystem based on the LE method are called Virtual
Plankton Ecosystems (VPE) to distinguish them from simpler (Eulerian) simulations that contain no infonnation about individual plankters. This distinction is
illustrated by the Azores VPE, derived from a simple food chain (NPZD) model.
The Azores VPE has a stable demographic attractor, which does not exhibit chaotic
fluctuations of the kind found in simulations by the Eulerian method. VPEs can be
used to parametrize Eulerian models (18.15). The computational cost of creating
VPEs was daunting in the last century, but the arrival of cheap parallel computers is
making it affordable in the 21 st century. You can explore the Azores VPE for yourself with a graphical analysis tool, called VEWdata, provided in a CD-ROM
(18.19).
The LE method was originally developed to simulate the plankton ecosystem
one-dimensionally in a vertical water column in the open ocean. It can be extended
to simulate the plankton ecosystem in shallow water (18.13) and in three-dimensions (18.14). The LE method is suitable for modelling a number ofphenomena of
practical interest in which plankton dynamics play an important role. These include
fisheries recruitment, pollution, epidemics of diseases that are transmitted to larger
animals, including humans, climate change by the greenhouse effect, and the
impact of enhanced UV radiation on life in the ocean. We now have mature technology that allows us to use biological primitive equation models to address practical problems.