Primitive Equation Modelling of Plankton Ecosystems
427
18.20 Conclusion
In this article I have shown the potential of primitive equation modelling in
plankton ecology, and described the technology needed to implement it. That technology was developed during the last twenty years of the 20 th century. It is an
example of individual-ba sed modelling, using Biological Lagrangian Integration.
The goal for the new modelling technology is to use the life histories of a large
number of individual plankton to compute (1) the demographic history of each
plankton population, and (2) its influence of the physical and chemi cal environment (i.e. biofeedback). These are the emergent properties of virtual ecosystems
created from biological primitive equation models.
The first technology to achieve that specification is my Lagrangian Ensemble
(LE) method, which simulates the life histories of millions of particles, each of
which behaves like an individual plankter, while carrying information about a
dynamic sub-population of identical plankton. Methods have been developed to
monitor and control the errors in the LE method. Tests show that it realistically
simulates competition between populations of different plankton, demonstrating
competitive exclusion and co-existence. And, unlike Eulerian simulations, Virtual
Plankton Ecosystems derived from the LE method are stable, with demographic
attractors that respond rationally to changes in external conditions. That stability
makes the method useful for practical application.
The LE technology is now fully engineered, and powerful parallel computers are
available to run it. The Virtual Ecology Workbench provides user-friendly tools for
the non-programmer wishing to use the method. The Lagrangian Ensemble technology is being adapted to address a wide range of practical problems requiring
ac curate prediction of the plankton ecosystem. These applications include fisheries
recruitment, pollution, epidemiology and climate change. Biological primitive
equation modelling promises useful predictions for these important problems.
The limiting factor is the availability ofbiological primitive equations. BPEs are,
by definition, derived from reproducible laboratory experiments with plankton.
The scarcity of true BPEs made it necessary to develop LE technology with
pseudo-BPEs similar those used in Eulerian modelling, but adapted for individualbased modelling. The success of LE technology opens up exciting prospects for
prediction. To exploit that opportunity it will be necessary to increase funding to
marine biologists, so that they can accelerate their programme of laboratory experiments designed to establish Biological Primitive Equations.
Acknowledgements
My research on the plankton ecosystem modelling has been supported by the
Deutsche Forschung Gemeinschaft, the UK Natural Environment Research CounciI, the Southampton Oceanography Centre, the International Marine Centre
(Oristano), the European Commission (MAST), and the Imperial College Parallel
427
18.20 Conclusion
In this article I have shown the potential of primitive equation modelling in
plankton ecology, and described the technology needed to implement it. That technology was developed during the last twenty years of the 20 th century. It is an
example of individual-ba sed modelling, using Biological Lagrangian Integration.
The goal for the new modelling technology is to use the life histories of a large
number of individual plankton to compute (1) the demographic history of each
plankton population, and (2) its influence of the physical and chemi cal environment (i.e. biofeedback). These are the emergent properties of virtual ecosystems
created from biological primitive equation models.
The first technology to achieve that specification is my Lagrangian Ensemble
(LE) method, which simulates the life histories of millions of particles, each of
which behaves like an individual plankter, while carrying information about a
dynamic sub-population of identical plankton. Methods have been developed to
monitor and control the errors in the LE method. Tests show that it realistically
simulates competition between populations of different plankton, demonstrating
competitive exclusion and co-existence. And, unlike Eulerian simulations, Virtual
Plankton Ecosystems derived from the LE method are stable, with demographic
attractors that respond rationally to changes in external conditions. That stability
makes the method useful for practical application.
The LE technology is now fully engineered, and powerful parallel computers are
available to run it. The Virtual Ecology Workbench provides user-friendly tools for
the non-programmer wishing to use the method. The Lagrangian Ensemble technology is being adapted to address a wide range of practical problems requiring
ac curate prediction of the plankton ecosystem. These applications include fisheries
recruitment, pollution, epidemiology and climate change. Biological primitive
equation modelling promises useful predictions for these important problems.
The limiting factor is the availability ofbiological primitive equations. BPEs are,
by definition, derived from reproducible laboratory experiments with plankton.
The scarcity of true BPEs made it necessary to develop LE technology with
pseudo-BPEs similar those used in Eulerian modelling, but adapted for individualbased modelling. The success of LE technology opens up exciting prospects for
prediction. To exploit that opportunity it will be necessary to increase funding to
marine biologists, so that they can accelerate their programme of laboratory experiments designed to establish Biological Primitive Equations.
Acknowledgements
My research on the plankton ecosystem modelling has been supported by the
Deutsche Forschung Gemeinschaft, the UK Natural Environment Research CounciI, the Southampton Oceanography Centre, the International Marine Centre
(Oristano), the European Commission (MAST), and the Imperial College Parallel