Long Term Simulation of Transport Using Spatially and Temporally
Aggregated Models
Brian Williams
University of Newcastle, Australia
In order to test nutrient control strategies, simulations of algal blooms
over planning periods of 20 years or more are desirable for the management
of catchments surrounding significant bodies of water. With current computing
technology, if transport and water quality processes are simulated
simultaneously within POM at hydrodynamic grid-scales, this is a massive
computational task.
For situations in which the transport processes are largely wind-driven,
it may be possible to use a hydrodynamic model such as POM to determine the
mass transport between moderate sized 'boxes' within the computational region
under a set of wind regimes. The mass transport, in the form of a matrix for
each wind condition, could then be used with a stand-alone water quality
model, which selects appropriate transport matrices in the course of its
simulation.
The major concern with this proposal is the extent of numerical diffusion,
which will take place in the spatial integration process, and ultimately how
this will affect prediction of algal blooms. The proposal in this project is
to compare the results from a water quality model using the integrated
transport process with a full resolution model, which is incorporated into
POM at the scale of the hydrodynamic grid.
This paper provides some preliminary results from an investigation of
appropriate time steps for the wind transport matrices and some comparisons
of short-term simulations of the transport model and the full resolution model.