|
The factors controlling the salinity distribution in Moreton Bay were investigated by a combination of mathematical models and filed measurements. The accuracy that can be realistically obtained from hydrodynamic models was investigated and the spatial resolution justified when limited data for boundary conditions is available was also investigated. A preliminary model of Moreton Bay was developed using data available at the commencement of this project. The results from this model were used in the planning of an extensive field programme and in the application of more sophisticated models. In the field programme, surface evaluations were measured at ten locations and the tidal and residual velocities measured at twenty locations in Moreton Bay. Where sufficient data was available, the principal tidal constants were evaluated. The Eulerian residual velocities are large up to 15 cm sec-1) and have a complex spatial structure. The Stokes drifts are less than 0.9 cm sec-1. A two-dimensional, vertically integrated hydrodynamic model was used to investigate the tidal and residual circulations. The average r.m.s difference between the computed and measured surface evaluations was 2.7 cm for a spring tide and 2.3 cm for a neap tide. When a 1 km grid was used, the agreement between the measured and computed tidal velocities and the tide-induced residual velocities was good at some locations even though it was poor near the boundaries of the model. While coarse grids adequately represent the tidal circulations, fine grids are necessary to investigate the tide- and wind-induced residual velocities. These computed residual velocities were as large as 15 cm sec-1 but the gravitational circulations evaluated with a two-dimensional, side elevation model were approximately 5 cm sec-1 (except near the mouth of the Brisbane R.). A vertically integrated, advection-diffusion model was developed. When tested against analytical solutions for the two-dimensional distribution of a tracer which moves sinusoidally with the tidal velocity, the maximum error expressed as a percentage of the maximum concentration was 1.8% and the error in the displacement of the tracer with the tidal velocity was 0.06%. This model was used to investigate the salinity distribution in Moreton Bay during a 65 day period. The final computed salinities were within 0.6°/o o(1.5%) of the measured salinities at all locations (except WIB). It was concluded that in the model both the residual circulations and the dispersion are important for the transport of salt in regions of complex bathymetry but that the dispersion is small at other locations.
|
