Keywords: adaptive mesh, multi-scale model, solute transport, coliforms.

For structured mesh models, multi-scale modelling is achieved by nesting higher resolution child grids within lower resolution parent grids. The nested grids provide selective high resolution in the areas of interest only, thus computational cost is significantly reduced. The most common nesting techniques employ static nested grids where the nested grid structure (i.e. location and extent) remains unchanged during the simulation. However, the use of static nested grids to model dynamic features such as solute plumes can still result in higher computational costs than are necessary. In the current approach an adaptive mesh multi-scale model has been developed allowing high resolution child grids to move throughout their lower resolution parent grids. The approach allows the nested grid structure to change during the simulation enabling it to track the movement of a feature of interest.

The paper describes an adaptive mesh, multi-scale model for tidal hydraulics and mass transport in coastal waters. Model performance and accuracy was assessed by simulating wastewater discharges in a large bay on the west coast of Ireland. Two different nested grid configurations were used to assess the performance and accuracy of the model. The first configuration, a single static child grid, was used to obtain a base-line child grid accuracy. The second configuration was an adaptive child grid. Grid adaptions were user-specified, with move-times and new grid extents determined from plume movements computed using the parent grid model. Both child grid solutions were found to have superior accuracy to the lower resolution parent grid solution and both showed excellent correlation with the reference solution and with measured data. High levels of child grid accuracy were observed for both hydrodynamics and coliform transport.

On the basis of the model results it was concluded that adaptive mesh nesting approaches can offer additional computational efficiencies over classical static grid nesting approaches; the static grid approach achieved 74% savings in computational cost while the adaptive mesh approach achieved 84% savings. However, adaptive mesh techniques should be used with caution. It is recommended that adaptions are user-specified, as opposed to computer-automated, and that they are planned using suitable nested boundary placement procedures based on parent grid accuracy. If not, then significant degradation of model accuracy could occur. For optimum model performance, changes in grid extents for adaptive meshes should be informed by analysis of hydrodynamic and water quality results from the lower resolution PG model.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £130 +P&P)