next up previous contents
Next: Accuracy estimation. Up: Grid refinement study, test Previous: Grid refinement study, test

  
The test setup.

Along the following chapters we will use the same test setup: a rectangular channel geometry, 100 km wide, 200 km long (sketched in figure 6.1 for the 4 blocks DD), 20 meter depth, varying the cell size ($\Delta x$) to 500, 1000 and 2000 meters and using a fixed time step ($\Delta t$) of 60 s. The employed meshes have then 80.000 (th80 grid), 20.000 (th20 grid) and 5.000 (th5 grid) grid points respectively.

A monochromatic semi-diurnal wave is imposed at the West border, using a similar period as the semi-diurnal astronomic tide M2 component (period T = 12 hours, amplitude A = 5 cm, wave celerity $c=\sqrt{gH}$ = 14 m/s and wave length $\lambda=c \; T$ = 600 km and in the East border, a low reflective open boundary condition allowing the wave to cross the channel with a small decrease in amplitude (figure 7.1), being the South and North borders of the non-flow type.


  
Figure 7.1: Simulated levels along the channel.
\begin{figure}
\begin{center}
\mbox{\epsfig{file=amp_dec.eps, width=8cm} } %
\par\end{center}\end{figure}



Elias Kaplan M.Sc.
1998-07-22