194 |
|
|
195 |
For the non-hydrostatic equations, dropping the thin-atmosphere |
For the non-hydrostatic equations, dropping the thin-atmosphere |
196 |
approximation re-introduces metric terms involving $w$ and are |
approximation re-introduces metric terms involving $w$ and are |
197 |
required to conserve anglular momentum: |
required to conserve angular momentum: |
198 |
\begin{eqnarray} |
\begin{eqnarray} |
199 |
{\cal A}_w \Delta r_f h_w G_u^{metric} & = & |
{\cal A}_w \Delta r_f h_w G_u^{metric} & = & |
200 |
- \overline{ \frac{ \overline{u}^i \overline{w}^k }{a} {\cal A}_c \Delta r_f h_c }^i \\ |
- \overline{ \frac{ \overline{u}^i \overline{w}^k }{a} {\cal A}_c \Delta r_f h_c }^i \\ |
258 |
``lat-lon'' grid for Laplacian viscosity. |
``lat-lon'' grid for Laplacian viscosity. |
259 |
\marginpar{Need to tidy up method for controlling this in code} |
\marginpar{Need to tidy up method for controlling this in code} |
260 |
|
|
261 |
It should be noted that dispite the ad-hoc nature of the scaling, some |
It should be noted that despite the ad-hoc nature of the scaling, some |
262 |
scaling must be done since on a lat-lon grid the converging meridians |
scaling must be done since on a lat-lon grid the converging meridians |
263 |
make it very unlikely that a stable viscosity parameter exists across |
make it very unlikely that a stable viscosity parameter exists across |
264 |
the entire model domain. |
the entire model domain. |
291 |
The no-slip condition defines the normal gradient of a tangential flow |
The no-slip condition defines the normal gradient of a tangential flow |
292 |
such that the flow is zero on the boundary. Rather than modify the |
such that the flow is zero on the boundary. Rather than modify the |
293 |
stresses by using complicated functions of the masks and ``ghost'' |
stresses by using complicated functions of the masks and ``ghost'' |
294 |
points (see \cite{Adcroft+Marshall98}) we add the boundary stresses as |
points (see \cite{adcroft:98}) we add the boundary stresses as |
295 |
an additional source term in cells next to solid boundaries. This has |
an additional source term in cells next to solid boundaries. This has |
296 |
the advantage of being able to cope with ``thin walls'' and also makes |
the advantage of being able to cope with ``thin walls'' and also makes |
297 |
the interior stress calculation (code) independent of the boundary |
the interior stress calculation (code) independent of the boundary |
308 |
|
|
309 |
In fact, the above discretization is not quite complete because it |
In fact, the above discretization is not quite complete because it |
310 |
assumes that the bathymetry at velocity points is deeper than at |
assumes that the bathymetry at velocity points is deeper than at |
311 |
neighbouring vorticity points, e.g. $1-h_w < 1-h_\zeta$ |
neighboring vorticity points, e.g. $1-h_w < 1-h_\zeta$ |
312 |
|
|
313 |
\fbox{ \begin{minipage}{4.75in} |
\fbox{ \begin{minipage}{4.75in} |
314 |
{\em S/R MOM\_U\_SIDEDRAG} ({\em mom\_u\_sidedrag.F}) |
{\em S/R MOM\_U\_SIDEDRAG} ({\em mom\_u\_sidedrag.F}) |
325 |
to the variable grid lengths introduced by the finite volume |
to the variable grid lengths introduced by the finite volume |
326 |
formulation. This reduces the formal accuracy of these terms to just |
formulation. This reduces the formal accuracy of these terms to just |
327 |
first order but only next to boundaries; exactly where other terms |
first order but only next to boundaries; exactly where other terms |
328 |
appear such as linar and quadratic bottom drag. |
appear such as linear and quadratic bottom drag. |
329 |
\begin{eqnarray} |
\begin{eqnarray} |
330 |
G_u^{v-diss} & = & |
G_u^{v-diss} & = & |
331 |
\frac{1}{\Delta r_f h_w} \delta_k \tau_{13} \\ |
\frac{1}{\Delta r_f h_w} \delta_k \tau_{13} \\ |
343 |
\tau_{33} & = & A_v \frac{1}{\Delta r_f} \delta_k w |
\tau_{33} & = & A_v \frac{1}{\Delta r_f} \delta_k w |
344 |
\end{eqnarray} |
\end{eqnarray} |
345 |
It should be noted that in the non-hydrostatic form, the stress tensor |
It should be noted that in the non-hydrostatic form, the stress tensor |
346 |
is even less consistent than for the hydrostatic (see Wazjowicz |
is even less consistent than for the hydrostatic (see |
347 |
\cite{Waojz}). It is well known how to do this properly (see Griffies |
\cite{wajsowicz:93}). It is well known how to do this properly (see |
348 |
\cite{Griffies}) and is on the list of to-do's. |
\cite{griffies:00}) and is on the list of to-do's. |
349 |
|
|
350 |
\fbox{ \begin{minipage}{4.75in} |
\fbox{ \begin{minipage}{4.75in} |
351 |
{\em S/R MOM\_U\_RVISCLFUX} ({\em mom\_u\_rviscflux.F}) |
{\em S/R MOM\_U\_RVISCLFUX} ({\em mom\_u\_rviscflux.F}) |