--- manual/s_algorithm/text/time_stepping.tex 2001/11/13 20:51:36 1.13 +++ manual/s_algorithm/text/time_stepping.tex 2001/11/14 21:07:13 1.14 @@ -1,4 +1,4 @@ -% $Header: /home/ubuntu/mnt/e9_copy/manual/s_algorithm/text/time_stepping.tex,v 1.13 2001/11/13 20:51:36 adcroft Exp $ +% $Header: /home/ubuntu/mnt/e9_copy/manual/s_algorithm/text/time_stepping.tex,v 1.14 2001/11/14 21:07:13 adcroft Exp $ % $Name: $ The equations of motion integrated by the model involve four @@ -558,19 +558,108 @@ connecting $u,v^n$ with $G_\theta^{n-1/2}$. The flow used to advect tracers around is not naturally located in time. This could be avoided by applying the Adams-Bashforth extrapolation to the tracer field -itself and advection that around but this is not yet available. We're -not aware of any detrimental effect of this feature. The difficulty -lies mainly in interpretation of what time-level variables and terms -correspond to. +itself and advecting that around but this approach is not yet +available. We're not aware of any detrimental effect of this +feature. The difficulty lies mainly in interpretation of what +time-level variables and terms correspond to. \section{Non-hydrostatic formulation} \label{sect:non-hydrostatic} -[to be written...] +The non-hydrostatic formulation re-introduces the full vertical +momentum equation and requires the solution of a 3-D elliptic +equations for non-hydrostatic pressure perturbation. We still +intergrate vertically for the hydrostatic pressure and solve a 2-D +elliptic equation for the surface pressure/elevation for this reduces +the amount of work needed to solve for the non-hydrostatic pressure. -Equation for $w^{n+1}$ will be here as will 3-D elliptic equations. -\label{eq:discrete-time-w} +The momentum equations are discretized in time as follows: +\begin{eqnarray} +\frac{1}{\Delta t} u^{n+1} + g \partial_x \eta^{n+1} + \partial_x \phi_{nh}^{n+1} +& = & \frac{1}{\Delta t} u^{n} + G_u^{(n+1/2)} \label{eq:discrete-time-u-nh} \\ +\frac{1}{\Delta t} v^{n+1} + g \partial_y \eta^{n+1} + \partial_y \phi_{nh}^{n+1} +& = & \frac{1}{\Delta t} v^{n} + G_v^{(n+1/2)} \label{eq:discrete-time-v-nh} \\ +\frac{1}{\Delta t} w^{n+1} + \partial_r \phi_{nh}^{n+1} +& = & \frac{1}{\Delta t} w^{n} + G_w^{(n+1/2)} \label{eq:discrete-time-w-nh} \\ +\end{eqnarray} +which must satisfy the discrete-in-time depth integrated continuity, +equation~\ref{eq:discrete-time-backward-free-surface} and the local continuity equation +\begin{equation} +\partial_x u^{n+1} + \partial_y v^{n+1} + \partial_r w^{n+1} = 0 +\label{eq:non-divergence-nh} +\end{equation} +As before, the explicit predictions for momentum are consolidated as: +\begin{eqnarray*} +u^* & = & u^n + \Delta t G_u^{(n+1/2)} \\ +v^* & = & v^n + \Delta t G_v^{(n+1/2)} \\ +w^* & = & w^n + \Delta t G_w^{(n+1/2)} +\end{eqnarray*} +but this time we introduce an intermediate step by splitting the +tendancy of the flow as follows: +\begin{eqnarray} +u^{n+1} = u^{**} - \Delta t \partial_x \phi_{nh}^{n+1} +& & +u^{**} = u^{*} - \Delta t g \partial_x \eta^{n+1} \\ +v^{n+1} = v^{**} - \Delta t \partial_y \phi_{nh}^{n+1} +& & +v^{**} = v^{*} - \Delta t g \partial_y \eta^{n+1} +\end{eqnarray} +Substituting into the depth integrated continuity +(equation~\ref{eq:discrete-time-backward-free-surface}) gives +\begin{equation} +\partial_x H \partial_x \left( g \eta^{n+1} + \widehat{\phi}_{nh}^{n+1} \right) ++ +\partial_y H \partial_y \left( g \eta^{n+1} + \widehat{\phi}_{nh}^{n+1} \right) + - \frac{\epsilon_{fs}\eta^*}{\Delta t^2} += - \frac{\eta^*}{\Delta t^2} +\end{equation} +which is approximated by equation +\ref{eq:elliptic-backward-free-surface} on the basis that i) +$\phi_{nh}^{n+1}$ is not yet known and ii) $\nabla \widehat{\phi}_{nh} +<< g \nabla \eta$. If \ref{eq:elliptic-backward-free-surface} is +solved accurately then the implication is that $\widehat{\phi}_{nh} +\approx 0$ so that thet non-hydrostatic pressure field does not drive +barotropic motion. + +The flow must satisfy non-divergence +(equation~\ref{eq:non-divergence-nh}) locally, as well as depth +integrated, and this constraint is used to form a 3-D elliptic +equations for $\phi_{nh}^{n+1}$: +\begin{equation} +\partial_{xx} \phi_{nh}^{n+1} + \partial_{yy} \phi_{nh}^{n+1} + +\partial_{rr} \phi_{nh}^{n+1} = +\partial_x u^{**} + \partial_y v^{**} + \partial_r w^{*} +\end{equation} + +The entire algorithm can be summarized as the sequential solution of +the following equations: +\begin{eqnarray} +u^{*} & = & u^{n} + \Delta t G_u^{(n+1/2)} \label{eq:ustar-nh} \\ +v^{*} & = & v^{n} + \Delta t G_v^{(n+1/2)} \label{eq:vstar-nh} \\ +w^{*} & = & w^{n} + \Delta t G_w^{(n+1/2)} \label{eq:wstar-nh} \\ +\eta^* & = & \epsilon_{fs} \left( \eta^{n} +P-E+R \right)- \Delta t + \partial_x H \widehat{u^{*}} ++ \partial_y H \widehat{v^{*}} +\\ + \partial_x g H \partial_x \eta^{n+1} ++ \partial_y g H \partial_y \eta^{n+1} +- \frac{\epsilon_{fs} \eta^{n+1}}{\Delta t^2} +& = & +- \frac{\eta^*}{\Delta t^2} +\label{eq:elliptic-nh} +\\ +u^{**} & = & u^{*} - \Delta t g \partial_x \eta^{n+1} \label{eq:unx-nh}\\ +v^{**} & = & v^{*} - \Delta t g \partial_y \eta^{n+1} \label{eq:vnx-nh}\\ +\partial_{xx} \phi_{nh}^{n+1} + \partial_{yy} \phi_{nh}^{n+1} + +\partial_{rr} \phi_{nh}^{n+1} & = & +\partial_x u^{**} + \partial_y v^{**} + \partial_r w^{*} \\ +u^{n+1} & = & u^{**} - \Delta t \partial_x \phi_{nh}^{n+1} \label{eq:un+1-nh}\\ +v^{n+1} & = & v^{**} - \Delta t \partial_y \phi_{nh}^{n+1} \label{eq:vn+1-nh}\\ +\partial_r w^{n+1} & = & - \partial_x u^{n+1} - \partial_y v^{n+1} +\end{eqnarray} +where the last equation is solved by vertically integrating for +$w^{n+1}$. @@ -623,7 +712,7 @@ This is known as the correction step. However, when the model is non-hydrostatic ($\epsilon_{nh}=1$) we need an additional step and an additional equation for $\phi'_{nh}$. This is obtained by substituting -\ref{eq:discrete-time-u}, \ref{eq:discrete-time-v} and \ref{eq:discrete-time-w} +\ref{eq:discrete-time-u-nh}, \ref{eq:discrete-time-v-nh} and \ref{eq:discrete-time-w-nh} into continuity: \begin{equation} \left[ {\bf \nabla}_h^2 + \partial_{rr} \right] {\phi'_{nh}}^{n+1}