Difference between revisions of "Equations of Motion"

From WikiROMS
Jump to navigationJump to search
(convert to TeX)   (change visibility)
Line 1: Line 1:
<div class="title">Equations of Motion</div>
<div class="title">Equations of Motion</div>
The primitive equations in Cartesian coordinates are shown here. The momentum balance in
<wikitex>
the <math>\,\!x</math>- and <math>\,\!y</math>-directions are:


:<math>\frac{\partial u}{\partial t} + \vec{v} \cdot \nabla u - fv = - \frac{\partial \phi}{\partial x} + {F}_u + {D}_u</math>
The primitive equations in Cartesian coordinates are shown here. The momentum balance in the $x$- and $y$-directions are:


:<math>\frac{\partial v}{\partial t} + \vec{v} \cdot \nabla v + fu = - \frac{\partial \phi}{\partial y} + {F}_v + {D}_v</math>
$$\frac{\partial u}{\partial t} + \vec{v} \cdot \nabla u - fv = - \frac{\partial \phi}{\partial x} + {\cal F}_u + {\cal D}_u$$


The time evolution of the
$$\frac{\partial v}{\partial t} + \vec{v} \cdot \nabla v + fu = - \frac{\partial \phi}{\partial y} + {\cal F}_v + {\cal D}_v$$
potential temperature and salinity fields, <math>\,\!T(x,y,z,t)</math> and
<math>\,\!S(x,y,z,t)</math>, are governed by the advective-diffusive equations:


:<math>\frac{\partial T}{\partial t} + \vec{v} \cdot \nabla T ={F}_T + {D}_T</math>
The time evolution of the potential temperature and salinity fields, $T(x,y,z,t)$ and $S(x,y,z,t)$, are governed by the advective-diffusive equations:


:<math>\frac{\partial S}{\partial t} + \vec{v} \cdot \nabla S ={F}_S + {D}_S</math>
$$\frac{\partial T}{\partial t} + \vec{v} \cdot \nabla T ={\cal F}_T + {\cal D}_T$$
 
$$\frac{\partial S}{\partial t} + \vec{v} \cdot \nabla S ={\cal F}_S + {\cal D}_S$$


The equation of state is given by:
The equation of state is given by:


:<math>\rho = \rho(T,S,P)\,\!</math>
$$\rho = \rho(T,S,P)$$


In the Boussinesq approximation, density
In the Boussinesq approximation, density variations are neglected in the momentum equations except in their
variations are neglected in the momentum equations except in their
contribution to the buoyancy force in the vertical momentum equation.  
contribution to the buoyancy force in the vertical momentum equation.  
Under the hydrostatic approximation, it is further
Under the hydrostatic approximation, it is further
Line 26: Line 24:
force:
force:


:<math>\frac{\partial \phi}{\partial z} = \frac{-\rho g}{\rho_o}</math>
$$\frac{\partial \phi}{\partial z} = \frac{-\rho g}{\rho_o}$$


The final equation expresses the continuity equation
The final equation expresses the continuity equation
for an incompressible fluid:
for an incompressible fluid:


:<math>\frac{\partial u}{\partial x} + \frac{\partial v}{\partial y} + \frac{\partial w}{\partial z} = 0</math>
$$\frac{\partial u}{\partial x} + \frac{\partial v}{\partial y} + \frac{\partial w}{\partial z} = 0$$


For the moment, the effects of forcing
For the moment, the effects of forcing
and dissipation will be represented by the schematic terms <math>F\,\!</math>
and dissipation will be represented by the schematic terms ${\cal F}$
and <math>D\,\!</math>, respectively.  The horizontal and vertical mixing will
and ${\cal D}$, respectively.  The horizontal and vertical mixing will
be described more fully in [[Horizontal Mixing]] and [[Vertical Mixing Parameterizations]].
be described more fully in [[Horizontal Mixing]] and [[Vertical Mixing Parameterizations]].
The variables used are shown here:
The variables used are shown here:
{| border="1" cellspacing="0" cellpadding="5" align="center"
{| border="1" cellspacing="0" cellpadding="5" align="center"
| <math>{D}_u, {D}_v, {D}_T, {D}_S\,\!</math>
| ${\cal D}_u, {\cal D}_v, {\cal D}_T, {\cal D}_S$
| diffusive terms
| diffusive terms
|-
|-
| <math>{F}_u, {F}_v, {F}_T, {F}_S\,\!</math>
| ${\cal F}_u, {\cal F}_v, {\cal F}_T, {\cal F}_S$
| forcing terms
| forcing terms
|-
|-
| <math>f(x,y)\,\!</math>
| $f(x,y)$
| Coriolis parameter
| Coriolis parameter
|-
|-
| <math>g\,\!</math>
| $g$
| acceleration of gravity
| acceleration of gravity
|-
|-
| <math>h(x,y)\,\!</math>
| $h(x,y)$
| bottom depth
| bottom depth
|-
|-
| <math>\nu, \kappa\,\!</math>
| $\nu, \kappa$
| horizontal viscosity and diffusivity
| horizontal viscosity and diffusivity
|-
|-
| <math>K_m, K_T, K_S\,\!</math>
| $K_m, K_T, K_S$
| vertical viscosity and diffusivity
| vertical viscosity and diffusivity
|-
|-
| <math>P\,\!</math>
| $P$
| total pressure <math>P \approx -\rho_o gz</math>
| total pressure $P \approx -\rho_o gz$
|-
|-
| <math>\phi(x,y,z,t)\,\!</math>
| $\phi(x,y,z,t)$
| dynamic pressure <math>\phi = \left(P/\rho_o \right)</math>
| dynamic pressure $\phi = \left(P/\rho_o \right)$
|-
|-
| <math>\rho_o + \rho(x,y,z,t)\,\!</math>
| $\rho_o + \rho(x,y,z,t)$
| total ''in situ'' density
| total ''in situ'' density
|-
|-
| <math>S(x,y,z,t)\,\!</math>
| $S(x,y,z,t)$
| salinity
| salinity
|-
|-
| <math>t\,\!</math>
| $t$
| time  
| time  
|-
|-
| <math>T(x,y,z,t)\,\!</math>
| $T(x,y,z,t)$
| potential temperature
| potential temperature
|-
|-
| <math>u,v,w\,\!</math>
| $u,v,w$
| the (<math>x,y,z\,\!</math>) components of vector velocity <math>\vec{v}\,\!</math>
| the ($x,y,z$) components of vector velocity $\vec{v}$
|-
|-
| <math>x,y\,\!</math>
| $x,y$
| horizontal coordinates
| horizontal coordinates
|-
|-
| <math>z\,\!</math>
| $z$
| vertical coordinate
| vertical coordinate
|-
|-
| <math>\zeta(x,y,t)\,\!</math>
| $\zeta(x,y,t)$
| the surface elevation
| the surface elevation
|-
|-
Line 96: Line 94:
The vertical boundary conditions can be prescribed as follows:
The vertical boundary conditions can be prescribed as follows:


:top (<math>z = \zeta(x,y,t))\,\!</math>:
:top ($z = \zeta(x,y,t))$:
::<math>K_m \, \frac{\partial u}{\partial z} = \tau^x_s (x,y,t)</math>
::$K_m \, \frac{\partial u}{\partial z} = \tau^x_s (x,y,t)$
:: <math>K_m \, \frac{\partial v}{\partial z} = \tau_s^y(x,y,t)</math>
::$K_m \, \frac{\partial v}{\partial z} = \tau_s^y(x,y,t)$
::<math>K_T \, \frac{\partial T}{\partial z} = {Q_T \over \rho_o c_P} + {1 \over \rho_o c_P} {dQ_T \over dT} (T - T_{\rm ref})</math>
::$K_T \, \frac{\partial T}{\partial z} = {Q_T \over \rho_o c_P} + {1 \over \rho_o c_P} {dQ_T \over dT} (T - T_{\rm ref})$
::<math>K_S \, \frac{\partial S}{\partial z} = {(E-P)S \over \rho_o}</math>
::$K_S \, \frac{\partial S}{\partial z} = {(E-P)S \over \rho_o}$
::<math>w = {\partial \zeta \over \partial t}</math>
::$w = {\partial \zeta \over \partial t}$


:and bottom (<math>z = -h(x,y)\,\!</math>):
:and bottom ($z = -h(x,y)$):
::<math>K_m \, \frac{\partial u}{\partial z} = \tau_b^x (x,y,t)</math>
::$K_m \, \frac{\partial u}{\partial z} = \tau_b^x (x,y,t)$
::<math>K_m \, \frac{\partial v}{\partial z} = \tau_b^y (x,y,t)</math>
::$K_m \, \frac{\partial v}{\partial z} = \tau_b^y (x,y,t)$
::<math>K_T \, \frac{\partial T}{\partial z} = 0</math>
::$K_T \, \frac{\partial T}{\partial z} = 0$
::<math>K_S \, \frac{\partial S}{\partial z} = 0</math>
::$K_S \, \frac{\partial S}{\partial z} = 0$
::<math>- w + \vec{v} \cdot \nabla h = 0</math>
::$- w + \vec{v} \cdot \nabla h = 0$


The new variables above are:
The new variables above are:
Line 115: Line 113:
! Description
! Description
|-
|-
| <math>E-P\,\!</math>
| $E-P$
| evaporation minus precipitation
| evaporation minus precipitation
|-
|-
| <math>\gamma_1, \gamma_2\,\!</math>
| $\gamma_1, \gamma_2$
| linear and quadratic bottom stress coefficients
| linear and quadratic bottom stress coefficients
|-
|-
| <math>Q_T\,\!</math>
| $Q_T$
| surface heat flux
| surface heat flux
|-
|-
| <math>\tau_s^x , \tau_s^y\,\!</math>
| $\tau_s^x , \tau_s^y$
| surface wind stress
| surface wind stress
|-
|-
| <math>\tau_b^x , \tau_b^y\,\!</math>
| $\tau_b^x , \tau_b^y$
| bottom stress
| bottom stress
|-
|-
| <math>T_{\rm ref}\,\!</math>
| $T_{\rm ref}$
| surface reference temperature
| surface reference temperature
|-
|-
Line 136: Line 134:


The surface boundary condition variables are defined in the table above.
The surface boundary condition variables are defined in the table above.
Since <math>Q_T\,\!</math> is a strong function of the surface
Since $Q_T$ is a strong function of the surface
temperature, it is also prudent to include a correction term for the
temperature, it is also prudent to include a correction term for the
change in <math>Q\,\!</math> as the surface temperature drifts away from the reference
change in $Q$ as the surface temperature drifts away from the reference
temperature that was used in computing <math>Q_T\,\!</math>.  On the variable bottom,
temperature that was used in computing $Q_T$.  On the variable bottom,
<math>z = -h(x,y)\,\!</math>, the horizontal velocity components are constrained to
$z = -h(x,y)$, the horizontal velocity components are constrained to
accommodate a prescribed bottom stress which is a sum of linear and
accommodate a prescribed bottom stress which is a sum of linear and
quadratic terms (actually one or the other, or a log layer, depending on the cpp options):
quadratic terms (actually one or the other, or a log layer, depending on the cpp options):


:<math>\tau_b^x = (\gamma_1 + \gamma_2 \sqrt{u^2 + v^2} ) u</math>
$$\tau_b^x = (\gamma_1 + \gamma_2 \sqrt{u^2 + v^2} ) u$$


:<math>\tau_b^y = (\gamma_1 + \gamma_2 \sqrt{u^2 + v^2} ) v</math>
$$\tau_b^y = (\gamma_1 + \gamma_2 \sqrt{u^2 + v^2} ) v$$


The vertical heat and salt flux may also be prescribed at the bottom,
The vertical heat and salt flux may also be prescribed at the bottom,
Line 157: Line 155:
included for open boundaries which may or may not work for your
included for open boundaries which may or may not work for your
particular application.  Appropriate boundary conditions are
particular application.  Appropriate boundary conditions are
provided for <math>u,v,T,S,\,\!</math> and <math>\zeta\,\!</math>.
provided for $u,v,T,S,$ and $\zeta$, as described in [[Boundary Conditions]].
At every timestep the subroutines '''xxxx''' and
'''yyyy''' are called to fill in the necessary boundary values.


The model domain is logically rectangular, but it is possible to
The model domain is logically rectangular, but it is possible to mask out land areas on the boundary and in the interior. Boundary conditions on these masked regions are straightforward, with a choice of no-slip or free-slip walls.
mask out land areas on the boundary and in the interior. Boundary
conditions on these masked regions are straightforward,
with a choice of no-slip or free-slip walls.


If biharmonic friction is used, a higher order boundary condition
If biharmonic friction is used, a higher order boundary condition must also be provided. The model currently has this built into the code where the biharmonic terms are calculated.  The high order  
must also be provided. The model currently has this built into the
boundary conditions used for $u$ are $\frac{\partial}{\partial x} \left(
code where the biharmonic terms are calculated.  The high order
\frac{h\nu}{mn} \frac{\partial ^2 u}{\partial x^2} \right) = 0$ on the  
boundary conditions used for <math>u\,\!</math> are <math>\frac{\partial}{\partial x} \left(
eastern and western boundaries and $\frac{\partial}{\partial y} \left(
\frac{h\nu}{mn} \frac{\partial ^2 u}{\partial x^2} \right) = 0</math> on the
\frac{h\nu}{mn} \frac{\partial ^2 u}{\partial y^2} \right) = 0$ on the  
eastern and western boundaries and <math>\frac{\partial}{\partial y} \left(
northern and southern boundaries.  The boundary conditions for $v,T,$
\frac{h\nu}{mn} \frac{\partial ^2 u}{\partial y^2} \right) = 0\,</math> on the
and $S$ are similar.  These boundary conditions were chosen because  
northern and southern boundaries.  The boundary conditions for <math>v,T,\,\!</math>
they preserve the property of no gain or loss of volume-integrated  
and <math>S\,\!</math> are similar.  These boundary conditions were chosen because
they preserve the property of no gain or loss of volume-integrated
momentum, temperature, or salt.
momentum, temperature, or salt.
</wikitex>

Revision as of 00:08, 10 July 2008

Equations of Motion

<wikitex>

The primitive equations in Cartesian coordinates are shown here. The momentum balance in the $x$- and $y$-directions are:

$$\frac{\partial u}{\partial t} + \vec{v} \cdot \nabla u - fv = - \frac{\partial \phi}{\partial x} + {\cal F}_u + {\cal D}_u$$

$$\frac{\partial v}{\partial t} + \vec{v} \cdot \nabla v + fu = - \frac{\partial \phi}{\partial y} + {\cal F}_v + {\cal D}_v$$

The time evolution of the potential temperature and salinity fields, $T(x,y,z,t)$ and $S(x,y,z,t)$, are governed by the advective-diffusive equations:

$$\frac{\partial T}{\partial t} + \vec{v} \cdot \nabla T ={\cal F}_T + {\cal D}_T$$

$$\frac{\partial S}{\partial t} + \vec{v} \cdot \nabla S ={\cal F}_S + {\cal D}_S$$

The equation of state is given by:

$$\rho = \rho(T,S,P)$$

In the Boussinesq approximation, density variations are neglected in the momentum equations except in their contribution to the buoyancy force in the vertical momentum equation. Under the hydrostatic approximation, it is further assumed that the vertical pressure gradient balances the buoyancy force:

$$\frac{\partial \phi}{\partial z} = \frac{-\rho g}{\rho_o}$$

The final equation expresses the continuity equation for an incompressible fluid:

$$\frac{\partial u}{\partial x} + \frac{\partial v}{\partial y} + \frac{\partial w}{\partial z} = 0$$

For the moment, the effects of forcing and dissipation will be represented by the schematic terms ${\cal F}$ and ${\cal D}$, respectively. The horizontal and vertical mixing will be described more fully in Horizontal Mixing and Vertical Mixing Parameterizations. The variables used are shown here:

${\cal D}_u, {\cal D}_v, {\cal D}_T, {\cal D}_S$ diffusive terms
${\cal F}_u, {\cal F}_v, {\cal F}_T, {\cal F}_S$ forcing terms
$f(x,y)$ Coriolis parameter
$g$ acceleration of gravity
$h(x,y)$ bottom depth
$\nu, \kappa$ horizontal viscosity and diffusivity
$K_m, K_T, K_S$ vertical viscosity and diffusivity
$P$ total pressure $P \approx -\rho_o gz$
$\phi(x,y,z,t)$ dynamic pressure $\phi = \left(P/\rho_o \right)$
$\rho_o + \rho(x,y,z,t)$ total in situ density
$S(x,y,z,t)$ salinity
$t$ time
$T(x,y,z,t)$ potential temperature
$u,v,w$ the ($x,y,z$) components of vector velocity $\vec{v}$
$x,y$ horizontal coordinates
$z$ vertical coordinate
$\zeta(x,y,t)$ the surface elevation

Vertical Boundary Conditions

The vertical boundary conditions can be prescribed as follows:

top ($z = \zeta(x,y,t))$:
$K_m \, \frac{\partial u}{\partial z} = \tau^x_s (x,y,t)$
$K_m \, \frac{\partial v}{\partial z} = \tau_s^y(x,y,t)$
$K_T \, \frac{\partial T}{\partial z} = {Q_T \over \rho_o c_P} + {1 \over \rho_o c_P} {dQ_T \over dT} (T - T_{\rm ref})$
$K_S \, \frac{\partial S}{\partial z} = {(E-P)S \over \rho_o}$
$w = {\partial \zeta \over \partial t}$
and bottom ($z = -h(x,y)$):
$K_m \, \frac{\partial u}{\partial z} = \tau_b^x (x,y,t)$
$K_m \, \frac{\partial v}{\partial z} = \tau_b^y (x,y,t)$
$K_T \, \frac{\partial T}{\partial z} = 0$
$K_S \, \frac{\partial S}{\partial z} = 0$
$- w + \vec{v} \cdot \nabla h = 0$

The new variables above are:

Variable Description
$E-P$ evaporation minus precipitation
$\gamma_1, \gamma_2$ linear and quadratic bottom stress coefficients
$Q_T$ surface heat flux
$\tau_s^x , \tau_s^y$ surface wind stress
$\tau_b^x , \tau_b^y$ bottom stress
$T_{\rm ref}$ surface reference temperature

The surface boundary condition variables are defined in the table above. Since $Q_T$ is a strong function of the surface temperature, it is also prudent to include a correction term for the change in $Q$ as the surface temperature drifts away from the reference temperature that was used in computing $Q_T$. On the variable bottom, $z = -h(x,y)$, the horizontal velocity components are constrained to accommodate a prescribed bottom stress which is a sum of linear and quadratic terms (actually one or the other, or a log layer, depending on the cpp options):

$$\tau_b^x = (\gamma_1 + \gamma_2 \sqrt{u^2 + v^2} ) u$$

$$\tau_b^y = (\gamma_1 + \gamma_2 \sqrt{u^2 + v^2} ) v$$

The vertical heat and salt flux may also be prescribed at the bottom, although they are usually set to zero.

Horizontal Boundary Conditions

As distributed, the model can easily be configured for a periodic channel, a doubly periodic domain, or a closed basin. Code is also included for open boundaries which may or may not work for your particular application. Appropriate boundary conditions are provided for $u,v,T,S,$ and $\zeta$, as described in Boundary Conditions.

The model domain is logically rectangular, but it is possible to mask out land areas on the boundary and in the interior. Boundary conditions on these masked regions are straightforward, with a choice of no-slip or free-slip walls.

If biharmonic friction is used, a higher order boundary condition must also be provided. The model currently has this built into the code where the biharmonic terms are calculated. The high order boundary conditions used for $u$ are $\frac{\partial}{\partial x} \left( \frac{h\nu}{mn} \frac{\partial ^2 u}{\partial x^2} \right) = 0$ on the eastern and western boundaries and $\frac{\partial}{\partial y} \left( \frac{h\nu}{mn} \frac{\partial ^2 u}{\partial y^2} \right) = 0$ on the northern and southern boundaries. The boundary conditions for $v,T,$ and $S$ are similar. These boundary conditions were chosen because they preserve the property of no gain or loss of volume-integrated momentum, temperature, or salt.

</wikitex>