Difference between revisions of "Vertical S-coordinate"

From WikiROMS
Jump to navigationJump to search
(Added s-coordinate figure)   (change visibility)
(convert to TeX)   (change visibility)
Line 1: Line 1:
<div class="title">S-coordinate</div>
<div class="title">S-coordinate</div>
 
<wikitex>
Following Song and Haidvogel (1994), the vertical coordinate has
Following Song and Haidvogel (1994), the vertical coordinate has
been chosen to be:
been chosen to be:


:<math>z = \zeta + \left(1 + {\zeta \over h} \right) \left[h_c s + (h - h_c) C(s)\right],
$$z = \zeta + \left(1 + {\zeta \over h} \right) \left[h_c s + (h - h_c) C(s)\right],
   \qquad \qquad -1 \leq s \leq 0</math>
   \qquad \qquad -1 \leq s \leq 0$$


where <math>h_c\!\,</math> is either the minimum depth or a shallower depth above which we wish to have more resolution. <math>C(s)\!\,</math> is defined as:
where $h_c$ is either the minimum depth or a shallower depth above which we wish to have more resolution. $C(s)$ is defined as:


:<math>C(s) = (1 - b) {\sinh (\theta s) \over \sinh \theta } +
$$C(s) = (1 - b) {\sinh (\theta s) \over \sinh \theta } +
   b { \tanh [\theta ( s + {1\over 2})] -
   b { \tanh [\theta ( s + {1\over 2})] -
   \tanh ( {1\over 2} \theta) \over  
   \tanh ( {1\over 2} \theta) \over  
   2 \tanh ( {1\over 2} \theta) }</math>
   2 \tanh ( {1\over 2} \theta) }$$


where <math>\theta\!\,</math> and <math>b\!\,</math> are surface and bottom control parameters. Their ranges are <math>0 < \theta \leq 20\!\,</math> and <math>0 \leq b \leq 1\!\,</math>,
where $\theta$ and $b$ are surface and bottom control parameters. Their ranges are $0 < \theta \leq 20$ and $0 \leq b \leq 1$, respectively.  The first equation leads to $z = \zeta$ for $s = 0$ and $z = h$ for $s = -1$.
respectively.  The first equation leads to <math>z = \zeta\!\,</math> for <math>s = 0\!\,</math> and <math>z = h\!\,</math> for <math>s = -1\!\,</math>.


Some features of this coordinate system:
Some features of this coordinate system:


*It is a generalization of the <math>\sigma\!\,</math>-coordinate system. Letting <math>\theta\!\,</math> go to zero and using L'Hopital's rule, we get:
*It is a generalization of the $\sigma$-coordinate system. Letting $\theta$ go to zero and using L'Hopital's rule, we get:
    
    
:<math>z = (\zeta + h)(1 + s) - h\!\,</math>
$$z = (\zeta + h)(1 + s) - h$$
   
   
which is the <math>\sigma\!\,</math>-coordinate.
which is the $\sigma$-coordinate.
* It has a linear dependence on <math>\zeta\!\,</math> and is infinitely differentiable in <math>s\!\,</math>.
* It has a linear dependence on $\zeta$ and is infinitely differentiable in $s$.
* The larger the value of <math>\theta\!\,</math>, the more resolution is kept above <math>h_c\!\,</math>.
* The larger the value of $\theta$, the more resolution is kept above $h_c$.
* For <math>b = 0\!\,</math>, the resolution all goes to the surface as <math>\theta\!\,</math> is increased.
* For $b = 0$, the resolution all goes to the surface as $\theta$ is increased.
* For <math>b = 1\!\,</math>, the resolution goes to both the surface and the bottom equally as <math>\theta\!\,</math> is increased.
* For $b = 1$, the resolution goes to both the surface and the bottom equally as $\theta$ is increased.
* For <math>\theta \neq 0\!\,</math> there is a subtle mismatch in the discretization of the model equations, for instance in the horizontal viscosity term.  We recommend that you stick with "reasonable" values of <math>\theta\!\,</math>, say <math>\theta \leq 5\!\,</math>.
* For $\theta \neq 0$ there is a subtle mismatch in the discretization of the model equations, for instance in the horizontal viscosity term.  We recommend that you stick with "reasonable" values of $\theta$, say $\theta \leq 5$.
* Some problems turn out to be sensitive to the value of <math>\theta\!\,</math> used.
* Some problems turn out to be sensitive to the value of $\theta$ used.


The following figure shows the <math>s\!\,</math>-surfaces for several values of <math>\theta\!\,</math> and <math>b\!\,</math> for one of our domains.  It was produced by a Matlab tool
The following figure shows the $s$-surfaces for several values of $\theta$ and $b$ for one of our domains.  It was produced by a Matlab tool written by Hernan Arango which is available from our web site.
written by Hernan Arango which is available from our web site.


[[Image:Scoord.png|vertical s-coordinate]]
[[Image:Scoord.png|vertical s-coordinate]]


Figure: The <math>s\!\,</math>-surfaces for the North Atlantic with (a) <math>\theta = 0.0001\!\,</math> and <math>b = 0\!\,</math>, (b) <math>\theta = 8\!\,</math> and <math>b = 0\!\,</math>, (c) <math>\theta = 8\!\,</math> and <math>b = 1\!\,</math>.  (d) The actual values used in this domain were
Figure: The $s$-surfaces for the North Atlantic with (a) $\theta = 0.0001$ and $b = 0$, (b) $\theta = 8$ and $b = 0$, (c) $\theta = 8$ and $b = 1$.  (d) The actual values used in this domain were $\theta = 5$ and $b = 0.4$.
<math>\theta = 5\!\,</math> and <math>b = 0.4\!\,</math>.


We find it convenient to define:
We find it convenient to define:
:<math>H_z \equiv {\partial z \over \partial s} </math>
$$H_z \equiv {\partial z \over \partial s}$$


The derivative of <math>C(s)\!\,</math> can be computed analytically:
The derivative of $C(s)$ can be computed analytically:


:<math>{\partial C(s) \over \partial s} = (1-b) {\cosh (\theta s) \over
$${\partial C(s) \over \partial s} = (1-b) {\cosh (\theta s) \over
   \sinh \theta} \theta + b {\coth ( {1 \over 2} \theta) \over
   \sinh \theta} \theta + b {\coth ( {1 \over 2} \theta) \over
   2 \cosh^2 [ \theta (s + {1\over 2})] } \theta </math>
   2 \cosh^2 [ \theta (s + {1\over 2})] } \theta$$


However, we choose to compute <math>H_z\!\,</math> discretely as <math>\Delta z/ \Delta s\!\,</math> since this leads to the vertical sum of <math>H_z\!\,</math> being exactly the total water depth <math>D\!\,</math>.
However, we choose to compute $H_z$ discretely as $\Delta z/ \Delta s$ since this leads to the vertical sum of $H_z$ being exactly the total water depth $D$.
</wikitex>

Revision as of 18:53, 10 July 2008

S-coordinate

<wikitex> Following Song and Haidvogel (1994), the vertical coordinate has been chosen to be:

$$z = \zeta + \left(1 + {\zeta \over h} \right) \left[h_c s + (h - h_c) C(s)\right],

  \qquad \qquad -1 \leq s \leq 0$$

where $h_c$ is either the minimum depth or a shallower depth above which we wish to have more resolution. $C(s)$ is defined as:

$$C(s) = (1 - b) {\sinh (\theta s) \over \sinh \theta } +

  b { \tanh [\theta ( s + {1\over 2})] -
  \tanh ( {1\over 2} \theta) \over 
  2 \tanh ( {1\over 2} \theta) }$$

where $\theta$ and $b$ are surface and bottom control parameters. Their ranges are $0 < \theta \leq 20$ and $0 \leq b \leq 1$, respectively. The first equation leads to $z = \zeta$ for $s = 0$ and $z = h$ for $s = -1$.

Some features of this coordinate system:

  • It is a generalization of the $\sigma$-coordinate system. Letting $\theta$ go to zero and using L'Hopital's rule, we get:

$$z = (\zeta + h)(1 + s) - h$$

which is the $\sigma$-coordinate.

  • It has a linear dependence on $\zeta$ and is infinitely differentiable in $s$.
  • The larger the value of $\theta$, the more resolution is kept above $h_c$.
  • For $b = 0$, the resolution all goes to the surface as $\theta$ is increased.
  • For $b = 1$, the resolution goes to both the surface and the bottom equally as $\theta$ is increased.
  • For $\theta \neq 0$ there is a subtle mismatch in the discretization of the model equations, for instance in the horizontal viscosity term. We recommend that you stick with "reasonable" values of $\theta$, say $\theta \leq 5$.
  • Some problems turn out to be sensitive to the value of $\theta$ used.

The following figure shows the $s$-surfaces for several values of $\theta$ and $b$ for one of our domains. It was produced by a Matlab tool written by Hernan Arango which is available from our web site.

vertical s-coordinate

Figure: The $s$-surfaces for the North Atlantic with (a) $\theta = 0.0001$ and $b = 0$, (b) $\theta = 8$ and $b = 0$, (c) $\theta = 8$ and $b = 1$. (d) The actual values used in this domain were $\theta = 5$ and $b = 0.4$.

We find it convenient to define: $$H_z \equiv {\partial z \over \partial s}$$

The derivative of $C(s)$ can be computed analytically:

$${\partial C(s) \over \partial s} = (1-b) {\cosh (\theta s) \over

  \sinh \theta} \theta + b {\coth ( {1 \over 2} \theta) \over
  2 \cosh^2 [ \theta (s + {1\over 2})] } \theta$$

However, we choose to compute $H_z$ discretely as $\Delta z/ \Delta s$ since this leads to the vertical sum of $H_z$ being exactly the total water depth $D$. </wikitex>