next up previous index
Next: Output Up: Model description Previous: Physics   Index

Numerics


\begin{picture}(18,0.12)
\thicklines
\put(0,0){\line(1,0){17}}
\put(0,0.1){\line(1,0){17}}
\end{picture}

         |    BSBT
PROP    <                     | Sec |
         |    GSE  [waveage] <  MIn  >
         |                    |  HR |
         |                    | DAy |

\begin{picture}(18,0.12)
\thicklines
\put(0,0){\line(1,0){17}}
\put(0,0.1){\line(1,0){17}}
\end{picture}

Command to choose:


BSBT the BSBT scheme will be used in the computations.  
GSE garden-sprinkler effect is to be counteracted in the S&L propagation scheme  
  (default for nonstationary regular grid computations) or in the propagation  
  scheme for unstructured grids by adding a diffusion term to the basic equation.  
  This may affect the numerical stability of SWAN (see Scientific/Technical  
  documentation).  
[waveage] the time interval used to determine the diffusion which counteracts the so-called  
  garden-sprinkler effect. The default value of [waveage] is zero, i.e. no added  
  diffusion. The value of [waveage] should correspond to the travel time of  
  the waves over the computational region.  

Notes:


\begin{picture}(18,0.12)
\thicklines
\put(0,0){\line(1,0){17}}
\put(0,0.1){\line(1,0){17}}
\end{picture}

NUMeric ( STOPC [dabs] [drel] [curvat] [npnts]                      &

                    | -> STAT [mxitst] [alfa] |
                   <                           >  [limiter]       ) &
                    |    NONSTAT [mxitns]     |

          ( DIRimpl [cdd]                                         ) &

          ( SIGIMpl [css] [eps2] [outp] [niter]                   ) &

          ( CTheta [cfl]                                          ) &

          ( CSigma [cfl]                                          ) &

          ( SETUP [eps2] [outp] [niter]                           )

\begin{picture}(18,0.12)
\thicklines
\put(0,0){\line(1,0){17}}
\put(0,0.1){\line(1,0){17}}
\end{picture}

With this optional command the user can influence some of the numerical properties of SWAN.

STOPC With this option the user can influence the criterion for terminating the  
  iterative procedure in the SWAN computations (both stationary and  
  nonstationary). The criterion make use of the second derivative, or curvature,  
  of the iteration curve of the significant wave height. As the solution of a  
  simulation approaches full convergence, the curvature of the iteration curve will  
  tend to zero. SWAN stops the process if the absolute change in Hs from one  
  iteration to the next is less than [dabs] or the relative change in Hs from one  
  iteration to the next is less than [drel] and the curvature of the iteration  
  curve of Hs normalized with Hs is less than [curvat]. Both conditions need to  
  be fulfilled in more than fraction [npnts]% of all wet grid points.  
[dabs] Default: [dabs] = 0.005 [-]  
[drel] Default: [drel] = 0.01 [-]  
[curvat] Default: [curvat] = 0.005 [-]  
[npnts] Default: [npnts] = 99.5 [-]  
STAT indicates the use of parameters in a stationary computation.  
[mxitst] the maximum number of iterations for stationary computations.  
  The computation stops when this number is exceeded.  
  Default: [mxitst] = 50.  
  Note that [mxitst] can be set to 0 if one wants to check the input to the  
  model without making computations.  
[alfa] proportionality constant used in the frequency-dependent under-relaxation  
  technique. Based on experiences, a suggestion for this parameter is [alfa] = 0.01.  
  In case of diffraction computations, the use of this parameter is recommended.  
  Default: [alfa] = 0.00.  
  NOT MEANINGFUL FOR NONSTATIONARY COMPUTATIONS.  
NONSTAT indicates the use of parameters in a nonstationary computation.  
[mxitns] the maximum number of iterations per time step for nonstationary computations.  
  The computation moves to the next time step when this number is exceeded.  
  Default: [mxitns] = 1.  
  Note that [mxitns] can be set to 0 if one wants to check the input to the  
  model without making computations.  
[limiter] determines, in both stationary and nonstationary runs, the maximum change per  
  iteration of the energy density per spectral ($ \sigma$,$ \theta$)-bin, given in  
  terms of a fraction of the omni-directional Phillips level (see Scientific/  
  Technical documentation).  
  Default: [limiter] = 0.1.  
DIRIMPL this option is used to influence the numerical scheme for refraction.  
[cdd] A value of [cdd]=0 corresponds to a central scheme and has the largest  
  accuracy (diffusion $ \approx$ 0) but the computation may more easily generate  
  spurious fluctuations. A value of [cdd]=1. corresponds to a first order  
  upwind scheme and it is more diffusive and therefore preferable if (strong)  
  gradients in depth or current are present.  
  Default: [cdd] = 0.5.  
SIGIMPL controls the accuracy of computing the frequency shifting and the stopping criterion  
  and amount of output for the SIP solver (used in the computations in the presence  
  of currents or time varying depth).  
CTHETA this option prevents an excessive directional turning at a single grid point or vertex  
  due to a very coarse bathymetry or current locally. This option limits the directional  
  turning rate c$\scriptstyle \theta$ based on the CFL restriction. (See Eq. 3.114 of Scientific/  
  Technical documentation).  
  Note that if this command is not specified, then the limiter is not activated.  
[cfl] upper limit for the CFL restriction for c$\scriptstyle \theta$. A suggestion for this parameter is  
  [cfl] = 0.9.  
  Default: [cfl] = 0.9 (when command CTHETA is activated).  
CSIGMA this option prevents an excessive frequency shifting at a single grid point or vertex  
  due to a very coarse bathymetry or current locally. This option limits the frequency  
  shifting rate c$\scriptstyle \sigma$ based on the CFL restriction.  
  Note that if this command is not specified, then the limiter is not activated.  
[cfl] upper limit for the CFL restriction for c$\scriptstyle \sigma$. A suggestion for this parameter is  
  [cfl] = 0.9.  
  Default: [cfl] = 0.9 (when command CSIGMA is activated).  
SETUP controls the stopping criterion and amount of output for the SOR solver in the  
  computation of the wave-induced set-up.  
[css] A value of [css]=0 corresponds to a central scheme and has the largest  
  accuracy (diffusion $ \approx$ 0) but the computation may more easily generate  
  spurious fluctuations. A value of [css]=1. corresponds to a first order upwind  
  scheme and it is more diffusive and therefore preferable if (strong) gradients in  
  depth or current are present.  
  Default: [css] = 0.5.  
[eps2] Relative stopping criterion to terminate the linear solver (SIP or SOR). The  
  criterion for the SIP solver is based on | A$ \vec{{N}}_{{k}}^{{}}$ - $ \vec{{b}}\,$|2 $ \leq$ [eps2] |$ \vec{{b}}\,$|2
  where A is a matrix, $ \vec{{N}}\,$ is the action density vector, $ \vec{{b}}\,$ is the right hand vector  
  and k is the iteration number.  
  The criterion for the SOR solver is based on |$ \overline{{\eta}}_{{{k+1}}}^{{}}$ - $ \overline{{\eta}}_{{{k}}}^{{}}$|$\scriptstyle \infty$ $ \leq$ [eps2] where  
  $ \overline{{\eta}}$ is the set-up.  
  Default: [eps2] = 1.e-4 in case of SIP and [eps2] = 1.e-6 in case of SOR.  
[outp] output for the iterative solver:  
  0 = no output  
  1 = additional information about the iteration process is written to the PRINT file  
  2 = gives a maximal amount of output concerning the iteration process  
  3 = summary of the iteration process  
  Default: [outp] = 0.  
[niter] maximum number of iterations for the linear solver.  
  Default: [niter] = 20 in case of SIP and [niter] = 1000 in case of SOR.  


next up previous index
Next: Output Up: Model description Previous: Physics   Index
The SWAN team 2017-03-23