Activation of physical processes

SWAN contains a number of physical processes (see Scientific/Technical documentation) that add or withdraw wave energy to or from the wave field. The processes included are: wind input, whitecapping, bottom friction, depth-induced wave breaking, dissipation due to vegetation, mud, sea ice or turbulence, obstacle transmission, nonlinear wave-wave interactions (quadruplets and triads) and wave-induced set-up. In version 41.41, another source term is introduced called the Bragg scattering, taking into account the scattering of waves by small-scale bottom features.


SWAN can run in several modes, indicating the level of parameterization. SWAN can operate in first-, second- and third-generation mode. The first- and second-generation modes are essentially those of Holthuijsen and De Boer (1988); first-generation with a constant Phillips "constant" of 0.0081 and second-generation with a variable Phillips "constant". An overview of the options is given in Table below.
Table 2.1: Overview of physical processes and generation mode in SWAN.
process authors generation                
mode                
1st 2nd 3rd                
Linear wind growth Cavaleri and Malanotte-Rizzoli (1981) $\times$ $\times$                  
  (modified)                      
  Cavaleri and Malanotte-Rizzoli (1981)     $\times$                
Exponential wind growth Snyder et al. (1981) (modified) $\times$ $\times$                  
  Snyder et al. (1981)     $\times$                
  Janssen (1989, 1991)     $\times$                
  Yan (1987) (modified)     $\times$                
Whitecapping Holthuijsen and De Boer (1988) $\times$ $\times$                  
  Komen et al. (1984)     $\times$                
  Janssen (1991)     $\times$                
  Alves and Banner (2003)     $\times$                
Quadruplets Hasselmann et al. (1985)     $\times$                
Triads Eldeberky (1996) $\times$ $\times$ $\times$                
  Booij et al. (2009) $\times$ $\times$ $\times$                
Depth-induced breaking Battjes and Janssen (1978) $\times$ $\times$ $\times$                
Bottom friction JONSWAP (1973) $\times$ $\times$ $\times$                
  Collins (1972) $\times$ $\times$ $\times$                
  Madsen et al. (1988) $\times$ $\times$ $\times$                
Obstacle transmission Seelig (1979), d'Angremond (1996) $\times$ $\times$ $\times$                
Wave-induced set-up   $\times$ $\times$ $\times$                
Vegetation dissipation Dalrymple (1984), $\times$ $\times$ $\times$                
  Jacobsen et al (2019)                      
Mud dissipation Ng (2000) $\times$ $\times$ $\times$                
Sea ice dissipation Collins and Rogers (2017), $\times$ $\times$ $\times$                
  Rogers (2019)                      
Turbulence dissipation   $\times$ $\times$ $\times$                
Bragg scattering Ardhuin and Herbers (2002) $\times$ $\times$ $\times$                

The processes are activated as follows: For the preliminary SWAN runs, it is strongly advised to use the default values of the model coefficients. First, it should be determined whether or not a certain physical process is relevant to the result. If this cannot be decided by means of a simple hand computation, one can perform a SWAN computation without and with the physical process included in the computations, in the latter case using the standard values chosen in SWAN.


After it has been established that a certain physical process is important, it may be worthwhile to modify coefficients. In the case of wind input one may at first try to vary the wind velocity. Concerning the bottom friction, the best coefficients to vary are the friction coefficients. Switching off the depth-induced breaking term is usually not wise, since this may lead to unacceptably high wave heights near beaches (the computed wave heights 'explode' due to shoaling effects).

The SWAN team 2024-09-09