Introduction
On this page we provide an overview of SWAN settings to help users to apply a proper set of source term parametrizations in order to perform their own calculations.
There is no such thing as one single list of parameter settings that can be used for various applications (e.g., hindcasts, forecasts, regional
climate studies, long-term trend analysis) as the suitability of the associated parametrizations varies in different environmental conditions.
For SWAN applications, a distinction is made between the open ocean, shelf seas and open coasts, which exhibit a characteristic sloping beach profile, on the one hand,
and (semi-)closed coastal basins with complex bathymetries, including flat and rippled beds, on the other. The related key parameters are further elaborated below.
The default physics of SWAN have been changed a few times over the years (see the seventh column in the first table below).
However, at any time, the user can choose its own preferred options other than the current ones (of version 41.45).
The two tables below can help you with this, if desired.
Tested source term packages suitable for regional-scale applications
SWAN includes deep water source terms with a variety of wind input and whitecapping parametrizations, which are summarized in the table below.
Furthermore, related to the wind input, wind roughness is accounted for by employing a wind drag parametrization.
The Scientific/technical documentation provides further theoretical background on these parametrizations.
Note that the input and dissipation terms of the WAM cycle 4 (Janssen, 1991)
have not been tested thoroughly. For this reason, it is recommended not to use this source term package.
Apart from the input-dissipation source terms, the nonlinear four-wave interaction is routinely modelled by the Discrete Interaction Approximation (DIA) of
Hasselmann et al. (1985).
The source terms that represent shallow water physics include surf breaking, nonlinear triad interaction and bottom friction.
These source terms can be specified separately at the discretion of the user.
Below an overview of the various tested source term parametrizations available for open water systems at the coastal and regional scale O(10-100 km).
| parametrizations |
input commands |
since version |
additional references |
| deep water physics |
shallow water physics |
| wind input + whitecapping |
wind drag |
surf breaking |
triads |
bottom friction |
| WAM cycle 3 ("Komen") |
Wu (1982) |
Battjes and Janssen (1978) |
original LTA |
JONSWAP for wind sea |
link |
40.11 |
Komen et al. (1984),
Eldeberky (1996),
Bouws and Komen (1983),
Booij et al. (1999) |
| modified Komen |
parabolic fit |
Battjes and Janssen (1978) |
consistent collinear LTA |
JONSWAP for swell |
link |
41.01 |
Rogers et al. (2003),
Zijlema et al. (2012),
Hasselmann et al. (1973),
Salmon et al. (2016) |
| Wu (1982) |
|
Madsen et al. (1988) |
link |
41.10 |
| ST6 physics |
Hwang (2011) |
Battjes and Janssen (1978) |
|
Madsen et al. (1988) |
link |
41.31 |
Rogers et al. (2012),
SWAN+ADCIRC page |
| Westhuysen |
Wu (1982) |
Battjes and Janssen (1978) |
collinear DCTA |
JONSWAP for swell |
link |
41.45 |
Van der Westhuysen et al. (2007),
Booij et al. (2009),
Doering and Bowen (1995) |
|
|
|
|
|
|
|
|
| Tested source term packages suitable for (semi-)enclosed basins with complex topography |
Numerous studies have demonstrated the high degree of success of the constant breaker index γBJ of
the depth-induced breaking model of Battjes and Janssen (1978). This is especially the case for coastal seas with sloping beds (see above).
By contrast, for coastal basins with varying topography, such as estuaries, tidal inlets, coastal bays, fjords, lagoons, reefs and shallow lakes,
a variable γBJ scaling
is required (e.g., Nelson, 1997,
Van der Westhuysen, 2010,
Salmon and Holthuijsen, 2015).
Typically, the value of γBJ is controlled by both local bottom slope
β and local characteristic wave number kd. Also, the effect of directional spreading is accounted for.
A similar modification concerns the estimate
of the biphase of the self-self interaction of spectral peak for the purpose of computing the nonlinear triad interaction. It is assumed that the biphase is scaled in terms of
local bed slope β and local peak period Tp (De Wit, 2022).
Finally, the impact of ambient currents on waves can also be dominant in semi-closed basins
(e.g., tidal inlets), including steepness dissipation of waves on opposing currents.
The table below presents the overall modification of the parameter settings that allows
SWAN to make accurate wave predictions over complex field sites.
|
| Westhuysen |
Wu (1982) |
β-kd scaling |
collinear DCTA using De Wit's biphase estimate |
JONSWAP for swell |
link |
41.45 |
Ris and Holthuijsen (1996),
Van der Westhuysen (2012),
Salmon et al. (2015),
De Wit (2022) |
|
