MIKE 21 BW, Руководство пользователя, Страница: 140 / 196

Reference Manual
140 MIKE 21 BW - © DHI
Moving shoreline using slot technique
The vertical cross section comprises a physical regime with clear water on
top of an artificial porous flow regime. The idealised vertical variation of the
porosity is unity in the physical domain and a very small porosity in the poros-
ity regime below. In order to avoid numerical instabilities it is necessary to
introduce a transition between the two regimes. Here an exponential transi-
tion has been used. For a thorough description of the determination of the
moving shoreline using slots reference is made to Madsen et al (1997a). This
paper is included in the scientific documentation.
Recommended moving shoreline parameters
The moving shoreline is calculated using the following parameters:

Slot depth
In practice the slot depth is normally chosen as the datum of the toe of
the slope. It is important to keep slot so deep that it is not dried out dur-
ing a down-rush.

Slot width
In principle the slot width (or the minimum porosity) should be as small
as possible in order to avoid a distortion of the mass balance and a dis-
turbance of the flow in the physical domain. On the other hand, the
numerical solution becomes unstable for extreme values of the slot
width. In practice, it turns out that the slot width should be chosen in
the interval of 0.01 to 0.001. For steep slopes an even higher slot width
is sometimes required. Too high values of the slot width may results in
an underestimation of the run-up.

Slot smoothing parameter
In principle the slot smoothing parameter should be as large as possi-
ble in order to avoid a distortion of the mass balance and a disturbance
of the flow in the physical domain. On the other hand, the numerical
solution becomes unstable for extreme values of the slot smoothing
parameter. In practice, it turns out that the parameter should be order
of 100. For steep slopes an a small slot smoothing parameter is some-
times required. Too small values of the parameter width may results in
an underestimation of the run-up.

Slot friction coefficient
In most applications this parameters should be set to zero. Non-zero
values may damp out high-frequency noise, but may also end up in an
underestimation of the run-up.
Please note
Numerical modelling of a moving shoreline is a very difficult task. You may
sometimes see numerical instabilities and eventually blow-up in some of your