302
T. Soomere
identification of semi-persistent patterns 8 but do not provide any information about
their existence in reality. In general, for each study area, circulation and trajectory
model, it is essential to evaluate the necessary spatial and temporal scales in order
to reach representative results.
Several parameters may have different optimum values depending on the particular problem addressed. For example, a time interval covering at least three years
is necessary to reach an appreciable approximation to the climatologically valid optimum fairway in the Gulf of Finland (Andrejev et al. 2011). For certain sea areas
(such as the vicinity of the Danish Straits) a climatological solution might be a poor
compromise and the optimum should be determined separately for dynamically different but irregularly occurring flow regimes such as the inflow and outflow conditions (Lu et al. 2012). The solutions may also exhibit substantial seasonal variations
Soomere et al. (2011a).
9.4 Simulations of Environmental Risks
9.4.1 The Coast as the Vulnerable Region
It is straightforward to specify the boundaries of the vulnerable regions that are located offshore where their boundaries do not modify the currents and current-driven
transport (Delpeche-Ellmann and Soomere 2013). The treatment of such boundaries
usually needs some care when the vulnerable area is at the coast or in the nearshore.
The circulation models usually prescribe that the velocity component normal to the
rigid boundaries (bottom and coast) vanishes in the boundary cells (Chaps. 3 and 4).
This feature, although physically correct, constrains the simulated flow to be largely
alongshore in the nearshore and in many cases suppresses the motions towards the
coast at a distance of a few grid cells.
The rate of suppression of the cross-shore motions depends on the details of
the circulation model. The relevant features are usually not explicitly described and
should be requested from the developers of the models. The problem becomes evident in modelling the beaching of oil spills. The exact current-driven trajectory of
an oil spill (or any other item) follows the velocity data and, in principle, cannot
reach the land even if it has been brought into a grid cell adjacent to land by, e.g., a
strong downwelling event. In particular, a coastal hit of a trajectory calculated using
TRACMASS without spreading is impossible (Chap. 7). A coastal hit may happen
if, e.g., a particular trajectory is constructed using a too long time step and low-order
difference scheme.
In the real ocean wind drag and wave-driven effects substantially contribute to
the beaching of different floating items. The process of beaching is usually solved
8 Here we have in mind patterns of currents or transport that are directly related neither to single
synoptic eddies, particular events of coastal (or otherwise topographically controlled) jet currents
nor to the mean circulation in the particular basin.
T. Soomere
identification of semi-persistent patterns 8 but do not provide any information about
their existence in reality. In general, for each study area, circulation and trajectory
model, it is essential to evaluate the necessary spatial and temporal scales in order
to reach representative results.
Several parameters may have different optimum values depending on the particular problem addressed. For example, a time interval covering at least three years
is necessary to reach an appreciable approximation to the climatologically valid optimum fairway in the Gulf of Finland (Andrejev et al. 2011). For certain sea areas
(such as the vicinity of the Danish Straits) a climatological solution might be a poor
compromise and the optimum should be determined separately for dynamically different but irregularly occurring flow regimes such as the inflow and outflow conditions (Lu et al. 2012). The solutions may also exhibit substantial seasonal variations
Soomere et al. (2011a).
9.4 Simulations of Environmental Risks
9.4.1 The Coast as the Vulnerable Region
It is straightforward to specify the boundaries of the vulnerable regions that are located offshore where their boundaries do not modify the currents and current-driven
transport (Delpeche-Ellmann and Soomere 2013). The treatment of such boundaries
usually needs some care when the vulnerable area is at the coast or in the nearshore.
The circulation models usually prescribe that the velocity component normal to the
rigid boundaries (bottom and coast) vanishes in the boundary cells (Chaps. 3 and 4).
This feature, although physically correct, constrains the simulated flow to be largely
alongshore in the nearshore and in many cases suppresses the motions towards the
coast at a distance of a few grid cells.
The rate of suppression of the cross-shore motions depends on the details of
the circulation model. The relevant features are usually not explicitly described and
should be requested from the developers of the models. The problem becomes evident in modelling the beaching of oil spills. The exact current-driven trajectory of
an oil spill (or any other item) follows the velocity data and, in principle, cannot
reach the land even if it has been brought into a grid cell adjacent to land by, e.g., a
strong downwelling event. In particular, a coastal hit of a trajectory calculated using
TRACMASS without spreading is impossible (Chap. 7). A coastal hit may happen
if, e.g., a particular trajectory is constructed using a too long time step and low-order
difference scheme.
In the real ocean wind drag and wave-driven effects substantially contribute to
the beaching of different floating items. The process of beaching is usually solved
8 Here we have in mind patterns of currents or transport that are directly related neither to single
synoptic eddies, particular events of coastal (or otherwise topographically controlled) jet currents
nor to the mean circulation in the particular basin.