5 European Semi-enclosed Seas
151
She et al. (2007) used a high resolution operational ocean model of the North
Sea–Baltic Sea region to study the transports in the Danish Straits and the sensitivity
of the bottom water salinity in the Arkona Basin to the bathymetry. By deepening
the bathymetry in the straits the realism of simulated bottom water salinity in the
Fehmarn Belt was substantially improved.
The development of modelling efforts on the dynamics of the Bosporus Straits
evolved from simple models (e.g., Oguz et al. 1990) to process modelling. An example is given by Simeonov et al. (1997) and Stanev et al. (2001). They used results from numerical simulations with a reduced gravity model of the Mediterranean
plume (active deep layer and motionless surface layer) intruding into the Black Sea.
The model, which had a horizontal resolution of 600 m, was used to study how the
water properties and thickness of the Mediterranean plume is modified by turbulent entrainment, bottom friction and stratification. This model was coupled with a
simple chemical model simulating the oxidation of H 2 S by O 2 .
The characteristics of the simulated outflow (Fig. 5.14) were comparable with its
observed features. A good illustration of the dominating spatial patterns is provided
by the simulated bottom concentration of oxygen (that is high in the area of the
strait and decreases when the depth increases). The flow first spreads on the shelf
in a funnel-like form, then after reaching the shelf edge (100 m isobath) cascades
down the very steep continental slope (blue colour in Fig. 5.14). It was found that
the mixing of the Mediterranean and Black Sea water, as well as the termination
depth of the plume, are very sensitive to specific combinations of the governing parameters. In the shown case the plume reaches a depth of ∼250 m. The relatively
large entrainment rate of about 10–12 compared to the one in the outflow from the
Strait of Gibraltar of about 2–3 (Baringer and Price 1997), the small deflection of
the outflow to the right caused by the Earth’s rotation compared to the angle in other
outflows, and the shallow penetration depth are explained as a result of specific
combinations of governing parameters, topography routing and ambient stratification.
Knowledge of the Bosporus outflow was further developed by Stashchuk and
Hutter (2003) who used a non-hydrostatic two-dimensional radial model of the
Mediterranean Sea dense plume with saline and warm water intruding into the less
dense Black Sea water. The model takes into account the typical vertical temperature and salinity distribution as well as the bottom relief. Similarly to the case of
Stanev et al. (2001), the Mediterranean water moves down the shelf slope and mixes
with Black Sea water. However, this model was superior because it simulated well
the separation from the bottom when buoyancy became neutral (see Konovalov et al.
2003). The level of the lens separation from the slope depended on the intensity of
the mixing processes.
One fundamental problem in the numerical modelling of the basin-wide circulation in the European semi-enclosed seas is the correct representation of the exchange
through the straits. For the Mediterranean Sea this is not so difficult to achieve because the Strait of Gibraltar is relatively wide and a reasonable resolution could
be applied. The case in the Danish Straits is similar. However, the situation in the
Black Sea is completely different. Models with structured grids can hardly resolve
151
She et al. (2007) used a high resolution operational ocean model of the North
Sea–Baltic Sea region to study the transports in the Danish Straits and the sensitivity
of the bottom water salinity in the Arkona Basin to the bathymetry. By deepening
the bathymetry in the straits the realism of simulated bottom water salinity in the
Fehmarn Belt was substantially improved.
The development of modelling efforts on the dynamics of the Bosporus Straits
evolved from simple models (e.g., Oguz et al. 1990) to process modelling. An example is given by Simeonov et al. (1997) and Stanev et al. (2001). They used results from numerical simulations with a reduced gravity model of the Mediterranean
plume (active deep layer and motionless surface layer) intruding into the Black Sea.
The model, which had a horizontal resolution of 600 m, was used to study how the
water properties and thickness of the Mediterranean plume is modified by turbulent entrainment, bottom friction and stratification. This model was coupled with a
simple chemical model simulating the oxidation of H 2 S by O 2 .
The characteristics of the simulated outflow (Fig. 5.14) were comparable with its
observed features. A good illustration of the dominating spatial patterns is provided
by the simulated bottom concentration of oxygen (that is high in the area of the
strait and decreases when the depth increases). The flow first spreads on the shelf
in a funnel-like form, then after reaching the shelf edge (100 m isobath) cascades
down the very steep continental slope (blue colour in Fig. 5.14). It was found that
the mixing of the Mediterranean and Black Sea water, as well as the termination
depth of the plume, are very sensitive to specific combinations of the governing parameters. In the shown case the plume reaches a depth of ∼250 m. The relatively
large entrainment rate of about 10–12 compared to the one in the outflow from the
Strait of Gibraltar of about 2–3 (Baringer and Price 1997), the small deflection of
the outflow to the right caused by the Earth’s rotation compared to the angle in other
outflows, and the shallow penetration depth are explained as a result of specific
combinations of governing parameters, topography routing and ambient stratification.
Knowledge of the Bosporus outflow was further developed by Stashchuk and
Hutter (2003) who used a non-hydrostatic two-dimensional radial model of the
Mediterranean Sea dense plume with saline and warm water intruding into the less
dense Black Sea water. The model takes into account the typical vertical temperature and salinity distribution as well as the bottom relief. Similarly to the case of
Stanev et al. (2001), the Mediterranean water moves down the shelf slope and mixes
with Black Sea water. However, this model was superior because it simulated well
the separation from the bottom when buoyancy became neutral (see Konovalov et al.
2003). The level of the lens separation from the slope depended on the intensity of
the mixing processes.
One fundamental problem in the numerical modelling of the basin-wide circulation in the European semi-enclosed seas is the correct representation of the exchange
through the straits. For the Mediterranean Sea this is not so difficult to achieve because the Strait of Gibraltar is relatively wide and a reasonable resolution could
be applied. The case in the Danish Straits is similar. However, the situation in the
Black Sea is completely different. Models with structured grids can hardly resolve