330
T. Soomere
is used with the time step of 150 s for the baroclinic and 15 s for the barotropic propagation. The output is stored once in six hours to keep the size of output information
at a reasonable level.
The RCO model was forced with meteorological data from a regionalization of
the ERA-40 reanalysis over Europe using a regional atmosphere model covering the
Baltic Sea with a horizontal resolution of 25 km during 1961–2007 (Samuelsson
et al. 2011). This data set contains wind at the 10 m level (adjusted using simulated
gustiness, Höglund et al. 2009), air temperature, relative and specific humidity at
the 2 m level, total precipitation and total cloudiness with a temporal resolution
of 3 hours, sea level atmospheric pressure, snow depth, actual albedo, short- and
long-wave radiation and evaporation.
10.4.2 The OAAS Model
The numerical model OAAS, constructed for use in basins with complicated bathymetry and hydrography (Andrejev and Sokolov 1989, 1990) and named after the
authors (Oleg Andrejev ja Alexander Sokolov), was used to simulate the currents
in the Gulf of Finland to the east of 23 ◦ 27 E. This free-surface, 3D baroclinic zcoordinate circulation model employs standard simplifications such as the Boussinesq and hydrostatic approximation, and the no-slip condition for the entire seabed.
It is based on the primitive equations of horizontal momentum balance, the continuity equation and equations for the transport of heat and salt in Cartesian coordinates.
Some of the model features (such as the equation of state) have been tuned for
the Baltic Sea conditions (Millero and Kremling 1976). The use of the governing equations in the flux form allows automatic maintaining of a number of integral constraints (Blumberg and Mellor 1987). The finite-difference method uses the
Arakawa C-grid (Mesinger and Arakawa 1976) and the method of splitting the time
step (Liu and Leendertse 1978). An overview of the model equations and the methods for their solving is presented in Andrejev and Sokolov (1989, 1990), Sokolov
et al. (1997), Andrejev et al. (2004a, 2004b, 2010).
As the winters during the period covered in simulations (1987–1991) were rather
mild and the Gulf of Finland was mostly free of ice, a simple parameterization
was used for ice phenomena. For water temperatures below freezing point, the wind
stress was decreased by a factor of 10 in order to mimic the presence of ice. At 0 ◦ C,
the vertical heat flux was stopped as long as cooling conditions prevailed. The loss
of heat during ice melting was approximated by decreasing the upward heat flux in
the early spring by a factor of four until the water temperature reaches +1 ◦ C.
The model resolution, originally restricted to 1 nm in order to match the available
bathymetric information (Seifert et al. 2001), has been increased to 0.25 nm for the
use of the technique in question (Andrejev et al. 2010). The model was applied in
three resolutions (2 nm, 1 nm and 0.5 nm) with otherwise identical set-up. The vertical resolution was 1 m in the entire water column (except for the uppermost layer
T. Soomere
is used with the time step of 150 s for the baroclinic and 15 s for the barotropic propagation. The output is stored once in six hours to keep the size of output information
at a reasonable level.
The RCO model was forced with meteorological data from a regionalization of
the ERA-40 reanalysis over Europe using a regional atmosphere model covering the
Baltic Sea with a horizontal resolution of 25 km during 1961–2007 (Samuelsson
et al. 2011). This data set contains wind at the 10 m level (adjusted using simulated
gustiness, Höglund et al. 2009), air temperature, relative and specific humidity at
the 2 m level, total precipitation and total cloudiness with a temporal resolution
of 3 hours, sea level atmospheric pressure, snow depth, actual albedo, short- and
long-wave radiation and evaporation.
10.4.2 The OAAS Model
The numerical model OAAS, constructed for use in basins with complicated bathymetry and hydrography (Andrejev and Sokolov 1989, 1990) and named after the
authors (Oleg Andrejev ja Alexander Sokolov), was used to simulate the currents
in the Gulf of Finland to the east of 23 ◦ 27 E. This free-surface, 3D baroclinic zcoordinate circulation model employs standard simplifications such as the Boussinesq and hydrostatic approximation, and the no-slip condition for the entire seabed.
It is based on the primitive equations of horizontal momentum balance, the continuity equation and equations for the transport of heat and salt in Cartesian coordinates.
Some of the model features (such as the equation of state) have been tuned for
the Baltic Sea conditions (Millero and Kremling 1976). The use of the governing equations in the flux form allows automatic maintaining of a number of integral constraints (Blumberg and Mellor 1987). The finite-difference method uses the
Arakawa C-grid (Mesinger and Arakawa 1976) and the method of splitting the time
step (Liu and Leendertse 1978). An overview of the model equations and the methods for their solving is presented in Andrejev and Sokolov (1989, 1990), Sokolov
et al. (1997), Andrejev et al. (2004a, 2004b, 2010).
As the winters during the period covered in simulations (1987–1991) were rather
mild and the Gulf of Finland was mostly free of ice, a simple parameterization
was used for ice phenomena. For water temperatures below freezing point, the wind
stress was decreased by a factor of 10 in order to mimic the presence of ice. At 0 ◦ C,
the vertical heat flux was stopped as long as cooling conditions prevailed. The loss
of heat during ice melting was approximated by decreasing the upward heat flux in
the early spring by a factor of four until the water temperature reaches +1 ◦ C.
The model resolution, originally restricted to 1 nm in order to match the available
bathymetric information (Seifert et al. 2001), has been increased to 0.25 nm for the
use of the technique in question (Andrejev et al. 2010). The model was applied in
three resolutions (2 nm, 1 nm and 0.5 nm) with otherwise identical set-up. The vertical resolution was 1 m in the entire water column (except for the uppermost layer