Case Studies - Coastal Lagoons: Ecosystem Processes and Modeling for Sustainable Use and Development - page 428

The data on the vertical profiles of salinity at the monitoring stations obtained
during 18 months of measurements were used to calibrate the advection-dispersion
module. Analysis indicated that the artificial Kaliningrad Navigation Channel, which
is twice as deep as the lagoon area, plays a key role in the water exchange processes
between the Baltic Sea, the northern part of the Vistula Lagoon, and the Pregel River.
This channel passes along the northern coast of the lagoon and is bounded by a set
of artificial islands.
The inadequate description of the spatial distribution of salinity in the eastern
part of the lagoon by the MIKE21 model was anticipated. This is because the model
uses only a two-dimensional horizontal approach and does not describe the two-layer
water dynamics that exists in the channel. It was decided to focus special attention
on this feature during simulations. Specific values of calibration coefficients have
been found that provided a simulation of the arrival of salt water in the eastern part
of the lagoon, which is the same as the real water exchange, by the deep Navigational
Channel. 25

9.4.2.3 Hydrodynamic Modeling (MIKE21 HD)

The time step in the hydrodynamic simulations was 300 s, the grid step was 1 km
per 1 km, and the period included all of 1994. The input data included wind speed
and direction as well as water-level variation at the lagoon entrance. Furthermore,
28 sources were defined: 20 were rivers and eight were wastewater discharge outlets.
The inflow from the Pregel River, the main river of the Vistula Lagoon, and the
Pasleka River, the main river in the Polish part, were based on weekly measurements
taken during 1994. For other rivers the mean annual discharge was estimated based
on 1994 measurements and historical data. The seasonal variations of flows in these
middle-sized rivers were assumed to be like those for the Prohladnaya River on the
Russian side and the Pasleka River on the Polish side, where measurements were
conducted.
Data were collected on the distribution of various kinds of bed sediment of the
Vistula Lagoon (granulation of the lagoon bottom as well as rooted vegetation) for
the lagoon bed resistance coefficient. However, after calculation using the HD model,
it was clear that the water-level variations and currents in the lagoon are not sensitive
to the variability of bed conditions, at least with the applied model resolution. For
all simulations, a bed resistance of 32 m 1/3 s −1 was applied.
Differing wind friction approaches were investigated. It was decided to apply
detailed simulation (200-m grid step). In general, a wind friction coefficient of 0.0015
was used, and another of 0.0003 was used in the southwestern part of the lagoon
for wind directions between 180 and 270 because of the influence of the nearby

hills, which reduce wind stress. For simulation with a 1-km grid step used as a basic
grid for eutrophication processes, a wind friction coefficient of 0.0017 was chosen.
Sensitivity analysis showed that the water levels were not sensitive to changes
in eddy viscosity, whereas current structure is a function of eddy viscosity. In the
case of currents, the best correspondence between simulations and measurements
was obtained with an eddy viscosity of 20 m 2 s −1 .

°