With respect to the standard version of POM an important change in the model structure has been implemented into this version of POM. The default POM centered difference scheme for the advection of tracers has been substituted with the Smolarkiewicz and Smolarkiewicz and Clark positive definite iterative scheme as coded into POM by Sannino.

The grid and bathymetry

Both the AREG and ASHELF-1 use grids with constant horizontal resolution. The AREG grid has a resolution of 5 km (approximately 1/20°). The Model domain encompasses the whole Adriatic basin and extends south of the Otranto channel into the northern Ionian Sea, where the only open boundary is located. Grid points located over the Thyrrenian Sea have been masked out

The ASHELF-1 grid has a resolution of 1.5 km (approximately 1/37°). The model domain is rotated by 67° with respect to the AREG grid and extends over the northern Adriatic Sea. The only open boundary cuts the basin across an ideal line spanning from the southern tip of the Istrian peninsula to the Italian coast at approximately 43.6° lat. N.

In the vertical, POM uses a bottom following, sigma-coordinate system. In ADRICOSM AREG has 21 vertical sigma levels, while 11 vertical sigma levels define the ASHELF-1 vertical resolution. In both models the sigma levels are more compressed (logarithmic distribution) near the surface and the bottom.

The bottom geometry for both models was obtained from the U.S. Navy 1/60° bathymetric database DBDB1, by bilinear interpolation of the depth data into the model grid. The minimum depth was set to be 10 m for AIM and 3 m for ASHELF-1.

AREG initial conditions

Temperature and salinity initial conditions for the AREG simulations were obtained from the Artegiani et al. dataset updated with stations having bottom depth shallower than 15 m. However, since this data set has no information south of the Otranto channel, in order to cover the Ionian sector of the model domain, it was merged with the temperature and salinity gridded (0.25°) monthly data available from the MED6 dataset. The resulting dataset was used to produce seasonal fields mapped on the AREG grid using objective analysis techniques. Seasons were defined as follows: winter, January, to April; spring, May to June; summer, July to October; autumn, November to December. The winter fields were used as AREG initial conditions.

ASHELF-1 Initial conditions

The ASHELF-1 initial fields were obtained from the last year of integration of the AREG simulations. The AREG results were averaged over 10 days and the averages corresponding to the last 10 days of December of the perpetual year were interpolated in the NASM grid in order to start the model in January.

Surface and bottom boundary conditions

Both AREG and ASHELF-1 have been forced with monthly varying fields of surface heat, water and momentum (wind stress) fluxes.

The surface salinity flux is composed by the balance of Evaporation (E), Precipitation (P) and River-runoff along with a flux correction term in order to impose a forcing producing sea surface salinities consistent with the seasonal climatology.

The monthly river runoff data was obtained from monthly climatology. The major Adriatic rivers were considered as point sources, while minor contributions were defined as distributed source functions. The Po River runoff in both AIM and NASM was distributed along more grid points, in order to represent the freshwater discharge of the various mouths of the delta.

Lateral open boundary conditions

The models are hierarchically connected to one another by a simple off-line, one way nesting technique. Time varying 10 day averaged temperature, salinity and velocity data obtained from the OGCM and AREG simulations were directly specified on the open boundary of the AREG and ASHELF-1 respectively, after interpolation on the corresponding grid. The OGCM-AREG and the AREG-ASHELF-1 nestings were designed in a way as to ensure that the volume transport across the open boundary of the finer resolution model matches the volume transport across the corresponding section of the coarser model. This involved the correction of the interpolated total velocity component normal to the boundary on the basis of the differences between the volume transport computed on the fine and coarse resolution grids.

For detailed information see Zavatarelli et al. 2001

ACOAST 1.1

The ocean model used in ACOAST-1.1 is MITgcm (MIT general circulation model), developed at MIT (Massachusetts Institute of Technology, Boston). It is a three-dimensional, finite volume (with partial cells capability), free surface numerical model, utilizing the Boussinesq approximation. It can be run in non-hydrostatic mode and allows the use of alternative numerical schemes for the advective and diffusive terms. The turbulence closure, for the horizontal component, is achieved with constant harmonic and biharmonic coefficients of eddy viscosity and diffusivity; the vertical processes of mixing and diffusion can be parametrized either with constant harmonic operators, or by the second-order turbulernce closure based on the KPP method. Density can be calculated by either the linear or the UNESCO equation of state. The model is designed to ensure efficient portability on several computational platforms and uses standard MPI calls for the parallel implementation.

The grid and bathymetry

The ACOAST-1.1 computational domain is an Arakawa-C grid with horizontal resolution of 250 m (88 x 128 cells) and 25 levels (1 m thick, with the opportunity of using partial cells on the bottom elements). The model domain encompasses the Gulf of Trieste: the open boundary is the imaginary vertical plane between Grado, on the italian coast, and Piran, on the slovenian coast. The bathymetry of the Gulf is rotated 30° clockwise to better fit the model grid.

a) Bathymetry of the Gulf of Trieste (rotated 30° clockwise). The position of the MAMBO meteomarine buoy (OGS) is indicated.
b) Model discretization: the horizontal grid (resolution: 250 x 250 m). In the vertical direction the model grid has 25 levels, 1 m thick each.

ACOAST-1.1 initial conditions

Temperature and salinity initial conditions for the ACOAST-1.1 simulations will be obtained from the results of the ACOAST-1.2 model. In the first preliminar experiments, initial condition data are taken from the MAMBO (OGS) buoy, deployed at 45°41.54'N, 13°42.30'E. This meteo-marine buoy has a vertical profiler for temperature and salinity.

Lateral open boundary conditions

ACOAST-1.1 is a high-resolution model fitted in the hierarchy of models defined by the ADRICOSM project. The one way nesting technique between ACOAST-1.2 and ACOAST-1.1 will guarantee (via the interpolation of thermohaline and hydrodynamic fields) a correct remote forcing for ACOAST-1.1, preserving the mass conservation across the open boundary. Any perturbation generated inside the domain (internal waves, freshwater or pollutant plumes) will also be radiated outside the basin through the open boundary.

Surface boundary conditions

The model will be forced with time and space varying fields of surface heat, water and momentum (wind stress) fluxes. In the case study shown below (coastal upwelling caused by a strong Bora wind event in Summer 2002), the model is wind driven only.

Boundary conditions: wind speed (hourly means) measured by CNR-ISMAR (Sezione di oceanografia chimica e fisica di Trieste; used as surface boundary condition for the simulations) and by ARPA FVG - OSMER (OSservatorio MEteorologico Regionale) in June 2002. Both the meteorologic stations are located in Trieste, near the coast. In this particular case study, the only external forcing is the wind stress. The wind coming from ENE is Bora, the typical wind that blows on the Gulf of Trieste.

Horizontal sections of temperature: three levels (2, 10, 15 m) at four snapshots (00:00, 08:00, 16:00, 24:00 of 25th June 2002). The Bora event breaks the initial stable horizontal stratification after 8 hours. The upwelling temperature structure remains almost unchanged for one day.

Horizontal sections of salinity: three levels (2, 10, 15 m) at four snapshots (00:00, 08:00, 16:00, 24:00 of 25th June 2002). The Bora event breaks the initial stable horizontal stratification after 8 hours. The upwelling salinity structure remains almost unchanged for one day.

Horizontal sections of velocity: three levels (2, 10, 15 m) at four snapshots (00:00, 08:00, 16:00, 24:00 of 25th June 2002). In every sketch the maximum and average velocity are indicated. The figure (expecially at 08:00) shows the influence of the Bora wind on surface circulation (d): warm water flows outside the domain. Conversely, to guarantee the mass conservation, an inshore current of deep cold water (e and f) is flowing in the bottom layer.

IRMA Models are:

Sewer Model

River Model

Coastal Model

The MOUSE model will represent the main combined and storm sewer collectors and overflows into the receiving waters. Once the hydrodynamic model is completed, the advection-dispersion of the pollution load in the sewer system will be analysed using the MOUSETRAP module, in order to establish the quality characteristics of the sewage which overflowed in the receiving bodies. Specific parameters such as the Biological Oxygen Demand (BOD), Chemical Oxygen Demand (COD), Suspended Solids (SS), ammonia or total nitrogen will be investigated.

River Model

The MIKE11 model simulates one dimensional, vertically homogeneous unsteady flows in rivers and channels as well as the variation of quality parameters along the entire river stretch. Following completion of the hydraulic model, the advection-dispersion module will be set-up in order to simulate the WQ processes along the simulated river.