378
J. Murawski and J. Woge Nielsen
not resolved by numerical ice models (Reed et al. 1999). This remains an unresolved
problem for high-latitude oil drift modelling.
11.3 Modelling of Circulation and Oil Spill
Oil drift modelling has been developed largely since the 1980s, and a number of
models are now on the market (Reed et al. 1999). Differences in forecasting quality
are mainly related to externalities, i.e., model set-up and forcing. Statistical results
of drift pattern analysis and ship routing design studies are expected to be resilient
to the choice of the oil drift and fate model.
In this study we use the operational oil dispersion model BSH dmod (Dick and
Soetje 1990). Both the hindcast and forecast of the oil drift substantially rely on the
information about various hydrophysical fields that are usually calculated using a
‘classical’ ocean circulation model. The Danish Meteorological Institute (DMI) uses
for this purpose an operational three-dimensional, free-surface, baroclinic ocean circulation and sea ice model that solves the primitive (Navier–Stokes) equations for
horizontal momentum and mass, and budget equations for salinity and heat on a
spherical grid that co-moves with the Earth’s rotation.
11.3.1 DMI’s Ocean Circulation and Oil Spill Modelling System
The DMI’s circulation model used to produce the necessary information for oil
spill tracking was originally developed by the Federal Maritime and Hydrographic
Agency (BSH) in Hamburg (called BSH cmod then), and has been applied to the
North Sea and Baltic Sea since the early 1990s (Kleine 1994; Dick et al. 2001).
Since the initial implementation at the DMI, the model (called DMI/BSHcmod) has
been continuously developed further (Larsen et al. 2007; Liu et al. 2009; She et al.
2007). The model code has been completely rewritten and redesigned conceptually,
which has led to an increase in numerical efficiency, in terms of correctness, portability and speed. Among the added components are, e.g., the implementation of a
k–ω turbulence scheme, and the development of a new mixing scheme (Berg 2012).
The development follows Fortran 95 standards and strict code styling rules (Berg
and Weismann Poulsen 2012; Weismann Poulsen and Berg 2012). The model became the official Baltic Sea model in the MyOcean framework and is called HBM
(HIROMB–BOOS Model) 1 since then.
The HBM model shares many features with the Rossby Centre Ocean Model
(RCO) that is described in detail in Chap. 4. It reproduces the hydrodynamic processes on spatial scales from local scales of some hundred meters to regional scales
1 HIROMB is an abbreviation for High Resolution Oceanographic Model for the Baltic. As the
name suggests, it is a circulation model with high resolution intended to be used for the Baltic Sea
(e.g., Funkquist 2001). BOOS stands for the Baltic Operational Oceanographic System.
J. Murawski and J. Woge Nielsen
not resolved by numerical ice models (Reed et al. 1999). This remains an unresolved
problem for high-latitude oil drift modelling.
11.3 Modelling of Circulation and Oil Spill
Oil drift modelling has been developed largely since the 1980s, and a number of
models are now on the market (Reed et al. 1999). Differences in forecasting quality
are mainly related to externalities, i.e., model set-up and forcing. Statistical results
of drift pattern analysis and ship routing design studies are expected to be resilient
to the choice of the oil drift and fate model.
In this study we use the operational oil dispersion model BSH dmod (Dick and
Soetje 1990). Both the hindcast and forecast of the oil drift substantially rely on the
information about various hydrophysical fields that are usually calculated using a
‘classical’ ocean circulation model. The Danish Meteorological Institute (DMI) uses
for this purpose an operational three-dimensional, free-surface, baroclinic ocean circulation and sea ice model that solves the primitive (Navier–Stokes) equations for
horizontal momentum and mass, and budget equations for salinity and heat on a
spherical grid that co-moves with the Earth’s rotation.
11.3.1 DMI’s Ocean Circulation and Oil Spill Modelling System
The DMI’s circulation model used to produce the necessary information for oil
spill tracking was originally developed by the Federal Maritime and Hydrographic
Agency (BSH) in Hamburg (called BSH cmod then), and has been applied to the
North Sea and Baltic Sea since the early 1990s (Kleine 1994; Dick et al. 2001).
Since the initial implementation at the DMI, the model (called DMI/BSHcmod) has
been continuously developed further (Larsen et al. 2007; Liu et al. 2009; She et al.
2007). The model code has been completely rewritten and redesigned conceptually,
which has led to an increase in numerical efficiency, in terms of correctness, portability and speed. Among the added components are, e.g., the implementation of a
k–ω turbulence scheme, and the development of a new mixing scheme (Berg 2012).
The development follows Fortran 95 standards and strict code styling rules (Berg
and Weismann Poulsen 2012; Weismann Poulsen and Berg 2012). The model became the official Baltic Sea model in the MyOcean framework and is called HBM
(HIROMB–BOOS Model) 1 since then.
The HBM model shares many features with the Rossby Centre Ocean Model
(RCO) that is described in detail in Chap. 4. It reproduces the hydrodynamic processes on spatial scales from local scales of some hundred meters to regional scales
1 HIROMB is an abbreviation for High Resolution Oceanographic Model for the Baltic. As the
name suggests, it is a circulation model with high resolution intended to be used for the Baltic Sea
(e.g., Funkquist 2001). BOOS stands for the Baltic Operational Oceanographic System.