Operational Oil Spill Modelling: From Science to Engineering Applications in the Presence of Uncertainty - Mathematical Modelling and Numerical Simulation of Oil Pollution Problems - page 99

Fig. 5.1 General structure of an oil spill forecast system. Geophysical models typically require
data from a coastal observation system (COS). River inflow forecasts can be considered part of
typical COS data. Individual components are discussed in text
advect and disperse oil at (or near) the water surface at relatively coarse grid scales,
so the near-field spill behaviour must be either modelled separately (particularly for
deep-water spills, [ 31 , 56 ]) or estimated based on previous experience with similar
spills.
5.2.2 Geophysical Forcing Models
Developing an integrated geophysical forecast system is a non-trivial effort. Ideally,
such a system should be in place and continuously running so that nowcast and
forecast winds, waves, tides, river inflows, and currents are always available. Such a
modelling system should be integratedwith COS data so that each successive forecast
uses the latest available observations and the latest forecast data. Where predictive
numerical models have not been developed and tested, it is possible to use COS
measurements and climatology data to build a forecast model based on statistics [ 2 ];
however, until such models are more comprehensively tested, they are best used in
response planning and management rather than for an actual event response.
5.2.3 Oil Spill Models
An oil spill model represents the physical and chemical processes governing advec-
tion, dispersion, and weathering of the spill. Oil advection is driven by water cur-
rents, wind, and waves [ 45 ]. The dispersion of the oil depends on wave conditions,