Shallow Water and Port Oil Spill Forecasting with Cosmos-Skymed
Lee, Steve
Stevenson Astrosat, UNITED KINGDOM

Modelling water surface (and subsurface) dispersion is commonly achieved by the solution of Saint Venant's equations. In shallow waters these solutions are derived from depth integrating the Navier–Stokes equations assuming that the horizontal length scale is much greater than the vertical length scale. Under this condition, conservation of mass implies that the vertical velocity of the fluid is small, the problem here however is that the assumption is made of hydrostatic conditions and that horizontal pressure gradients are due to the displacement of the surface only implying that the horizontal velocity field is constant throughout the depth of the fluid. Vertically integrating allows the vertical velocity to be removed from the equations, which suits satellite feeds for modelling where only the average surface wave data can be provided.

The shallow water equations are thus derived and can "predict" surface and hence oil spill dispersion, simple models are therefore easily deployed but they are inaccurate and diverge rapidly from real cases in ports or near coasts. In reality the conditions in ports and coasts are far from hydrostatic and indeed often not shallow (or flat) enough to use the simplifications used above. For accurate modelling of a spill the flow of water has to be calculated in 3-dimensions taking into account many variables derived from topographic models, water in-flow (and out-flow) knowledge and pivotaly real time data on the surface wind speed/wave height and oil spill volumes (as input conditions.)

We will present a solution to this problem allowing for oil spill propagation and forecasting in shallow waters or ports.

Our solutions uses various computational products which use controlled volume finite element methods on meshes which can be unstructured in all three dimensions to accurately define the port conditions (and vertical topography) over time and which can also adapt to optimally resolve solution dynamics based on real input data from subsequent satellite passes (ideal with Sky-Med and the e-GEOS aggregated return times.)

Our system will pre-model a clients' port or geography then, during an oil spill, run iterative propagation models based on the above data provided as a direct input from the e-GEOS system.

This value add - will allow for much more efficient and rapid deployment of oil spill recovery strategies for a port. It is hoped that the costs of oil spill monitoring and modelling from our combined solutions will be off-set by savings from a more rapid clean up and recovery to full operation, as well as mitigation of environmental fines and compliance fees. It is hoped that the tools will allow for much more selective closure of areas in a port rather than complete shut down as would often be the case when oil spill is detected late or detected but not modeled correctly.

We hope to present the results of trials conducted throughout 2013 in partnership with, Stevenson Astrosat, large port operators and Vega Space UK.