Are oil spill models really reliable?

Oil spills introduce volumes of oil into the atmosphere and top layers of the water where natural weathering begins immediately. Depending on the water temperatures and composition, gases vent off, the remaining solids are subject to movement by tides and currents and the oil may begin to be broken down by the slower processes of bacteria and other life forms that metabolize the hydrocarbons in oil.

Add in dispersants that can break apart the components of the oil, separating the heavier components into “tarballs” which sink to the bottom and into “slicks” which ride on the surface. In between are medium weight components which can float underwater in microscopic or larger form. Dispersants rapidly “fling” the molecules far and wide, often in every direction at once, leading to dispersal that can go on for miles.

All of the time, water currents and tides and the wind work to move the oil mass in ways that are not easy to forecast or to predict on the large scale or in unique physical environments.

Models are algorithms that factor in many complex factors, measurements and data to provide an expectation of oil movement that depends upon natural and man made contributions to the movement and dispersal of the oil or its components. A visual representation of all of the inputted information is created. These days the visual representations are in the form of animations, some of which can even show the movement of oil in three dimensions as it moves at all levels in the water.

One 2-D animated model is owned by the National Oceanic and Atmospheric Administration (NOAA). It is the General NOAA Operational Modeling Environment (GNOME). NOAA’s Emergency Response Division uses GNOME to predict:

The ways in which wind, currents and other factors will cause oil to move or spread.

Study the ways in which the inexactness and uncertainty come from existing measures and forecasts of winds and currents.

Studying and examining predictions of how the oil will weather, or be broken down in the atmosphere, when it sits on the water.

GNOME is accessible to the public. The user downloads tide and current information for the location in question, then enters it into the GNOME program.  A “movie” results, showing the forecasted trajectory of the oil!  

The fact that GNOME recognized inexactness and uncertainty in the current modeling technology indicates that models are not completely dependable because of the uncertainties that come from atmospheric measuring and tide/current prediction models.

With the ongoing argument and disagreement concerning global warming, it is clear that no existing system for measuring temperature, weather and tides is unassailable. Therefore the certainty of modeling output that is based on existing measurements and data collection is also in question, and oil trajectory models become some of the most dicey propositions of all.

Dr. Pootjitha D. Yapa and others have been working on various models for predicting the trajectory of oil spills under various conditions and in various presentations of data.

This group has come up with several models:

Animated 3-D model: One model, developed at the Clarkson University Engineering Department, focuses on predicting the trajectory of the BP oil spill based on physical and chemical processes that oil goes through in the water. There is evaporated, dispersed, dissolved, surface and water column oil. These and other factors are input to produce an animated 3-D model of the predicted oil trajectory throughout the water, not just at the surface.

There is a model for analyzing the spread and movement of oil in ice covered waters.

The DEPOSE oil trajectory model looks at dissolution, evaporation, photo-oxidation, sedimentation, and emulsification

As shown, there are many models for projecting the movement, transport, trajectory and dissolution or breakdown of oil in various waters of the world. But such models are only as good as the data that is collected. It is impossible to know exactly how dispersants are doing in every square millimeter of water and oil. The presence of oil that has broken down into microscopic particles is creating new challenges. There is no way to measure the atmospheric and water temperatures in an equally comprehensive way. There are issues that approach chaos theory in their nonlinear and other effects.

As a result, the models are dependable in the gross sense of being able to predict the general transport of huge masses of oil over large areas, but the exact movement, timing of movement and qualities of the oil, dispersant and oil components is impossible to predict with complete and detailed accuracy.

The accuracy required for litigation, for example, is just not there. Litigation requires definitive answers to definitive questions and this science cannot yet provide them. In litigation, the “most reasonable” or “most likely” type of thought may be required in order to adjudicate or to argue matters.

For disaster response, economic, business, social and riparian recovery and long term bio remediation, the general predictions and trajectories provided by today’s models, with understanding of the uncertainty involved, will more than serve the needs of program developers, budget personnel, lawmakers, local and regional authorities and response forces. These agencies are built to plan based on predictions then to adapt based on reality.

One thing is sure: Scientists are working very hard and with lifelong preparation and dedication to produce atmospheric, oceanic and planetary models that are improving all of the time.