Original URL: https://www.theregister.com/2010/06/07/ncar_oil_slick_sim/

Supercomputer oil slick sims predict greasy Atlantic

Deepwater doo-doo

By Timothy Prickett Morgan

Posted in HPC, 7th June 2010 21:21 GMT

Supercomputers are good for more than just designing nuclear weapons or making doomsday predictions about climate change. They can depress us in other ways, like showing us the extent of the damage that could be done by BP's Deepwater Horizon offshore rig spewing oil into the Gulf of Mexico.

The National Center for Atmospheric Research, which is funded by the US National Science Foundation to do climate modeling, has borrowed some computing capacity at the New Mexico Computer Applications Center and the Oak Ridge National Laboratory, loaded up the Parallel Ocean Program (a part of the Community Climate System Model simulation created by NCAR and the US Department of Energy), and dropped some simulated dye in the simulated Gulf of Mexico and Atlantic Ocean and watched how it disperses over time.

"I've had a lot of people ask me, 'Will the oil reach Florida?'" Synte Peacock, one of the NCAR researchers who did the simulation, said in a report on the early findings released by NCAR. "Actually, our best knowledge says the scope of this environmental disaster is likely to reach far beyond Florida, with impacts that have yet to be understood."

NCAR Oil Slick Simulation

The static image of the simulation (above) illustrates where dye had dispersed – and how densely – 132 days after being released into the upper 65 feet of the ocean, assuming a continuous spilling of dye at the same rate and location as the oil from April 20 through June 20. NCAR researchers warn that the simulated dye in the water has the same viscosity of water, unlike the oil that is actually spreading through the Gulf of Mexico.

Oil can clump up and moves in different ways. This simulation, which you can view in animated form here, does not show how the dye-as-oil-proxy is moving in the colder depths of the ocean, but NCAR used its ocean models to simulate the dispersal of dye at six different layers of the Gulf of Mexico and the Atlantic to get a sense of how the oil might move. (You can view all six simulations here).

The simulations show that once the dye (and hence the oil slick) hits the Loop Current, a fast-moving, clockwise whirl linking the tip of the Yucatan Peninsula to the western edge of the Florida Panhandle, it can quickly run up the Atlantic Coast as far north as Cape Hatteras in North Carolina before being sucked out into the Atlantic where it disperses across a wide area. So not only will the BP slick reach Florida, it will absolutely engulf it. The Loop Current runs at about 40 miles per day, but the Gulf Stream current in the Atlantic can do 100 miles per day, so the oil could really move fast once it rounds the Panhandle, heading east.

The BP slick could have a dramatic effect on Europe as well. Martin Visbeck, of the University of Kiel in Germany, is part of the team trying to figure that out using these simulations. "Our assumption is that the enormous lateral mixing in the ocean together with the biological disintegration of the oil should reduce the pollution to levels below harmful concentrations." Visbeck explained in the NCAR statement. "But we would like to have this backed up by numbers from some of the best ocean models."

While it is wonderful that the United States has the supercomputing capacity to do such elegant simulations, it is astonishing that such simulations are not part of the environmental impact statement for such deep-water rigs and that such simulations have not been tweaked to simulate hypothetical oil rather than dye, which doesn't necessarily behave like spilled oil will. The NCAR researchers know this, and they are aware that oil is subject to breaking down and being eaten by bacteria.

This simulation takes into account normal weather patterns for the time show and assumes that the oil stops bubbling out of the Deepwater Horizon's busted pipe on June 20.

Imagine what might happen if the leak continues beyond that date, and then the hurricanes start coming. The National Oceanographic and Atmospheric Association has just released its predictions for the hurricane season, which started on June 1. NOAA is predicting that there is a 70 per cent chance that there will be 14 to 23 named storms (which means they have top winds of 39 miles per hour or higher), with 8 to 14 storms qualifying as hurricanes (top winds of 74 miles per hour or higher) and 3 to 7 being major hurricanes (with winds above 111 miles per hour).

The NCAR simulations do not take into account what might happen if a hurricane rips through the Gulf and then ricochets out to the Atlantic across the Panhandle. ®