The Observer, July 2006

return to this edition

Methodology in dispersant tests is still questionable

A pair of citizens’ council observers has found continuing problems in how chemical oil-spill dispersants are tested at a federal facility in New Jersey.

The council team – Dr. Merv Fingas, a Canadian dispersants expert, and Elise DeCola, an Alaska researcher – traveled to the facility to observe cold-water dispersant testing in February and March of this year.

The first of two reports on their observations indicates that dispersant tests at the facility – known as OHMSETT, for Oil and Hazardous Materials Simulated Environmental Test Tank – still suffer from many of the same problems that council observers identified in tests conducted in 2002 and 2003. OHMSETT’s procedures in the earlier tests were also criticized by the National Academies of Science, and this year’s test were in part an effort to address some of those criticisms.

The facility is owned by the federal Minerals Management Service and operated for the agency by a contractor, MAR, Inc. The OHMSETT tank is 667 feet long by 65 feet wide, and contains 8 feet of sea water. For dispersant tests, a traveling bridge moves along the length of the tank. Oil is sprayed onto the water from the leading edge of the bridge; a few seconds later, dispersant sprayed from the trailing edge of the bridge hits the oil.

Fingas and DeCola, in the first of their reports to the council, said OHMSETT’s 2006 tests appeared to resolve some of the problems identified in the earlier tests. Those included applying dispersants to heated oil rather than oil that had been allowed to cool to ambient air temperature; the use of booms to contain the oil during the tests; and testing dispersants on artificially weathered oil rather than oil that had been allowed to weather naturally, as would occur after an actual spill.

But several other major concerns did not appear to have been addressed in the latest tests, according to the observers. Among them were:

Resurfacing of oil. This refers to the fact that oil dispersed into the water column tends to un-disperse and return to the surface in a relatively short time. According to Fingas and DeCola, an estimated 90 percent of the North Slope crude in one test resurfaced within about two hours of being sprayed with Corexit 9527, the primary dispersant stockpiled for use on Prince William Sound spills.

Herding with fire hoses. As the final stage of each test, fire hoses are used to ‘herd’ the remaining surface oil for collection and measurement. “This was observed to have the unintended consequence of temporarily re-dispersing oil,” the observers noted.

Tank contamination. The OHMSETT tank is so large that it cannot feasibly be cleaned between tests. As a result, leftover oil and dispersant from one test are still present when the next test begins. “Contamination continued to be an issue,” the observers wrote, “and sheening was visible throughout the test period from the prominent oil stain around the edges of the tank.”

Use of oils not transported through Prince William Sound. OHMSETT gets its samples of North Slope crude from Pump Station One at the start of the trans-Alaska pipeline. However, refineries along the pipeline right-of-way take some crude out for refining, then pump the refinery remnants back into the pipeline stream. “Therefore,” the Observers wrote, “the test oils vary in composition from oil that might actually be spilled in Prince William Sound. . . . In the future, test oils should be obtained from the Valdez Marine Terminal.”

Uncontrollable natural forces. Because OHMSETT is an outdoor facility located on the New Jersey seashore, it is not possible to control such important variables as air temperature, sunlight, and wind speed, or to factor in their influence on dispersion rates. The council observers reported that, during their visit, air temperatures ranged from the low 30s to the low 40s, while winds varied from 10 mph to 40 mph. On two of the four days the observers were there, winds blew constantly at 15-20 mph, with occasional gusts near 40 mph.

Unrealistic time frame of dispersant application. At OHMSETT, dispersant is sprayed onto the oil a few seconds after the oil hits the water in the tank, long before a slick has time to form. In a real spill, several hours would almost certainly pass before dispersant could be applied. By then, the oil would form a slick.

Unrealistic amounts of dispersant. The usual recommendation is that dispersant be applied at a ratio of 1:20 to the oil. For example, if 100 gallons of oil were spilled, 5 gallons of dispersant should be applied. However, at OHMSETT, dispersant-to-oil ratios appeared to range as high as 1:4.7, according to the operator logs from the facility. For a 100-gallon spill, that would work out to about 21 gallons of dispersant. At that ratio, treating an 11 million gallon spill like the Exxon Valdez would require 2.3 million gallons of dispersant.

As a result of the problems they saw at OHMSETT, wrote Fingas and DeCola, “we do not believe that accurate quantification of dispersant effectiveness is possible under the observed experimental design.”

Their report, titled “Observers’ Report – MMS Cold Water Dispersant Tests,” is available on the council website at www.pwsrcac.org/docs/d0028000.pdf or by request to either council office.

The second report by Fingas and DeCola is titled “Analysis of Oil Spill Dispersant Testing at OHMSETT.” It will provide a technical analysis of the experimental design and testing protocols for this year’s OHMSETT tests and should be available soon.