Description:
About petroleum spills in Washington State:
In 1997, the Washington State Department of Ecology published an historical analysis of oil spills, that assembled the available information on spill type, size and method of occurrence. The Department of Ecology paper summarized oil spills going back to the 1960's, however it was not until the early 1990's that systematic reporting of all spills, regardless of size, took place. Before 1990, only larger spills were systematically reported, which makes a statistical analysis of the older data very problematic.
The data include both marine and land-based spills (underground leaking storage tanks are excluded), and include a wide variety of petroleum byproducts such as crude oil, fuel oil, lubricating oils, diesel fuel, jet fuel and gasoline (but no vegetable or animal oils). While all of these petroleum products are serious contaminants, the heavier hydrocarbons (crude oil and fuel oil) are much less volatile and much more persistent, even in high energy coastal environments such as Prince William Sound, Alaska. Heavy hydrocarbons that settle into beach deposits can be excavated by storms, re-contaminating the nearshore environment for decades following the spill.
Because the Washington State petroleum spills in the data set range from 30 gallons to 600,000 gallons, over 4 orders of magnitude, the logarithm of the spill size is calculated, binned and plotted against its frequency. As expected, small spills are the most abundant, whereas large spills are rather infrequent. Many phenomena such as landslides, earthquakes, volcanic eruptions, nuclear accidents, auto wrecks, and wildfires show a similar distribution of sizes; many small events, some medium sized events, and few large events. Size versus frequency for these phenomena often exhibit a power law relationship. If petroleum spills in Washington State follow this same relationship, then the Department of Ecology's compilation under-represents the smallest spills, which should be 2-3 times more frequent. This under-representation is consistent with the historical lack of comprehensive reporting.
Since the period of record is known (5 years, from 1991-1996), the recurrence interval for a spill of a given size can be calculated. Extrapolation to larger sizes using the power law can yield recurrence intervals for these rare events. On New Year's day in 1972, the General M. C. Meigs, a US Navy supply vessel, spilled 2.3 million gallons of heavy fuel oil into the Straights of Juan de Fuca between Washington State and Canada. What is the recurrence interval for an event of this size?
