The OSRA model lives deep within the Department of the Interior and the acronym stands for Oil Spill Risk Analysis.
For an organizational map, there is the Department of the Interior, a secretariat and cabinet level agency that is broadly charged with protecting our environment, managing our natural resources, protecting tribal and cultural interests, and providing energy.
The Department of the Interior has several agencies and bureaus, the most famous being the Bureau of Indian Affairs and the National Park Service. The bureau that is the home of OSRA is the Bureau of Ocean Energy Management, Regulation, and Enforcement, or BOEMRE. This bureau used to be part of the now disbanded Minerals Management Service.
Now that we have made our way to BOEMRE, let's take a look at OSRA. This analytical system evaluates the probability of oil spill contacts and occurrances based on information about past oil spills and other probability calculations.
The initial model was developed in 1975 by the Department of the Interior. Specifically mentioned are Smith, et al (1982) and Anderson and LaBelle (1985). The Minerals Management Service contributed databases on oil spills, originally with the Spills and Summary database of 1964-1995 and currently with the 1996-2011 Outer Continental Shelf Related Incidents summary and statistics database.
The model was revised in 2000 by Anderson and LaBelle, using extensive spill data that goes up to 1999. Areas of data included: US OCS pipeline and tanker spills, worldwide tanker spills and US barge spills.
Normalization was a process to relate volume of the crude oil that was being handled to the number of spills.
Restriction was a process to exclude spills of less than 1000 barrels.
Various exposure variables are considered.
Confidence Intervals were set at 95 percent.
Peer review of the work of Anderson and Labell and others was part of the process.
The bottom line is that this complicated model will give many a student of statistics and probability a run for their money as they go through the various reasonings that are and were used to predict the likelihood of oil spillage in the Outer Continental Shelf. How this model will be useful, given the increase in deep water oil extraction is unknown, but the recent BP oil disaster will certainly skew the data and will cause many to pay attention to the probability of other spills of the deep water extraction variety.