leaks are lumped together with breaks in pipelines caused by commercial 

 fishing pier damages, the dragging of anchors across pipelines, and struc- 

 tural failures, as well as with chronic leaks due to mechanical problems or 

 material failures. Thus, it is not possible to directly assess the impacts 

 of oil leaks on natural resources until the level of OCS activity is known, 

 production platforms placed, and pipelines and gas plants located. 



It is important to be aware of the decreased likelihood of major spills 

 if oil is brought ashore via pipeline rather than by tanker, as highlighted 

 in the conclusions of the M.I.T. study - Analysis of Oil Spill Statistics 

 (39, part 2), which are reproduced here (the purpose of the M.I.T. analysis 

 was to utilize past spill experience to estimate the likelihood of spills 

 along the Atlantic continental shelf): 



1. It has been determined that the size range of an individual 

 oil spill is extremely large--eight orders of magnitude. The 

 great majority of all spills are at the lower end of this range. 

 But most of the oil is spilled in a few very large spills. 



2. For all the reasons given in 1, point estimates of spillage and 

 spillage rates are practically meaningless. Further, from the 

 biological points of view, the frequency and magnitude of indiv- 

 idual spills is at least as important as total spillage. There- 

 fore, an estimate of the probability densities of the number of 

 spills of a given category which will occur from a given hypo- 

 thetical development and the probability density of the size of 

 these spills are both of particular significance. These esti- 

 mates are broken into six categories and result from the appli- 

 cation of an assumed spill incidence. 



Tanker/Barge 

 Platform 

 Offshore Pipel ine 



46 



