AN ANALYSIS OF OIL-SPILLS FROM VESSEL TRAFFIC ACCIDENTS

In addition to the traditional activities such as fishing and transpor tation, a steady growth in world trade, recreation, and exploration of underwater resources has caused a rapid growth in vessel traffic for several decades. This has increased the potential risk of marine casua lties involving loss of life and damage to vessels, cargo, navigationa l aids, and structures (e.g. bridges). Various Asian, European, and No rth American countries have enacted a wide array of safety measures to reduce the casualties. For instance, the U.S. Coast Guard requires al l incoming vessels to have a local pilot on board, have an officer on the bridge, and that the vessel operator should establish radio commun ications with the Coast Guard after getting on board. In addition, tra ffic separation lanes and convergence points have been established to improve traffic flow. Navigational aids assist the vessel operators by sensing oncoming traffic hazards such as bridges and shallow waters. The U.S. Army Corp of Engineers keeps ship channels open by regularly dredging waterways that are used for commercial traffic. In many count ries, offshore deep water ports similar to the loop in New Orleans, Lo uisiana, have been constructed to reduce traffic near shallow and cong ested ports. In spite of all these safety measures, accidents do happe n. In this paper, we study the historic data related to the vessel acc idents over a period of several years. In particular, we focus on the distribution of sizes of the oil spills that have occurred during thes e accidents. The sizes of the oil spills are ranked from largest to sm allest. The regression analysis showed a strong linear relationship be tween the natural logarithm of size of casualty and the natural logari thm of its rank. The regression coefficients may indicate the 'degree of damage' in a particular year or for a specific geographic region un der investigation. It could be argued that they could be used to asses s an aggregated impact of the implemented safety measures. (C) 1997 El sevier Science Ltd.