A statistical analysis of 45 oil spills shows a weak log-log correlation between spill volume and numbers of seabirds killed. This relationship cannot be used to predict mortality and loses its significance if one extreme case is omitted. The data show the wide variance in mortality in spills of all sizes. A loose 'rule-of-thumb' that is often used in poorly documented spills is that the overall mortality is ten times the actual body count. There is no justification for this notion. The mean estimate used is 4-5 times the body count, but each spill should be examined independently.
All data in the Integrated Marine Information System (IMIS) is subject to the VLIZ privacy policy
