Morphometric comparison of Australian Shy and New Zealand White-capped Albatrosses

Albatrosses are frequently killed by longline and trawl fishery operations but the relative impact of such activities at the species or population level are largely unknown. Such information requires the widespread presence of fishery observers and an ability to identify accurately the species and provenance of all albatrosses killed by fishing vessels. In this study we investigate the use of morphometric measurements to identify Shy (Thalassarche cauto) from White-capped (T. steadi) Albatrosses, two taxa recently suggested to be separate species. Measurements were taken from a collection of 103 Shy and White-capped Albatrosses killed by longline vessels within the Australian Fishing Zone between 1988 and 2000 and identified to species level using a recently developed DNA-based test. Within-sex comparisons of Shy and White-capped Albatrosses found that six of the 10 measurements were significantly different for both sexes. However, all measurements showed considerable overlap and no single measurement separated the two taxa. Discriminant classification functions based on wing chord, maximum head width and two bill measurements were able to simultaneously identify the species and sex of approximately 84% of bycatch specimens (n = 70). The discriminate classification functions for species identification alone correctly assigned approximately 89% of bycatch specimens. When this classification method was applied to measurements taken from live specimens a similar level of accuracy was achieved (82%, n = 17).