SHORT COMMUNICATION Heterogeneous zooplankton distribution in a ship’s ballast tanks

ships’ ballast water has apparently enabled numerous species to expand their ranges to environments in which they did not evolve (Carlton 1998; Cohen and Carlton, 1998; Ruiz et al., 2000). In normal trading operations, ships load seawater ballast in one port, then discard it at another in exchange for cargo. Modern international voyages span just a few days or weeks, affording entrained organisms a unique opportunity for transport to biogeographically remote locations which would otherwise be rarely accessible. Severe, and potentially irreversible, impacts of introduced aquatic organisms have been demonstrated in many parts of the world, for example, the zebra mussel Dreissena polymorpha in the Great Lakes of North America (Roberts, 1990; Strayer et al., 1998), and the toxic dinoflagellate Gymnodinium catenatum in southern Australia (Bolch and Hallegraeff, 1990). The identification of ballast water as a probable dispersal mechanism for these and numerous other exotic organisms has prompted numerous international agencies to set-up ballast water sampling programmes. However, recent surveys reveal the use of a wide range of zooplankton sampling equipment and methodologies (Sutton et al., 1998; Gollasch, in preparation); including pump sampling via sounding pipes (Hay et al., 1996), bucket hauls (Gollasch et al., 2000) and full-depth (Wonham et al., 2001) and near-surface (Ruiz et al., 2000; Smith et al., 2000) plankton net tows. The assessment of risk associated with ballast water discharge depends on reliable knowledge of the identity, viability and quantity of its inhabitants (Hayes and Hewitt, 1998). Implicit in any estimation of the quantity of organisms in a ballast tank is the assumption that the samples provide an unbiased estimate of the zooplankton densities throughout the tank. While the difficulties inherent in obtaining quantitative and representative samples of ballast organisms have long been recognized (Carlton, 1982, 1985; Gollasch et al., 2000), even basic assumptions relating to zooplankton distribution have remained untested. Vertical concentration gradients of zooplankton are well studied in open water [e.g. (Pearre, 1973; Lampert and Taylor, 1985)]. However, several factors peculiar to ballast tank environments might be expected to distinguish the distribution of their zooplankton populations— first, the complete absence of light; second, a potentially diminishing food supply imposed by the isolation of the tank from external inputs and third; relatively rapid changes in water temperature associated with voyages between different latitudes [e.g. (Carlton, 1982; Gollasch et al., 2000)]. We assessed vertical trends in zooplankton abundance within the ballast tanks of the MV ‘Iron Sturt’ against the null hypothesis that zooplankton are randomly distributed in order to identify potential problem areas for quantifying zooplankton in ballast water. Ballast water sampling was conducted on the bulk carrier MV ‘Iron Sturt’ during two commercial voyages