Removal of natural populations of marine plankton by a large-scale ballast water treatment system

Large-scale experiments using a hydrocyclone, a self-cleaning 50 mum screen, and a UV unit were undertaken to evaluate the treatment efficiency of these commercially available units for preventing the transfer of unwanted species via ships' ballast water. The water flow through the treatment system was approximately 5.7 m(3) min(-1). The effect of increased suspended solids on these processes was the focus of this research. During each experimental run, 760 1 samples were obtained and passed through 35 pm plankton nets for zooplankton collection. Samples were also collected for phytoplankton, microbiological, ATP and protein analyses. After the initial samples were obtained, a second set of samples was held for 18 h to determine the effects of storage on the effectiveness of treatment processes. Screening the seawater at 50 mum removed most of the zooplankton and a small percentage of the microphytoplankton, but hydrocyclonic separation was not effective. Initially, UV treatment reduced the viable count of microorganisms to an undetectable level; however, bacterial regrowth was observed in the samples held for 18 h. Statistical evaluation showed that increased turbidity (5 to 90 nephelometer turbidity units; NTU) had no effect on the treatment regime, even on the UV unit. At the highest turbidity (90 NTU), the UV dose was lowered to approximately 35 MW S cm(-2); however, this dose was still sufficient to inactivate microorganisms. Overall, it was observed that only the 50 mum screen was effective in the removal of organisms, especially potential invading organisms such as large zooplankton or invertebrate larvae.