Phytoplankton and bacterial assemblages in ballast water of U.S. military ships as a function of port of origin, voyage time, and ocean exchange practices

We characterized the physical/chemical conditions and the algal and bacterial assemblages in ballast water from 62 ballast tanks aboard 28 ships operated by the U.S. Military Sealift Command and the Maritime Administration, sampled at 9 ports on the U.S. West Coast and 4 ports on the U.S. East Coast. The ballast tank waters had been held for 2-176 days, and 90% of the tanks had undergone ballast exchange with open ocean waters. Phytoplankton abundance was highly variable (grand mean for all tanks, 3.21 × 10⁴ viable cells m^-3; median, 7.9 × 10³ cells m^-3) and was unrelated to physical/chemical parameters, except for a positive relationship between centric diatom abundance and nitrate concentration. A total of 100 phytoplankton species were identified from the ballast tanks, including 23 potentially harmful taxa (e.g. Chaetoceros concavicornis, Dinophysis acuminata, Gambierdiscus toxicus, Heterosigma akashiwo, Karlodinium veneficum, Prorocentrum minimum, Pseudo-nitzschia multiseries). Assemblages were dominated by chain-forming diatoms and dinoflagellates, and viable organisms comprised about half of the total cells. Species richness was higher in ballast tanks with coastal water, and in tanks containing Atlantic or Pacific Ocean source waters rather than Indian Ocean water. Total and viable phytoplankton numbers decreased with age of water in the tanks. Diversity also generally decreased with water age, and tanks with ballast water age >33 days did not produce culturable phytoplankton. Abundance was significantly higher in tanks with recently added coastal water than in tanks without coastal sources, but highly variable in waters held less than 30 days. Bacterial abundance was significantly lower in ballast tanks with Atlantic than Pacific Ocean source water, but otherwise was surprisingly consistent among ballast tanks (overall mean across all tanks, 3.13 ± 1.27 × 10¹¹ cells m^-3; median, 2.79 × 10¹¹ cells m^-3) and was unrelated to vessel type, exchange status, age of water, environmental conditions measured, or phytoplankton abundance. At least one of four pathogenic eubacteria (Listeria monocytogenes, Escherichia coli, Mycobacterium spp., Pseudomonas aeruginosa) was detected in 48% of the ballast tanks, but toxigenic strains of Vibrio cholerae were not detected. For ships with tanks of similar ballasting history, the largest source of variation in phytoplankton and bacteria abundance was among ships; for ships with tanks of differing ballasting histories, and for all ships/tanks considered collectively, the largest source of variation was within ships. Significant differences in phytoplankton abundance, but not bacterial abundance, sometimes occurred between paired tanks with similar ballasting history; hence, for regulatory purposes phytoplankton abundance cannot be estimated from single tanks only. Most tanks (94%) had adequate records to determine the source locations and age of the ballast water and, as mentioned, 90% had had ballast exchange with open-ocean waters. Although additional data are needed from sediments that can accumulate at the bottom of ballast tanks, the data from this water-column study indicate that in general, U.S. Department of Defense (DoD) ships are well managed to minimize the risk for introduction of harmful microbiota. Nevertheless, abundances of viable phytoplankton with maximum dimension >50 μm exceeded proposed International Maritime Organization standards in 47% of the ballast tanks sampled. The data suggest that further treatment technologies and/or alternative management strategies will be necessary to enable DoD vessels to comply with proposed standards.