Microbial responses to experimental sea-ice formation: Implications for the establishment of Antarctic sea-ice communities

Abstract

The fate of algae and bacteria during the transition from open water to early stages of sea-ice formation was investigated under simulated conditions in the laboratory. Distribution patterns and metabolic activities of three common Southern Ocean diatoms ( Nitzschia curta, Thalassiosira tumida, Chaetoceros sp.) and an Antarctic bacterial community were determined after 3 and 14 days of incubation in an insulated 301 plastic vessel. Activity measurements suggest that a close coupling existed between these two groups of organisms prior to ice formation. After 3 days of freezing at −5 °C, cell densities and biomasses of algae increased in pore water within an ice-pancake that formed during this period. Accumulation in the pore water exceeded the concentration effect caused by freezing out of water. Bacteria showed similar increases in the presence of algal cells during freezing. As the ice incorporated bacterial populations experienced a strong metabolic inhibition, bacterial growth as a reason for enhanced cell numbers in the pore water seems to be unlikely. Reduced metabolic activities were also recorded for the algal species, most pronounced in T. tumida which showed lowest cell-specific assimilation rates accompanied by a high cell mortality after 3 days of freezing. It is hypothesized that scavenging of algal cells by ice crystals in conjunction with attachment of bacteria onto algal cells was predominantly responsible for the observed enrichment patterns. Different capacities of the algae to concentrate bacterial cells in the pore water were related to differences in algal surface area available for bacterial colonization, which varied among the three species due to different morphologies and cell concentrations. After 2 weeks of incubation under simulated ice conditions and at salinities of 50%., activity of algae and bacteria increased again. In contrast to observations made in open water prior to freezing, no influence of algal species on bacterial activity was recorded after this 2-week period. It is concluded that a bacterial community different to that of the open water had developed which needs a longer time span than employed in the present experiment to establish close metabolic coupling between algae and bacteria as recorded for microbial communities of thick pack and fast ice.