While it is known that Antarctic sea ice biomass and productivity are highly variable over small spatial and temporal scales, there have been very few measurements from eastern Antarctic. Here we attempt to quantify the biomass and productivity and relate patterns of variability to sea ice latitude ice thickness and vertical distribution. Sea ice algal biomass in spring in 2002, 2003 and 2004 was low, in the range 0.01-8.41 mg Chl a m -2 , with a mean and standard deviation of 2.08 ± 1.74 mg Chl a m -2 (n = 199). An increased concentration of algae at the bottom of the ice was most pronounced in thicker ice. There was little evidence to suggest that there was a gradient of biomass distribution with latitude. Maximum in situ production in 2002 was approximately 2.6 mg C m -2 h -1 with assimilation numbers of 0.73 mg C (mg Chl a) -1 h -1 . Assimilation numbers determined by the 14 C incubations in 2002 varied between 0.031 and 0.457 mg C (mg Chl a) -1 h -1 . Maximum fluorescence quantum yields of the incubated ice samples in 2002 were 0.470 ± 0.041 with E k indices between 19 and 44 lmol photons m -2 s -1 . These findings are consistent with the shade-adapted character of ice algal communities. In 2004 maximum in situ production was 5.9 mg C m -2 h -1 with an assimilation number of 5.4 mg C (mg Chl a) -1 h -1 . Sea ice biomass increased with ice thickness but showed no correlation with latitude or the time the ice was collected. Forty-four percent of the biomass was located in bottom communities and these were more commonly found in thicker ice. Surface communities were uncommon.

Iron availability affects the growth of not only phytoplankton, but also sea ice diatoms. Iron availability had a clear effect on the growth rates of Fragilariopsis cylindrus and F. curta. Maximum growth rates were 0.57 d -1 for F. cylindrus and 0.28 d -1 for F. curta; K m (half-saturation growth constant) was 0.51 × 10 -12 and 1.3 × 10 -12 M for F. cylindrus and F. curta, respectively. For both F. cylindrus and F. curta, F v /F m (quantum yield of fluorescence) was highest and lowest for the cultures grown with the highest and lowest concentrations of iron, respectively. For both species there was also a reduction in both rETR max (maximum relative electron transfer rate) and α (photosynthetic efficiency) with decreasing iron concentration. For F. cylindrus grown with the least iron, rETR max was half of the iron-replete value, while α was reduced by 65%. Changes in E k (light-adaptation parameter) were not well defined. Immunoassays were developed for the proteins ferredoxin and flavodoxin in Antarctic pack ice. Iron availability had different effects on the expression of flavodoxin and ferredoxin in the 2 Fragilariopsis species tested. Cultures of F. cylindrus grown at high iron concentration produced predominantly ferredoxin, with a small amount of flavodoxin. Ferredoxin was sequentially replaced by flavodoxin for cultures grown with less iron, although the response was not a simple switch from one protein to the other. The ability to produce ferredoxin is apparently absent in F. curta, with relatively constant levels of flavodoxin produced at all iron concentrations. These results strongly imply that the presence of flavodoxin alone cannot be used as evidence of ironlimited growth.

Sea ice algal communities are naturally exposed to very high concentrations of dissolved oxygen, which are likely to lead to increasing stress levels and declines in productivity. To test this hypothesis, cultures of Fragilariopsis cylindrus (Grun?) Hasle, Pseudo-nitzschia sp., Fragilariopsis curta (Van Heurch), Porosira glacialis (Grunow), and Entomoneis kjellmannii (Cleve) from Antarctic sea ice and Nitzschia frigida from Arctic sea ice were exposed to elevated dissolved oxygen levels, and their growth, maximum quantum yield, relative maximum electron transport rate, and photosynthetic efficiency were measured. At oxygen concentrations equivalent to approximately four times air saturation (89% oxygen), the growth rate and maximum quantum yield were significantly reduced in all taxa. When the oxygen concentration was regularly allowed to drop, the effect on growth and quantum yield was reduced. At lower dissolved oxygen concentrations (52%), the declines in growth and quantum yield were reduced but were still mostly significantly different from the controls (21% oxygen). It is likely that the generation of excess active oxygen radicals in the presence of free oxygen is responsible for most of the decline in growth, maximum quantum yield, relative maximum electron transport rate, and photosynthetic efficiency in all species.Abbreviations: a, photosynthetic efficiency; dF/F m 0 , effective quantum yield; E k , minimum saturating irradiance; F v /F m , maximum quantum yield; NPQ, nonphotochemical quenching coefficient; rETR max , relative maximum electron transport rate; RLC, rapid light curve High dissolved oxygen concentrations under sea ice were first reported by McMinn et al. (2000). They noted that on bright sunny days, oxygen production by algae under the ice far exceeded its rate of removal by diffusion, leading to extensive oxygen bubble formation. Because of the increased solubility of oxygen in seawater at low temperatures, these organisms are likely to experience some of the highest dissolved oxygen concentrations on the planet, which is likely to be toxic to the sea ice algae. The question of how these algae cope with these high oxygen levels and other extreme conditions was also raised and discussed by Thomas and Dieckmann (2002). However, elevated dissolved oxygen concentrations are not unique to sea ice habitats but are found in the boundary layers of most microbial mat communities and frequently on the stapes or leaves of macrophytes and sea grasses. Oxygen concentrations as high as 560% air saturation, for instance, have been reported in boundary layers associated with seaweed stipes (Irwin and Davenport 2002), and similar dissolved oxygen concentrations have been observed above sediments dominated by diatoms (Raven 1991).Elevated oxygen levels are known to have an inhibitory effect on the metabolism and growth of plants (Raven 1991, Raven et al. 1994). There are two mechanisms thought to be largely responsible for these deleterious effects: the competitive effect of O 2 on RUBISCO (photorespiratio...

While the growth of Southern Ocean phytoplankton is often limited by iron availability, there are no comparable experiments on sea-ice algae. Here we assess the use of ferredoxin and flavodoxin to investigate the iron nutritional status of sea-ice algae and describe the development of a quantitative immunoassay for both proteins in marine diatoms. High-affinity monoclonal antibodies toward both proteins were produced from Cylindrotheca closterium (Ehrenb.) J. M. Lewin et Reimann, and these were used to develop Western blots. Western blots run on whole protein extracts detected both proteins with little cross-reactivity toward other proteins. The two proteins could be successfully quantitated when applied to gels at between 5 and 50 ng in a volume of 25 μL (0.2-2 μg · mL(-1) ). Flavodoxin and ferrodoxin expression was examined in the Antarctic diatoms Entomoneis kjellmannii (Cleve) Poulin et Cardinal, Navicula directa (W. Sm.) Ralfs, Fragilariopsis curta (Van Heurck) Hust., Pseudo-nitzschia sp., Porosira glacialis (Grunow) E. G. Jørg., Fragilariopsis cylindrus (Grunow) Willi Krieg., Fragilariopsis sublinearis (Van Heurck) Heiden et Kolbe, C. closterium, Nitzschia lecointei Van Heurck, and the dinoflagellate Polarella glacialis Montresor, Procaccini et Stoecker. Two Arctic isolates were also examined, Nitzschia frigida (Grunow) and Fragilariopsis oceanica (Cleve) Hasle. Significant heterogeneity of protein expression was observed despite all cultures being grown in iron-replete f/2 medium. Only one species, F. cylindrus, displayed the expected expression of ferredoxin only in iron-replete medium. Four were observed to produce both proteins under iron-replete conditions. Ferredoxin was not detected at all in F. curta and Pseudo-nitzschia sp., but distinct flavodoxin bands were observed in both of these organisms. All species examined were observed to express either flavodoxin or ferredoxin or both of the proteins as determined by Western immunoblotting.

No abstract