Annual and interannual variability in phytoplankton at a permanent station off Kerguelen Islands, Southern Ocean

From November 1992 to February 1995 a quantitative and qualitative phytoplankton study was conducted at a permanent station (Kerfix) southwest off the Kerguelen Islands, in the vicinity of the Polar Front (50°40′S–68°25′E). Phytoplankton populations are low in this area both during summers and winters. They consist, in order of decreasing cell abundance, of pico- and nanoflagellates (1.5–20 μm), coccolithophorids (<10 μm), diatoms (5–80 μm) and dinoflagellates (6–60 μm). Flagellates form the dominant group throughout the year and attain the highest summer average of 3.0 × 10⁵ cells l⁻¹. Next in abundance year-round are coccolithophorids with the dominant Emiliania huxleyi (highest summer 1992 average 1.9 × 10⁵ cells l⁻¹), diatoms (summer 1992 average 1.0 × 10⁵ cells l⁻¹) and dinoflagellates (average 3.8 × 10⁴ cells l⁻¹). Winter mean numbers of flagellates and picoplankton do not exceed 8.4 × 10⁴ cells l⁻¹; those of the three remaining algal groups together attain 2 × 10⁴ cells l⁻¹. Summer peaks of diatoms and dinoflagellates are mainly due to the larger size species (>20 μm). The latter group contributes most to the total cell carbon biomass throughout the year. Dominant diatoms during summer seasons include: Fragilariopsis kerguelensis, Thalassionema nitzschioides, Chaetoceros dichaeta, C. atlanticus, Pseudonitzschia heimii, and P. barkleyi/lineola. This diatom dominance structure changes from summer to summer with only F. kerguelensis and T. nitzschioides retaining their first and second positions. Any one of the co-dominant species might be absent during some summer period. The variable diatom community structure may be due to southward meandering of the Polar Front bringing “warmer” species from the north, and to the mixing of the water masses in this area. The entire community structure characterized both during summer and winters by the dominance of flagellates can be related to deep mixing (ca. 40–200 m) of the water column as the probable controlling factor. Received: 13 November 1997 / Accepted: 11 May 1998