Interannual variability in the distribution of the phytoplankton standing stock across the seasonal sea-ice zone west of the Antarctic Peninsula

The spatial distribution of phytoplankton cell abundance, carbon (C) biomass and chlorophyll a (Chl a) concentration was analysed during three summers (1996, 1997 and 1999) in a seasonal sea-ice area, west of the Antarctic Peninsula. The objective of the study was to assess interannual variability in phytoplankton spatial distribution and the mechanisms that regulate phytoplankton accumulation in the water column. Phytoplankton C biomass and Chl a distributions were consistent from year to year, exhibiting a negative on/offshore gradient. The variations in C concentration had a close and non-linear relationship with the upper mixed layer depth, suggesting that the vertical mixing of the water column is the main factor regulating phytoplankton stock. The magnitude of C gradients was 5-fold higher during 1996 than during 1997 and 1999. This was ascribed to interannual variations in the concentration of diatom blooms in the region influenced by sea-ice melting. Vertical distribution of the phytoplankton, as estimated from Chl a profiles, also varied along an on/offshore gradient: Chl a was distributed homogeneously in the upper mixed layer in coastal and mid-shelf stations and concentrated in the deep layer (40-100 m) occupied by the winter waters (WW, remnants of the Antarctic surface waters during summer) in more offshore stations. The region with a deep Chl a maximum layer (DCM layer) was dominated by a phytoplankton assemblage characterized by a relatively high concentration of diatoms. The extent of this region varied from year to year: it was restricted to pelagic waters during 1996, extended to the shelf slope during 1997 and occupied a major portion of the area during 1999. It is hypothesized that iron depletion in near surface waters due to phytoplankton consumption, and a higher concentration in WW, regulated this vertical phytoplankton distribution pattern. Furthermore, we postulate that year-to-year variations in the spatial distribution of the DCM layer were related to interannual variations in the timing of the sea-ice retreat. The similarity between our results and those reported in literature for other areas of the Southern Ocean allows us to suggest that the mechanisms proposed here as regulating phytoplankton stock in our area may be applicable elsewhere.