Search

Spatial and interannual variations of phytoplankton diversity were characterized along the French Atlantic Coast in relation to physical factors and large-scale climatic indices using phytoplankton surveys conducted from 1993 to 2010 in four geographical areas. This study relates phytoplankton diversity to oyster recruitment success by comparing a 'specific' versus a 'functional' diversity approach. Functional diversity was represented by functional groups of different phytoplanktonic functional traits (size, shape, toxicity), relevant for oyster feeding during growth, reproduction and larval development. Phytoplankton diversity patterns along the French Atlantic coast corresponded to a geographical distribution mainly driven by a latitudinal gradient and hydrodynamic features as confirmed by the functional characteristics of the indicator species recorded for each geographical area (C-S-R strategies of Reynolds, specific habitat). This geographical typology was less clear for functional diversity. Only few functional groups were explained by physical factors: the non-toxic small and solitary cells were in particular related to temperature, with high densities observed in the southern areas. Even if temporal variation was less important in explaining the phytoplankton diversity patterns, the main drivers explaining the interannual pattern were the large-scale climatic indices, mainly the Atlantic Multidecennal Oscillation. Functional groups were readily explained by climatic indices than species even if the relationships cannot be applied generally because of the non-linearity of the correlations (local and temporal variations). However, the potentially toxic, small and linear phytoplankton were anticorrelated to NAO. The functional approach thus brings constructive elements concerning the relationships between the prey assemblage of oysters and their physical drivers. Redundancy and co-inertia analyses appeared as complementary analyses in investigating phytoplankton pattern of variation, being particularly useful in analyzing geographical and temporal diversity fluctuation, respectively. (C) 2012 Published by Elsevier Ltd.