Dinoflagellate cysts as environmental indicators in surface sediments from the Congo deep-sea fan and adjacent regions

Forty-nine surface sediment samples from the Congo deep-sea fan and adjacent hydrographic environments were prepared palynologically and analyzed for fossilizable dinoflagellate cysts. The main objective was to investigate the extent to which cysts reflect present-day hydrography, and therefore may be used downcore as paleoceanographic indicators. The region sampled covers a range of marine environments including the Congo River plume, the area of pronounced upwelling off West-Central Africa, and a nearshore–offshore gradient from coastal to oceanic waters. This is the first study of the effect of a major tropical river plume on associated cyst assemblages. Cyst counts were low, particularly near the river mouth and furthest offshore. A combination of statistical treatments was used to analyze the surface cyst assemblages, including cluster analysis, correspondence analysis and canonical correspondence analysis. This helped to identify five distinctive sample groupings representing the main ecological signals in the cyst assemblage data. Group A, dominated by spherical brown protoperidinioid cysts, is an upwelling signal. Group B, dominated by the cosmopolitan species O. centrocarpum, is the main signal for the river plume effect, interpreted as reflecting the environmental instability this creates. Group C, dominated by S. delicatus with accessory O. centrocarpum, characterizes the outer plume, somehow reflecting the mixing of progressively diluted river plume water and the surrounding oceanic waters. Group D, with high proportions of Impagidinium and Nematosphaeropsis together with O. centrocarpum and S. delicatus, characterizes the location of the Angola–Benguela Front. Group E, with mainly Impagidinium species, is an oceanic signal. The identification of nutrients and water stability as main determining factors influencing cyst signals has implications for cyst-based paleoenvironmental interpretations using transfer functions, which are developed almost exclusively from comparisons with more easily available data for sea-surface temperature, salinity, and sea ice.