We present a three-dimensional physical-biological modelling study aiming to infer the effect of freshwater-associated bulk turbidity on the Gulf of St. Lawrence (Canada) plankton ecosystem. Bulk turbidity is parameterised using an inverse relationship derived from an extensive in situ dataset linking salinity to the diffuse attenuation coefficient of photosynthetically available radiation (PAR) due to nonchlorophyllous matter. Embedding bulk turbidity in the model led to shallowing of the photic zone in the estuarine plume in accordance with coincident observations and allowed a better discrimination between Case 1 (chlorophyll-mediated variability of the photosynthetic available radiation attenuation) and Case 2 waters. The spring bloom was delayed. and primary and secondary production rates as well as the export of biogenic matter at depth decreased in the freshwater-influenced subregions. Comparisons with literature and coincident in situ measurements showed that nitrates were over- and underestimated in the run with and without bulk turbidity, respectively. A sensitivity analysis was performed with a relatively simple but robust parameterisation of photoacclimation, i.e. the adjustment of the phytoplankton photosynthetic efficiency to local underwater light conditions. Photoacclimation allowed simulated chlorophyll and nitrate concentrations as well as lateral fluxes of nitrate to achieve the best agreement with coincident measurements and literature estimates, respectively. This study showed that accounting for the freshwater-associated bio-optical variability and phytoplankton response in terms of photosynthetic efficiency improved the model's ability to predict the plankton ecosystem dynamics and associated biogeochemical fluxes in the river-influenced Gulf of St. Lawrence. (C) 2010 Elsevier B.V. All rights reserved.