For several decades now, species distribution models (SDMs) have been a promising area of ecological research. The aim of the present study is to define optimal ecological niches and habitat suitability for the population of the bivalve Cerastoderma edule in the Seine estuary. The method involved applying quantile regression to a 20-year biological dataset at the scale of the estuary, coupled with a hydro-morpho-sedimentary model data set validated over a longer period (25 years) also at the scale of the estuary, using 100-m mesh cells. This study was carried out to describe biological responses to environmental factors involved in defining an optimal ecological niche, using the bifactorial Gaussian equation using physical forcings (tidal currents, bed shear stress, etc.) as explanatory factors. On the basis of a preliminary multivariate analysis of the physical descriptors, a comparison was made between three different types of equation (linear, B-spline and Gaussian) in four sets of paired environmental factors: daily maximum current speed & inundation time, daily salinity range & temperature, daily salinity range & bathymetry, daily maximum bed shear stress & mud content. The non-linear quantile regression with a bifactorial Gaussian equation produced the best description of habitat suitability and optimal niches, at the 95th centile and using the biomass (gAFDW/m² - Ash Free Dry weight). Daily maximum current speed & inundation time and daily salinity range & temperature were the most pertinent SDMs. The optimal ecological niche for C. edule appeared to be lower intertidal marine areas, with temperate and low dynamic waters, settled in muddy sand sediment of the tidal flats of Seine estuary. Using this technique, the calculation of optimal niches in this ecosystem was explored over a period of 25 years and analysed in isolated sectors and can now be applied in different scenarios related to the global warming. We propose several reliable models, so that different kinds of prediction can now be applied according to the context of future scenarios.