Organismal fecundity (F) and its relationship with body size (BS) are key factors in predicting species distribution under current and future scenarios of global change. A functional trait-based dynamic energy budget (FT-DEB) is proposed as a mechanistic approach to predict the variation of F and BS as function of environmental correlates using two marine bivalves as model species (Mytilus galloprovincialis and Brachidontes pharaonis). Validation proof of model skill (i.e., degree of correspondence between model predictions and field observations) and stationarity (i.e., ability of a model generated from data collected at one place/time to predict processes at another place/time) was provided to test model performance in predicting the bivalve distribution throughout the 22 sites in the Central Mediterranean Sea under local conditions of food density and body temperature. Model skill and stationarity were tested through the estimate of commission (i.e., proportion of species' absences predicted present) and omission (i.e., proportion of presences predicted absent) errors of predictions by comparing mechanistic predicted vs. observed F and BS values throughout the study area extrapolated by lab experiments and literature search. The resulting relationship was reliable for both species, and body size and fecundity were highly correlated in M. galloprovincialis compared to B. pharaonis; FT-DEB showed correct predictions of presence in more than 75 % of sites, and the regression between BS predicted vs. observed was highly significant in both species. Whilst recognising the importance of biotic interactions in shaping the distribution of species, our FT-DEB approach provided reliable quantitative estimates of where our species had sufficient F to support local populations or suggesting reproductive failure. Mechanistically, estimating F and BS as key traits of species life history can also be addressed within a broader, scale-dependent context that surpasses the limitations related to correlative species distribution models.