Ecological carryover effects occur in any situation in which an individual’s previous history and experience (e.g., pelagic larval stages) influence performance in a subsequent state (e.g., benthic or demersal juveniles). Using a combination of geometric morphometrics, otolith microstructure analysis, satellite data, and numerical modeling, we studied the potential carryover effect between environmental conditions, larval growth, and the morphology of rockfish Sebastes oculatus settlers in a fjord-type inlet in southern Chile. We detected two sympatric morphotypes of young-of-the-year rockfish, one with a robust head, taller preoperculum, larger mouth, and frontally inserted pectoral fins, and a second with a slender head, shorter preoperculum and mouth, and distally inserted pectoral fins. Body conditions, otolith size, and age were similar between morphs. When past growth was analyzed, however, different increment width trajectories were found, particularly between 40- and 60-days post-hatch. The combination of otolith growth trajectories from each morph with the spatial heterogeneity of satellite data (seawater temperatures between 9 and 15 °C, chlorophyll-a between 4 and 10 mg m−3), within the inner sea of northern Chilean Patagonia, along with particle tracking and back-tracking hydrodynamical models, suggests that the focal site (i.e., where YOY were collected) was supplied by discrete cohorts of larvae with different patterns of phenotypic variation, which may have traveled through distinct water masses. Therefore, these results suggest ecological carryover effects of the growth history of larval stages on the phenotype of rockfish settlers. The identification of carryover effects at different temporal scales of the life history of demersal fish may lead to a better understanding of the factors affecting variation in individual fitness.