Understanding the environmental processes shaping connectivity can greatly improve management and conservation actions which are essential in the trailing edge of species’ distributions. In this study, we used a dataset built from 32 populations situated in the southern limit of the kelp speciesLaminaria digitata. By extracting data from 11 microsatellite markers, our aim was to (1) refine the analyses of population structure, (2) compare connectivity patterns and genetic diversity between island and mainland populations and (3) evaluate the influence of sampling year, hydrodynamic processes, habitat discontinuity, spatial distance and sea surface temperature on the genetic structure using a distance-based redundancy analysis (db-RDA). Analyses of population structure enabled to identify well connected populations associated to high genetic diversity, and others which appeared genetically isolated from neighboring populations and showing signs of genetic erosion verifying contrasting ecological (and demographic) status in Brittany and the English Channel. By performing db-RDA analyses on various sampling sizes, geographic distance appeared as the dominant factor influencing connectivity between populations separated by great distances, while hydrodynamic processes were the main factor at smaller scale. Finally, Lagrangian simulations enabled to study the directionality of gene flow which has implications on source-sink dynamics. Overall, our results have important significance in regard to the management of kelp populations facing pressures both from global warming and their exploitation for commercial use.