Résumé: The scleractinian coral Lophelia pertusa has been the focus of deep-sea research since the recognition of the vast extent of coral reefs in North Atlantic waters two decades ago, long after their existence was mentioned by fishermen. These reefs where shown to provide habitat, concentrate biomass and act as feeding or nursery grounds for many species, including those targeted by commercial fisheries. Thus, the attention given to this cold-water coral (CWC) species from researchers and the wider public has increased. Consequently, new research programs triggered research to determine the full extent of the corals geographic distribution and ecological dynamics of “Lophelia reefs”. The present study is based on a systematic standardised sampling design to analyze the distribution and coverage of CWC reefs along European margins from the Bay of Biscay to Iceland. Based on Remotely Operated Vehicle (ROV) image analysis, we report an almost systematic occurrence of Madrepora oculata in association with L. pertusa with similar abundances of both species within explored reefs, despite a tendency of increased abundance of L. pertusa compared to M. oculata toward higher latitudes. This systematic association occasionally reached the colony scale, with “twin” colonies of both species often observed growing next to each other when isolated structures were occurring offireefs. Finally, several “false chimaera” were observed within reefs, confirming that colonial structures can be “coral bushes” formed by an accumulation of multiple colonies even at the inter-specific scale, with no need for self-recognition mechanisms. Thus, we underline the importance of the hitherto underexplored M. oculata in the Eastern Atlantic, reestablishing a more balanced view that both species and their yet unknown interactions are required to better elucidate the ecology, dynamics and fate of European CWC reefs in a changing environment.

Résumé: Aim: To infer cold-water corals' (CWC) post-glacial phylogeography and assess the role of Mediterranean Sea glacial refugia as origins for the recolonization of the northeastern Atlantic Ocean. Location: Northeastern Atlantic Ocean and Mediterranean Sea. Taxon: Lophelia pertusa, Madrepora oculata. Methods: We sampled CWC using remotely operated vehicles and one sediment core for coral and sediment dating. We characterized spatial genetic patterns (microsatellites and a nuclear gene fragment) using networks, clustering and measures of genetic differentiation. Results: Inferences from microsatellite and sequence data were congruent, and showed a contrast between the two CWC species. Populations of L. pertusa present a dominant pioneer haplotype, local haplotype radiations and a majority of endemic variation in lower latitudes. Madrepora oculata populations are differentiated across the northeastern Atlantic and genetic lineages are poorly admixed even among neighbouring sites. Conclusions: Our study shows contrasting post-glacial colonization pathways for two key habitat-forming species in the deep sea. The CWC L. pertusa has likely undertaken a long-range (post-glacial) recolonization of the northeastern Atlantic directly from refugia located along southern Europe (Mediterranean Sea or Gulf of Cadiz). In contrast, the stronger genetic differentiation of M. oculata populations mirrors the effects of long-term isolation in multiple refugia. We suggest that the distinct and genetically divergent, refugial populations initiated the post-glacial recolonization of the northeastern Atlantic margins, leading to a secondary contact in the northern range and reaching higher latitudes much later, in the late Holocene. This study highlights the need to disentangle the influences of present-day dispersal and evolutionary processes on the distribution of genetic polymorphisms, to unravel the influence of past and future environmental changes on the connectivity of cosmopolitan deep-sea ecosystems associated with CWC.