Résumé
Deep ocean hydrothermal vent ecosystems face physical disturbances from naturally occurring volcanic and tectonic activities and are at increasing risk of mineral resource exploitation, raising concerns about the resilience of endemic biological communities. Following destructive events, efficient and rapidly applicable surveys of organisms are required to monitor the state, evolution and a possible return of these ecosystems to their original baseline status. In this study, we explored the environmental DNA (eDNA) approach as a tool (1) to assess biodiversity of benthic communities associated with deep-sea hard substrata and (2) tracked the recolonization dynamics of benthic invertebrate communities living on the Montségur edifice within the Lucky Strike vent field (Mid-Atlantic Ridge), after an induced disturbance that consisted of faunal clearance within experimental quadrats. Hard substratum samples were collected prior to and one year after the disturbance, for eDNA metabarcoding using one marker of the mitochondrial Cytochrome Oxidase I (COI) gene and three markers of the nuclear 18S ribosomal RNA (18S) gene. We also generated a DNA barcoding inventory that consisted of taxa physically collected from Montségur and morphologically identified. This inventory contained amplified barcodes from COI, 18S and the nuclear large subunit ribosomal RNA (28S) gene. The resulting sequence information from the COI and 18S were used for eDNA taxonomic assignment. The eDNA datasets uncovered a high diversity of metazoan OTUs, which included macro- and meiofauna common to Lucky Strike. Baseline data collected at the start of the experiment identified higher OTU richness at sites peripheral to the active edifice, as well as at inactive sites. One year following the initial disturbance, analysis of recolonization data found no statistical difference in presence/absence from baseline communities. The eDNA protocols provide a reproducible strategy to quickly assess biodiversity associated with deep sea hard substratum, enabling comparisons across various habitats. To follow recolonization dynamics at small spatial scales, however, we recommend an approach that uses both molecular and morphological-based traditional methods. Finally, we present original data on the “unseen” diversity of the fauna inhabiting the poorly studied inactive sites, locations that are targeted by commercial mining. Continued monitoring of these sites is currently ongoing and will bring new insight on recovery potential over time, with the ultimate goal of informing conservation and management decisions in relation to the protection of hydrothermal vent ecosystems.