Advancing species distribution modeling with multispecies eDNA metabarcoding: the case of the Atlantic bluefin tuna (Thunnus thynnus) in the Gulf of St. Lawrence (Canada)

Quantifying a species' distribution and identifying its ecological drivers are essential for effective ecosystem monitoring, assessment, and management. In marine ecosystems, conventional fishery-based methods are often inadequate for producing indicators of the abundance and distribution of large, fast-moving pelagic predators such as Atlantic Bluefin Tuna (ABFT, Thunnus thynnus ). In the Gulf of St. Lawrence (GSL), Canada, fisheries provide limited presence-only data due to technical and economic limitations, while ABFT predation on key preys has likely intensified in recent years. To enhance our understanding of ABFT distribution and its ecological context in the GSL, we developed a spatial presence-absence model using multispecies environmental DNA (eDNA) detections for ABFT and its primary preys, alongside environmental covariates. Water samples were collected at over 300 locations across four years. ABFT occurrence probability was best predicted by maximum depth and Atlantic mackerel ( Scomber scombrus ) presence, but not sea surface temperature. Model-predicted ABFT presence diverged from distribution inferred from fishery catch data and that predicted by basin-scale habitat suitability models, suggesting that conventional methods may underestimate ABFT presence in certain areas. These discrepancies highlight the value of eDNA as a complementary ecological indicator for elusive or under-sampled marine species. Despite inherent uncertainties, as any method, eDNA-based inference can fill critical gaps in biodiversity monitoring. Simultaneous detection of predators and prey offers a novel avenue for tracking trophic interactions and ecosystem dynamics. We propose integrating eDNA-based indicators into science-based management frameworks, particularly for mobile marine predators and complex oceanic environments. • eDNA complements fisheries for monitoring elusive marine predators. • Spatial model integrates eDNA detections of ABFT, preys, and environment. • ABFT occurrence linked to depth and mackerel presence, not SST. • Simultaneous predator–prey detection offers insights into trophic dynamics. • eDNA predictions diverge from fisheries, improving ecosystem management.