A Framework to Predict and Mitigate Bycatch for Vulnerable Marine Species

Abstract

Bycatch is a pressing concern impeding the sustainability of fisheries. Yet, bycatch is often not well mitigated due to incomplete monitoring of at-sea. This thesis aims to inform management strategies for bycatch mitigation using fisheries-independent data. I utilized spatiotemporal modeling of survey data to predict high-risk regions for three threatened skates in Atlantic Canada caught as bycatch in bottom-trawl fisheries. I identified bycatch risk hotspots across the Scotian Shelf, and independently validated bycatch risk through skate presence in at-sea observer datasets. Bycatch risk estimated by species co-occurrence, modelled from fisheries-independent data, was predictive of species presence in observed fishing sets. I then evaluated relative reductions in bycatch risk to be expected by closing zones to bottom-trawl fishing. When closures are precise, a 50% reduction in bycatch risk for skates would displace less than 10% of bottom-trawl landings by weight (4.9 ± 2.45%). I discuss the need for new approaches to mitigate bycatch of vulnerable species, and how these tools can help to meet regulatory requirements for bycatch reduction at low cost.