Inferring parental areas of juvenile mussels using hydrodynamic modelling

Mussel aquaculture is heavily reliant on wild mussel populations that supply juveniles (spat) for seeding farms. However, little is often known about parent populations, representing a risk for the sustainability of the industry. We used hydrodynamic back-tracking models to identify potential parental areas that provision green-lipped mussel (Perna canaliculus) spat across a range of settlement sites in New Zealand's largest aquaculture area. Median parental area varied considerably between 19 km² for sites located in enclosed bays and a maximum of >1150 km² for sites located in open bays. Median distance to parent populations ranged between 1.8 and 21.4 km, with a maximum larval dispersal estimated to be ca. 100 km. Small seasonal variations in parental area and dispersal distance were detected in some regions, whereas inter-annual variability was relatively minor. Regional connectivity between settlement and parental regions ranged between a minimum of 45% of larvae originating in the same parental region, to maximum retention rates of 99.9% for sites in enclosed bays, implying a considerable regional variation in the potential for self-seeding and exporting mussel larvae other areas. Our results also delineate areas that support spatfall by identifying likely locations for wild or farmed parental populations, and by establishing the spatial extent where mussel reproduction and larval development through to settlement take place. These dispersal and connectivity patterns are crucial to support management decisions for the conservation and restoration of parental populations, and other environmental constraints, such as water quality, which are necessary to ensure the sustainability of spat catching operations that enable shellfish farming.