89-12 Estimating Atlantic Salmon Survival in Estuaries and Bays of Maine
We have monitored Atlantic salmon smolt emigration from select Maine rivers using smolt trapping and acoustic telemetry since 1996. Initial studies consisted of active tracking of smolts and have since evolved to monitoring through extensive fixed-position arrays. Through these studies, we have 1) quantified survival; 2) identified zones of high mortality: and 3) documented individual fates, including confirmation of predator species. We examined the survival of Atlantic salmon in coastal Narraguagus and Dennys Rivers populations (< 4,000 smolts) and in the larger Penobscot River population (> 500,000 smolts). Each system and population presents different monitoring/sampling and analysis challenges. We modeled survivorship and detection-efficiency probabilities between monitoring sites using the Cormack-Jolly-Seber release-recapture model from detection histories at multiple sites using Program MARK. To geo-reference fish losses, complex individual-based analyses were done to understand fish reversals (upstream movements during a downstream migration). We used both manual and automated analyses to describe routes and behaviors of individuals. We discuss the evolution of our analyses and the utility of focused studies that partition emigration survival by habitat. We found that average survival was highest in the Narraguagus River and lowest in the Dennys River populations with the Penobscot River population intermediate. Our results, suggest that losses during the short estuary transit are higher than the monthly average losses at sea. These findings can be used in stage-based salmonid life history models to inform mitigation and management prioritization.