Modeling the impacts of dams and stocking practices on an endangered Atlantic salmon (Salmo salar) population in the Penobscot River, Maine, USA

Abstract

Dams challenge Atlantic salmon (Salmo salar) conservation, while hatcheries are a common but poorly evaluated recovery tool. We developed a spatially explicit smolt survival model for the Penobscot River, Maine, USA, population. By partitioning survival through dams (with flow dependency), free-flowing river reaches, and the estuary (with dam dependency), the model quantified how these factors influenced the number of fish entering the ocean. Given historical impounded conditions, 74%–22% of hatchery smolts released entered the ocean annually from 1970 to 2012. Of 19.7 million smolts stocked, 7.7 million entered the ocean (39%). Survival was most variable at dams (range 95% to 63%), followed by in-river (range 98% to 70%) and estuary (range 88% to 82%). Overall, lower-river stocking sites resulted in significantly higher numbers at ocean entry because of fewer dam encounters and shorter migrations. Higher flows also resulted in reduced losses. By reconstructing these freshwater and estuary dynamics, the model provides a more accurate estimate of ocean recruitment annually and can be used for scenario planning of future stocking locations relative to predicted flows while being adaptable to new survival rates.