Retrospective analysis of marine growth and relationships to return rates of Penobscot River Atlantic salmon

Beginning in the 1980s, return rates of Atlantic salmon to the Penobscot River, Maine U.S.A. declined and have persisted at low levels. This downturn coincided with similar declines in North American and European Atlantic salmon stocks and with changes in the Northwest Atlantic ecosystem. Previous studies investigated whether early marine growth explained the declines, but results varied, with decreased growth associated with declines in European stocks but not North American stocks. In this study, we evaluate whether growth over the entire marine stage is related to Atlantic salmon marine survival. We constructed a growth time series from scales of returned Penobscot River Atlantic salmon spanning periods of varying marine survival. We used ANOVA and post-hoc tests to quantify seasonal growth increment differences and principal component analysis to characterize variability among the suite of growth increments. We observed reduced growth during the second winter and second marine year starting in the 1990s, with compensatory seasonal growth relationships. These results indicate that diminished growth during late marine stages is associated with low return rates in this population.

Differential impacts of freshwater and marine covariates on wild and hatchery Chinook salmon marine survival

Large-scale atmospheric conditions in the Northeast Pacific Ocean affect both the freshwater environment in the Columbia River Basin and marine conditions along the coasts of Oregon, Washington, and British Columbia, resulting in correlated conditions in the two environments. For migrating species, such as salmonids that move through multiple habitats, these correlations can amplify the impact of good or poor physical conditions on growth and survival, as movements among habitats may not alleviate effects of anomalous conditions. Unfortunately, identifying the mechanistic drivers of salmon survival in space and time is hindered by these cross-habitat correlations. To address this issue, we modeled the marine survival of Snake River spring/summer Chinook salmon with multiple indices of the marine environment and an explicit treatment of the effect of arrival timing from freshwater to the ocean, and found that both habitats contribute to marine survival rates. We show how this particular carryover effect of freshwater conditions on marine survival varies by year and rearing type (hatchery or wild), with a larger effect for wild fish. As environmental conditions change, incorporating effects from both freshwater and marine habitats into salmon survival models will become more important, and has the additional benefit of highlighting how management actions that affect arrival timing may improve marine survival.

Zooplankton variability in the Strait of Georgia, Canada, and relationships with the marine survivals of Chinook and Coho salmon

The Strait of Georgia, Canada, has complex interactions among natural and human pressures that confound understanding of changes in this system. We report on the interannual variability in biomass of 12 zooplankton taxonomic groups in the deep (bottom depths greater than 50 m) central and northern Strait of Georgia from 1996 to 2018, and their relationships with 10 physical variables. Total zooplankton biomass was dominated (76%) by large-sized crustaceans (euphausiids, large and medium size calanoid copepods, amphipods). The annual anomaly of total zooplankton biomass was highest in the late 1990s, lowest in the mid-2000s, and generally above its climatological (1996–2010) average after 2011, although many individual groups had different patterns. Two latent trends (derived from dynamic factor analyses) described the variability of annual biomass anomalies underlying all zooplankton groups: a U-shaped trend with its minimum in the mid-2000s, and a declining trend from 2001 to 2011. Two latent trends also described the physical variables. The variability represented by these four latent trends clustered into two periods: 1996–2006, with generally declining zooplankton biomass and increasing salinities, and 2007–2018, with increasing zooplankton biomass and decreasing salinities. ARIMA modelling showed sea surface salinity at Entrance Island in the middle Strait of Georgia, the Pacific Decadal Oscillation, and the peak date of the spring phytoplankton bloom were significantly related to the two latent zooplankton trends. ARIMA models comparing zooplankton and physical variables with the marine survivals of four salmon populations which enter the Strait as juveniles (Chinook: Cowichan River, Puntledge River, Harrison River; Coho: Big Qualicum River) all included zooplankton groups consistent with known salmon prey; prominent among the physical variables were sea surface salinity and variables representing the flow from the Fraser River. These regressions explained (adjR2) 38 to 85% of the annual variability in marine survival rates of these salmon populations over the study time period. Although sea temperature was important in some relationships between zooplankton biomass and salmon marine survival, salinity was a more frequent and more important variable, consistent with its influence on the hydrodynamics of the Strait of Georgia system.