2005 Annual Survey of Juvenile Salmon and Ecologically Related Species and Environmental Factors in the Marine Waters of Southeastern Alaska
September 24, 2005
Juvenile Pacific salmon (Oncorhynchus spp.), ecologically-related species, and associated biophysical data were collected by the Southeast Coastal Monitoring Project along primary marine migration corridors in the southern and northern regions of southeastern Alaska. Up to 17 stations were sampled in four time periods (40 sampling days) from May to August 2005. This survey marked the ninth consecutive year of systematic monitoring of how juvenile salmon interact in marine ecosystems, and was implemented to identify the relationships among biophysical parameters that influence the habitat use, marine growth, predation, stock interactions, and year-class strength of salmon. Typically, at each station, fish, zooplankton, physical profile data, and water samples were collected using a surface rope trawl, conical and bongo nets, a conductivity-temperature-depth profiler, and a water sampler during daylight. Surface (3-m) temperatures and salinities ranged from 9.3 to 15.7 ºC and 13.8 to 31.5 PSU over the season. A total of 6,874 fish and squid, representing 19 taxa, were captured in 92 rope trawl hauls from June to August. Juvenile salmon comprised 96% of the total fish and squid catch in each region. Juvenile salmon occurred frequently in both regions, with pink (O. gorbuscha), chum (O. keta), sockeye (O. nerka), and coho (O. kisutch) occurring in 63-86% of the trawl hauls, and juvenile Chinook salmon occurring in 20-25% of the trawl hauls. Of the 6,651 salmonids caught, over 99% were juveniles. In both regions, only two non-salmonid species represented >1% of the catch: market squid (Loligo spp.) in the southern region (2%) and crested sculpin (Blepsias bilobus) in the northern region (2%). Temporal and spatial differences were observed in the catch rates, size, condition, and stock of origin of juvenile salmon species. Catch rates of juvenile salmon were highest in June for all species except pink salmon, which had the highest catch rates in August. Size of juvenile salmon increased steadily throughout the season; mean fork lengths in June, July, and August were, respectively: 92, 127, and 170 mm for pink; 108, 124, and 191 mm for chum; 115, 123, and 180 mm for sockeye; 184, 207, and 239 mm for coho; and 205, 245, and 255 for Chinook salmon. Coded-wire tags were recovered from 17 juvenile coho, 6 juvenile Chinook, and 2 immature Chinook salmon; all but six of these fish were from hatchery and wild stocks of southeastern Alaska origin. The non-Alaska stocks were juvenile coho and Chinook salmon originating from Oregon and Washington. Alaska enhanced stocks were also identified by thermal otolith marks from 53% of the chum, 18% of the sockeye, 9% of the coho, and 50% of the Chinook salmon. Onboard stomach analysis of 63 potential predators, representing eight species, revealed one predation instance on juvenile salmon by a spiny dogfish (Squalus acanthias). Forecasting models using catch-per-unit effort (CPUE) of juvenile pink salmon in strait habitat of the northern region in 2003 and 2004 produced accurate predictions of southeastern Alaska pink salmon harvests in 2004 and 2005. However, the models using 2005 CPUE as a predictor overestimated harvest of pink salmon in 2006, indicating that CPUE alone is not sufficient to consistently predict year class strength. These results suggest that in southeastern Alaska, juvenile salmon exhibit seasonal patterns of habitat use and abundance, and display species- and stock-dependent migration patterns. Long-term monitoring of key stocks of juvenile salmon, on both intra- and interannual bases, will enable researchers to better understand ecological interactions that affect interannual variation in salmon abundance and the role that salmon play in North Pacific marine ecosystems.
The Southeast Coastal Monitoring Project (SECM), a long-term fisheries oceanography study in southeastern Alaska, was initiated in 1997 to annually study the early marine ecology of Pacific salmon (Oncorhynchus spp.) and ecologically related species, and to better understand effects of environmental change on salmon production. Salmon are a keystone species that constitute important ecological links between marine and terrestrial habitats, and therefore play a significant, yet poorly understood, role in marine ecosystems. Fluctuations in the survival of this important living marine resource have broad ecological and socioeconomic implications for coastal localities throughout the Pacific Rim. Increasing evidence for relationships between production of Pacific salmon and shifts in climate conditions has renewed interest in processes governing salmon year-class strength (Beamish 1995). In particular, climate variation has been associated with ocean production of salmon during El Niño and La Niña events, such as the recent warming trends that benefited many wild and hatchery stocks of Alaskan salmon (Wertheimer et al. 2001). However, research is lacking in areas such as the links between salmon production and climate variability, between intra- and interspecific competition and carrying capacity, and between stock composition and biological interactions. Past research has not provided adequate time-series data to explain such links (Pearcy 1997). Because the numbers of salmonids produced in the region have increased over the last few decades (Wertheimer et al. 2001), mixing between stocks with different life history characteristics has also increased. The consequences of such changes for the growth, survival, distribution, and migratory rates of salmonids remain unknown.
One SECM goal is to identify mechanisms linking salmon production to climate change using a time series of synoptic data that combines stock-specific life history characteristics of salmon with ocean conditions. Until recently, stock-specific information relied on labor-intensive methods of marking individual fish, such as coded-wire tagging (CWT; Jefferts et al. 1963), which could not practically be applied to all of the fish released by enhancement facilities. However, mass-marking with thermally induced otolith marks (Hagen and Munk 1994) is a technological advance that is currently implemented in many parts of Alaska. The high incidence of these marking programs in southeastern Alaska (Courtney et al. 2000) offers an opportunity to examine growth, survival, and migratory rates of specific salmon stocks during a period of high levels of regional hatchery production of hatchery chum salmon (O. keta) and historically high returns of wild pink salmon (O. gorbuscha). In 2005 for example, over 400 million chum salmon were released from hatcheries in southeastern Alaska (White 2006). Of those releases, over 340 million were otolith-marked juvenile chum salmon released by three private non-profit enhancement facilities. Consequently, over the past decade, commercial harvests of adult chum salmon in the common property fisheries in southeastern Alaska have averaged about 11.7 million fish annually (ADFG 2006). These harvests are represented by a high proportion of fish released from regional enhancement facilities. In 2005 for example, 61% of the chum salmon harvested in southeastern Alaska was comprised of enhanced fish (White 2006). In addition to chum salmon, sockeye salmon (O. nerka), coho salmon (O. kisutch), and Chinook salmon (O. tshawytscha) are also otolith-marked by some enhancement facilities. Therefore, examining the early marine ecology of marked stocks along with unmarked stocks provides an opportunity to study stock-specific abundance, distribution, and species interactions of juvenile salmon that will later recruit to the fishery.
Increased hatchery production of juvenile salmon in southeastern Alaska has raised concern over potential hatchery and wild stock interactions during their early marine residence. A recent study using a bioenergetics approach and SECM data from Icy Strait concluded that hatchery and wild stocks consumed only a small percentage of the available zooplankton (Orsi et al. 2004a); this study also suggested that abundant vertically-migrating planktivores (e.g., walleye pollock (Theragra chalcogramma)) could have a greater impact on the zooplankton standing stock than hatchery stock groups of chum salmon. These findings stress the importance of examining the entire epipelagic community of ecologically-related species in the context of trophic interactions (Park et al., 2004; Sturdevant et al. 2004, 2005; Orsi et al. 2006, in press; Brodeur et al. in press; Weitkamp et al. in press).
To broaden the SECM research scope in southeastern Alaska, sampling was expanded in 2005 to include strait habitats within the southern region. This new focus on regional comparisons is supported by funding from the Northern Fund of the Pacific Salmon Commission over a 3-year period, and emphasizes 1) forecasting of adult pink salmon returns from juvenile pink salmon abundance, and 2) understanding differences in trophic dynamics using bioenergetics models.
The Northern Fund forecasting component will develop and test forecast models for southeastern Alaska pink salmon using juvenile catch-per-unit-effort (CPUE) data. Because of poor pre-season forecasting success and large uncertainty in estimating escapement numbers, the Alaska Department of Fish and Game (ADFG) no longer uses a spawner/recruit approach to forecast southeastern Alaska pink salmon, but instead predicts future harvests from the time series of prior harvest using an exponential smoothing model (Plotnick and Eggers 2004; Eggers 2005). Because mortality of juvenile pink salmon is high and variable during their initial marine residency, it may be a major determinant of year-class strength (Parker 1968; Mortensen et al. 2000; Willette et al. 2001). Therefore, sampling juveniles after the period of high initial mortality may provide information that can be used with associated environmental data to forecast abundance. Wertheimer et al. (2006) found that abundance of juvenile pink salmon from 1997 to 2004 in the strait habitats of the northern region sampled by SECM was highly correlated with the subsequent year’s catch in southeastern Alaska, and had promise as a forecast tool for pink salmon.