Numbers of wild anadromous Atlantic salmon (Salmo salar) have declined demonstrably throughout their native range. The current status of runs on rivers historically supporting salmon indicate widespread declines and extirpations in Europe and North America primarily in southern portions of the range. Many of these declines or extirpations can be attributed to the construction of mainstem dams, pollution (including acid rain), and total dewatering of streams. Purported effects on declines during the 1960s through the 1990s include overfishing, and more recently, changing ocean conditions, and intensive aquaculture. Most factors affecting salmon numbers do not act singly, but rather in concert, which masks the relative contribution of each factor. Salmon researchers and managers should not look for a single culprit in declining numbers of salmon, but rather, seek solutions through rigorous data gathering and testing of multiple effects integrated across space and time.

Winter is a dynamic period. Effects of the winter regime on northern streams and rivers is extremely variable and characterized by dramatic alterations in physical habitat to which Atlantic salmon (Salmo salar) must acclimate and adapt to survive. In this paper, we synthesize recent advances in the biological and hydrologic/ geomorphic disciplines, with specific reference to Atlantic salmon overwintering in the freshwater portions of those running waters subject to freezing water temperatures. The specific requirements and adaptations for surviving winter at the three distinct life-stages in freshwater (egg, parr, kelt) are identified in relation to the characteristics of three biophysical phases: early winter (temperature decline and freeze-up), midwinter (ice growth and habitat reduction), and the break-up/warming phase. In a case study of Catamaran Brook (New Brunswick), a hydro-ecological analysis was used to explain interannual variability in juvenile abundance, especially for young-of-the-year salmon. A strong relation was found between winter discharge and interstage survival (egg to 0+, 0+ to 1+, 1+ to 2+) in 5 of 6<~>years. That is, juvenile salmon abundance in summer was highest following winters with high streamflow, presumably a function of habitat availability, especially beneath ice cover. However, the lowest measured egg-0+ survival (9.2%) was related to an atypical midwinter, dynamic ice break-up triggered by a rain-on-snow event that resulted in severe scouring of the stream-bed and redds. Thus, interannual variability in Atlantic salmon parr abundance from 1990 to 1996 was largely explained by density-independent (environmental) constraints to winter survival. The complexity of stream processes during winter underscores the need for interdisciplinary research to quantify biological change.

Declines in the populations of salmonid fishes have generated major interest in conservation and restoration of wild populations and river habitats. We used a foraging‐based model, combined with field observations and surveys, to predict individual habitat use, and to assess the effects of stream habitat conditions and management practices on the potential for reestablishing Atlantic salmon, Salmo salar. Using a model based on a simple trade‐off between increasing prey encounter rate and decreasing salmon capture success with increasing stream current velocity, we predicted favorable foraging locations for salmon in their first (age‐0) spring and summer. We tested, in six streams, whether (1) salmon preferred locations (=microhabitats) that were predicted to yield high consumption rates, (2) salmon growth and survival was greater in streams with a greater proportion of preferred, profitable, microhabitats, and (3) stream habitat remediation (introduction of large in‐stream structures such as large woody debris) increased the availability of microhabitats found to be preferred by salmon, and energetically profitable. Salmon early in their first season (May–June) were predicted to obtain the highest consumption rates (within 10% of maximum) in microhabitats with a narrow range of relatively slow current velocities (0.08–0.18 m/s). In contrast, later in the season (July–August) fish were predicted to obtain highest consumption rates over a wide range of fast current velocities (0.21–0.57 m/s). Salmon in both the early and late seasons showed strong preferences (use in proportion to availability) for microhabitat in velocity categories predicted to provide high consumption. Streams with the greatest proportion of preferred early‐season, but not late‐season, microhabitats retained a higher proportion of salmon as measured at the end of the first summer. Stream habitat remediation increased the amount of preferred early‐season microhabitat and did not negatively affect invertebrate prey abundance, or the amount of preferred late‐season microhabitats. Thus, the availability of favorable foraging areas for juveniles significantly improves the retention of salmon during the critical first summer, and stream remediation provides better foraging habitat during this important period. Our results are encouraging for broader application to identify sites that show promise for salmon reintroduction, and to help guide restoration of particular sites to provide suitable habitat.

Paired day-night underwater counts of juvenile Atlantic salmon (Salmo salar) were completed on tributaries of the West River, Vermont, U.S.A., between 28 August and 10 September 1995. At water temperatures ranging from 13 to 23°C, the relative count of juvenile salmon was greater at night. Nocturnal counts differed for young-of-the-year and post-young-of-the-year (PYOY) salmon, with PYOY exhibiting almost exclusive nocturnal activity. Nocturnal activity in late summer may enable salmon to maintain population densities when space and suitable feeding areas may be limited. Nocturnal activity of juvenile salmon should be considered in studies of habitat use, competition, time budgets, and associated bioenergetic processes.

Spatial and temporal variation in growth conditions for young juveniles may determine the ultimate success of salmonid populations. To assess this aspect of habitat quality, we developed a spatially explicit bioenergetics model to predict age-0 Atlantic salmon Salmo salar growth rate potential (GRP) in rearing streams of the Connecticut River, from the time of stocking in the spring, to the end of the summer. During the first month after stocking, there appears to be a paucity of suitable habitat. Most available habitat is predicted to result in low foraging success of small fish and to be energetically stressful because of the combination of high spring discharge and low water temperature. Although less than 14% of available habitat was predicted to support positive growth in the spring, 47% of the fish we observed occupied microhabitats predicted to yield positive growth, indicating the importance of habitat selection. In contrast, from mid-June to August, 67% of available habitat was predicted to yield positive growth, and 92% of all fish occupied positive growth microhabitats. Consistent with these results, sites with higher salmon GRP in the early season, but not in the mid-or late season, had higher final salmon densities by the end of August. Hydroclimatic regimes characteristic of more southerly rearing streams in the Connecticut River basin were predicted from our model to cause a potential shift from earlyseason to late-season habitat-related growth constraints along this environmental gradient. This work demonstrates the value of applying a bioenergetics approach to issues related to conservation of Atlantic salmon and provides a framework for future research on early life history energetic determinants of habitat quality.