Autumnal changes in behavior and distribution of three age-classes of juvenile Atlantic salmon (Salmo salar) were determined during 3 yr in the Little Sevogle River of northeastern New Brunswick. In summer, salmon were always observed above the streambed, each holding a station over a single, unshaded stone. About 84% of the entire population occurred in the run habitat-type, 12% in riffles and 5% in pools. Underwater visual censuses showed the salmon to be continuously numerous in summer, but, as soon as water temperature fell to or below 10 °C in autumn, they disappeared from their stations and their visible population decreased by 92–98%. Thereafter, the salmon were found almost exclusively in sheltered substrate chambers beneath surface streambed stones. However, salmon distribution among runs, riffles, and pools (77, 18, and 5%, respectively) did not differ significantly from that in summer. Trapping, marking, and absolute population estimates indicated neither dwindling nor egress of the resident population. Thus, instead of suddenly leaving the river area or moving to another habitat-type at low temperatures, the salmon merely moved from their unsheltered summer stations to sheltered winter stations within the streambed. The results indicate the importance of winter habitat management in managing juvenile Atlantic salmon populations.

Over three summers we used direct underwater observation to examine the summer to autumn differences in seven microhabitat properties of three age-classes of juvenile Atlantic salmon (Salmo salar) in the Little Sevogle River of northeastern New Brunswick. Salmon of all three age-classes occupied a wide range of water depths during summer, but were concentrated mainly in depths of 24–36 cm. In autumn, they occurred in this range almost exclusively. The streambed stones most closely associated with the individual positions of all ages were always <20 cm in summer and mostly (84–92%) <10 cm in diameter. In autumn, all ages were associated with home stones up to 40 cm in diameter, with 65–83% of the stones exceeding 20 cm; the size of home stones selected increased with fish age in autumn. There was no apparent relationship between the water depth and home stone size distributions occupied by all age-classes and available in the stream during either summer or autumn. Summer focal water velocity (velocity at the fish's snout) was predominantly 10–30 cm∙s−1 for 0+, 10–40 cm∙s−1 for 1+, and 30–50 cm∙s−1 for 2+ salmon, but during autumn it was almost always <10 cm∙s−1 for all ages. The bottom and surface water velocities as well as the maximum water velocity within 1 m of fish stations increased with fish age during summer and autumn. At the summer–autumn transition, 0+ salmon selected higher bottom, surface, and maximum water velocities, 2+ salmon selected lower velocities, but selection by 1+ salmon remained unchanged. We view substrate size followed by water depth as the primary properties influencing stream suitability for juvenile Atlantic salmon in autumn.

No abstract

The locations of tagged parr and fin-clipped fry were observed from summer through early winter in a 1000-ft (328-m) study area of a small coastal stream. Tagged parr were usually found in or near places of original capture which are herein designated as homes. Parr whose homes were in pools, which appeared to be as suitable habitats as riffles, and those in adjacent riffles appeared to stay in their respective habitats. Some returned to their homes after having been moved as much as 700 ft (213 m) upriver or downriver. Fry were most numerous in shallow riffles and appeared to remain within small areas of the stream during summer but moved into parr habitats, the pools and deep riffles in autumn.

Measurements of routine and standard rates of oxygen consumption of various sized cod at temperatures between 3 and 15 °C revealed a well-marked size effect; small cod consume oxygen at a greater rate per unit weight than do large ones. Increases in temperature raise oxygen consumption in starved and fed fish. The increase in rate of oxygen consumption of starved fish between 3 and 10° is proportionately greater than that between 10 and 15 °C. Feeding of cod which have previously been starved increases the rate of oxygen consumption by 40–90%. The rate subsides to the starvation level in 4–7 days depending on temperature and amount of food eaten. Handling cod causes them to increase their rate of oxygen consumption; rates return to normal levels in 3–5 hours. Crowding reduces the rate of oxygen consumption apparently by reducing the space for movement and thus restricting activity. Reducing the ambient oxygen from about 10 to 3 mg/l lowers the rate of oxygen consumption slightly, but the respiratory volume (volume of water pumped over the gills per unit time) is markedly increased. This suggests there is added stress because the increased metabolic cost of irrigating the gills is not met by increased rates of oxygen consumption.