Reddin et al Diurnal and nocturnal temperatures for Salmo solar 



421 



15 



10 

 5^ 

 0- 



'-' l^l'Iii'r'' i\"',', I' 





1 1 1 1 1 1 1 1 1 r 



May 28 Jun 4 Junll Jun 18 Jun 25 Jul 2 Jul 9 Jul 16 Jul 23 Jul 30 



Figure 3 



Hourly temperatures recorded by data storage tags on Atlantic salmon (Salmo salar) 

 postsmolts versus date for tags 45 and 267 applied at Campbellton River, 2002. Also 

 shown are water temperatures recorded at the trap in the Campbellton River counting 

 fence and in the sea near the Campbellton River estuary, 2002. Arrows indicate time of 

 departure from and entry to freshwater. Periods selected for further analysis are shown 

 and labelled alphabetically from A to F and are shown in detail in Figure 4. 



pronounced at greater depths. Because sea temperatures 

 are lower at night than during the day and because the 

 water temperatures experienced by the fish are warmer 

 at night, it is likely that the fish are actively changing 

 depths. Also, there were times when water temperatures 

 declined or increased by 5°C but the temperature of the 

 fish remained steady; this feature would indicate that 

 the fish was actively seeking specific temperatures and 

 changing position in the water column in order to find 

 these temperatures. 



The results of the mixed-effects ANOVA with data 

 for weeks 24 to 29 from fish tags 45 and 267 indicated 

 significant differences in temperature for week (Fc, 

 jgg7=251.82, P<0.0001) and night-versus-day effects 

 (Fj 239,3=54.53, P<0.0001). Both smolts spent about 90% 

 of the time in water temperatures ranging from 8° to 

 about 15°C. Overall, mean temperature for day activi- 

 ties for both fish combined was 11.6°C and for night it 

 was 12.5°C. Therefore, for all weeks for both fish, tem- 

 peratures experienced during the day were generally 

 lower than during the night. Whether these differences 

 were due to choice or simply because the fish were in 

 different locations and experiencing local water condi- 

 tions is unknown. 



The frequency distribution of water temperatures while 

 the fish were in the sea showed a wide range, from below 

 0° to about 20°C (Fig. 6). This distribution was wider and 

 included temperatures that were both colder and warmer 

 than temperatures available in the estuary off Camp- 

 bellton River at 8 m or at Comfort Cove at 10 m. The 

 warmer temperatures in the estuary than those recorded 

 by the sea recorders are possibly due to the fish being 

 near the surface or shore (or to both factors). The DST 

 temperatures were lower than freshwater temperatures, 

 reflecting the sea temperatures rather than freshwater 

 temperatures. The pattern for each fish presumably re- 

 flects a combination of individual preferences and the 

 availability of specific water temperatures depending on 

 the location of the fish at sea. The temperatures recorded 

 by each tag reflect the time (season) of entry into the sea 

 and the time of recapture, especially for those fish caught 

 in fisheries while the fish were still at sea. 



Discussion 



The transition from freshwater to ocean life for Atlantic 

 salmon (whether as smolts or kelts) can have a serious 



