424 



Fishery Bulletin 104(3) 



over 14°C, indicating that there may be environmental 

 windows for successful transition of smolts into the 

 sea at locations other than at the extreme edges of the 

 temperature ranges documented by Power et al. (1987). 

 Reddin and Friedland (1993) noted that during colder 

 years with heavy ice flows along the northeast coast 

 of Newfoundland, commercial catches and returns of 

 salmon to rivers were lower, even though freshwater 

 warming was occurring normally and they also noted 

 that these low catch rates and returns may have been 

 due to the mortality of postsmolts early in the marine 

 phase. Our results from DSTs applied to Campbellton 

 River smolts indicated that temperature-related mortal- 

 ities, at least in 2002, were unlikely because the smolts 

 spent little time in water colder than 7°C. Furthermore, 

 because the only colder water present was deeper in the 

 water column, salmon postsmolts were able to avoid it 

 by remaining in the warmer surface waters. In 2002, 

 there was little or no inshore ice at the time of the 

 Campbellton River smolt migration as is sometimes 

 present in other years when mortalities from cold water 

 temperatures could occur. 



It is interesting to note that temperatures recorded 

 by the DSTs for the first few weeks after the smolts 

 entered the sea were very similar to that experienced by 

 salmon kelts tagged with DSTs also from Campbellton 

 River (Reddin et al., 2004). In addition, the tempera- 

 tures from tagged kelts when compared to temperatures 

 from tagged smolts showed similar periods of stability 

 just after entry of the salmon into the sea, as well as 

 abrupt changes in temperatures consistent with deep 

 diving activities. Although we did not know the precise 

 geographic location of either group of fish, it appears 

 that kelts and postsmolts are found initially in water of 

 similar temperatures and behave in similar ways with 

 respect to temperature. Because neither the postsmolts 

 nor kelts started their deep diving activities for four 

 weeks after sea entry, it may be that they were recover- 

 ing from the transition process or that they may have 

 been near shore where deeper depths are not available 

 (or both these reasons may apply). Resolution of these 

 questions will have to await the results from application 

 of geolocation and depth-sensing tags. 



Colbourne et al.'", Friedland et al. (2003), Beamish 

 and Bouillon (1993), and Downton and Miller (1998) 

 and many others have examined relationships between 

 environmental variables and the abundance of Pacific 

 and Atlantic salmon and other fish species with a view 

 of providing forecasts of future abundance but without 

 any knowledge of what thermal regimes the fish actu- 

 ally use. Colbourne et al.'^ indicated that the goal of 

 searching for relationships is important in order that 



1*^ Colbourne, E. B., E. G. Dawe, D. G. Parsons, E. F. Murphy, 

 W. R. Bowering, E. L. Dalley, J. T. Anderson, J. B. Demp- 

 son, D. Orr, D. E. Stansbury, and G. P. Ennis. 2002. A 

 preliminary review of environmental-stock relationships 

 for some species of marine organisms in NAFO waters 

 of the Northwest Atlantic. NAFO SCR Doc. 02/34, 21 

 p. Northwest Atlantic Fisheries Organization, P.O. Box 

 638, Dartmouth, Nova Scotia, Canada B2Y 3Y9. 



20 22 24 25 



Water temperature (-C) 



Figure 6 



Frequency distributions for all temperatures recorded 

 hourly at sea at 10 m depth (Sea temp at 10 m) and at 

 8 m depth (Sea temp at 8 m), and for salmon in fresh- 

 water from the DSTs (Salmon in fresh), and for salmon 

 from DSTs while in the sea (Salmon in sea). All sea 

 temperatures from the DSTs were used and the fresh- 

 water and marine temperatures are for the same time 

 periods that the DST-tagged fish were in the sea. 



influences of the physical ocean environment can be 

 used to provide more accurate forecasts of stock abun- 

 dance and ultimately so that they can be used to create 

 management plans for various fisheries (Bisbal and Mc- 

 Connaha, 1998; Friedland, 1998). These predictions are 

 currently used for both North American and European 

 Atlantic salmon, for which abundance has been fore- 

 casted partly on the basis of environmental information 

 (Anon.'^). Physical conditions in the ocean have been 

 shown to be related to mortality and growth of some 

 other species (Brander, 1995; Dutil et al., 1999; Wata- 

 nabe and Yatsu, 2004). Blackbourn (1993) and Down- 

 ton and Miller (1998) have suggested that freshwater 

 survival rates for some species of Pacific salmon are 

 even related to SSTs experienced by potential spawners 

 while still at sea, shortly before their return to fresh- 

 water. These studies and their importance clearly show 

 the need for studies on the physical oceanography and 

 ecology of fish at sea. For Atlantic salmon, Reddin and 

 Friedland (1993) created a profile of sea temperatures, 

 based on research vessel catch rates, that indicated 

 that Atlantic salmon were commonly found in water 

 at temperatures from 4° to 10°C. Reddin et al. (2004) 



