713 



Straying of adult sockeye salmon^ 

 Oncorhynchus nerka, 

 entering a non-natal hatchery 



Jason N. Griffith 

 Andrew P. Hendry 

 Thomas P. Quinn 



School of Fisheries 



University of Washington 



Box 357980, Seattle, Washington 98195 



E-mail address (for T P Quinn, contact author) tquinnin'tisti wastiinqton edu ) 



Salmonid fishes tend to return to 

 their natal site ("home") for repro- 

 duction (topic reviewed by Quinn 

 and Dittman, 1992; Quinn, 1993). 

 Odors learned by juveniles during 

 freshwater residence and down- 

 stream migration guide adults dur- 

 ing the final stages of homing mi- 

 gration (Hasler and Scholz, 1983: 

 Dittman and Quinn, 1996), al- 

 though adults also respond to 

 spawning site characteristics ( Blair 

 and Quinn, 1991) and odors of con- 

 specifics (Newcombe and Hartman, 

 1973; Honda, 1982; Groot et al., 

 1986). However, some salmonids do 

 not return to their natal site, but 

 instead "stray" and spawn else- 

 where (Quinn, 1993). 



The terms "homing" and "stray- 

 ing" are defined by the endpoints 

 of migration (i.e. the natal or non- 

 natal site, respectively), but during 

 their migration some salmonids as- 

 cend one stream, only to later leave 

 and spawn elsewhere (Ricker and 

 Robertson, 1935; Ricker, 1972). 

 Mature salmonids respond to the 

 stimulus of imprinted odors with 

 positive rheotaxis and move down- 

 stream when they no longer detect 

 home odors (Johnson and Hasler, 

 1980); therefore movement up a 

 non-natal stream does not neces- 

 sarily mean that the fish will spawn 

 there. It may be natural for salmon 

 migrating up complex river sys- 

 tems to ascend non-natal streams 



for a brief distance before the ab- 

 sence of homestream odors triggers 

 negative rheotaxis. However, the 

 behavior of some fish is not ad- 

 equately explained by such "prov- 

 ing" behavior. For example, 16*^^^ of 

 the sockeye salmon {Oncorhynchus 

 nerka ) radio-tagged by Burger et al. 

 (1995) in Tustumena Lake that 

 eventually spawned in a particular 

 tributary, initially entered a differ- 

 ent tributary and stayed there for 

 up to one week. In addition, 21^^^ of 

 the fish that spawned on the shore- 

 line of the lake had previously en- 

 tered a tributary stream. These fish 

 might be "exploring" (actively seek- 

 ing different sites and comparing 

 their attributes) or "wandering" 

 ( searching in the absence of stimuli ). 

 Unfortunately, the principal 

 methods for studying homing fail 

 to distinguish straying from explor- 

 ing or other related behavior pat- 

 terns. One approach is to collect adult 

 salmon fi-om spawning grounds, dis- 

 place them to another spawning site 

 or to a nonspawning area, and 

 monitor their subsequent move- 

 ments. Most of the fish returned to 

 the site of their capture (e.g. Hart- 

 man and Raleigh, 1964; McCart, 

 1970; Varnavskiy and Varnavskiy, 

 1985; Blair and Quinn, 1991), and 

 the authors assumed but did not 

 verify that the capture site was 

 home. Another approach is to cap- 

 ture and track adults as thev move 



upstream to their spawning site 

 (e.g. Berman and Quinn, 1991; 

 Burger et al., 1995). In these stud- 

 ies, the final spawning location is 

 assumed to be the natal one but 

 this also is not verified. 



The primary alternative ap- 

 proach to capturing and marking 

 adult salmon is to mark them as 

 juveniles and monitor the locations 

 where they subsequently spawn 

 (e.g. Quinn and Fresh, 1984; Quinn 

 et al., 1991; Pascual and Quinn, 

 1994; Vander Haegen and Doty, 

 1995). The origin of these salmon 

 is known, but once they enter a 

 hatchery, they cannot leave and are 

 only identified as strays after they 

 are killed. Much of what we know 

 about the frequency of straying is 

 based on data from hatchery popu- 

 lations, but these data reveal little 

 about the processes of homing and 

 straying and may not represent 

 wild populations (Quinn, 1993). 

 Information on the extent to which 

 salmon that enter a non-natal 

 hatchery would leave, if given the 

 opportunity, would provide insights 

 into migratory behavior and the 

 potential biases of estimating the 

 extent of straying from hatchery 

 populations. 



The University of Washington 

 hatchery (UWH) pi'ovides an excel- 

 lent opportunity to study homing, 

 exploring, and straying of salmon. 

 The UWH releases chinook (O. 

 tshawytscha ) and coho (O. kisutch) 

 salmon smolts each spring. Sock- 

 eye salmon are abundant elsewhere 

 in the Lake Washington watershed, 

 and a few enter the UWH although 

 they are not reared there. In our 

 study, all adult sockeye that en- 

 tered the UWH were tagged and 

 released just outside the hatchery. 

 If they were proving or exploring, 

 we would not expect them to re-en- 

 ter the UWH. However, if they re- 

 turned repeatedly after release, it 

 would indicate that they would 



Manuscript accepted 15 September 1998. 

 Fish. Bull. 97:713-716(1999). 



