FISHERY BULLETIN: VOL. 86, NO. 2 



Table 7. — Average catches of juvenile coho salmon in purse 

 seines sets in coastal waters along the west coast of North America 

 during July 198412 



Area 



No. per set 



No. per km2 



Sitka-Juneau 



Ketchikan-Sitka 



Cape Scott-Dixon Entrance 



Vancouver Is. 



Washington 



Oregon No. California 



1.29 

 0.58 

 1.90 

 1.80 

 3.76 

 2.59 



68 



3 



100 



95 

 198 

 136 



'Cruise Report. Drum Seiner FV Bering Sea, Coastwide NWAFC/OSU 

 Cooperative Study, Ecology of Juvenile Salmon in Coastal and Inside Waters 

 of Soutfieast Alaska, 28 June-26 July 1984, NWAFC Auke Bay Laboratory, 

 National Marine Fisfierles Service, NCAA, P.O. Box 115, Auke Bay, AK 

 99821 . 



2Fisher and Pearcy (1984). 



large as in some earlier years owing to low sur- 

 vival (Fig. 2; Pearcy 1984; Fisher and Pearcy in 

 press). 



Comparisons of the estimates of total juvenile 

 yearling coho salmon abundances off Oregon and 

 Washington with the production of coho smolts in 

 the Oregon Production Area (Columbia River to 

 California) also suggests that many juvenile coho 

 resided off Oregon and Washington during the 

 summer. By expanding our catches per m^ to the 

 region sampled, we estimated that the numbers of 

 juvenile yearling coho salmon in areas surveyed 

 by our purse seine sampling during August or 

 September 1981-84 were 6.3%, 6.5% 5.1%, and 

 5.2%, respectively, of the numbers of hatchery 

 and wild smolts released in the Columbia River 

 and in Oregon (T. Lichatowich^). The areas in- 

 cluded in these estimates were roughly 83%, 62%-, 

 51%, and 68% of the total area from Cape Flattery 

 to Cape Arago out to 37 km offshore. Recognizing 

 that the entire area was not sampled, that year- 

 class strength of coho salmon in this region is 

 probably established soon after ocean entrance 

 (Fisher and Pearcy in press), and that early 

 marine mortality may be inversely related to size 

 (Parker 1968; Ricker 1976) so that much of the 

 ocean mortality has occurred by late summer, 

 these percentages probably represent a substan- 

 tial portion of the surviving OPI coho smolts. In 

 fact, they are several times higher than the 

 smolt-to-adult survival of 1.3-2.8% for OPI public 

 hatchery coho salmon (excluding Rogue River 

 and California hatcheries) during 1981-84 (R. 

 Kaiser^). 



5T. Lichatowich, Oregon Department of Fish and Wildlife, 

 P.O. Box 59, Portland, OR 97207, pers. commun. September 

 1987. 



6R. Kaiser, Oregon Department of Fish and Wildlife, Hatfield 

 Marine Science Center, Newport, OR 97265, pers. commun. 

 September 1987. 



We conclude, therefore, that a major fraction of 

 the juvenile coho salmon from Oregon and Wash- 

 ington hatcheries did not undertake distant mi- 

 grations into the Gulf of Alaska in recent years. 

 This is not necessarily in conflict with Hartt and 

 Dell's (1986) data, since they established the pres- 

 ence of Oregon and Washington coho salmon in 

 northern waters but not the proportion of total 

 production that undertakes this migration. On 

 the other hand, neither the stocks of coho nor 

 oceanographic conditions have remained con- 

 stant over the period from 1956 to 1985 when 

 these two studies were conducted. Wild coho 

 smolts exceeded hatchery smolts in the Oregon 

 Production Area before 1961 (Nickelson 1986) but 

 comprised <12% of the smolts in 1980-85 (R. 

 Kaiser fn. 6). Perhaps wild smolts from the OPI 

 area had different migratory patterns than hatch- 

 ery smolts do today and migrated rapidly into 

 northern waters soon after ocean entrance. This 

 may explain Nickelson's (1986) finding that sur- 

 vival of hatcherv, but not wild coho smolts, was 

 significantly correlated with coastal upwelling off 

 Oregon. 



Ocean conditions have also changed over this 

 period. The late 1960's were accompanied by 

 strong upwelling along the coast compared to 

 weak upwelling in the early 1980's (Nickelson 

 1986; Mason and Bakun 1986). McLain (1984), 

 Norton et al. (1985), and Royer (1985) illustrated 

 that sea surface temperatures and sea levels in- 

 creased in the northeastern Pacific between 1976 

 and 1984. These factors and associated changes in 

 ocean circulation could explain differences in mi- 

 gratory behavior of coho salmon between 1960's 

 and 1980's. If currents provide orientational cues 

 to migration, cues facilitating northerly move- 

 ments may be reduced during years of weak up- 

 welling and weak Ekman transport from the 

 north. Ocean conditions may have modified mi- 

 gratory patterns, as they possibly have for the 

 migration of Fraser River sockeye salmon around 

 Vancouver Island (Groot et al. 1984; Hamilton 

 1985). 



ACKNOWLEDGMENTS 



We thank D. Larden and his crew of the FV 

 Pacific Warwind and Bering Sea for their cooper- 

 ation and competence during purse seining opera- 

 tions at sea; A. Chung, R. Brodeur, J. Shenker, 

 W. Wakefield, D. Gushee, C. Banner, J. Long, K. 

 Krefft, and C. Wilson for their help on cruises; 

 J. Norton and the Oregon Department of Fish and 



192 



