Fisher and Pearcy: Distribution, migration, and growth of Oncorhynchus tshawytscha 



279 



1981, 1982, and 1983 for individuals from groups of 

 CWT fish sampled 75 km upstream from the ocean 

 (rkm 75) during their downstream migration were 

 estimated by (FL 1 - FL Q )/d\ where FL = mean FL of 

 the tag group when sampled in the river (from 

 Dawley et al., 1985b), FL^ = length of the individual 

 fish when caught at sea, and d = days (>10) between 

 the date when half the fish in the tag group had 

 passed rkm 75 and the date of capture in the ocean. 

 (See Dawley et al., 1985, a and b; Dawley et al. 4 for 

 detailed information on in-river sampling.) 



Net migration speed of individual CWT smolts in 

 the lower Columbia River estuary, downstream of 

 rkm 75, and in the ocean prior to capture was esti- 

 mated by dividing the distance between rkm 75 and 

 the point of capture in the ocean (straight line seg- 

 ments) by the number of days (>10) between the date 

 when half the fish in the CWT group had passed rkm 

 75 during downstream migration (Dawley et al., 

 1985b), and the date when the individual CWT fish 

 was caught in the ocean. 



In September 1983 we caught a large number of 

 juvenile chinook salmon, none of which were CWT'd. 

 We examined scales from 128 of these fish in order 

 to estimate their size at time of ocean entry and the 

 length of time they had been in the ocean. Scale ra- 

 dii were measured from the focus to what we inter- 

 preted to be the ocean entrance mark (the last abrupt 

 change in circulus spacing, often accompanied by a 

 few narrowly spaced circuli). Fork length offish at 

 ocean entry was backcalculated by using a geomet- 

 ric mean regression (Ricker, 1973) of FL and scale 

 radius for juvenile chinook salmon (length range 

 approximately 100-350 mm FL) caught by us in the 

 ocean (1981-84) and in four Oregon estuaries by the 

 Oregon Department of Fish and Wildlife (FL lmm) = 

 2.3772 x Scale radius [mmatS8x) + 0.08, n=202,r 2 =0.88). 

 We roughly estimated the length of time these fish 

 had been in the ocean by dividing their ocean growth 

 (FL at capture minus back-calculated FL at time of 

 ocean entry) by an assumed ocean growth rate of 1.5 

 mm-d -1 ; a growth rate common for juvenile coho salmon 

 during their first summer in the ocean (Fisher and 

 Pearcy, 1988 ) and similar to the estimated mean growth 

 rate downstream of rkm 75 of five CWT Columbia River 

 chinook salmon in the same year (this study). 



Results 



Catch per unit of effort 



A total of 2,132 juvenile chinook salmon were caught 

 in 880 purse-seine sets during our ocean sampling 

 1981-85. A consistent seasonal trend among years 



in mean CPUE (all sampling areas combined) of ju- 

 venile chinook salmon was lacking (Table 2). In 1982, 

 CPUE was considerably higher in May and June than 

 in September, but in 1983, CPUE in September was 

 higher than it was earlier in May and June. In 1981 

 and 1984 little change occurred in CPUE between 

 early and late summer. By far the largest CPUE was 

 during latitudinally restricted sampling off the Co- 

 lumbia River mouth in May 1985 (Table 2). 



Origin and age of CWT fish 



CWT fish represented 8.7% ( 185) of juvenile chinook 

 salmon caught in purse seines 1981-85. The percent- 

 age of CWT fish during the different cruises ranged 

 from 0.0% in September 1983 to 15.6% in late May 

 1985 off the mouth of the Columbia River (Table 2). 



Most CWT fish caught in the ocean originated in 

 the Columbia River basin (Table 3). CWT Columbia 

 River fish represented 92% (170) of all CWT fish 

 caught in the ocean 1981-85. CWT fish from coastal 

 Oregon hatcheries and from coastal Washington 

 hatcheries represented 6.4% (12) and 1.1% (2) of the 

 total catch of CWT fish, respectively. No CWT fish 

 that originated in British Columbia or Puget Sound, 

 Washington, and only one CWT fish that originated in 

 a California hatchery (Klamath R. system) was caught. 



Most of the CWT fish that we caught in the ocean 

 were released from hatcheries as yearling fish the 

 same year we recovered them in the ocean (age 1.0, 

 Table 3). In addition, two CWT fish caught in the 

 ocean were released from hatcheries as subyearling 

 fish, but overwintered in freshwater before entering 

 the ocean (based on their size and scale characteris- 

 tics at time of capture). Age-1.0 fish represented 

 90.8% (168) of the catch of CWT fish 1981-85. Sub- 

 yearling fish released from hatcheries in the spring 

 or summer, a few months or less prior to capture in 

 the ocean (age 0.0 at ocean capture), accounted for 

 only 3.7% (7) of the catch of CWT fish between 1981 

 and 1985. Fish that overwintered in the ocean after 

 being released as subyearlings in the summer, fall, 

 or winter of the year prior to capture in the ocean 

 (age 0.1 at ocean capture) accounted for only 5.4% 

 (10) of the catch of CWT fish 1981-85 (Table 3). 



Yearling (age-1.0) chinook salmon smolts from the 

 Columbia River basin were the predominant group 

 of CWT fish in the May and June samples in most 

 years. Most of these were spring run fish, but fall 

 run yearling fish were also abundant in May and 

 June 1985, and fall, summer, and mixed stock year- 

 lings were also present in some years (Table 3). 



Age-1.0 CWT fish from the Columbia River basin 

 (all runs combined) accounted for 4.4%, 5.1%, 5.5%, 

 and 15.5% of the total ocean catch in May 1981, 1982, 



