

17 18 19 20 



Figure 10. — Numbers of days between release and passage through 

 the Chignik River weir for sockeye salmon tagged and released at 

 Chignik Lagoon, 1962-66 (Dahlbcrg 1968). 



Table 5.— Average duration of time from release to passage through 

 the weir for sockeye salmon tagged in Chignik Lagoon and sockeye 

 salmon tagged immediately below the Chignik River weir (combined 

 data from 1949, 1962-66) (Dahlberg 1968). 



correction factor for the migration time between the 

 lagoon and weir. 



Separation of stocks by time of entry. — For manage- 

 ment purposes, the sockeye salmon runs to the Chignik 

 watershed are considered to be comf)osed of two stocks: 

 Chignik Lake stock and Black Lake stock (Dahlberg and 

 Phinney'*). Counting the catch and escapement of the 

 two stocks is complicated. The two stocks travel through 

 the same fishing area and trunk stream on their spawn- 

 ing migration, and their times of entry overlap. They do 

 segregate on the spawning grounds, and their peak 

 spawning activity generally occurs at different times. 



Results of preliminary tagging studies in 1952, 1959, 

 1960, and 1961 indicated 1) that most of the fish entering 

 Chignik Lagoon in early June were bound for the spawn- 

 ing areas of Black Lake and Black River tributaries, and 

 2) that fish entering in late July were bound for the 

 spawning areas of Chignik Lake and certain Black River 

 tributaries (Narver and Dahlberg 1964; Roos, see foot- 

 note 6). 



The FRI conducted more intensive tagging studies 

 from 1962 to 1966 to determine the consistency of timing 

 of the runs and to measure overlap in entry time be- 

 tween the two stocks. Colored, 25 mm diameter disk tags 

 were used in all the experiments; a different color com- 

 bination was used during each tagging session. Recovery 

 was by foot survey of each major spawning area at the 

 peak of spawning (Table 6). 



"Dahlberg, M. L., and D. E. Phinney. 1967. Studies of mature 

 sockeye salmon at Chignik, 1966. Univ. Washington, Fish. Res. Inst. 

 Circ. 67-7, 41 p. 



Table 6.— Summary of tagging experiments to determine 

 time of entry of the sockeye salmon stocks of Chignik, Alaska, 

 l%2-66 (Dahlberg 1968). 



Total Total 



flsh fish Percentage Tagging 



Year tagged recovered recovered location 



12 



Lagoon and weir 

 Weir 



Lagoon and weir 

 Lagoon and weir 

 Lagoon and weir 



I classified the tagged fish into early-season and late- 

 season spawners and calculated their relative percent- 

 ages for each tagging date (early spawning peaked in the 

 first week of August, and late spawning peaked in the 

 last week of August and early September). Because tag- 

 ging sites varied between the lagoon and weir, all tagging 

 dates were standardized to correspond with releases at 

 the weir. Dates of tagging at the lagoon were adjusted to 

 compensate for the 2-day migration between the lagoon 

 and weir. 



To obtain a quantitative estimate of the proportion of 

 each stock present in the catch and escapement on a 

 given day, I sought a suitable mathematical model. One 

 approach was to consider the proportion of a stock pres- 

 ent on a given day as a quantal response to the inde- 

 pendent variable (time) and proceed with probit analysis 

 (Finney 1952). However, another approach to analysis of 

 data of this nature was proposed by Moore and Zeigler 

 (1967), namely: 



P = 



1 



1 + e- 



where P = proportionof early spawners, 

 \-P = proportion of late spawners, 

 e = base of Napierian system of logarithms, 

 t = coded time in days measured from 15 

 June (day 1), 

 a, 6 = parameters estimated from tagging 

 studies, 

 i = error term, assumed random. 



A method of nonlinear least squares which utilized the 

 techniques of steepest descent and linearization (Gales 

 1964) was used to fit curves to data collected in each year 

 and to all tagging data combined (Fig. 11, Table 7). 



The entry pattern was nearly the same from year to 

 year, but the time of entry changed slightly (the data for 

 1962 all fall to the left, or earlier entry; whereas, the data 

 for 1965 and 1966 are to the right, or later entry). Time of 

 entry might be expected to vary with annual fluctu- 

 ations in environmental conditions, such as, weather and 

 water temperatures; nonetheless, the curves can be used 

 to separate approximately the early and late spawners by 

 entry time and to indicate the daily proportion of each 

 stock in the fishery and escapement. 



10 



