FISHERY BULLETIN: VOL. 74, NO. 4 



Our experiment provides four unobserved 

 selections from the underlying probabilities of 

 survival, four estimates of binomial variation, and 

 four estimates of variance in estimated survival 

 due to errors in estimates of eggs and fry. Aver- 

 ages of the four estimates for the second and third 

 sources are used to estimate oj and aj. The sample 

 variance of s is used to estimate of. The estimate of 

 of is obtained by subtraction. 



RESULTS 



First, we evaluate the effectiveness of the Auke 

 Creek Hatchery by comparing survival of hatchery 

 and creek fishes at different life stages: egg to 

 emergent fry, fry to returning adult, and egg to 

 returning adult. Next we estimate survival from 

 marking. Then we compare size, stage of develop- 

 ment, and emergence timing of hatchery fry with 

 creek fry. Finally, we compare size and time of 

 return of hatchery fish and creek fish as adults. 



Survival from Egg to Fry 



We estimated survival from potential egg 

 deposition to fry for creek fry as the ratio of an 

 estimate of the alevins in the spawning area of the 

 creek above the weir in the spring of 1973 (just 

 before emergence) to an estimate of the potential 

 egg deposition. Because 459 females spawned in 

 the stream above the weir, we estimate potential 

 egg deposition as (459)(2,035) = 934,065 [SE: 

 (459)(93.27) = 42,811 eggs]. 



On 20 and 21 March 1973, we determined the 

 number of live alevins in each of 86 O.l-m^ units of 

 a simple random sample from the 8,600 such units 

 making up the spawning area above the weir. The 

 average number of alevins per unit was 8.593 (SE: 

 3.814). Hence, total live alevins in the spawning 

 area was estimated to be (8,600)(8.593) = 73,900 

 alevins [SE: (8,600)(3.814) = 32,800 alevins]. 



Survival to time of sampling is estimated as the 

 ratio of estimated total alevins to estimated 

 potential egg deposition or 73,900/934,065 = 0.079 

 (SE: 0.035). 



In the gravel incubators, estimated survival 

 from live egg to fry was calculated as the ratio of 

 estimated total emigrants to initial numbers of 

 live eggs. The sums of the daily numbers of fry in 

 subsamples from the four incubators were as 

 follows: (A) 5,960; (B) 5,792; (C) 5,153; and (D) 6,692. 

 Total emigrations from the incubators and corre- 

 sponding standard errors were estimated using 



966 



the calibration results: (A) (5,960)(24.537) = 

 146,240 fry [SE: (5,960)( 1.072) = 6,389 fry]; (B) 

 142,118 (SE: 6,209); (C) 126,439 (SE: 5,524); and (D) 

 164,202 (SE: 7,174). The grand total of fry emigrat- 

 ing was 579,000 (SE: 25,296). 



Estimates of survival from live eyed eggs to fry 

 and the standard errors of these estimates were as 

 follows: (A) 146,240/150,000 = 0.975 (SE: 0.043); 

 (B) 0.947 (SE: 0.041); (C) 0.843 (SE: 0.037); and (D) 

 1.039 (SE: 0.045). The estimate for incubator D is 

 not feasible, but since it lies within a standard 

 error of the feasible range, we do not suspect 

 errors in data recording or calculations. The mean 

 of the survival estmates was 0.951, and the sample 

 variance of the estimates was 0.00667. 



This variance estimate is divided into three 

 components-af, 65, a§-representing variation in 

 underlying survival probabilities, binomial varia- 

 tion, and variation due to errors in estimating 

 eggs and fry respectively. The estimates are as 

 follows: a\ = 0.00493, oi is negligible, and 

 03 = 0.00174. Therefore, most of the total variance 

 of survival estimates among the four incubators 

 seems due to variation in underlying survival 

 within the incubators rather than binomial varia- 

 tion or variation in egg or fry counts. 



The incubator survival rates are from live eyed 

 egg to fry. The creek survival rate is from poten- 

 tial egg deposition to fry. To make the survival 

 rates comparable, we adjust the incubator survival 

 to that from potential egg deposition to fry. The 

 proportion of potential egg deposition which 

 develops to the eyed stage in the hatchery is 

 estimated as the ratio of total estimated eyed eggs 

 obtained from the 386 females artificially spawned 

 to estimated potential egg deposition by the 

 females, or 614,000/785,510 = 0.782 (SE: 0.036). 

 The adjusted incubator survival rate from poten- 

 tial egg deposition to fry is (0.782)(0.951) = 0.743 

 (SE: 0.047). 



Survival from Fry to Returning Adult 



Although most of the marked returning adults 

 were recovered at the weir in Auke Creek below 

 their point of origin, some were recovered from 

 sport and commercial fishermen and from the 

 intertidal spawning area of Auke Creek (Table 2). 

 Total recoveries from all sources were used to 

 estimate relative survival from fry to returning 

 adult: 667 of the marked hatchery fish and 74 of the 

 marked creek fish were recovered. Estimated 

 survival of hatchery fry to returning adults is 



