Table 12.. --Estimated number of yearling male seals and kill 

 of 3-year-old male seals, year classes 1961-62, 

 and 1964-66, St. Paul Island 



The estimated regression is K :! = O.6I6Y2. 

 For the 1967 year class, Y 2 = 46.1 and K 3 

 = 28.4; the standard error of this forecast 

 is 5.7 thousand. In calculating- the standard 

 error, we have considered the error in esti- 

 mating the regression equation and Y 2 . The 

 latter error is approximated by comparing 

 estimates of the number of yearlings from re- 

 coveries at age 3 with those from combined 

 recoveries. The standard error may in fact 

 be slightly higher than 5.7 thousand, because 

 the series of comparisons is extremely short 

 (1961-65, except 1963). 



Some current estimates of the yearling pop- 

 ulation differ from those used in 1968 because 

 we now have more data with which to deter- 

 mine the true age composition of seals selected 

 as yearlings on the basis of size. 



Estimate of the kill of 3-year-old male seals 

 from the count of dead pups and the pup 

 population estimate. — Earlier in this section, 

 we gave the survival equation 



S = 37.9— 0.2D, 



where 



S = percent survival from age (August) 

 to the male harvest (combined kill 

 ages 2 to 5), 



D = dead pup count (thousands), St. Paul 

 Island. 



For 1967, D = 17.4, so that § = 34.4. We did 

 not estimate the fall pup population for all 

 rookeries in 1967, but we can reasonably as- 



sume that the population levels of recent years 

 have been similar and that variations in the 

 estimates have reflected variability in sam- 

 pling. It is valid, therefore, to consider the 

 average of recent estimates an estimate of 

 the pup population in August 1967. Also, the 

 variance of recent estimates may be considered 

 a measure of variability in population sizes. 

 The average of five available estimates (1963- 

 66 and 1969) for St. Paul Island is 243.0 thou- 

 sand (standard deviation 25.4 thousand). If 

 we apply the estimate of S to half this average 

 (only half are males), we estimate the male 

 kill at ages 2 to 5 from the 1967 year class 

 as 41.8 thousand. Of this number, 2.6 thou- 

 sand were taken as 2-year-olds in 1969. If 

 we estimate that 1,400 will be taken as 5-year- 

 olds and allocate the remainder to ages 3 and 4 

 on a 2:1 basis, the estimated 3-year-old kill 

 for St. Paul in 1970 by this method is 25.2 

 thousand. We must again neglect the short- 

 ness of the series involved in these calculations 

 but by methods similar to those we applied in 

 the last estimate, i.e., allowing for error in S 

 and the pup population estimate, we calculate 

 a standard error of 10.6 thousand. 



Combined estimates of the kill of 3-year-old 

 male seals on St. Paul Island. — The several 

 estimates and their standard errors are: 



Method or basis 



Estimate 



Standard 

 error 



Thousands Thousands 



Air temperature and pup 

 weight regression 



Yearling male seal popula- 

 tion estimate 



Survival estimate deter- 

 mined from dead pup 

 count applied to August 

 pup population estimate . 



Weighted average 



28.9 

 28.4 



25.2 

 27.8 



18.9 



5.7 



10.6 

 4.9 



Forecast of the Total Kill of 

 Male Seals 



Table 23 gives the forecast of the total kill 

 of male seals for both islands, by age. The 

 estimated kill at ages 2 and 5 is the average 

 of recent years. The extrapolation to St. 



24 



