142 



Fishery Bulletin 



1990 



I I ' I ! I 



jq MARCH - ) .3 APRIL 



T' ' r ' I '  I ' ' I   



I 4 - 29 APRIL 

 I I ' I I 1 



Figure 5 



Centroids of walleye pullnek egg dis- 

 tributions in Shelikof Strait by age 

 group (see Table 5) during time inter- 

 vals 2 (29 Mar.-13 Apr.) and 3 (14-29 

 Apr.). 



reflect the extent of intra- and interannual variations 

 in geographic patterns of distribution; and since the 

 ellipses overlap each other, it appears that there is little 

 variation in occurrences of the eggs through the season 

 or among years. It should be noted, however, that the 

 first time interval was sampled only in 1981; at least 

 5 years were sampled for all other time intervals. 



By age group The trend in abundances over the vari- 

 ous age groups of eggs varies with the time interval 

 (Table 5). The young ages (groups 1-3) are much more 

 abundant than the older ages (groups 4-6) in time in- 

 tervals 1 and 2. This is due to the combined effects of 

 mortality reducing the number of older eggs, and an 

 increasing spawning rate producing young eggs at a 

 higher rate than the older eggs were produced. Dur- 

 ing time intervals 3, 4, and 5, the abundances of young 

 and old eggs are nearly equal. This is probably due to 

 a declining spawning rate producing fewer yoimg eggs; 

 additionally the mortality rate might be lower than dur- 

 ing time intervals 1 and 2. 



Centroids of eggs by age group for the five time in- 

 tervals indicated that in most cases random events and 



sampling error may have masked patterns of changes 

 in spawning distribution and advection and dispersion 

 of the eggs. The ellipses around the centroids of the 

 six age groups generally overlapped one other within 

 a time interval. It appears that distribution of spawn- 

 ing has more influence on distribution of eggs than 

 advection, even though the eggs have a 2-week incuba- 

 tion period. Only the centroids of the six age groups 

 of eggs for the second and third time intervals, when 

 most eggs occurred, were analyzed (Fig. 5). The loca- 

 tions of centroids from the second time interval show 

 that eggs in age groups 4-6 are close to one other and 

 located south of eggs in age groups 1-3. This demon- 

 strates that spawning during this interval may occur 

 in two distinct but nearby areas. Spawning in the south- 

 ern area would take place earlier than in the northern 

 area, as indicated by the location of the centroid of all 

 egg ages from the first time interval (Fig. 4). During 

 the third time interval, the distribution of ages is the 

 opposite of the pattern seen in the second time interval, 

 with the older eggs located to the north of the younger 

 eggs. This pattern indicates that the spawning adults 

 were moving to the south during this time interval. 



