176 



Fishery Bulletin 103(1) 



1.0 



8 



06 



0.4 



0.2 



0.0 



1.0 



0.8 



0.6 



0.4 



0.2 



0.0 



1.0 l 



0.8 



0.6 



0.4 



02 



0.0 



1.0 n 



0.8 



0.6 



0.4 



0.2 



0.0 



JL 



1994-1995 

 n = 47 



1995-1996 

 n= 146 



1996-1997 

 n = 74 



6 7 8 

 Age (yr) 



9 10 11 12 13 14 



1.0 



0.8 



0.6 



0.4- 



0.2 



0.0 



1.0 



0.8 



0.6 



0.4 



0.2 



0.0 



1.0 i 



0.8 



0.6 



0.4 



0.2- 



0.0 



- I 1 



1998-1999 

 n = 207 



1999-2000 

 n=90 



12 3 4 5 



6 7 8 

 Age (yr) 



9 10 11 12 13 14 



Figure 4 



Age-frequency distribution (based on biological year. October-September) for striped trumpeter (Latris line- 

 ata) from 1994 through 2001. Gillnet-caught fish are represented by unshaded columns, hook-caught fish 

 are represented by shaded columns. Arrows represent the progression of the cohort spawned in 1993. 



Precision of the second reader's age estimates when 

 compared with those of the primary reader were also 

 satisfactory, yielding an APE of 1.59%, and no signifi- 

 cant bias was revealed at any age class (Table 2). 



The maximum observed ages for males and females 

 were 29 and 43 years, respectively. From the available 

 data, it is unclear whether apparent differences in lon- 

 gevity between the sexes are representative because 

 very few individuals over the age of 25 were sampled. 

 However, there was no significant difference in the age- 

 frequency composition of the pooled samples based on 

 sex (Kolmogorov-Smirnov; Z=1.05 P=0.22). 



Growth modeling 



The strong 1993 cohort, allowed us to closely monitor the 

 actual length at age of striped trumpeter. Average size 

 increased from 190 mm (1.3 years) in January 1995 to 

 300 mm (2.1 years) by November 1996 (Fig. 3) and 420 

 mm (4.0 years) by November 1997. The seasonal VBGF 



model indicated that the majority of observed growth 

 in this cohort occurred between January and May (late 

 austral summer through autumn) and that there was 

 little growth apparent between June and December 

 (Fig. 5). The sine wave representing seasonal fluctua- 

 tions indicated that the peak growth rate occurred in 

 May. Comparing this sine function with that derived for 

 SST (Fig. 6), we identified a first-order serial correla- 

 tion — the strongest correlation identified when a 34-day 

 lag period was incorporated in the growth phase. 



The parameters of the VBGF S and VBGF TP fitted 

 to the aged individuals are presented in Table 3. The 

 VBGF TP gave the more parsimonious fit to the pooled 

 length-at-age data according to the deterministic AIC 

 value and underestimated L x in relation to L max to a 

 lesser extent than the VBGF S (Table 3), reflecting a 

 better fit to the data in the older age classes. In con- 

 junction with a visual assessment of residuals, it was 

 apparent that the VBGF S underestimated length at age 

 above 20 years (Fig. 7). The better fit by the VBGF TP 



