Newman and Dunk: Age validation, growth, mortality, and additional population parameters of Piistipomoides multidens 



119 



The von Bertalanffy growth curves for both sexes were 

 compared with the likelihood ratio test of Cerrato (1990). 



Estimates of the instantaneous rate of total mortal- 

 ity (Z) were obtained from catch-at-age data from the 

 NDSF. Annual catch in weight was converted to annual 

 catch in numbers-at-age by the use of age-frequency data 

 standardized by fishing effort to obtain catch-per-age 

 class. Catch in weight was converted to catch in numbers 

 based on the mean weight of P. multidens observed in the 

 sampled catch each year. Mortality estimates were then 

 derived between successive years by obtaining the natural 

 logarithm of the catch per age class (e.g. age 7) in year 

 t and subtracting the natural logarithm of the catch per 

 age class (e.g. age 8) in year ^ + 1 for all fully recruited 

 age classes. Mean total Z was then calculated across all 

 fully recruited age classes. Instantaneous natural mortal- 

 ity rates (M) were derived by using the general regression 

 equation of Hoenig (1983) for fish, where logg Z = 1.46 - 

 1.01 logg t,fi(^jf- itf^nx=^^^ maximum age in years). The 

 Hoenig equation has been shown to provide a reasonable 

 approximation of M in tropical demersal fishes (Hart and 

 Russ, 1996; Newman et al., 1996; 2000b). 



The annual percentage removal was estimated by annual 

 percentage = [F/Z (l-e-^)! x 100%. Exploitation rates (E) 

 were derived from the estimates of Z and F as defined by 

 the equation E = F/Z iF=the instantaneous rate of fishing 

 mortality derived from the relationship F=Z-M). Reference 

 points for target (optimal) and limit fishing mortality rates 

 (F , and F,,„„,) were calculated for P. multidens by using 

 the estimate of natural mortality (M), because F,^, = 0.5 M 

 (Walters, in press) and F,,„,„ = 2/3 M (Patterson, 1992). 



Results 



A total of 3833 P. multidens (ranging in size from 80 to 701 

 mm FL [10.6-5770 gTW] ) were examined for age analysis. 

 Of the fish collected, 2063 were males ranging from 245 

 to 671 mm FL and from 296 to 5195 g TW, and 1751 were 

 females ranging from 284 to 701 mm FL and from 450 to 

 5770 g TW. Length conversion equations were derived for 

 total length: TL = (1.12xFL) -t- 21.84 (n=2137, 7--^=0.995); 

 fork length: FL = (0.89xrL) - 16.61 (/!=2137, r2=0.995); 

 FL = (1.12xSL) -I- 6.44 (;?=2148, /-=0.992); and standard 

 length: SL = (0.89xFL) - 2.14 (n=2148, r-=0.992). 



Length-weight models 



Length-weight relationships were calculated separately 

 for males, females, and for both sexes combined (Table 1). 

 The relationship between TW and FL is presented in 

 Figure 2. ANCOVA of TW-at-FL and CW-at-FL were both 

 significantly different between sexes (TW: F=42.56; df: 1, 

 3234; P<0.001; CW: F=94.29; df: 1, 2652; P<0.001); males 

 were larger than females. The length-frequency distribu- 

 tion for male and female P. multidens is shown in Figure 3. 

 Temporal trends were evident in the mean length and 

 weight of P. multidens over time. Mean FL was signifi- 

 cantly different among years from 1995 to 1999 (ANOVA; 

 F=31.29; df: 1, 4193, P<0.001), with (1995=1996=1997) > 

 (1998=1999). Mean TW was also significantly different 

 among years from 1995 to 1999 (ANOVA: F=89.33; df: 1, 

 3295, P<0.001), with 1995 > 1996 > 1997 > (1998=1999). 



Age validation 



Otoliths displayed alternating opaque and translucent 

 zones. A consistent annual trend was evident; the trans- 

 lucent zone was laid down from January to May and 

 the opaque zone formed from June to December. The 

 trend in thin opaque zone formation in June and July 

 was replicated in both 1997 and 1998. Figure 4 clearly 

 demonstrates that the opaque and translucent zones are 

 laid down once a year and represent valid annual growth 

 increments. Because the marginal increment analysis 

 involved random sampling across all age classes in the 

 sampled population, the validation of annual growth 

 increments can be expected to hold across all age classes. 

 In addition, the formation of the translucent zone in the 

 sagittal otoliths of P. multidens and the annual cycle of 

 sea surface temperatures in the Kimberley region of 

 northwestern Australia were found to be closely related 

 (Fig 5). 



Otolith structure, analysis, and functionality 



The sagittae of P multidens are somewhat laterally com- 

 pressed, elliptical structures. The distal surface is concave 

 and the rostrum and postrostrum are somewhat pointed. 

 The sagittae are characterized by variable growth reticu- 

 lations along the dorsal edge from the postrostrum to 



