POLOVINA: CATCHABILITY VERSION OF LESLIE MODEL 



Table 1 .—Species composition of bottom fish catch at Pathfinder 



Reef. 



line-hours. As is indicated in Figure 1, the circum- 

 ference of the reef can be divided into three 

 segments— north, west, and south-southeast, each 

 having similar species composition (Table 2). Further, 

 an attempt was made daily to allocate a consistent 

 proportion of the day's fishing effort to each seg- 

 ment. The proportion allocated to each segment was 

 influenced by the length of each segment and wind 



Table 2. — Species composition for the three segments of the cir- 

 cumference of Pathfinder Reef (see Figure 1). 



South- 

 Southeast 



North 



West 



Species 



No. 



% 



No. 



% 



No. 



Pristipomoides 



zonatus 

 P. auricilla 

 Etelis carbunculus 



358 51 



170 24 



171 25 



160 68 

 37 16 

 39 17 



223 



55 



104 



58 

 14 

 27 



and current conditions. On the average, the propor- 

 tion of the total daily effort allocated to each seg- 

 ment was 0.45 on the south-southeast, 0.21 on the 

 north, and 0.34 on the west. A chi-squared test ap- 

 plied to the daily allocation of fishing effort indicates 

 that there was no significant departure (P = 0.89) 

 from this allocation during the course of the fishing 

 experiment. Since the effort was reasonably con- 

 stant over the duration of the experiment and the 

 entire reef was fished each day, catch, effort, and 

 CPUE computed on a daily basis were used in the 

 analysis. An adjustment to cumulative catch sug- 

 gested by Chapman (1961) was subsequently shown 

 to improve the model fit in the Delury model 

 (Braaten 1969). This adjustment computes cumula- 

 tive catch for interval i as the cumulative catch to 

 interval i plus one half the catch during interval i. 

 This adjustment compensates for the decline in 

 CPUE within each time interval. The adjusted 

 cumulative catch is used as the independent variable 

 in all subsequent analyses (Table 3). 



Plots of CPUE against adjusted cumulative catch 

 for each of the three species of snappers show a 

 decline in CPUE for P. zonatus, a slight decline for 

 E. carbunculus, and an increase for P. auricilla (Fig. 

 2). A regression line fitted to these data results in 

 negative slopes for P. zonatus (P = 0.0007) and E. 

 carbunculus (P = 0.05) and a positive slope for P. 

 auricilla (P = 0.008). The constant catchability 

 Leslie model fitted the P. zonatus data well and 

 resulted in an R 2 of 0.71 and a pattern of residuals 

 which supports the linear model. The estimates of 

 N(0) and q for P. zonatus from this fit are 1,066 fish 

 and 0.0038 per line-hour. Due to the selectivity of 

 the fishing gear, N(0) estimated from this intensive 

 fishing data does not represent total population size 



Table 3.— Daily catch, effort, catch per unit of effort (CPUE), and adjusted cumulative catch for Pristipomoides zonatus, P. auricilla, and 



Efe//'s carbunculus 



425 



