THORNE: DIGITAL DATA-PROCESSING SYSTEM 



3 6 9 



INTEGRATION (ANALOG) 



Figure 5. — Relation between catch of Pacific hake and 

 integrated voltage from analog echo integrator. 



The relative error of the beta coefficient was 

 8.0% for the analog integration data and 9.4% 

 for the DDAPS, where relative error is defined 

 as the ratio of the standard deviation of beta to 

 beta expressed as a percentage. 



Population Estimate 



The number of integrated voltages and cor- 

 responding density observations from the 

 DDAPS varied from 125 to 145 per survey. A 

 weighted mean density of Pacific hake was 



determined for each survey and for all surveys 

 combined. Each obseiTation was weighted in 

 proportion to the associated number of samples 

 in the DDAPS output (Figure 2), which de- 

 pended on the number of pings and the depth 

 interval (10 m or less, depending on depth of 

 bottom). The population of Pacific hake in Port 

 Susan was estimated by multiplication of the 

 estimated mean density of Pacific hake by the 

 volume of water covered by the various surveys, 

 expressed as the surface area at the 60-m con- 

 tour (50 X 10" m'-) times the average depth 

 interval. The average depth interval was ob- 

 tained directly from the number of samples. 

 Since the DDAPS reads the integrated voltages 

 at a rate of 10,000 per second, or each 0.073 m 

 of depth, the average depth surveyed during a 

 transect series was the average number of 

 samples per ping times 0.073 m. The weighted 

 mean densities, associated variances, volumes, 

 and estimated total biomasses for the four series 

 and for all series combined are given in Table 

 1. The population estimates for the four series 

 ranged from 15.9 to 18.8 x 10" lb. These esti- 

 mates are higher than the preliminary results 

 reported in Moose, Thorne, and Nelson (1971) 

 because of the greater area involved (50 x 10** 

 m2). 



A weighted mean density was also determined 

 for the survey analyzed with the analog echo 

 integrator. The densities along each transect 

 were weighted by the duration of the transect. 

 Since depth was not determined directly, the 

 volume was assumed to be the same as that 

 determined with the DDAPS analysis of the 

 same sui'V'ey. The results of the analysis are 

 also given in Table 1. The population estimate 

 was 18.2 X 10« lb. 



Table 1. — Results of data analysis for various transect series. 



841 



