4-87 



APPENDIX 4-1 



COMPARABILITY OF ZOOPLANKTON COLLECTION METHODS 



Scatter plots (Appendix Figure 4-1) and correlation coeffi- 

 cients (Appendix Table 4-1) demonstrate that there was little consistent 

 relationship between density estimates derived from the Clark-Bumpus (CB) 

 and 0.5 m net (1/2-meter) methods. It is our opinion that the larger 

 diameter, larger-meshed (158y) 1/2-meter nets used since June 1975, pro- 

 vided estimates more representative of true population than did the 

 previous method (12.5 cm, 76ym Clarke-Bumpus) . The small' CB net yielded 

 different and erratic quantitative estimates compared to the 1/2 meter 

 net. The larger net sampled larger volumes and was less susceptible to 

 mesh clogging, which impairs filtering and flowmeter accuracy. The 

 larger net was still sufficiently fine (158ym) to capture small zooplank- 

 ters, such as veliger larvae of molluscs and copepod and barnacle 

 nauplii . 



Scatter plots and regression analyses for each of the 9 

 selected abundant species provide some information useful in comparing 

 the validity of the two methods. Scatter plots demonstrate the lower 

 limits of detection by each method and the frequency and densities at 

 which one method was effective while the other was ineffective at catching 

 each species. From the regression equation, a point of equivalence was 

 calculated: this was the point at which both methods should provide 

 equivalent density estimates (in the regression line equation, the case 

 where x = y) . For 1/2-m net density estimates below the equivalence 

 point, comparable CB densities would be higher, while above this point 

 CB estimates should be lower than 1/2-m density estimates. In general, 

 the 1/2 m net had lower thresholds of detection by ten to one hundred 

 times. Consistently with this, the 1/2-m net indicated presence while 

 the CB net indicated absence more often than the reverse, for most 

 species. Each species is discussed with respect to these factors. 

 Considerations are based on Appendix Figure 4-1 and Appendix Table 4-1. 



Acartia spp. copepodites were captured in the 1/2 m net on 

 numerous occasions when they were absent from the CB sample. The CB 

 net never captured Acartia copepodites in densities less than 125/m ; 

 the larger net yielded density estimates as low as 3/m . According to the 

 regression equation, the CB estimate would be greater than the estimate 

 provided by 1/2 m net at all times except during peaks of abundance 

 (>8604/m ) . The scatter plot, however, showed that the regression pre- 

 diction was a poor approximation of the relationship between the methods, 

 with an actual distribution spanning greater than two orders of magni- 

 tude around the regression line. 



Acartia hudsonica was also more effectively captured by the 

 1/2 m net than by the CB; both capture rate and detection thresholds were 

 facorable for the 1/2 m net. According to the regression equation, esti- 



