O'BRIEN AND MAYO: CPUE OF YELLOWTAIL FLOl'NDER 



Table 2. — Geometric mean CPUE^ (landings per day fished, metric tons) for Georges Bank, Southern New England, and Cape Cod 



yellowtail flounder trips by vessel tonnage class, 1964-83. 



^Calculated as exp 



1 nX In 



' landings I 

 effort I 



20nly one trip by tonnage classes 32 and 33 in 1970 on Cape Cod grounds. 



u 



a + 2! lP#(/] + e 



where U = \n CPUE, 



a = intercept estimate, 



P,Y = model parameter estimates in loga- 

 rithmic units for category 7 for ton- 

 nage class, season, and year, 



X,j = dummy variable for tonnage class, 

 season, and year ( = 1 when category 

 j occurs; = otherwise), and 



e — error term. 



All tests for significance of main effects were 

 based on the above model without interaction. 

 Separate ANOVAs were also performed to exam- 



ine first order interactions. Parameter estimates 

 obtained from the model without interaction were 

 retransformed following methods described by 

 Bradu and Mundlak (1970) to derive unbiased 

 fishing power, seasonal, and annual coefficients. 

 Annual coefficients corresponding to the 1965-83 

 period were multiplied by the reference year 

 CPUE to compute annual standardized CPUE es- 

 timates. 



RESULTS 



Smaller vessels (TC 21-24) generally exhibited 

 the lowest CPUE indices in all three areas, al- 

 though TC 21-23 vessels were not represented on 

 Georges Bank (Table 2). Catch rates of medium 

 vessels (TC 25; 31-33) were similar to each other, 

 and were generally greater than those for TC 21- 

 24 vessels. Mean CPUE indices for the largest 

 vessels (TC 41 and 42) were more variable, but 

 generally were equal to or greater than those 



99 



