with the Jolly composite index of abundance until 1982 (Table 8; Fig. 4). The CPUE in 1982 (42.6) was 

 nearly the same as in 1981 (43.4), but the abundance index increased 72% (28,693 to 49,439). However, 

 the Jolly estimate during 1982 was relatively imprecise with a large confidence interval (± 16,666). The 

 decline in CPUE for fish larger than 20 cm from 1983 (22.1) to 1984 (12.8) and 1985 (12.6) was much 

 greater than for the abundance index (29,912 to 29,282 and 21,632). The JoUy index for 1986 (8,252) 

 declined greatly from 1985, but the CPUE decreased by only about 20%. The CPUE for 1984-86 indicated 

 population levels about one-half of that during 1976-80, which also contradicted the Jolly abundance 

 indices. The apparent differences between abundance indices were evaluated below by examining meth- 

 odologies used in obtaining the estimates. 



Data from the trawl monitoring program for the five stations outside the Niantic River were used in 

 calculating a median CPUE during January through April of each year. This period overiapped the river 

 spawning and allowed for an increase in available data. The median CPUE showed relatively low densities 



Table 7. Compo.site index of abundance for Niantic River winter flounder from 



1976 through 1986. 



Year No. of Composite abundance Adjusted 



values used — 2 standard errors^ abundance 



49,439 - 16,666 



29,912 ± 7,042 34,997 



29,282 ± 10,518 36,310 



21,632 ± 9^345 26,175 



8,252 ± 2,723 9,663 



For winter flounder larger than 15 cm during 1976-82 and 20 cm thereafter. Abundance 

 adjusted to all fish larger than 15 cm for 1983-86. 



b 



Only N] excluded. 



No values of N excluded. 



30 



