RESULTS AND DISCUSSION 

 Adult abundance studies 



Abundance in the Niantic River 



The Niantic River winter flounder population is demographically open and therefore subject to 

 immigration, emigration, natural death, and removal by fishermen. Although attempts to estimate the 

 abundance of Niantic River winter flounder using mark and recaptures techniques began in 1973, the first 

 reliable estimates were not made until 1976. The sampling intensity was too low and effort spread across 

 too much area and time in 1973 and 1974 for estimates to be made by any means. In 1975, a more 

 intensive sampling program was undertaken in the river. Fish were marked by fin clips, but recaptures 

 were not marked. Nevertheless, the Jolly (1965) model, a multiple mark and recapture method, was applied 

 to the data. Multiple recaptures were mathematically simulated, but the resulting estimates were judged 

 to be inadequate because of large errors in parameter estimates. In addition, the 1975 survey data were 

 not similar to later years for calculation of CPUE or other relative abundance indices. The March 31 

 start meant that the survey missed most of the spawning and was not temporally comparable to later 

 years. Most tows were very brief (average of 5 min) and defmition of stations differed from later years. 



Surveys expressly designed to estimate abundance of open populations usmg the stochastic model of 

 .lolly began in 1976. The Jolly model is an extremely powerful general formula that uses all the information 

 provided by the mark and recapture experiment and provides the most efiicient estimates as long as failures 

 do not occur in basic assumptions (Cormack 1968; Southwood 1978; Begon 1979). The second measure 

 of abundance for winter flounder over the past 1 1 yr was the CPUE during a 4-wk period from mid-March 

 through early April, the only time including comparable data for all surveys. The median CPUE was 

 used as the most appropriate catch statistic as the trawl catch data were not normally distributed and were 

 positively skewed. 



Annual Jolly composite abundance indies were calculated for 1976-86 using the mark and recapture 

 data (Table 6). The 1976-82 surveys extended into May, but later ones were made only during the period 

 when most spawning had occurred and ended by mid-April. Consequently, the earlier data were re-analyzed 



28 



