AUSTIN and HICKEY PREDICTING ABUNDANCE OF STRIPED BASS 





Figure 5. — Actual New York commercial landings of striped 

 bass from 1954 through 1974 with calculated landings through 

 1975 sujjerimposed using: A) 4-yr mean modal sizes of age II fish; 

 B) 5-yr mean modal sizes of age II fish. 



weight 8.1 lb) outnumbered the age VII fish by 2:1, 

 but outweighed them by only 1.3:1. 



It was apparent that during some years the New 

 York harvest of striped bass may be dominated by 

 five age-groups rather than four, as suggested by 

 Schaefer (1972). Another correlation analysis 

 was, therefore, performed between the New York 

 landings (Y) and a 5-yr mean of the computed 

 modal sizes of age II + fish 1 to 5 yr prior to harvest 

 iX). This 5-yr function was expressed as a linear 

 relationship (7 = 17,315,491 - 53,810X) (Figure 

 4B) with a correlation coefficient of -0.83 

 (r^ = 0.69), significant at the 0.001 probability 

 level it, =5.05, n =12). Although this 

 coefficient was reduced slightly from that of the 

 4-yr function above (r = -0.86), the fit of 

 estimated-to-actual landings (with 95^f confi- 

 dence limits) was better for many years (Figure 

 5B) and was closer to the actual landings than 



the calculated predictions of Schaefer (1972) 

 (Table 2). 



Size, Age, and Migration 



As stated, age II + modal sizes may be computed. 

 Another method of size determination at age II + is 

 by back calculation of scale radii from larger, older 

 fish. Although no age II modal sizes determined by 

 this method were used in the predictive models, 

 our attempts to do so produced some interesting 

 information. Mansueti (1961) described the body 

 length-scale length relationship of striped bass as 

 an allometric linear function, permitting the back 

 calculation of size at each year of age using the 

 scale radii method. Scales from 142 age III striped 

 bass captured in eastern Long Island waters dur- 

 ing 1973 (year class 1970) were made available to 

 the authors by the New York State Department of 

 Environmental Conservation. The ages were re- 

 checked and the fork lengths at age II determined 

 by back calculation from body length:scale radii 

 ratios. The length-frequency distribution of back- 

 calculated data was bimodal, with equal peaks at 

 205 mm and 235 mm. The second peak was 4.1% 

 lower than the observed unimodal size of 245 mm. 

 Although the back calculated values were slightly 

 lower than the observed, the fit suggests that back 

 calculations may be used for obtaining age II sizes 

 of striped bass during years when these data are 

 lacking. 



Unpublished length-frequency data for age II 

 striped bass of the year classes 1968, 1969, and 

 1971 taken in the Virginia rivers of the 

 Chesapeake Bay System were made available to 

 the authors by John V. Merriner of the Virginia 

 Institute of Marine Science. These data showed 

 bimodal distributions similar to that of the back- 

 calculated age II lengths above (Merriner pers. 

 commun.). Merriner suggested that multimodal 

 frequencies occurred because the fish were from 

 different river systems. Merriner's data were from 

 Virginia rivers while our data (Austin and Hickey 



Table 2. — Comparison of actual and calculated commercial 

 landings of striped bass in the State of New York 1972-75. 



'Courtesy Fred Blossom, National Marine Fisheries Service, NOAA, Patch- 

 ogue. NY 



^Forecasts using the linear regression formulae discussed in the text. 



471 



