Evaluation of growth rate estimation from fish otoliths using a striped bass bioenergetics model

Michael Jerome Schirripa, Florida International University

Abstract

Three different assumptions of the otolith-somatic growth relation were simulated using an individual-based striped bass bioenergetics model. The LINEAR model assumed that otolith growth was directly related to somatic growth; the TEMP-WT model assumed that otolith growth was a function of ambient temperature and the weight of the fish; and the O2CON model assumed that otolith growth was a function of oxygen consumption. Differences in the geometry between the fish body and otolith resulted in differences in the somatic length-otolith radius growth rates, despite the fact that in one case the somatic weight-otolith radius growth rates were identical. This implies that a comparison between somatic and otolith growth needs to consider and specify the dimensions involved. Ten different back-calculation equations, two different regression models of radius-length, and two schemes of annulus selection were examined for a total of twenty different methods to estimate size at age from simulated data sets of length and annulus measurements. The best back-calculation technique was directly related to how well the otolith radius-total length relation (OR-TL) model fit. An individually corrected Weibull cumulative distribution function is presented and, although not always resulting in the best outcome, proved to be very flexible in fitting different shaped OR-TL. Four gear selectives were examined and used to produce fishery-dependent samples for statistical analysis. Samples taken with selective gears produced OR-TLs with significantly different y-intercepts than those from an unfished population; significant differences were also found between gear selectivities. The OR-TL y-intercept of a non-selectively fished population was found to be systematically influenced downward by total mortality as a result of a decrease in the number of larger fish. These results imply that, rather than a fixed value, the y-intercept or the OR-TL is condition dependent, and, is dependent upon how the population is defined.

Subject Area

Aquaculture|Fish production|Biostatistics|Zoology

Recommended Citation

Schirripa, Michael Jerome, "Evaluation of growth rate estimation from fish otoliths using a striped bass bioenergetics model" (1996). ProQuest ETD Collection for FIU. AAI9625359.
https://digitalcommons.fiu.edu/dissertations/AAI9625359

Share

COinS