Osteometry and size reconstruction of the Indian and Pacific Oceans’ Euthynnus species, E. affinis and E. lineatus (Scombridae)

Length estimation of Atlantic bluefin tuna (Thunnus thynnus) using vertebrae

“…It is acknowledged that using such few reference specimens (compared to usually around 20-70 individuals in other species; Jelu et al, 2021;Marrast & Béarez, 2019;Thieren et al, 2012) may underestimate error in the current study. It is possible that the full extent of intraspecific variation might not have been observed and, therefore, caution should be taken in that the prediction error observed herein should be interpreted as an absolute minimum upon which further studies with greater sample sizes should elaborate on.…”

“…Some studies (e.g., Desse & Desse-Berset, 1996;Jelu et al, 2021;Lidour et al, 2018) have only reported R 2 values, to which our R 2 values of >0.97 and >0.98 compare favorably. Others (Gabriel et al, 2012;Marrast & Béarez, 2019;Martínez-Polanco & Béarez, 2020;Rurua et al, 2020;Thieren et al, 2012) report standard error of estimate (SEE) values without defining how they are calculated, which limits comparisons with our RSE values. In any case, R 2 and standard error values are prone to be skewed if sampling is uneven across a given size range and if error is not normally distributed.…”

“…The best performing model was applied to each type as judged from Table S2 [Colour figure can be viewed at wileyonlinelibrary.com] species, meager (Argyrosomus regius; Gabriel et al, 2012). Studies on the majority of other fishes must necessarily use particular vertebrae or sections of vertebral columns (Jelu et al, 2021;Marrast & Béarez, 2019;Martínez-Polanco & Béarez, 2020;Rurua et al, 2020).…”

“…When dealing with vertebrae rank or type, the use of power regression equations has become common in estimating body length, where model fit is often assessed by the coefficient of determination (R 2 ) and standard error values (Gabriel et al, 2012;Jelu et al, 2021;Marrast & Béarez, 2019;Martínez-Polanco & Béarez, 2020;Rurua et al, 2020). Because fish growth is considered allometric, that is, the relationship between body length and vertebrae dimensions is not linear, power regression models are optimal because they account for this (Reitz et al, 1987).…”