Main content area

Using geometric morphometrics for sex determination on adult shortnose sturgeon (Acipenser brevirostrum)

Balazadeh, Keyvan, Litvak, Matthew K.
Aquaculture 2018 v.487 pp. 89-96
Acipenser brevirostrum, adults, algorithms, aquaculture, discriminant analysis, ecology, equations, geometry, gonads, morphometry, multivariate analysis, sex determination, sex ratio, sexual dimorphism, sturgeon, ultrasonography, United States
Sturgeons are the most vulnerable vertebrate group in the world, yet little is known about their biology and ecology. A ban of most of their wild harvest has generated a dramatic increase of interest in production of sturgeon from aquaculture. Shortnose sturgeon, Acipenser brevirostrum, are listed as a species of special concern in Canada, listed as endangered in the US, and has been under aquaculture development in Canada for >15years. It is important to be able to delineate sex for both aquaculture production and determination of sex ratios in the wild. Sturgeons are thought to be sexually monomorphic. Currently, sturgeons are sexed using surgically invasive methods and/or expensive ultrasonography. We used geometric morphometrics (a combination of Procrustes superimposition, canonical variate analysis and MANOVA) on cephalic homologous landmarks to test if shortnose sturgeon were sexually dimorphic. A total of 110 shortnose sturgeon (ranging in size from 3–6kg and 6–8years of age) provided by Breviro Caviar, New Brunswick, Canada, were imaged and then sexed using a combination of ultrasound, terminal sampling and visual examination of gonads to create a working set of fish of known sex. Each fish was imaged dorsally, laterally and ventrally. We found that shortnose sturgeon are sexually dimorphic and not monomorphic. The sorting accuracy of the algorithm was tested using a jackknife test and was found to be 99% accurate when three views were in agreement. Using the landmarks from geometric morphometrics we developed an equation using discriminant function analysis to sort sex based on angular relationships (accuracy of 83%). Geometric morphometrics is a method that is accessible to any user and can provide useful metrics regarding sex identification for aquaculture production.