Comparing geometric morphometrics and outline fitting methods to monitor fish shape variability of Diplodus puntazzo (Teleostea: Sparidae)

Abstract

The definition of a remote system in the monitoring of fin fish shape variability relies on the appropriate shape analysis tools. In this paper, different methods are proposed and compared. Growth trajectories, computed using geometric morphometrics, allow the pattern of size-related shape change and visualize growth allometry to be analyse. Geometric morphometrics is used to describe shape change in three samples (507 5100 days after hatching) of 119 individuals of the Sparid Diplodus puntazzo. The results are potentially interesting but the method still relies on the manual detection of homologous landmarks (coordinates of points located unambiguously on each specimen’s profile or structure) which are basic for such analysis. In an attempt to proceed towards the remote estimation and the automation of data collection, two alternative approaches are tested. In particular, Elliptic Fourier analysis and Bezier function were applied to the same samples, and results compared with geometric morphometrics. Fourier analysis’ growth curves are very similar to geometric morphometrics’ ones and allows for an appropriate visualisation of global shape changes. At the same time, Fourier methods can be applied to remotely and automatically detected profiles. This method is thus considered as the best candidate in the context of automated fish shape monitoring. The Bezier method performed poorly, both in terms of statistical output and shape visualization.