Joint multivariate and functional modeling for plant traits and reflectances

White, Philip A.; Christensen, Michael F.; Frye, Henry; Gelfand, Alan E.; Silander, John A.

doi:10.1007/s10651-023-00574-7

Cited by 2 publications

(1 citation statement)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Outline analysis using the FDA approach on the images can be considered for future studies to improve classification accuracy. It is also of interest to overcome the challenge highlighted by White et al (2023) relating multivariate data response specifications for traits to functional data response specifications to allow relational inference between responses in the search of factors in analysing shapes.…”

Section: Discussionmentioning

confidence: 99%

A three-dimensional functional data geometric morphometrics approach for exploring shape variation

Pillay,

Dabo-Niang,

Pathmanathan

2023

Preprint

View full text Add to dashboard Cite

This research introduces a new method for analysing shape variation for 3D landmark coordinate data, called functional data geometric morphometrics (FDGM). FDGM uses functional data analysis (FDA) to treat landmark coordinates as continuous curves or functions. This allows for a more exhaustive description and analysis of shape variation compared to geometric morphometrics (GM), which treats landmark coordinates as discrete points. A simulation study was conducted to demonstrate the general effectiveness of FDGM compared to the GM. Principal component analysis (PCA) and linear discriminant analysis (LDA) were applied to both the landmark coordinates and the functional form of the landmark coordinates. The analyses favoured FDGM. The reconstruction error for FDGM was smaller when smoothed data was considered in generating the data. FDGM and GM were then applied to distinguish dietary categories of kangaroos (omnivores, mixed feeders, browser, and grazer) using landmarks obtained from crania of 41 kangaroo extant species. The results demonstrate that FDGM is a powerful method for analysing shape variation in 3D landmark coordinate data. FDGM can substantially enhance the domain of morphometrics, providing a valuable resource for driving future progress within this realm.

show abstract

Section: Discussionmentioning

confidence: 99%