Morphometrical methods as tools for identifying field maple (Acer campestre L.) trees. Issue 1 (9th March 2020)

• Record Type: Journal Article
• Authors: Wahlsteen, Eric
• Is Part Of: Feddes repertorium. Volume 131:Issue 1(2020)
• Page Start: 72
• Page End: 81
• Publication Date: 2020-03-09
• Subjects: Acer campestre -- discriminant analysis -- morphometrics -- principal component analysis -- samara; Botany -- Periodicals; Botany -- Classification -- Periodicals; Phytogeography -- Periodicals; 581.05
• DOI: 10.1002/fedr.201900015
• Languages: English
• ISSNs: 0014-8962

Abstract

Abstract : Several morphometrical methods were evaluated with the purpose of distinguishing individual maple trees by their samaras and to reveal shape trends. Landmarks for geometric morphometrics gave 97–98% correctly classified trees in a jackknifed confusion matrix from discriminant analysis, fan‐positioned landmarks and semilandmarks 92%, linear data 79% and elliptic Fourier analysis 65% correctly classified trees. Several principal component analyses (PCA) where conducted revealing shape trends in the investigated samaras. The linear data revealed change in length and the geometric and Fourier data change in the direction of the samara wing. MANOVA and NPMANOVA revealed significant differences between several group means. The results suggest that: 1) landmarks may be an adequate method for distinguishing different individuals of maples 2) the main shape change concerns the direction of the samara wing, which may be horizontal, curved upwards or downwards.