Expanding dendrochronology to palms: A Bayesian approach to the visual estimate of a palm tree age in urban and natural spaces

Abstract

The age of trees and palms is fundamental with respect to their probability of survival, the quality and quantity of their production and their value as unique specimens. Determining these ages is necessary in different contexts (natural, forest, agriculture, urban trees and landscaping). Dendrochronology makes it possible to determine the age of trees, but for palms (Arecaceae) it is still lacking. Here we present and use a method based on the study of whole palm tree images and linear regression of stem/crown ratio and age in years, created with individuals of known age, and posterior probability distribution functions using Bayesian and Monte Carlo methods. This methodology is applicable to the estimate of adult palm individuals of different Arecaceae genera that reach the maximum dimensions of crown once became adult, provided an ensemble of individuals with known age is available for comparison. This approach is here applied to the estimation of the age of Canary Islands palm trees. The proposed methodology shows that the age in years of a Canary Islands palm tree is 28.33 × stipe (S)/crown (C) ratio + 7.03 ± s. The application of the methodology allowed the discovery of a dispersal event around 1840–1845, unknown until now, and revealed two palms from Tenoya (Gran Canaria, Spain) as the oldest known living Canary Islands palms, with an estimated age of over three hundred years.

Keywords

• Canary palm
• Bayesian analysis
• Palm tree age methodology