Diameters of phloem sieve elements can predict stem growth rates of woody plants. (26th February 2022)

• Record Type: Journal Article
• Authors: Tang, Yunjia; Yin, Shijiao; Pace, Marcelo R; Gerolamo, Caian S; Nogueira, Anselmo; Zuntini, Alexandre R; Lohmann, Lúcia G; Plath, Martin; Liesche, Johannes
• Editors: Holtta, Teemu
• Is Part Of: Tree physiology. Volume 42:Number 8(2022)
• Page Start: 1560
• Page End: 1569
• Publication Date: 2022-02-26
• Subjects: carbon allocation -- forest ecology -- machine learning -- plant anatomy -- sugar transport -- tree physiology -- xylem; Trees -- Physiology -- Periodicals; 582.16
• DOI: 10.1093/treephys/tpac022
• Languages: English
• ISSNs: 0829-318X

Abstract

Abstract: Understanding forest dynamics is crucial to addressing climate change and reforestation challenges. Plant anatomy can help predict growth rates of woody plants, contributing key information on forest dynamics. Although features of the water-transport system (xylem) have long been used to predict plant growth, the potential contribution of carbon-transporting tissue (phloem) remains virtually unexplored. Here, we use data from 347 woody plant species to investigate whether species-specific stem diameter growth rates can be predicted by the diameter of both the xylem and phloem conducting cells when corrected for phylogenetic relatedness. We found positive correlations between growth rate, phloem sieve element diameter and xylem vessel diameter in liana species sampled in the field. Moreover, we obtained similar results for data extracted from the Xylem Database, an online repository of functional, anatomical and image data for woody plant species. Information from this database confirmed the correlation of sieve element diameter and growth rate across woody plants of various growth forms. Furthermore, we used data subsets to explore potential influences of biomes, growth forms and botanical family association. Subsequently, we combined anatomical and geoclimatic data to train an artificial neural network to predict growth rates. Our results demonstrate that sugar transport architecture is associated with growth rate to a similar degree as water-transport architecture. Furthermore, our results illustrate the potential value of artificial neural networks for modeling plant growth under future climatic scenarios.