Detection of acoustic events in Lavender for measuring the xylem vulnerability to embolism and cellular damages

TL;DR

This study investigates acoustic emissions in lavender during dehydration, identifying two phases linked to hydraulic failure and cellular damage, and develops a method to distinguish these signals for assessing drought vulnerability.

Contribution

The paper introduces a novel approach to differentiate hydraulic and cellular acoustic signals in plants, improving understanding of drought-induced injuries.

Findings

Two distinct acoustic emission phases identified during dehydration.

Hydraulic failure occurs before cellular damage in lavender.

A new algorithm discriminates between hydraulic and cellular acoustic signals.

Abstract

Acoustic emission analysis is a promising technique to investigate the physiological events leading to drought-induced injuries and mortality. However, the nature and the source of the acoustic emissions are not fully understood and make the use of this technique difficult as a direct measure of the loss of xylem hydraulic conductance. In this study, acoustic emissions were recorded during severe dehydration in lavender plants and compared to the dynamics of embolism development and cell lysis. The timing and characteristics of acoustic signals from two independent recording systems were compared by principal component analysis. In parallel, changes in water potential, branch diameter, loss of hydraulic conductance and electrolyte leakage were measured to quantify drought-induced damages. Two distinct phases of acoustic emissions were observed during dehydration: the first one associated with a rapid loss of diameter and a significant increase in loss of xylem conductance (90%) and the second one with a significant increase in electrolyte leakage and slower diameter changes. This second phase corresponds to a complete loss of recovery capacity. The acoustic signals of both phases were discriminated by the third and fourth principal components. The loss of hydraulic conductance during the first acoustic phase suggests the hydraulic origin of these signals (i.e. cavitation events). For the second phase, the signals showed much higher variability between plants and acoustic systems suggesting that the sources of these signals may be plural, although likely including cellular damage. A simple algorithm was developed to discriminate hydraulic-related acoustic signals from other sources, allowing the reconstruction of dynamic hydraulic vulnerability curves. However, hydraulic failure precedes cellular damage and lack of whole plant recovery is associated to these latter.