Page 50 - WINTER2019

Plant Bioacoustics
Probing Plant Physiology
Plants can transmit and reflect animal sounds, but they also produce sounds as well as a result of various physiological processes. Bending and drying wood produces microfractures and other displacements that can generate acoustic energy, a process dubbed acoustic emission (AE). As far back as 1933, Kishinouye (1990) conducted what may have been the first AE experiment by recording the sounds made by a piece of wood under bending stress. Active research continues into using AE techniques to check the quality of lumber being dried on an industrial scale as well as being a host of other inorganic struc- tures and materials.
However, the most consistent acoustic study of plant physi- ology began in 1966 (Milburn and Johnson, 1966), when researchers observed that dehydrating plants produce ultra- sonic cavitation sounds. We all know that plants need water to survive, but what we generally do not realize is that less than 1% of the water consumed by a plant is used to grow via photosynthesis.
Instead, more than 99% of the of water absorbed by plant roots rises through the stem or trunk along a thin cylindrical layer called the xylem, on its way to being transported toward stomata embedded in the leaves where it eventually transpires into the atmosphere. This internal water flow transports nutrients to the leaves and cools and maintains the shape and structure of the plant on both a macro- and microscopic level. As the water evaporates from the leaves, the high viscosity (and thus surface tension) of the water causes it to be wicked up from the roots, and thus the various conduits conducting water through the xylem are under hydraulic pressure. When water becomes scarce in the ground, the conduit tension increases, just like the tension you feel in your mouth when trying to suck a particu- larly thick milkshake through a small-diameter straw. When the milkshake is nearly gone, air bubbles enter the straw and create loud sounds that siblings have used to annoy each other for generations.
Similarly, if the hydraulic tension in the xylem of the plant gets too large, gas will begin entering the conduits, forming cavitation bubbles that, if they grow too numerous and large, will impair sap flow and eventually wilt and kill the plant. As these bubbles form and cavitate, they radiate acoustic energy, as any frustrated underwater propeller engineer will tell you. These ultrasonic signals can be picked up by transducers placed in the bark and can thus reveal whether an in situ plant is drought stressed.
The possibility of using noninvasive, nondestructive acoustic methods to measure the drought stress of plants in the field has fascinated scientists for decades. This is because standard methods for measuring hydraulic conductivity involve needles and other invasive poky things, which tend to inject air bubbles, creating all sorts of artifacts in the resulting measurements.
However, multiple challenges face acoustic measurements. Plants produce bubbles in a multitude of ways beyond dehy- dration, including fiber cavitation, mechanical strain and fracturing, rewatering, freezing, and thawing. Progress on this front has interesting parallels to signal classification in animal bioacoustics; increasing signal-processing capabili- ties has allowed more detailed features of cavitation signals to be extracted and correlated with different source mecha- nisms. For example, recent work has found that dehydration cavitation generally produces signals with peak frequencies between 100 and 200 kHz, whereas other bubble-formation processes tend to produce signals with lower peak frequen- cies (De Roo et al., 2016). Hindering further progress is a lack of knowledge about the detailed micromechanics of how cavi- tation bubbles form. And just as the case with insect sound transmission, the ultrasonic cavitation signals experience absorption and dispersion as they work their way through the complex structure of a plant, complicating signal feature extraction and classification.
Even more troublesome is that all acoustic metrics to date are nonlinearly related to dehydration level (formally defined as percentage loss of hydraulic conductivity), in part because acoustic cavitation is related to changes in air embolism (dehydration) rather than the absolute level of drought stress itself. Raw counting of cavitation signal rates turns out to be an inconsistent indicator of drought stress, so more recent work has added measurements of cumulative energy (or sound exposure), which has led to some improvements (De Roo et al., 2016). The prospect of measuring drought stress in the field with acoustics remains tantalizing.
A completely independent line of research into plant pho- tosynthesis has emerged in underwater acoustics, where multiple researchers have investigated how underwater photosynthesis in seagrass and, more recently, marine algae can be detected using sonar. The work by Freeman et al. (2018) on algae has even found that photosynthesis can be measured with passive acoustics by detecting the
“ringing” of the oxygen bubbles as they separate from the plant and drift toward the surface. Tank measurements of
50 | Acoustics Today | Winter 2019