Acoustic Emission Signal Detection in Drought-Stressed Trees: Beyond Counting Hits

Tree survival and fitness under changing environmental conditions with increasing drought stress is currently of high interest. When trees experience water shortage, the whole water transport system experiences strong tensions, which involves a risk of air-embolism formation in sap-conducting conduits. The phenomenon, in which air bubbles are formed in water under tension, is called ‘cavitation’ and has been associated with the generation of acoustic emission signals. Cavitation reduces the hydraulic conductivity of the water transport system and hampers the hydraulic performance, which may eventually lead to tree mortality. Although the significant role of cavitation during drought stress has been widely accepted, the processes of embolism formation and repair are still poorly understood. Detection of cavitation by acoustic emission might be a powerful nondestructive method to investigate this phenomenon. It has been shown that the cumulative number of hits gives an indication of the degree of cavitation, but quantification in the field is still doubtful with this empirical method. Moreover, it is still not known which processes are generating the acoustic emission signals in drought-stressed trees. We used broadband sensors to collect acoustic emission signals in dehydrating branches and examined the waveforms to get more insight into the source mechanisms of the respective acoustic emission signals. A pattern recognition algorithm was applied to identify natural clusters in the signals and the obtained waveform types were interpreted in terms of possible source mechanism and source location. Occurrence of cavitation was validated using X-ray computed tomography.