Abstract
The increasing demand for tree and forest health monitoring due to ongoing climate change requires new future-oriented and nondestructive measurement techniques. Electrical resistivity (ER) tomography represents a promising and innovative approach, as it allows insights into living trees based on ER levels and ER cross-sectional distribution patterns of stems. However, it is poorly understood how external factors, such as temperature, tree water status, and electrode installation affect ER tomograms. In this study, ER measurements were carried out on three angiosperms (Betula pendula, Fagus sylvatica, Populus nigra) and three conifers (Larix decidua, Picea abies, Pinus cembra) exposed to temperatures between -10 and 30°C and to continuous dehydration down to -6.3 MPa in a laboratory experiment. Additionally, effects of removal of peripheral tissues (periderm, phloem, cambium) and electrode installation were tested. Temperature changes above the freezing point did not affect ER distribution patterns but average ER levels, which increased exponentially and about 2.5-fold from 30 to 0°C in all species. In contrast, freezing of stems caused a pronounced raise of ER, especially in peripheral areas. With progressive tree dehydration, average ER increased in all species except in B. pendula, and measured resistivities in the peripheral stem areas of both angiosperms and conifers were clearly linearly related to the tree water status. Removal of the periderm resulted in a slight decrease of high ER peaks. Installation of electrodes for a short period of 32-72 h before conducting the tomography caused small distortions in tomograms. Distortions became serious after long-term installation for several months, while mean ER was only slightly affected. The present study confirms that ER tomography of tree stems is sensitive to temperature and water status. Results help to improve ER tomogram interpretation and suggest that ER analyses may be suitable to nondestructively determinate the hydraulic status of trees. They thus provide a solid basis for further technological developments to enable presymptomatic detection of physiological stress in standing trees.