Landslide-induced changes in tree-ring anatomy: A new dendrogeomorphic avenue?

Abstract

Listen

Translate article

The dendrogeomorphic dating of past landslides can be limited by the speed or magnitude of movements, as they are sometimes too small to induce growth responses in trees standing on unstable slopes. As a result, movements of low magnitude are generally assumed to remain undetectable in standard macroscopic analysis of growth disturbances in tree-ring series. To close this gap and to enlarge the suite of approaches available for the dendrogeomorphic detection of landslide activity in growth-ring records of trees, this study focuses on the detection of wood-anatomical growth responses in trees following low-magnitude landslides.Anatomical growth responses of five individuals of common spruce (Picea abies (L.) Karst.) growing on a monitored landslide have been analyzed in this study on a total of 200 tree rings from 40 increment cores extracted at two vertical levels of tree stems. Analysis included seven anatomical parameters commonly used in quantitative wood anatomy, with a focus on the year 2017 year during which the landslide body moved by 8 mm (prior to the start of the growing season) according to extensometer records. We report that mean tracheid lumen area (TLA) decreased significantly due to landsliding, whereas the mean number of traumatic resin ducts (TRD) and rays increased significantly after the occurrence of the landslide. By contrast, the intensity of stem tilting was visibly too small to induce the formation of compression wood but strong enough to induce larger number of rays, which are known to enhance stability as well. We also realize that the low intensity of the landslide movement also resulted in rather localized anatomical anomalies along the stem, with detected responses decreasing in intensity with increasing distance from the ground. We conclude that anatomical responses of P. abies trees can be used to detect past activity in low-magnitude landsliding and that this new proxy can thus extend the applicability of tree-ring approaches to landslide research and to the detection of events that have hitherto remained unidentified in reconstructions.