Is tree-ring anatomy the key? searching for an optimal dendrogeomorphic approach to distinguish shallow creep and true slide movements

Abstract

Tree-ring based chronologies of slide movements can serve as an important basis for determining landslide activity, assessing risk and triggers, or potentially predicting future landslide behaviour. Unfortunately, however, the landslide areas studied are often affected by another type of slope movement, namely creep, which can introduce a considerable amount of noise into the resulting chronology. Thus, this study focused on testing different dendrogeomorphic approaches for the extraction of landslide signal from tree-ring series with respect to their ability to filter out the influence of creep movements and slide movements. For this purpose, 45 trees (Fagus sylvatica L. and Acer pseudoplatanus L.) growing on the steep slope of a sub-landslide block whose movement has been monitored for a long time and for which a single landslide event is known (second half of 2010) were sampled.The first tested approach was based on changing tree-ring eccentricity values and has shown its very high sensitivity to all slope movements and is therefore not applicable for creep filtering and slide movements unsuitable on steep slopes. The second approach, using macroscopic identification of the tension tree, proved to be the most suitable, accurately detecting the slide event and successfully filtering out all creep-induced noise. A final approach based on the detection of sudden anatomical changes in tree rings (independent of the presence of tension wood) also captured high amounts of noise and also slide event. Its use would be possible in the case of a suitably set event-response index threshold. The study also includes a discussion of the general effect of creep movements on tree growth, the limitations and advantages of each approach, and recommendations for future research.