Dendrogeomorphology; Dormant landslide; Spatio-temporal analysis; Compression wood; Growth suppression; Tree ring eccentricity

Landsliding is a major natural hazard; therefore, understanding its activity is an important objective worldwide. For the investigation of the current landslide events, dendrogeomorphic methods are commonly used as they allow quite a precise dating of individual events. Nevertheless, there is still a question of whether dendrogeomorphic methods can successfully work for landslides with low-magnitude movements. To determine their usability and shortcomings, four commonly used dendrogeomorphologic approaches were put to the test: Approach I is based on the detection of abrupt growth suppression; Approach II is focused on the determination of compression wood; and Approaches III and IV that work with eccentric increments. For the detection of landslide events itself, It index thresholds and spatial statistics were used. In total, 329 individuals Picea abies (L.) Karst. Growing on a seemingly dormant landslide in the Outer Western Carpathians were sampled and processed. Overall, obtained landslide chronologies varied considerably although the same trees were used, which allowed pinpointing of the main limitations of each approach. Approach I showed a high sensitivity to water shortages, causing false noise signals. Approach II was not sensitive enough to low-magnitude movements. In contrast, Approaches III and IV recorded many possible landslide events, but most of the events were just noise signals induced by creep movements (and non-geomorphological influences). These conditions made it impossible to filter the real landslide events based on It index thresholds; however, as a substitute, spatial statistics combined with a detailed analysis of real landslide morphology were successfully used. Last but not least, a sensitivity of trees to record possible landslide movements in various stages of their lives was analysed for each approach.