Search

Flow-like landslides are a dangerous landslide type. They often express gradual movement or seeming dormancy, but occasional reactivation can, in extreme cases, result in catastrophic events. To predict their future behaviour, knowledge of past spatio-temporal development and relationships with hydrometeorological triggers is crucial. Moreover, regional data are more robust than case studies. Dendrogeomorphic (tree-ring-based) methods are a very precise approach for reconstructing past landslide behaviour. Nevertheless, regional reconstructions are very rare, which is probably due to their time-consuming procedures. This paper presents the results of a regional tree-ring-based reconstruction of the spatio-temporal development of flow-like landslides in a selected region in the Outer Western Carpathians. Six selected landslides were studied via analysis of 614 increment cores that came from 307 disturbed trees. The reconstruction provided data for approximately 70 individual landslide reactivation phases that were distributed in 44 event years. Events with regional extension (at least half of the studied landslides were active) were detected in six years (1940, 1941, 1953, 1961, 1985, and 1997). Periods of increased (1950s, 1990s) as well as decreased (1940s, 1970s, 2010s) landslide activity were reconstructed. The use of tree-ring data enabled the construction of landslide probability maps. Based on this analysis, all studied landslides exhibit extremely high probabilities of reactivation during a temporal horizon of 100 years, but even over shorter periods (5 and 20 years), their probability of reactivation is very high. Finally, analysis of meteorological triggers suggests the positive effect of precipitation in May (and possibly in September) to activate landslides with regional extent. Extreme short-duration (1-day) precipitation events probably do not play a role in landslide triggering. Moreover, gradual increases in precipitation totals during periods of at least one-half year preceding the event years were detected.

Our study analyzes the growth response (release or suppression) of Norway spruce trees growing along a landslide zone in eastern Czech Republic. A total of one hundred and eighty-six increment cores were extracted from Norway spruce (Picea abies (L.) Karst.) individuals, which were affected by two different disturbances, the Girova landslide in May 2010 and an anthropogenic cut-off in the neighborhood of the landslide later the same year. Growth changes were analysed in three zones that were defined according to the type of effect the disturbance had on the surrounding vegetation. The aims of our study are to (i) detect growth changes in trees that survived the disturbances in 2010, (ii) evaluate the delay time of the growth reaction to the disturbances, (iii) compare how growth changes differed in response to the two different disturbance types (natural landslide vs. anthropogenic harvesting and deforestation), and (iv) investigate spatio-temporal differences in the growth changes. Our results indicate that tree growth changed in response to the altered environmental conditions following the 2010 disturbances. The changes vary depending on the intensity of the disturbance and its effect on the forest stand. Trees in Zone 1 responded with a slight growth release (14.6 % of responding trees), which was strongest after 2013. However, growth suppression (85.4 %) was the dominant reaction in Zone 1; with the majority of trees showing growth suppression in 2014, four years after the events. The strongest and most abrupt growth release (66.1 %) occurred in Zone 2 in 2011, after a one-year delay. Since 2010, we have observed a trend of growth suppression (33.9 %) in this zone. In Zone 3, following a one-year delay, growth release has occurred gradually in about a third of the trees (37.0 %). The majority of trees in this zone have responded with growth suppression (63.0 %) in 2010 and with a three-year delay after 2013. We also found differences in how tree growth responds to anthropogenic and landslide disturbances, as well as spatio-temporal differences related to the extent of post-disturbance changes.