Abstract
The Czech part of the Outer Western Carpathians (COWC) formed by flysch rocks is among the most landslide-prone areas in Europe. Recent LiDAR-based inventory mapping reveals that the area of the COWC encompasses more than 13,500 landslides, with some geological domains exhibiting > 20% of the surface area covered by landslides. Although mass movements represent crucial geomorphic agents in the COWC, their timing and especially lifespan remain elusive. In this study, we focused on rockslides affecting the flanks of the flysch anticline developed within Paleogene sandstones. The area ("Hradisko" ridge) is well known as the most extensive rock city in the COWC. Structural-geological investigation, kinematic analysis and the application of near-surface geophysics (ERT, GPR and seismic refraction) suggest that rockslides accompanied by toppling and lateral spreads are predisposed by the architecture of bedding planes, joints and inherited faults cross-cutting the anticlinal ridge. Based on (10)Be exposure dating of five scarps and rockslide boulders (a total of 25 cosmogenic-dated samples), we were able to reconstruct the long-term history of rock slope failures. Although the obtained ages of headscarp exposures reveal consistent patterns suggesting major mass movement phases at approx. 18 ka and approx. 12 ka, the interpretation of boulder exposure ages is ambiguous. The study area has not been affected by catastrophic slope failure in historic times, but dendrogeomorphic analysis reveals surprisingly strong tree ring signals of mass movements within the last 150 years. These findings suggest that progressive failure is developing within some parts of the rockslide and/or slope portions above the major scarp. We conclude the following: (1) some rockslide spots within the COWC might express a very long history, encompassing the full Late Glacial-Holocene Period; (2) major mass movement activity in the study site temporally coincided with the major climatic changes; (3) rockslides with a very long history still represent potential hazards, although evidence of their active movement is not detected by standard geomorphic mapping techniques and (4) dating of the scarps should be preferred as even a higher number of boulders might be unreliable.