Are the historically known landslide events really the biggest? An insight using dendrogeomorphology

Abstract

The construction of the magnitude-frequency relationship is one of the key bases for the estimation of landslide hazard. However, data on past landslide occurrence are mostly from archival sources, and it is assumed that only the largest events are recorded. This situation may result in a critical underestimation or overestimation of the resulting landslide hazard. Therefore, in this study, chronological and spatial parameters were verified using dendrogeomorphic methods for six selected landslides for which at least one reactivation was historically known. The selected tree-ring-based parameters were thus compared between historically known events and other events that were identified using dendrogeomorphic reconstruction. In total, disturbed trees were analyzed, and growth disturbances were identified from their tree-ring series, from which landslide reactivations (including historically known ones) were reconstructed. The results suggest that for half of the landslides studied, the historically known event can indeed be assumed to have been the largest in area over the time period covered by the tree-ring data. However, for some of these landslides, the results suggest that the magnitude of movement (the rate of rotation of the landslide blocks) was significantly smaller than historically known. Conversely, for landslides whose reactivated area was smaller than that of historically known landslides for all unknown events, the results suggest that the magnitude of movement was significantly larger. In the case of only one landslide, the results suggest that the historically known event was one of the smallest, and most of the previously unknown reactivations (revealed by dendrogeomorphic analysis) were larger in area with greater rates of movement. The results thus provide new and original insight into the issue of the size of historically known landslide reactivations.