Experimental study on the formation of landslide dams by fragmentary materials from successive rock slides

Abstract

Landslide dams formed by successive rock slides in the same location may cause severe and long-term destruction to affected areas. However, studies on the formation of landslide dams that result from successive rock slides are still at early stages. In this paper, a flume test that could model rock fragmentation and the flow of fragmentary material is used to study the formation of landslide dams triggered by successive long-runout rock slides. The tests assess the effects of number of blocks, the number of times the blocks are released, and the initial arrangement of blocks before release on the dam geometry. The maximum and minimum landslide dam heights rise with the increasing number of blocks in each series of tests, while there is minimal or even no growth in the dam length for some series. Additionally, more block releases produce larger minimum dam height but smaller maximum dam height. Together, these results indicate that more releases are prone to fully block the flow channel by forming a steep dam. Furthermore, the initial block arrangement is also found to influence the landslide dam formation. In this study, landslide dams caused by successive rock slides are divided into two types: the stable type, namely, the landslide dam(s) already formed by the previous slide event(s) maintains general stability before the next slide lands on the dam surface; the unstable type, namely, the landslide dam(s) by the previous slide episode(s) has failed before the next damming slide occurs. Field investigations showed that slide episodes in both types of successive landslide dams are influential for the dam formation and river morphology.