Towards understanding the hydraulic and topographic controls of large boulders movement during glacial lake outburst floods (GLOFs)

Extreme glacial lake outburst floods (GLOFs) are characterized by flow velocities and peak discharges far exceeding those of “classical” floods. As such, GLOFs are frequently associated with extraordinary geomorphic impacts and remobilization of large volumes of material. However, surprisingly little is known about specific hydraulic and topographic conditions that drive and facilitate the erosion, transport and deposition of very large boulders (diameter > 3 m) during GLOFs. To bridge this gap, we analyzed examples of major GLOF events from around the globe and compiled the information about the remobilization of large boulders, using the analysis of time series of very high-resolution satellite images. Based on the interpretation of visual changes between pre- and post-event images, we distinguish: (i) eroded boulders (i.e., those only present in the pre-event images, not traceable in the post-event images); (ii) deposited boulders (i.e., those only present in the post-event images, not traceable in the pre-event images); and (iii) transported boulders (i.e., those traceable in both pre-and post-event images). We characterize each boulder (shape, dimensions, location, distance from the lake), its trajectory and surrounding topography (travel distance, minimum and mean slope of the trajectory, valley width) as well as the causal GLOF (GLOF mechanism, peak discharge). Our preliminary findings suggest that: (i) major GLOFs in mountain regions are capable transporting boulders exceeding 10 m in diameter; (ii) these boulders typically originate from a breached moraine dam or colluvial valley infill; (iii) the deposition of large boulders clusters in locations where the valley widens and/or the slope of the trajectory decreases. Since dimensions of transported boulders are linked to flood hydraulics, large boulders can be used as indicators of GLOF magnitude and can help to define boundary conditions for GLOF modelling studies. Our ongoing work covers a development of empirical relationships between the characteristics of mapped boulders, topography and GLOF characteristics, and the confrontation of observations with hydraulic theory and modelling studies.