Controls on the grain size distribution of landslides in Taiwan: the influence of drop height, scar depth and bedrock strength

Odin Marc,Patrick Meunier,Jens Turowski

doi:10.5194/egusphere-egu21-10795

Abstract

&lt;div&gt;The size of grains delivered to rivers by hillslopes processes is thought to be a key factor to better understand sediment transport, long-term erosion as well as sedimentary archives. Recently, models have been developed for the grain size distribution produced in soils, but they may be irrelevant to active orogens where high erosion rates on hillslopes are driven by landsliding. Still, until now relatively few studies have focused on measuring and explaining the variability of landslide grain size distributions.&lt;/div&gt;&lt;div&gt;Here we present grain size distribution obtained by the grid-by-number method on 17 recent landslide deposits in Taiwan, and we compare it to the geometrical and physical properties of the landslides, such as their width, area, rock-type and strength, drop height and estimated depth. All landslides occurred in slightly metamorphosed sedimentary units, except two which occurred in younger unmetamorphosed shales, with rock strength expected to be 3 to 10 times weaker from their metamorphosed counterparts. We found that 4 deposits displayed a strong grain size segregation on their deposit with grains at the toe (downslope) of the deposit 3 to 10 times coarser than the one at the apex. In 3 cases, we could also measure the grain size distribution inside the landslides that presented percentiles 3 to 10 times finer than the surface of their deposits. Both observations could be due to either kinetic sieving or deposit reworking after the landslide failure but we could not explain why only some deposits had a strong segregation.&lt;/div&gt;&lt;div&gt;Averaging this spatial variability we found the median grain size (D50) of the deposits to be strongly negatively correlated to drop height, scar width and depth. However, previous work suggests that regolith particlesvand bedrock blocks should become coarser with increasing depth (Cohen et al., 2010; Clarke and Burbank, 2011), opposite to our observation. Accounting for a model of regolith coarsening with depth, we found that the ratio of the original bedrock blocksize and the D50 was proportional to the potential energy of the landslide normalized to its bedrock strength. Thus the studied landslides agree well with the simple fragmentation model from Locat et al. (2006), even if it was calibrated on much larger and much stronger rock avalanches. This scaling may thus serve for future model of grain size transfer from hillslopes to river, trying to better understand landslide sediment evacuation and the coupling between hillslopes and river erosional dynamic.&lt;/div&gt;&lt;div&gt;&amp;#160;&lt;/div&gt;&lt;div&gt;

Highlights

Grain size is an essential parameter for understanding sediment transport and associated processes in river evolution or in hazards related to sediment pulses
We formulate two hypotheses: first, we suggest that Eq (1) could be generalized to landslides of intermediate size and depth and, that the landslide deposit D50 should increase with rock strength, σc, and the source material’s median size, Di, but decrease with drop height, H ; second, we hypothesize that materials mobilized by shallow landslides coarsen with the landslide scar thickness, T (i.e., Di increases with T ), due to a reduction with depth of the fracture density of the bedrock (Clarke and Burbank, 2011) and/or of the degree of physical and chemical weathering experienced by particles (Cohen et al, 2010; Anderson et al, 2013; Sklar et al, 2017)
We propose that the variability in landslide D50 can be reconciled with the fragmentation scaling of Locat et al (2006) (i.e., D50 decreases with the ratio of drop height to bedrock strength, as in Eq 1), when accounting for regolith coarsening with depth (e.g., Cohen et al, 2010)

Summary

Introduction

Grain size is an essential parameter for understanding sediment transport and associated processes in river evolution or in hazards related to sediment pulses. There are many processes that control the grain size distribution (GSD) delivered to rivers, and they are poorly understood (Allen et al, 2015). Models have been proposed that describe how weathering in the critical zone reduces the original size distribution of bedrock before the grains reach the surface In active orogens with high erosion rates (> 0.5 mm yr−1 ), landslides are likely the main providers of sediments to rivers (Hovius et al, 1997; Struck et al, 2015; Marc et al, 2019), and a large fraction of sediment may reach the river only partially weathered.

Results

Discussion

Conclusion