Quantifying sediment mass redistribution from joint time-lapse gravimetry and photogrammetry surveys

Abstract

Abstract. The accurate quantification of sediment mass redistribution is central to the study of surface processes, yet it remains a challenging task. Here we test a new combination of terrestrial gravity and drone photogrammetry methods to quantify sediment redistribution over a 1-km2 area. Gravity and photogrammetry are complementary methods. Indeed, gravity changes are sensitive to mass changes and to their location. Thus, by using photogrammetry data to constrain this location, the sediment mass can be properly estimated from the gravity data. We carried out 3 joint gravity-photogrammetry surveys, once a year in 2015, 2016 and 2017 over a 1-km2 area in southern Taiwan featuring both a wide meander of the Laonong River and a slow landslide. We first removed the gravity changes from non-sediment effects, such as tides, groundwater, surface displacements and air pressure variations. Then, we inverted the density of the sediment, with an attempt to distinguish the density of the landslide from the density of the river sediments. We eventually estimate an average loss of 4.7 ± 0.4 × 109 kg of sediment from 2015 to 2017, mostly due to the slow landslide. Although the gravity devices used in this study are expensive and need week-long surveys, new instrumentation progresses shall enable dense and continuous measurements at lower cost, making this method relevant to improve the estimation of erosion, sediment transfer and deposition in landscapes.

Gravity as a tool to improve the hydrologic mass budget in karstic areas

“…Another possible source of uncertainty comes from the role of sediment transport which is neglected in our modelling approach but can be an important flux mass source as testified by Mouyen et al (2020). The sediment mass affects the volume estimation by increasing the average density of the fluid and consequently reducing the water volume needed to explain the gravity anomaly.…”

Beyond 2D inventories : synoptic 3D landslide volume calculation from repeat LiDAR data

“…Addressing these three sources of uncertainty -volume-area scaling uncertainty, landslide amalgamation and the underdetection of landslide reactivation -requires to explore new approaches to obtain and analyse landslide inventories. In the last decade, the increasing availability of multi-temporal high resolution 3D point cloud data and digital elevation models (DEM), based on aerial or satellite photogrammetry and Light Detection and Ranging (LiDAR), has opened the possibility to better quantify landside volume and displacement (Bull et al, 2010;Mouyen et al, 2019;Okyay et al, 2019;Passalacqua et al, 2015;Ventura et al, 2011). The most commonly used technique is the difference of DEM (DoD) which consists in computing the vertical elevation differences between two DEMs of different time (Corsini et al, 2009;Giordan et al, 2013;Mora et al, 2018;Wheaton et al, 2010).…”