Assessment of erosion and settling properties of fine sediments stored in cobble bed rivers: the Arc and Isère alpine rivers before and after reservoir flushing

Abstract

Cohesive sediment dynamics in mountainous rivers is poorly understood even though these rivers are the main providers of fine particles to the oceans. Complex interactions exist between the coarse matrix of cobble bed rivers and fine sediments. Given that fine sediment load in such environments can be very high due to intense natural rainfall or snowmelt events and to man‐induced reservoir or dam flushing, a better understanding of the deposition and sedimentation processes is needed in order to reduce ecohydrological downstream impacts. We tested a field‐based approach on the Arc and Isère alpine rivers combining measurements of erosion and settling properties of river bed deposits before and after a dam flushing, with the U‐GEMS (Gust Erosion Microcosm System) and SCAF (System Characterizing Aggregates and Flocs), respectively. These measurements highlight that critical shears, rates of erosion, settling velocities and propensity of particles to flocculate are highly variable in time and space. This is reflective of the heterogeneity of the hydrodynamic conditions during particle settling, local bed roughness, and nature and size of particles. Generally the deposits were found to be stable relative to what is measured in lowland rivers. It was, however, not possible to make a conclusive assessment of the extent to which the dynamics of deposits after reservoir flushing were different from those settled after natural events. The absence of any relationships between erosion and deposition variables, making it impossible to predict one from another, underlined the need to measure all of them to have a full assessment of the fine sediment dynamics and to obtain representative input variables for numerical models. While the SCAF was found to be effective, an alternative to the U‐GEMS device will have to be found for the erodibility assessment in cobble bed rivers, in order to make more rapid measurements at higher shears. Copyright © 2017 John Wiley & Sons, Ltd.

Numerical modelling of the suspended particulate matter dynamics in a regulated river network

“…2). Cohesion is neglected since organic content is very low (< 5%, typically) as well as cohesiveness (Legout et al, 2018). 18…”

An autonomous and low-power instrument platform for monitoring water and solid discharges in mesoscale rivers

“…From the slopes of multiple isoabsorbance lines, settling velocity distributions (SVD) of suspended solids can be calculated, as well as an indicator of the propensity of particles to flocculate (Wendling et al 2015). The SCAF is able to operate for a wide range of SSC (from one to several tens of g/l) and settling regimes (free, flocculated, and hindered settling regimes) as illustrated in Legout et al (2018). The data produced by the SCAF are useful for the understanding and the modelling of the suspended sediment transport in rivers.…”

“…Measurements are acquired every 100 ms each value being the average of 10 measurements. The SCAF measurements typically last over 5 hours as shown in Legout et al (2018). Each SCAF unit is working separately, waiting for the various impulses sequence of the data logger.…”

An autonomous low-power instrument platform for monitoring water and solid discharges in mesoscale rivers

“…The first variable that has been added to hydrometric stations is velocity. During the last decade, the introduction of fixed flow velocity monitoring systems (transit time ultrasonic flowmeters, acoustic Doppler current profiler horizontal (H-aDcp) and vertical (V-aDcp), surface velocity radars) into hydrometric station equipment has gradually become more widespread (Costa et al, 2006;Levesque and Oberg, 2012;Nord et al, 2014;Thollet et al, 2017). These systems all provide a more or less local monitoring of flow veloc-ity.…”

“…Thus the development of innovative methods is necessary to estimate the mean channel velocity from surrogate monitoring. The first attempts concern the index velocity method, but this approach is still largely based on the use of gauging (Levesque and Oberg, 2012;Morlock et al, 2002). Continuous monitoring of the crosssectional flow area by a direct method is not possible to date because of the lack of dedicated technology.…”