Vous êtes


Réduire la taille texte Rétablir la taille du texte par défaut Agrandir la taille du texte Partager cette page Favoris Email Imprimer

Natural Risks in a Mountain Environment

During flood events, tonnes of sediment stripped from high-altitude mountain slopes are carried by torrential streams to the valleys.

During flood events, tonnes of sediment stripped from high-altitude mountain slopes are carried by torrential streams to the valleys. This is called bedload sediment transport or suspension of sediments. The disrupted sediment balance may have serious consequences downstream, such as destabilisation of banks, collapse of dikes, bridge blockages, etc.

Research at the crossroads between fundamental research and targeted research carried out at CEMAGREF (the French National Centre of Agricultural Machinery, Agricultural Engineering, and Water and Forests) in Grenoble, has enabled the study of the bedload transport phenomenon from particle-level to river-level scale.

Improving the measurement techniques

From particle-level scale to river-section scale
Find out more

Torrential bedload transport and natural risks: how are they related?

What researchers are talking about:

© M. Chiari

Improving the Measurement Techniques

To this day, the study of sediment transport has remained a real scientific challenge. Although a network of hydrological stations can now measure the flow rate of most streams, measurement of the sediment transport during floods has remained technically impossible. Engineers must rely on the predictions of transport formulas. Research conducted at CEMAGREF aims at improving the reliability of these formulas, understanding the fluctuations of bedload sediment flows and quantifying them. These studies are based on laboratory testing using an experimental stream channel. This involves reconstructing a river bed in an inclined channel with a slope of up to 20% bounded by transparent walls.

This sedimentary bed is fed upstream by a constant flow of water and particles, and its behaviour is monitored in both space and time. Various techniques of high-speed video image analysis are used to obtain high-resolution measurements of the variations of bedload sediment transport.

Camera system capable of measuring the sediment flow rate at the exit of the experimental channel

From Particle-Level Scale to River-Section Scale

At particle level

How do flow parameters – such as flow rate and slope gradient or streambed roughness parameters – influence patterns of sediment transport? How do the transported particles interact among themselves as well as with the moving gravels forming the river bed? To answer these questions, researchers have designed a narrow experimental channel at particle scale enabling the study of sediment flow rates. This is a complex procedure. Initially, research focused on a single particle size using glass beads of 6 mm in diameter, whose movement was filmed by a high-speed camera on a stationary or moving background. By zooming in on the sediment transport process, scientists have identified a collective motion of rolling particles over very short and regular time intervals. Currently, research encompasses two particles sizes, i.e., 4 mm and 6 mm. The main challenge is to understand the mechanism of sediment segregation according to grain size, which is a major process occurring during sediment transport.

At river-section scale

The effects of grain size separation on bedload sediment transport were also studied on a small-scale model of a river section, represented by an experimental channel measuring 6 m in length and 0.25 m in width. These studies, carried out at the standard scale of hydraulic studies, have witnessed a rapid expansion at CEMAGREF through the development of original devices. To control sediment inflows, scientists have designed a distributor that delivers a constant flow of sediment particles of different sizes. A Sediment Flow Granulometry (GDS) device placed at the channel exit performs almost-instant measurement of the sediment flow and grain size based on high-frequency image analysis. Tests carried out using different mixtures of non-uniform materials have highlighted periodic fluctuations of the sediment flow rate, of the river bed state and its average slope. Two types of fluctuation have been noted: short fluctuations related to the movement of the thrust sheets and long fluctuations, which have a profound influence on bed morphology.

These studies, currently at different scales of experimental work, have opened up new avenues for research intended to integrate the obtained results into a single model. These experimental channel tests are further enriched by measurements currently carried out at several Alpine rivers under the GESTRANS project (Management of torrential flood risks by better understanding and control of sediment drift) conducted by the National Research Agency (ANR).

© J.R. Malavoi

Torrential bedload transport and natural risks: how are they related?
In areas with a steep gradient (above 2%), the streambed is constantly shifting. Driven by the current, sediment moves by rolling, sliding or hopping over short distances, remaining very close to the bottom. This is referred to as bedload sediment transport. During torrential rain, sediment transport becomes a major natural risk factor. Upstream, the energy of flows with a light load of solid matter promotes intensive erosive processes. Downstream, as the slope decreases, excess sediment settles, thereby raising the level of the streambed. Erosion, or sediment settlement at the opposite end, may sometimes cause shifts that are several meters in height!

A Cyclic Streambed Pattern

Toward New Decision Support Tools

Testimonial by Alain Recking
Research unit: Torrent, erosion, snow and avalanches, Grenoble

"Under constantly changing regulations (Framework Directive 2000/60/EC), watercourse management is becoming more and more difficult, as it must take into consideration at the same time both the short-term risks of flooding and the long-term environmental objectives. In many cases, these often-conflicting objectives render decision making a difficult process. In addition, engineers and decision-makers in the field still lack the tools to assess sedimentary dynamics. In an attempt to provide concrete answers to this complex problem with transversal implications, the GESTRANS project (2010-2012) has brought together the expertise of engineers, geographers, geophysicists and sociologists. The work is conducted both in the laboratory and at various Alpine sites, using state-of-the-art technologies, such as direct measurement of bedload sediment by hydrophone, use of LIDAR technology to monitor streambed morphological changes, image analysis for the study of streambed grain size, etc. The sociological component of the project seeks to analyse the behaviour of populations and decision-makers in light of risks related to sedimentary deposits and their long-term management at river scale. Led by CEMAGREF, the project brings together four scientific laboratories, the Isère Department Land Directorate (Direction départementale des territoires de l'Isère) and the University of British Columbia."
















Testimonial by Vito Bacchi
Research unit:
Torrent, erosion, snow and avalanches, Grenoble

sédiments utilisés en canal expériementalTypes of sediment used in the experimental channel testing

"The main objective of my thesis is to understand the function of grain size separation in flows and sediment transport on steep slopes (around 10%). Since it is difficult to describe and quantify these phenomena directly in the streams, my work is exclusively carried out based on the experimental channel.

Upstream, the channel is fed by a constant flow of particles mixed with water. I use sediments of variable sizes falling into six size categories, from 1 mm to 25 mm. Each particle size can be identified by the naked eye according to its colour (red, green and blue for the largest pebbles, with variations of natural shades for the others). Using cameras positioned above, on the sides and at the exit of the channel, I analyse the morphological evolution of the streambed. Surface particles form characteristic structures, such as "step-pools". Over time, these structures slowly disaggregate and then reform in a cyclic manner at different locations of the channel bed. Another aspect is related to surface particle size analysis. This is an important study parameter, as the movement of this surface layer plays a fundamental role in sediment transport phenomena."