Thèse de doctorat en Sciences de la terre et de l'univers, et de l'environnement
Soutenue le 28-11-2012
à Grenoble , dans le cadre de École doctorale terre, univers, environnement (Grenoble) , en partenariat avec Institut national de recherche en sciences et technologies pour l'environnement et l'agriculture (France). Centre de Grenoble (laboratoire) .
Le président du jury était Frédéric Dufour.
Le jury était composé de Nicolas Rivière.
Etude expérimentale de l'influence des structures de protection sur les avalanches et pressions d'impact
L'auteur n'a pas fourni de résumé en français
Experimental study of the influence of protection structures on avalanches and impact pressures Abstract: In the frame of snow avalanche protection, the optimisation of defence structure design depends on the understanding of the flow dynamics and on a exhaustive knowledge of the flow-obstacle interaction. The study presented here utilises a mainly experimental approach. Small-scale laboratory tests were combined with field measurements and observations. Dense snow avalanches are modelled by granular materials. Dry cohesionless and mono-dispersed glass beads are released down an inclined channel. Reference tests (with no obstacles) were carried out in order to characterise the flow dynamic properties, and an obstacle was then mounted and force measurements were taken. The geometry of two obstacles was tested: a large wall spanning the whole width of the flow and a narrower wall allowing lateral flows. Results showed that an influence zone forms uphill from the obstacle and plays an important role in the flow dynamics. An analysis of this zone was carried out, together with precise measurements of the flow depth (laser technique), surface velocity (PIV) and impact forces (force sensors). In relation to density currents, powder snow avalanches are modelled by a dyed salt solution flowing down an inclined channel immersed in a water tank. We investigated the influence of two different catching-dam-type obstacles on the flow behaviour with respect to reference conditions. The maximum flow height and its front and core velocity were measured by means of image processing and ultrasound Doppler velocimetry. Results mainly showed the higher effectiveness of a dam with vertical uphill face rather than inclined and underscored the importance of the velocity norm in the computation of the total incoming flow velocity. In-situ full-scale measurements complement small-scale laboratory tests. A new full-scale experimental site was implemented on the existing avalanche defence system of Taconnaz (Chamonix, France). Three breaking mounds were equipped with velocity and pressure sensors. The reference event, the rough site conditions, the need for data synchronization and remote access defined the design constraints. In December 2010, the first event was recorded, which proved that the conceived system works effectively and also provided the first set of data. Preliminary results showed very high pressure peaks with high impact pressures even for low velocity regimes and thus a drag coefficient which increases when the Froude number decreases. Isolated rocks or ice blocks struck the sensors, contributing significantly to the total energy released by the avalanche. This work provided effective inputs for numerical and analytical models and enhanced the current knowledge of avalanche dynamics in order to optimise the future design guidelines for avalanche protection structures. Keywords: Snow avalanches, impact pressure, laboratory experiments, dense avalanches, granular flows, influence zone, dead zone to granular jump transition, powder avalanches, density currents, ultrasound Doppler velocimetry, full-scale measurements.