Laboratory experiments on the dynamics of powder snow avalanges

Abstract

Long powder snow avalanches are simulated in the laboratory as a chute flow from a steady source of a turbulent mixture of polystyrene particles suspended in water. Using the ultrasonic measuring technique described in Scheiwiller et al (1985, 1986, 1987) and employing ensemble averaging techniques we present results for the particle concentration and velocity in profiles of the steady tail as well as in the transient head of the avalanche. The initial conditions are held fixed, but the inclination angle of the chute and the measuring positions are varied. EXPERIENCES EN LABORATOIRE CONCERNANT LA DYNAMIQUE DES AVALANCHES DE POUDREUSE RESUMEE Les longes avalanches de poudreuse sont simulees au laboratoire en forme d'une suspension de particles de polystyrene dans l'eau. Cette mixture turbulente coule dans une chute dont l'inclination peut varier. Ayant pris une technique ultrasonique, descrite dans Scheiwiller e.a. (1985,86,87) et calculant des moyennes, nous presentons maintenant des resultats venant du corps et de la tete de l'avalanche. Les conditions initiales sont toujours les memes sauf l'inclination de la pente et les positions ou nous avons pris les dates. INTRODUCTION The two limiting flow configurations in real snow avalanches are the flow avalanches and the powder snow avalanches. They can essentially be characterized as laminar granular flows and turbulent two phase flows of snow particles suspended in air. Powder snow avalanches are much lesser known than flow avalanches, because they are rare phenomena, which only occur a few times in winter, preferably after heavy snow falls when observation is hindered. The particle concentration in powder snow avalanches is one to two orders of magnitude smaller and velocities are about three to five times larger than in flow avalanches, so pressures are comparable. Measurements in the field are difficult to perform, firstly, because the powder form of an artificially released avalanche is not predictable, but secondly also because observations ought to be performed remotely in view of the dangers involved. Moreover, corro431 432 Felix Hermann et al. boration of a theoretical model by experiments tends to be demanding in the amount of necessary data from controlled conditions which is hardly possible with real avalanches. Hence a laboratory simulation is needed. This was done by us in an extensive laboratory study starting in 1981. The details of the simulation, the experimental setup, the equipement, and the measuring method are described in detail by Scheiwiller, Bucher and Hermann (1985), a summary of the method including first results is given by Scheiwiller (1986) and Scheiwiller, Hutter and Hermann (1986). For convenience we briefly outline here the simulation and measuring method. In the laboratory simulation the air is replaced by water and the snow by a suitably chosen polystyrene powder, so that the laboratory two-phase flow behaves similarly to the powder snow avalanche in nature. The similarity between the real powder snow avalanche and that in the laboratory is basically due to the conservation of dimensionless quantities such as the velocity ratio U/W, a densimetric Froude number and the Reynolds number (see Table 1). The latter is in both cases very high and a dependence does not arise. In a large water tank a chute was installed permitting inclinations from 0 to 90 degrees (Fig.l), a specially designed feed apparatus allows the preparation of a concentrated water-powder mixture and guarantees for controllable constant flux and concentration at the' inlet orifice. This nozzle is placed at the chute bottom midway between the chute walls. The spreading zone where the flow is established extends over 0.5 to 0.7m; quasi-twodimensional flow confined between the two chute walls is thus well established lm downstream of the inlet orifice where first TRANSDUCER WATERPROOF BOX WITH STEP-DRIVE MOTOR MICROPROCESSOR CHUTE TANK 0 © (S) C-0 © BOTTOM OUTLET GANGV/AY FEED APPARATUS INTERIOR STEEL FRAME EXTERIOR STEEL FRAME © 0 © @ ® Fig.l Experimental setup for laboratory powder snow avalanches with polystyrene particles. The dimensions of the tank are 5 * 4 * 1.5m. Laboratory experiments on the dynamics of powder snow avalanches 433