A v a l a n c h e

D y n a m i c s

a n d

D e f e n ce

a consolidated body. Hence, it is reasonable to say that the
size of the head gives a strong effect on the flow velocity
changelistedinFigure10.Inaddition,itshouldbe
noteworthythattheheadand tailstructureshownthis
experiments had been often observed not only the snow
avalanche butalsootherlarge-scale naturalgeophysical
flows in nature.

CONCLUSIONS

We have started the avalanche experimentswiththe ski
jump,because it seems the longest inclined plane we are
available under the controlled condition.In winter, natu-
ral snow 300 kg in weight at maximum was released and
flow velocities, impact pressures, inducedwind velocities,
and dynamic friction coefficients were measured. Instead
of snow itself, in summer, wehaveusedup to 300,000 ping-

Figure 9. 250,000 ping-pongball flow along theski jump.

pongballs.Althoughsomemeasurementslikeinsnow
experimentswerecarried out aswell, inthis paper wehave
just concentrated on introducingthe leading edgevelocity
as functions of the runout distanceandthe number of ping-
pong balls. As mentioned earlier, a number of theories and
theirnumericalsimulationhavebeen carriedouttode-
scribe the granular flow. However,all of them are still too
simplifiedtodraweven the figure oftheping-pong ball
flow in this study. At least the model should be extended
tothree-dimensionandtheairdrageffecthastobe
included.Computersimulationsincludingaboveeffects
are currently in progress.
Our snow experiments proved that the measurement of
static depression was the reliable and useful tool to inves-
tigate the air movement in the snow avalanche. At Kurobe
Canyon where asystematic investigation of natural powder
snowavalanche is underway since 1989,an ultra-sonic
anemometerhas been utilized, but so far we could obtain
theairmovementdataforonesmallavalancheonce
(Nishimura et al., 1993);others showed just the abnormal
signals dueto the fairly largesnow concentrations. Securing
the foothold in this study, in the winter from 1995 to 1996
we installed the tube measuring the static pressure depres-
sion. During the observation period we had larger amount
ofsnowaccumulationthanaverageandseveralsnow

2 0

1 5

1 0

5

0

Figure 8. The leading edge velocity of the ping-pong ball flow. Here
250,000 balls were released from the point of 15 m.

208