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

while it took about a half of second to cover the top sensor.
This was due to deceleration of the slide.
The second slide recorded was on 3/19/96.It was on a
sunny day with approximately20cm ofnew snow from
twodaysprior.Thewarmweatherwasstabilizingthe
snowpack.The shearplate and the optical velocitysen-
sors were the onlyinstrumentation applied to this slide.
Fourpounds of explosive were used to trigger a relatively
small slide traveling at around6 m/s with very small pow-
dercloud. The bulk of the slide passed the shed in about
1.3seconds,reachinganestimatedmaximumdepth of
about 30cmbefore stalling out high inthe runout zone.
Theestimatedflowdepthforthisslidewastakenfrom
video and checkedwith measurements of the normal stress
combined with an approximation of the slide averageden-
sity.In this case a smallersluff arrived at the instrument
shed first and was overtaken by the main slide.This can
be seen in the plate data (Figure 4).The plate data also
showsthatboththenormaland shearstressescontinue
decreasing once theslide has come to a stop.It is believed
that this is caused by the plate slowly slipping back to its
unstrainedpositionandsinteringofdepositedsnow
bridging the plate.It can be seen in Figure 5 that the shear
to normalstress ratio(S/N) remains high throughout the
slide,averaging about.85.Approximately15-20cmof

Figure 4

to the applied stresses.Priorto the slide the plate surface
is preparedwith alayer of snow and calibratedwith several
known weights.
Ideally, the shear angle of the snow will be equal to but
not less than the slope angle keepingdeposition to a mini-
mum. A lower shearanglewouldcauseerosion of theslope,
and since the slope is artificially prepared, erosion would
degrade the results.
Error is introducedinthe dataas theslide depositssnow
ontheplate.Thiserrorisduetotheshearandnormal
componentsofthedepositedsnow'sweight.Thisalso
introduces errorin the flow depth measurement because
the depth gaugeincludes in its measures the height of the
depositedsnowbeneath theslide.Itisassumedthat if
depositionisminimizedand limitedtothatdue tofric-
tional deceleration then these errors are small.

RESULTS

During the winter of 1996 implementation ofnew instru-
mentationwasperformedforseveralavalanches.The
optical sensors were used on all of the slides to measure
velocities,butattemptstocalibratedthemfordensity
measurements was only attempted once.The instrumen-
tation for capacitanceprobe was only availablefor oneslide
this last winter.The shear plate was installed forourlast
two slides,while the depth gaugewas only used forone.
Priortothefirstslideon3/27/96,acoldfrontcame
thoughand deposited 76cmofnewsnowand dropped
temperatures below 0 ^oF with high easterly winds.Ideal
avalancheconditionsallowed us to releaseour largest slide
of the season.With 6lbs.of explosivesa large slab was
released running approximately 1.5 m deep and traveling
up to 8 m/s.The capacitance probe and the optical sen-
sorsmeasuringvelocity,were theonlyinstrumentation
applied to this slide.The capacitance sensors measured
densities at only1 cmand 6 cm above the slide running
surface.The data shows that the density at 1 cm above the
runningsurfacewas in the rangeof 450 Kg/m³and quickly
decreasing to about 300 Kg/m³at 6 cm.This indicates that
the density of the slide quickly decreases in the first few
centimeters and then decreases very little toward the sur-
faceoftheslide.Althoughona wholethere waslittle
deposition on the runningsurface with this slide, the data
indicates that the sensors were quickly buried.The bot-
tomsensorwas covered in less than a tenth ofa second,

Figure 6

205