|
 Dept. of Civil Engineering,University of Calgary, 2500 University Dr. NW
Calgary AB T2N 1N4, Canada. fax: 1 403 282 7026, e-mail: jschweiz@acs.ucalgary.ca
|
|
|
Keywords: snow mechanics, snowstrength, snow tempera-
ture, snow failure, fracture
ABSTRACT.
Snow samples (115 mmin diameter, 16-18 mmin height)
were taken from the field and shortly afterwards tested in
a cold laboratory using a direct simple shear apparatus of
Norwegian manufacture. The effectsof strain rate andtem-
perature onsnowstrength, stiffnessand toughnesswere
studied.The transitionfromductiletobrittlebehaviour
was found to be at a strain rate between10-4and 10-3s-1for
thesnowtypesand temperaturestested.Stiffness upto
peakstressprovedtobehighlytemperaturedependent.
Toughnessisalsosignificantlytemperaturedependent,
while snow strengthseems to dependonly slightly on tem-
perature.Thedependenceoftheshearstrengthon
temperatureseems partly hidden by the scatter in strength
data due to variations inherent in sampling and testing.
INTRODUCTION
The behaviour of snow under shear is believed to be one of
the most important factors fordescribing slab avalanche
formation and assessing the snow stability. Although most
previouscontrolledshearexperimentswere doneusing
slow displacement rates, fieldtests of snow strength or slab
stability such as the rutschblock test involve rapid but less
controlledloadingandstrainrates. Previoustestingin shear
underlaboratoryconditionswasdonebyBallardetal.
(1965),McClung (1977)and Fukuzawa(1993).Some field
studies:Roch(1966),Perla(1977),Föhn(1987)and
Jamieson(1995).RecentlyFöhnand Camponovo(1996)
havetriedtomeasureshearstrengthinsituunder
controlled conditions.
METHODS
Testswere performed with a Norwegiantype directsimple
shear apparatus in the cold laboratory at Rogers Pass, Gla-
cier National Park (British Columbia, Canada) (Fig. 1). The
effects of shearing rate and temperature were studied.In
particular,to establish the ductileto brittle transition, rapid
tests at strain rate larger than 10-4s-1were performed.
Specimens (115 mm in diameter,16-18 mm in height)
weretakenfromrelativelyhomogeneouslayersoffine
grained snow with densities between 220 and 340 kg/m3 ,
and handhardness index: 3 (1 finger) to 4 (pencil). These
were tested at -5,-10 and -15 ºC forsix different displace-
mentrates:about0.0073 mm/min,0.074mm/min,0.17
mm/min,0.29 mm/min,0.50 mm/min,and 4.2 mm/min.
For eachof the displacement rates andtemperatures about
5 tests were performed, resultingin a total of about 90 tests.
Typically, samples were tested ata certain displacement
rate on three consecutive days at the three temperatures.
Accordingly, samples were stored typically between 1 and
5days.Thisproceduremayhaveslightlyaffectedthe
results due to ongoing age-hardening.
|
|
|
|
Thenormal force applied was 4.9 N,which corresponds
toastressof470Pa,andwaskeptconstantduringall
experiments.Duringthetestshorizontalandvertical
displacements,andtheapplied horizontal(shear)force
were measured and recorded. Due to hard-/software limi-
tationsthemaximumpossiblescanfrequency wasonly
10 Hz,which proved to be insufficient forthe most rapid
displacement rates (strain rates between 10-3and 10-2s-1 ),
so the numberofmeasurements is generally too small to
establish a reliable stress-strain curve.
The shear strength (failure or peak stress) is defined as
the maximumstress,the peak,on the stress-strain curve.
The initialslope of the stress-strain curve iscalled stiff-
ness,and representsinitialresistance tosheardeforma-
tion.Stiffnessasgivenbelowistheslopeofthesecant
intersectingthe stress-straincurve at80%of the peakstress.
Strain rates, calculated from the time-displacement (or
strain) curve,are typicallynotconstant during the tests,
since horizontaldisplacement(andaccordinglystrain)does
not increase linearly with time.The strain rates given in
the following sections are mean values calculatedfrom the
deformation at failure,the time to failure and the sample
thickness.
Effectof strain rate
Typical stress strain-curves forthree different strain rates
are shown in Fig.2.Curve (a) showsthe response of the
stressed snow sample fora strain rate of 6.1 = 10-5s-1 ;test
duration was 80 min.The curve shows typical strain sof-
tening behaviourwith a ductile type of failure character-
izedbylargedeformationandhigh toughness orenergy
absorption capability asevidenced by the area under the
stress-straincurve.Curve(b)istypicalforthe testsper-
formed with strain rates of about 10-4s-1and is believed to
|
|
1
2
3
