S n o w

C o v e r

S t a b i l i t y,

A v a l a n c h e

I n i t ia t i o n

a n d

F o r e c a s t i n g

SYNTHESIS AND DISCUSSIONOF OBSERVATIONS^{thanthecompressivemotion(ona36oslope)andthe}
Wesuspect the apparently anomalous behavior discussed^{resultantdirectionofmotionisclosetovertical.The}
aboveiscausedbymetamorphicprocesses.Although^{component of motion in the downslope directiondepends}
metamorphism is fairly well understood on the grain scale^{on the slopeangle.Most of the curvaturein thecreepprofile}
(eg. Arons and Colbeck, 1995; Dash et al, 1996), the impact^{isaresultofchanges inthe rate of(vertical)settlement}
on macroscale behavior is less well understood. Below we^{throughthe depthofthesnowpack.Therate ofmotion}
discuss observed macroscale behaviorin context of grain^{decreasesrapidlywithdensity;largegradientsinthe}
scale processes.^{velocityprofileoccurbecauseofcontrastsinlayer}
properties.
SHRINKAGE AND SINTERING^{The result suggests that the usual assumptions used to}
Thermodynamicprocessesacttominimizethesurface^{describe the mechanicalbehavior of snow are incomplete.}
energyofsnowgrains.The preferred shape isasphere,^{Following on from our previous discussion, we believethat}
hence grainsbecomemoreroundedwithtime.Concur-^{metamorphic and/orcapillary strain contribute to the ob-}
rently,adjacent particlessinterbondwhich reduces the
surface energy by removing free surfaces.In dry, natural
snowpackstherateofroundingandsinteringdepends
mainly on microscaletemperaturegradients(Colbeck, 1980,
1983).Thesaturationvaporpressureoverice increases
exponentially with temperaturewhichmeans the sintering
rateshouldincreaser apidlywithtemper atur e.
Experimental evidence (eg. Gubler, 1982;de Montmollin,
1982;Dashetal,1996)confirmsthatsinteringratesin-
crease rapidly astemperaturesincrease aboveabout-10
o _C.
Introduction of liquid water not only enhances the rate
of mass transport on the grain scale but also formsliquid
bridgesbetweengrainsresultingina"capillarystrain"
(Hwang et al., 1987).Capillary forces in snow in excess of
4000Pa have been measured in partially saturated snow
(Colbeck, 1974; Wankiewicz, 1979). Such forces are much

higher than gravitational forces expected in the near-sur-
face snow and cause rapiddensification independentlyof
gravity. The highcapillary pressures causegrains to cluster
(Colbeck,1982).We haveconductedexperimentsto
investigateratesof shrinkagewhenwater is first introduced
to snow. Figure 4a shows an experiment in which snow ( r
= 90 kg m^-3 ) was collected in a can 140 mm high and 100
mmdiameter.The sample was kept at -4^o C for20 hours
during which time the snow settled 1 cm.Introduction of
liquid water(300 ml at 0^o C) caused the sample to shrink
both vertically and from the sides (figure 4a). Weinterpret
theshrinkage,inparticularthelateralshrinkage,tobe
evidenceofcapillarystrain.Therateofdensification
increasedbyfourordersofmagnitudeattheonsetof
warmingandwetting(figure 4b). The rate decreasedrapidly
as the density approachedthe final (dry) density of about
360kg m ^-3 ;the rate of densification depends on density.
Highest rates are expected during first wetting of very low
density snow. The experiment clearly illustrates that snow
can undergosubstantialdeformationwithoutsignificant
changes in the gravitational stress.

CREEP BEHAVIOR OF ALPINE SNOW

The usual approach to modelling creep behavioris to ap-
ply a constitutive law that relates the bulk response of the
snowtoanapplied stress.Followingtheusual assump-
tions, for snow on a 36 ^ºslope (see Perla (1980) for a review
ofpreviouswork)weexpect the shearingcomponent of
motion would be about 50% greater than the compressive
component (in the range of the shaded area of figure 3b).
This contrasts with ourmeasurements and some by Perla
(1971) that indicate the shearing component is often less

78