I n s t r u m e n t s

a n d

M e t h o d s

Fig.2: Some typical wind situations without significant snowdrift.

the particles collector given by calculations were conven-
ientfordailymeasurements (fig.3).Theexperimentsite
was located in a high mountain area (Weissfluhjoch, Swit-
zerland).Anautomaticwindmeasurement stationwas
availableonthe samesiteas well.So itwaspossibleto
evaluatesnowconcentrations frommassfluxtobesure
that experimental conditionswere correct. Thesetests have
shownthatthisprincipleofseparationisefficient.The
userhas simply to weigh the caughtsnow to get anindex
(not an actual value!) of the snowdrift intensity.

1.2.DETERMINATION OF SNOWDRIFT DIRECTION

Thelocalavalanche forecastersneed toknowthe snow-
drift directionbecause the locationof snow accumulations
directly depends on it.But it makes no sense to want to
knowwhetherthe snowdriftdirection is180or181de-
grees. So it was decidedthat the snowdrift direction indi-
cated by the driftometershould be a sectorof 45 degrees:
N orNW orW orSW orS orSE orE orNE.
Themostsimplesolutionforknowingthesnowdrift
sector is to put one snow particle collector in each of the 8
sectors:iftheNorthcollectorcatchessnowparticles,it
means that the snowdrift direction is more orless North.
Unfortunately, if the snowdrift direction is North then the
collectors North-East and North-Westwill also catch snow
particles! So we designeda system of deflectors (fig.4) and
tested it in the wind-tunnelat CEMAGREF with laser visu-
alization.With thissystem,onlyonecollectorcatches a
significant mass of snow if the wind direction is constant:
so it ispossible to know snowdrift direction for24 hours
without any automatic recording instrumentation.
Following the validation of the principle in the wind-
tunnel, the last prototype was built and installed in a high
mountain ski resort (Alpe d'Huez, France) to be tested in
actual conditions by practitioners. According to their sug-
gestions, some last technicalimprovements weremade and
the plans of the definitive instrument were drawn up.

130

The driftometer (fig.5) is now manufactured and sold by a
small French company called ISER'OUTIL (price approxi-
mately$200), and used daily by the numerous ski resorts
which work with the avalancheforecasting system NxLog
(Bolognesi,1994).Thedriftometerisalsoused forsome
scientificstudiesaboutsnowslabsformation(Duclos,
1995).

2. VALIDATION
   
   Every snowfallis more orlessresponsible foravalanche
   occurrence: forecasters usually consider the heightof new
   snow as a predictive variable foravalanche activity. So it
   seems reasonable to think that the snowdrift, which is sim-
   ply a 'horizontal snowfall', may also be a predictive vari-
   able foravalancheactivity.