I n s t r u m e n t s

a n d

M e t h o d s

Manysensorshavebeendeveloped inordertoprovide
frequent and accuratemeteorological data. But case-based
forecasting systems alsoneed avalanche data, which are
often difficult to collect,even with many human observ-
ers,because offog,storms,ornight.Butwhat cannot be
seen may be heard...
Appliedacousticshasrecentlycomeupwithanew
means ofautomaticallydetectingavalanches:ARFANG
system(cf.§2) is already able to deliver real time informa-
tion about avalanche activity overmountain ranges.
Ourobjective isnow toconnect ARFANG tothe fore-
castingexpertsystemNXLOG(Bolognesi,Buser,Good,
1994) in order to try to create an efficient chain.

2. THE ACOUSTIC SYSTEM
   
   Dur ingtheirmovement,mostavalanchesproduce
   infrasounds, i.e. acoustic signals within approx. 1- 20 Hz
   (Chritin, Rossi, 1995). These inaudibleandlong-rangelow-
   frequency sounds(fig.2)aresuitableforautomaticand
   real-timedetection and localisationoftheavalanche ac-
   tivityoveran extended area of several square km.
   The ARFANG experimental system(fig.3)consists of
   four special outdoor microphones combined in sucha way
   as toconstituteanacousticgoniometer(instrumentto
   measure angles). The fourmicrophones are set up in the
   form ofa cross orstareitheron masts orunderthe snow
   coverin a 20 to 50 square meter area.
   The system determines automaticallythe incident di-
   rection-geographical azimuthandelevation -ofsound
   wavesfromtheircalculated arrivaltimedelays between
   pairsofmicrophones.Dedicatedmicrophones(Rossi,
   Chritin,1995),named ECHO and optimised to the moun-
   tain wintermeteorologicaland topographical conditions
   were designed, built and installed.
   Apart from avalanches, some otherinfrasound sources
   exist - natural(storms, waterfalls, etc.)or man-made(planes,
   industry, etc.) - all of which can also potentially be piched
   by the microphones. The question is therefore to recognize
   avalanche infrasounds from all otherinfrasounds present
   in theenvironment.Themethodisdevelopedonthe
   principleoftheacousticformanalysis(fig.5)ofthe
   incomingacousticevents.Thisanalysis isbasedonthe
   comparisonofthetimeevolutionofintensityvaluesin
   appropriate frequency bands.
   (fig. 6) ifboth the output values ofthe intensity analysis
   and of the goniometerfullfill avalanches criteria charac-
   teristics.

Fig. 2: Sound wave formrecorded during an artificial release of ava-
lanche.The signalrepresents thetimeevolution (abscissa) of the acous-
tic pressure in the (low) frequency range 1- 20 Hz(vertical axis).

Fig. 3: Principle of avalanche localisation by acoustic goniometry

130