Réf. Baggi & Schweizer 2009 - A

Référence bibliographique complète

BAGGI, S., SCHWEIZER, J. 2009. Characteristics of wet-snow avalanche activity: 20 years of observations from a high alpine valley (Dischma, Switzerland). Natural Hazards, 50:97–108, DOI 10.1007/s11069-008-9322-7.

Abstract: The occurrence of wet-snow avalanches is, in general, poorly understood. For 20 years (winters of 1975–1976 to 1994–1995), the avalanche activity has been observed in the Dischma valley near Davos (Eastern Swiss Alps). The study area comprises a large starting zone of north-easterly aspect (2,300 m a.s.l.) with several avalanche paths. The authors have analyzed the occurrence data in combination with meteorological and snowpack data collected at an elevation of 2,090 m a.s.l. During the 20-year observation period, almost 800 wet-snow avalanches were observed, about 4.5 times more loose snow avalanches than slab avalanches. Considering both types of avalanches jointly, snow depth, precipitation and air temperature showed the highest correlation with avalanche activity. Most loose snow avalanches occurred when air temperature was high and/or after a precipitation period. Slab avalanches occurrence was primarily related to warm air temperatures and snowpack properties such as the isothermal state and the existence of capillary barriers. Radiation did not show up as a significant variable. The results suggest that in a transitional snow climate wet-snow avalanches are, as dry snow avalanches, often related to precipitation events, and that wet slab instability strongly depends on snowpack properties in relation to warming of the snowpack and melt water production.

Mots-clés
Snow avalanche, Wet snow, Snow cover, Avalanche forecasting

Organismes / Contact

WSL Institute for Snow and Avalanche Research SLF, Flüelastrasse 11, 7260 Davos Dorf, Switzerland - schweizer@slf.ch


(1) - Paramètre(s) atmosphérique(s) modifié(s)
(2) - Elément(s) du milieu impacté(s)
(3) - Type(s) d'aléa impacté(s)
(3) - Sous-type(s) d'aléa
    Avalanches Loose snow avalanche, Slab avalanche

Pays / Zone
Massif / Secteur
Site(s) d'étude
Exposition
Altitude
Période(s) d'observation
Switzerland Eastern Swiss Alps Stillberg study area in the Dischma valley (near Davos) generally north-east aspect 2,000-2,300m a.s.l. (avalanche starting zones) 1975-1995

(1) - Modifications des paramètres atmosphériques
Reconstitutions
 
Observations
 
Modélisations
 
Hypothèses
 

Informations complémentaires (données utilisées, méthode, scénarios, etc.)
 

(2) - Effets du changement climatique sur le milieu naturel
Reconstitutions
 
Observations
 
Modélisations
 
Hypothèses
 

Sensibilité du milieu à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)
 

 


(3) - Effets du changement climatique sur l'aléa
Reconstitutions
 
Observations

The authors have explored a large dataset of weather, snowpack and avalanche occurrence data in order to get insight into wet-snow avalanche activity. For the first time, they have considered variables describing snowpack characteristics that have been proposed as relevant for wet slab avalanche release such as capillary barriers. Though the results might be partly site specific and influenced by the generally north-east aspect of the starting zones, important conclusions on, for example, the timing of wet-snow slab avalanches should not be site specific since they depend on the state of the snowpack. Though the isothermal state will be reached earlier on south-facing slopes, the mechanism involved in triggering might still be the same. Armstrong (1976) has pointed out the sequence of release as a function of aspect.

In general, a large variability in wet-snow avalanche occurrence over the 20 winters was observed. There was a slight trend of increased wet slab avalanche activity in winters with weak basal layers. This coincides with a common hypothesis that in winters with significant depth hoar formation at the beginning of the season, more wet slab avalanches in spring have to be expected. However, the trend was statistically not significant.

Wet loose snow avalanche activity peaked in May, towards the end of the melting season, whereas wet-snow slab avalanches were mostly observed in March at the beginning of the melting season and subsequently when the snowpack had become isothermal. Avalanche activity in general was often related to precipitation events, in particular rain.

Loose snow avalanches were most frequently observed with warm air temperatures and/ or rain. Whereas, high air temperatures shortly after the entire snowpack had reached 0°C (isothermal snowpack) and the existence of capillary barriers characterized the wet slab avalanche days.

The present analysis suggests that for the release of wet slab avalanches snowpack parameters (including stratification) might be more important than commonly assumed. They might even be as important as in the case of dry-snow slab avalanche formation. Hence, considering snowpack properties might also be the key for the forecasting of wet-snow slab avalanches. However, snowpack monitoring (manual and automatic) as well as simulation is certainly more challenging for wet-snow slab avalanche forecasting since the relevant processes, in particular, water infiltration, are highly non-linear and variable in space and time. The ‘‘right’’ conditions for the release of a wet slab avalanche might prevail for a shorter time (hours) than in the case of dry-snow slab avalanches which may release on weak layers that persist for days, or even weeks.

Supposedly, climate change will affect wet-snow avalanche activity—possibly in respect to timing and location/elevation of occurrence. However, given the considerable lack of understanding of wet-snow avalanche formation it seems premature for any firm statement on possible consequences of climate change on avalanche activity.

Finally, based on the analysis, the authors suggest three triggering mechanisms for wet-snow avalanches: (1) loss of strength due to water infiltration and storage at capillary barrier, (2) overloading of partially wet (and weakened) snowpack due to precipitation and (3) gradual weakening of (basal) snowpack due to warming of snowpack to 0°C and eventual failure of basal layers. Obviously, combinations of these three mechanisms may exist.

Modélisations
 
Hypothèses
 

Paramètre de l'aléa
Sensibilité des paramètres de l'aléa à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)
Avalanche occurrence: daily avalanche activity index

Various studies at the site have been performed since the 1950s with the primary focus on reforestation and, in general, on the interaction between forest, snow cover and avalanches. In the course of these projects, avalanche activity has consistently been observed for many years. At 2,090 m a.s.l., an automatic weather station (AWS) is located that since 1975 records—among other elements—air temperature, soil surface temperature, snow temperatures, relative humidity, precipitation, snow depth, global radiation (incoming and reflected short wave radiation), wind speed and wind direction. Near the AWS, bimonthly manual snow profiles have been recorded including grain type and size, snow hardness index, snow temperature, estimated liquid water content (wetness), snow density and ram hardness profile (Colbeck et al. 1990). Avalanche activity was observed on a daily basis. Records include date (and if possible time), location, elevation, aspect, estimated fracture depth, estimated avalanche width and length, and avalanche characteristics according to the international morphological classification of avalanches (de Quervain et al. 1973).

Statistical analysis Meteorological and snowpack data were related to avalanche occurrence data applying standard univariate and multivariate statistics. The non-parametric Mann–Whitney U-test was used to relate single variables to avalanche occurrence. For categorical variables, the data were cross-tabulated and a Yates’ corrected Pearson χ² statistic was calculated (Spiegel and Stephens 1999). For the multivariate analyses primarily, the classification tree method was used (Breiman et al. 1998). The classification tree method was previously applied for avalanche forecasting, for example, by Davis et al. (1999), and by Hendrikx et al. (2005). A level of significance p = 0.05 was chosen to decide whether the observed correlations or differences were statistically significant.

For the analysis, the authors have chosen 20 years of observations from 1975–1976 to 1994–1995 (October–June), since for this period data from the AWS were available, avalanche occurrence observations were considered as consistent (they become less so towards the end of the 1990s) and the period seemed long enough to cover a variety of snow and avalanche conditions. [see details of data in the study]


(4) - Remarques générales
 

(5) - Syntèses et préconisations
 

Références citées :

Armstrong RL (1976) Wet snow avalanches. In: Armstrong R, Ives JD (eds) Avalanche release and snow characteristics, San Juan Mountains, Colorado. Sun Juan Avalanche Project, Final report 1971–1975, Occasional paper no. 19. Institute of Arctic and Alpine Research, University of Colorado, Boulder, CO, USA, pp 67–82

Breiman L, Friedman JH, Olshen RA, Stone CJ (1998) Classification and regression trees. CRC Press, Boca Raton, USA, p 368

Colbeck SC, Akitaya E, Armstrong R, Gubler H, Lafeuille J, Lied K, McClung D, Morris E (1990) The international classification of seasonal snow on the ground. International Commission on Snow and Ice (ICSI), International Association of Scientific Hydrology, Wallingford, Oxon, UK, p 23

Davis RE, Elder K, Howlett D, Bouzaglou E (1999) Relating storm and weather factors to dry slab avalanche activity at Alta, Utah, and Mammoth Mountain, California, using classification and regression trees. Cold Reg Sci Technol 30(1–3):79–89. doi:10.1016/S0165-232X(99)00032-4

de Quervain MR, de Crecy L, LaChapelle ER, Losev K, Shoda M (1973) Avalanche classification. Hydrol Sci Bull 18(4):391–402

Hendrikx J, Owens I, Carran W, Carran A (2005) Avalanche activity in an extreme maritime climate: the application of classification trees for forecasting. Cold Reg Sci Technol 43(1–2):104–116. doi: 10.1016/j.coldregions.2005.05.006

Spiegel MR, Stephens LJ (1999) Schaum’s outline of theory and problems of statistics. McGraw-Hill, New York, p 538