Analysis of snow avalanche potential in Bâlea glacial area-făgăraş massif, Southern Carpathians (Romanian Carpathians) with 15 figures and 5 tables

Transcription

1 Analysis of snow avalanche potential in Bâlea glacial area-făgăraş massif, Southern Carpathians (Romanian Carpathians) by MIRCEA VOICULESCU, FLORINA ARDELEAN, ALEXANDRU ONACA and MARCEL TÖRÖK-OANCE with 15 figures and 5 tables Abstract. Based on a case study, this paper is intended to be a first attempt, in the Romanian geography, to analyze the snow avalanche potential in one of the most representative glacial areas from the Făgăraş massif (Southern Carpathians- Romanian Carpathians), respectively Bâlea glacial area, located on the northern slope of the Făgăraş massif, in the central glacial sector. Although this area is subject to a permanent snow avalanche activity and a potential risk of snow avalanches, it also benefits from significant winter tourist activities. For completing our analysis, we took into consideration the terrain factors and, by means of GIS Softwares, we drew up the appropriate maps underlying the elevation, slope and aspect. Based on the topographic map, satellite images, our observations and the information of members of Bâlea Lake Programme of Nivometeorology and members of Mountainers Rescuers, we identified 29 snow avalanche paths with different locations: 21 on the eastern slope, 5 on the western slope and 3 in the glacial cirque. Using the climatic variables we managed to characterize and define the snow avalanches winter regime. We also managed to set forth an altitudinal and morphological classification of the snow avalanche paths and to identify the most frequent snow avalanches. In this context, our study can be the basis for future studies involving other areas of the Făgăraş massif or from the Romanian Carpathians related to the zoning of the snow avalanches and a management of hazard and risk of snow avalanches. Résumé. Basé sur une étude de cas, notre travail représente une première tentative dans la géographie roumaine, d analyser le potentiel avalancheux d une région montagneuse. Il s agit de la région glaciaire Bâlea, située dans le Massif Făgăraş sur ubac dans son secteur glaciaire centrale. Bien que cette région est soumise pendant la saison froide à une activité avalancheuse très active, presque 7-8 mois/année elle bénéficie également des activités touristiques d'hiver très importantes. Pour mettre en évidence le potentiel avalancheux, nous avons envisagé tant les facteurs de terrain que les variables climatiques. En utilisant les programmes SIG nous avons fondé les cartes thématiques, c'est-à-dire la carte hypsométrique, la carte de la déclivité et la carte de l aspect. En même temps, en utilisant les cartes topographiques et les images satellitaires, comme nos observations sur le terrain et les informations obtenues par les membres du Programme de Nivométéorologie du Bâlea Lac, situé dans le cirque glaciaire Bâlea à 2040 m altitude et par les membres du Service Publique Salvamont, nous avons identifié 29 couloirs d'avalanche: 21 sur le versant de l est de la vallée glaciaire Bâlea, 5 sur le versant de l ouest et 3 dans le cirque glaciaire Bâlea. En utilisant les variables climatiques, nous avons défini et caractérisé le régime d'hiver des avalanches de neige et la fréqunce des avalanches de neige. Nous avons également crée une classification altitudinale et morphologique des couloirs d'avalanche, représentative pour le Massif Făgăraş et en même temps pour les Carpates Roumaines. Dans ce contexte, notre étude peut être utilisé dans les futures analyses sur le potentiel avalancheux du Massif Făgăraş ou des Carpates Roumaines, dans le zonage des avalanches de neige et dans la gestion des dangers et de risque des avalanches de neige. 1 Introduction Our study intends, for the first time in the Romanian research, to give an accurate evaluation of snow avalanche potential in the Bâlea area, which is one of the most representative glacial cirque and valley in the Southern Carpathians-Romanian Carpathians, and whose slopes are incised by snow avalanche paths, underlying thus one of the basic characteristics of this glacial area. Within this geomorphic context, our study is valuable based on two significant points of view. Bâlea glacial area is known for the Transfăgărăşan highway, built between , being located at the highest altitude in the Romanian Carpathians (2070 m) and which connects the northen slope with the southern slope. From October to June, the Transfăgărăşan highway is closed due to the severe climate and snow avalanche activity. On the other hand, the Bâlea glacial area holds a significant potential for mountain winter sport activities (alpine skiing, snowboarding, telemark skiing, skitouring, freeride and freestyle, climbing), and this is the place where, through the years, there was recorded the highest number of

2 human victims caught by snow avalanches. Therefore, it is imperative to fully understand the snow avalanche potential and to supervise the snow avalanche activity. For a better understanding of the snow avalanche potential and snow avalanche activity we should identify, based on the spatial and temporal criteria, the study site or the avalanche sites (ANCEY & CHARLIER 1996, ANCEY 1998, BUTLER 1979, SCHAERER 1967). The site with snow avalanches is represented by the smallest geographical unit where snow avalanches occur, irrespective of their size degree. Any such site is made up of the so-called subunits, also known as areas, which, later on, will be the basis of zoning the snow avalanches. We took into consideration this aspect, considering also the geographical location of the area, the overall morphology and the existence of 3 different sectors with snow avalanche activity: the western slope, the eastern slope and the glacial cirque. According to MCCLUNG & SCHAERER (1993) the recognition and identification of snow avalanche paths is possible taking into consideration certain procedures, such as using specialized maps, satellite images and observations collected in the field and from local observers, the assessment of the climate, the vegetation, the erosional material deposited by the snow avalanches, the snow avalanche landforms, and the runout distance. At the same time, the study of snow avalanche potential should also consider the analysis dedicated to the morphology of the snow avalanche paths and the morphometric characteristics and types of the starting, track and runout zones (BUTLER 1979, BUTLER & MALANSON 1985a, 1985b; BUTLER & MALANSON 1992, MCCLUNG & SCHAERER 1993, MCCLUNG 2001, SCHAERE 1967, SMITH & MCCLUNG 1997B, WEIR 2002). The main control factor of the snow avalanche potential, terrain factors and climatic variables are well known (CAMPBELL et al. 2007, MAGGIONI & GRUBER 2003, LUCKMAN 1992, MCCLUNG & SCHAERER 1993, PERLA & MARTINELLI 1976, SCHAERER 1977, SMITH & MCCLUNG 1997a, 1997b). In the same context, any study is incomplete without a description of these two factors (SMITH 1995, p. 29). On the other hand, an inter-connecting relationship takes place between the climate and the snow avalanche activity or snow avalanche occurrence (BIRKELAND 1997, BIRKELAND & MOCK 2001, JOMELLI et al. 2007, MOCK & BIRKELAND 2000) and which can be useful for snow avalanche prediction (JOMELLI et al. 2007). 2 Study area The Făgăraş massif is located in Southern Carpathians, at the intersection of the parallel of N and the E meridian (fig. 1): Fig. 1. Făgăraş massif location

3 The Făgăraş massif is approximately 1500 km 2 in area, and is like a huge ridge (70-80 km long) with an east-west orientation from which two slopes detach - the northern and the southern one. The Făgăraş massif is distinguished by the greatest mass and the highest altitudes in all of the Romanian Carpathians, notably with the following peaks: Moldoveanu (2544 m) and Negoiu (2535 m). They also show the most important inherited glacial relief in the Southern Carpathians and present-day periglacial processes. Bâlea glacial area is located in the central glacial sector of the Făgăraş massif, on its northern slope (see fig. 1). Between its highest peaks and the tree line, located at about m, the length of the valley, in the alpine and subalpine level, covers almost 2 km, and its overall length reaches even 6.9 km. The valley originates from a Bâlea glacial cirque. The cirque is flanked by peaks of over m height and which culminates, in the eastern part, in Vânătoarea lui Buteanu Peak (2507 m). After it overpasses the hummocky areas, the valley becomes wider, reaching, in certain sectors, even 2 km in width. After it passes by the Bâlea water fall, which itself represents a glacial threshold, the valley enters the forestry level, becoming a fluvial valley. Between the glacial peak level located above the Bâlea glacial cirque and tree line, the valley s both sides are dominated by higher and sharp peaks, cut up by numerous snow avalanche paths that underline the versants dynamics, giving a particular geomorphic authenticity to this area. Our area is characterized by the predominance of the metamorphic rocks, generally represented by micaschists, paragneisses and sericito-chlorite schists, and sometimes, crystalline limestone and amphibolites scattered along and which render a petrographic monotony to this entire area. 3 Materials and methods 3.1. Terrain factors The spatial distribution of the snow avalanches is mainly controlled by topographic factors, especially by slopes, altitude, concave relief shapes, rocks, snow avalanche paths and by the trees cover (ANCEY et al. 2004, BUTLER & MALANSON 1992, LUCKMAN 1977, 1978, 1992; SCHÖNENBERGER et al. 2005). In the same context, the slope incline allows snow avalanches to start and accelerate and the terrain factor is represented by orientation to wind and sun, forest cover, ground surface roughness, slope dimension and configuration and elevation (MCCLUNG & SCHAERER 1993, p. 491). At the same time, the snow avalanche paths are located in association with geological structure, rock types and stream channels (BUTLER 1979, BUTLER & WALSH 1990, WALSH et al. 1990). Using the 1: topographic map, 1:5000 orthophotoplans acquired in 2006 and the GIS Softwares (Idrisi Andes, ArcGIS 9.1), all of these were georeferenced with stereographic projection (Stereo 70 is the projection specific for romanian territory) and then we have digitize a vectorial layer of polygon type that shows the extension of snow avalanche paths. We have also used GPS points collected in the field at the edge of these avalanche paths (points that were accessible for us taking into account the roughness of topography) in order to realize accuracy assessment of the extension of these paths. Being aware of the fact that the snow avalanches follow well defined paths (BUTLER & MALANSON 1992, BUTLER 2001) we identified 29 snow avalanche paths, whose main characteristics are presented in table 1. In order to calculate these characteristics of the snow avalanche paths, a digital elevation model of the valley was used. The resolution of the cells was 10 m and was obtained by digitizing isolines from the topographical map 1: that we mentioned before. If we take into account the tectonic origin of the Făgăraş massif, as a fault-shaped scarp, we can consider that the snow avalanche paths are, firstly, of tectonic origin and secondly, in this respect, have a nival erosive corrosion origin (PEEV 1966). This process continues up to the late summer, and the snow deposited in the snow avalanche paths represents the basis of the active nivation processes and which, by the snowmelt process leads to the supply of the hydrological system (DE SCALLY 1992, 1996), as is the case at the Făgăraş massif. Considering the lithologic structure that should be taken into consideration while analyzing the snow avalanche paths (BUTLER 1979), and which, in this case,

4 proves to be a friable structure, the snow avalanche activity provides to the slopes, year by year, a significant spatial dynamic. Table 1 Features of snow avalanche paths Number of snow avalanche paths Altitude (m) Surface (km 2 ) Average Slope ( o ) Sinuosity index Max. Min. Vertical drop (starting zone) (runout zone) This is underlined by the debris transfer on steep slopes or at the base of the snow avalanche paths (BELL et al. 1990, GARDNER 1970, LUCKMAN 1977, 1978, 1988, 1992, PEEV 1966). The removal of both the material and the snow from the snow avalanche paths is also facilitated by the value of the sinuosity index, calculated as a ratio between the line along which the snow is moving and the straight length of the snow avalanche paths, within the snow avalanche paths. Basically, the sinuosity index of the snow avalanche paths reaches values very close to 1. Basically, the sinuosity index of the snow avalanche paths reaches values very close to 1 and this is the explanation for their longitudinal profile and the rapid removal of the erosion material from the path, especially on the eastern slope and in Bâlea glacial cirque. Several snow avalanche paths are characterized by a smaller index (5, 7, 12, 16, 21, 22), being thus frequently met on the western slope. The disposition of the snow avalanche paths in Bâlea glacial area is quite different. Within the glacial cirque we found 3 snow avalanche paths, located in the alpine level. On the eastern slope we found 21 snow avalanche paths, from which 4 show a starting zone in alpine level and runout zone in subalpine level, 5 have their starting zone in subalpine level and runout zone in forestry level and 12 are entirely placed in the subalpine level. The western slope underlines 5 snow avalanche paths whose starting zone is located in the alpine level and runout zone in subalpine level (fig. 2). To highlight the characteristics of the terrain factors (MCCLUNG 2001, SCHAERER 1977) in the Bâlea glacial area, we made thematic maps (hypsometry, slopes, aspect) using GIS Software. The large vertical relief of the Bâlea glacial area is covered by several snow avalanche paths. The hypsometric map (VOICULESCU, 2009a) analysis outlines high elevation in this area, with values over m (fig. 3). It is easy to remark that maximum altitude is 2507 m, and minimum over 660 m. The mean value of the study area is m. Our study area spreads along the alpine level, subalpine level and upper part of forestry level.

5 Fig. 2. Location of the starting and runout zones of snow avalanches Fig. 3. Bâlea glacial area - hypsometric map (Voiculescu, 2009a) Snow avalanche paths of Bâlea glacial area have variable lengths: 2 are less than 200 m, located in the Bâlea glacial cirque, 16 have lengths between m, all of them being located on the western slope, 2 have lengths between m, 5 have lengths between m and 4 are characterized by lengths exceeding 600 m. The last two classes of snow avalanche paths are located both on the western slope near the tree line and on the upper sector of the forestry level but on the eastern slope in the upper and median sector of the Bâlea glacial valley. However, the longest snow avalanche paths lay on the largest areas of spatial manifestation.

6 The declivity map (VOICULESCU, 2009a) points out the great high degree of declivity of the study area (fig. 4). All values between 1 o -15 o represent 6.9% (3.4 km 2 ) from the surface of Bâlea glacial area. The values between 15 o -25 o represent 18.1% (9 km 2 ). The values between 25 o -35 o represent 11.1% (5.52 km 2 ), the values between 25 o -35 o represent 33.2% (16.4 km 2 ), the values between 35 o -45 o represent 31.8% (15.8 km 2 ) and the values above 45 o represent 9.5% (4.7 km 2 ) from the surface of Bâlea glacial area. All values show the very good potential for snow avalanche activity. Fig. 4. Bâlea glacial area - declivity map (Voiculescu, 2009a) The slope plays an important role in snow avalanche activity and is the primary variable in avalanche terrain (MAGGIONI & GRUBER 2003, LUCKMAN 1977, 1978, MCCLUNG & SCHAERER 1993). Optimal slopes are between 25 o and 50 o (ANCEY 2001, ARMSTRONG et al. 1994, EMBLETON 1979, MARTINELLI 1974, MCCLUNG and SCHAERER 1993, TREMPER 2001, SCHAERER 1977). Snow layer thickness also contributes to snow avalanche occurrences and so the following categories have been established for corresponding slope degrees and snow thickness: 50 o for 5 cm of snow; 30 o for 15 cm of snow; 22 o for 50 cm of snow (PISSART 1987). Considering that the optimal slopes of snow avalanche manifestation are between especially between (CIOLLI et al. 1998, MCCLUNG & SCHAERER 1993, MARTINELLI 1974, LUCKMAN 1977), we found that all snow avalanche paths comply with these values. The average slope is between 27.3 o and 42.6 o, indicating a high potential of snow avalanches activity. The average values are recorded on the eastern slope of the valley, where 6 snow avalanche paths exceed 40 o. The snow avalanche paths located on the western slope show close average values, between 32.3 o and 35.8 o. According to EMBLETON (1979), at these slope values, snow avalanches are frequent and according to ARMSTRONG et al. (1994), at these slope values snow avalanches have a high to

7 moderate strength. This is necessary to mention if we take into account the fact that in this area snow avalanches are not surveyed for frequency and strength. The orientation of slopes to the sun plays a strong influence on the stability of snowpack (ANCEY 2001). The radiation of the sun controls snow surface temperature more than air temperature (TREMPER 2001) and plays a very important role affecting snow instability (MCCLUNG & SCHAERER 1993) and determining the snow avalanches type. Taking into account that the snow avalanche was produced in spring, the temperature increase enhances stability of snowpacks on shady slopes and instability on sunny slopes (ANCEY 2001, pp. 3). The aspect map (VOICULESCU, 2009a) shows that most of the areas are covered by the western slopes by 2.80 km 2 (29.4%) and by the north-eastern slopes, by 2.04 km 2 (21.5%) (fig. 5). Higher values are recorded also in the eastern slopes with 1.22 km 2 (12.9%) and the north-western slopes with 1.40 km 2 (14.7%). The lowest values are recorded in the south-west oriented slopes with 0.64 km 2 (6.8%), south-east oriented slopes with 0.19 km 2 (2.03%) and southern slopes with 0.14 km 2 (1.57%). Analyzing the map aspect and according to BUTLER (1979) we found that 13 snow avalanche paths have a north-eastern disposition, 9 are oriented towards east, 4 are oriented towards west, 2 are placed towards north-west and 1 with a northern disposition (fig. 6). N 15 NW 10 NE 5 W 0 E SW SE S Fig. 5. Bâlea glacial area - aspect map (Voiculescu, 2009a) Fig. 6. Star diagram illustrating the number of snow avalanche paths and their corresponding slope aspects 3.2. Climatic variables The relation between the climate and the snow avalanche activity has been studied by several authors, especially for North America (ARMSTRONG & ARMSTRONG 1987, BIRKELAND 1997, BIRKELAND & MOCK 2001, BUTLER 1986, FITZHARRIS 1987, GERMAIN et al. 2009, HÄGELI & MCCLUNG 2003, 2004, LACHAPELLE 1966, MCCLUNG & SCHAERER 1993, MOCK & KAY 1992, MOCK 1996, MOCK & BIRKELAND 2000). However, for the mountains of European, the scientific preoccupations focused upon the classification of the snow climates (BJÖRNSSON 1980; FITZHARRIS & BAKKEHØi 2007, HÖLLER 2009, LATERNSER & SCHNEEBELI 2002, 2003).

8 Although the climatic variables give us indirect evidence regarding the snow stability, LACHAPELLE (1980) has set forth three data variables, like meteorology, snow structure and snow mechanics which can be held responsible for the occurrence of the snow avalanches while MCCLUNG & TWEEDY (1994) have divided the climatic variables as follows: the stability factors, the snowpack parameters and the meteorological parameters. Romania is located in the temperate - continental climate, which, together with the maritime and transitional climate could be defined by significant snowfalls and snow avalanches (BIRKELAND & MOCK 2001, HÄGELI & MCCLUNG 2004, MCCLUNG & SCHAERER, 1993). Depending on its geographical position, 7 types of climatic influences have been identified in Romania. Among these, two such climatic influences act upon the Făgăraş massif: the northen slope where Bâlea glacial area is located, is under the influence of the western, marine and wet air masses, while the southern slope is under the influence of warm air from the Mediterranean Sea. Therefore, the regional climate is determined by the manifestation of the snow avalanches (SMITH 1995), being strongly influenced by the following factors: solar radiation, temperature, snowfall intensities and the periods of melt (LUCKMAN 1978, MCCLUNG & SCHAERER, 1993, ZINGG 1966, WEIR 2002). The air temperature varies depending on the altitude, with a powerful impact upon the type of snow, and also affecting the type of avalanche. We have used the meteorological data collected from three meteorologic stations in order to complete the synthetic definition of the altitudinal climate of the Făgăraş massif (table 2): Table 2 Climatic characteristics of the Făgăraş massif (average annual values) Meteo station (m) Lat. N. Long. E. T o C Ann. Min. Max. Pp (mm) Air humidity (%) Days with snow Days with snow cover Depth of snow (cm) Sunny days while there is snow cover Vf. Omu ,9 5, > Bâlea Lake ,4 8,8 1246,2 83 > 96 > Cozia ,3 12,3 844,2 - > The meteorogical stations which provided such data are the following: Vf. Omu, positioned near the Făgăraş massif, in the Bucegi Mountains, located at 2505 m altitude, being representative for the highest alpine peaks; Bâlea Lake, located at about 2070 m altitude, on the northern slope of the Bâlea glacial cirque and Cozia, positioned on the Southern slope, at 1577 m altitude, close to the tree line. According to the thermic perspective, the alpine level, between 2200 (2300) m - and the highest peaks - is characterized by a cold climate (<-2 o C) while the subalpine level, between tree line ( m) and 2200 (2300) m is characterized by a rather chilly climate (between 2 C and -0 o C). In order to define the pluviometric particularity, we calculated the nival coefficient, expressed as a ratio between the quantity of precipitation fallen as snow and the overall quantity of precipitations. Therefore, the alpine level underlines a moderate nival characteristic, with the nival coefficient exceeding 60%, while the subalpine level is characterized as a nivo-pluvial feature, according to the nival coefficient which fluctuates between 50% and 60%. The snows represent a phenomenon which characterizes the Făgăraş massif. The snows can be classified according to their frequency, period and quantity, reaching their highest values between January (February) and April (May). However, in the Făgăraş massif as well as in other worldwide mountain areas, the heavy snowfalls are the bases of the disastrous avalanches (LATERNSER & SCHNEEBELI, 2003). The main causes for the frequent occurrence of the snowfalls in the Făgăraş massif are as follows: the movement of the polar air masses, reduction of the temperature below 0 C and, on the other hand, the local relief features, such as the altitude, the general orientation of the mountains, and the specific site exposure. The snow cover is generated by snowfalls, expressed as snowfall days. We can talk about the good correlation between the snowfalls and the altitude (fig. 7). The snowfall days fluctuate on an altitudinal-basis, between tree line and the highest peaks of the study area, from 190 days to 270 days. The fluctuation of this coefficient (fig. 8) results in both the manifestation and the frequency of the snow avalanches whose frequency is higher as we go upward towards the peaks.

9 Alt (m) y = x R 2 = Cozia Vf. Omu Bâlea Lake Alt (m) y = x R 2 = Cozia Vf. Omu Bâlea Lake snowfalls (days) snow cover (days) Fig. 7. Relationschip between altitude and snowfalls days Fig. 8. Bâlea glacial area - map of number of days with snow falls According to EGLI (2008) and HÖLLER (2009) we used, for our study, the thickness of the snow cover. We process all values as the average monthly decade (table 3): Table 3 The snow depth, average monthly decade values Meteo st. (m) J F M A M J J A S O N D Year Vf. Omu Bâlea Lake Cozia The thickness of the snow cover fluctuates both on a monthly and a decadal basis, depending on the altitude; it reaches the highest values between February and March, for the lowest altitudes, and for the higher altitudes, the maximum values are reached from January to April (May) (fig. 9): snow depth (cm) A S O N D J F M A M J J month Vf. Omu Bâlea Lake Cozia Fig. 9. Snow depth variation Here, we have to note that values, rather small, of the snow cover recorded for the highest altitudes within the area of the Vf. Omu meteorological facility are owed to the fact that this facility is located on an open-area, being thus affected by the continental climatic influences and by the strong winnds that hinder the snow accumulation. The snow avalanche activity has been well monitored by members of Work Laboratory of the Programme of Nivometeorology, which was set within the National Administration of Meteorology R.A. (NAM) in partnership with Météo France, Centre d Études de la Neige-Grenoble and by members of Bâlea Lake Mountainers Rescuers. The main purpose of the programme is to study snow and its future evolution, as well as avalanche triggering conditions (ADMINISTRAŢIA NAŢIONALĂ DE METEOROLOGIE, R.A ). The snow avalanche activity has noticeably grown, and when snow avalanches triggering conditions are met, snowpack thickness is regarded as a causal variable in

10 snow avalanche production (BIRKELAND & MOCK 2001; Fitzharris 1987). Taking into account the fact that the winter of , respectively the winter of was distinguished by massive snowfalls, which led to an increased snow avalanche activity. A good correlation exists between snow layer thickness and the frequency of snow avalanches (fig. 10): Fig. 10. The variation in snow depth and the frequency of snow avalanches in Bâlea glacial area (on the left) and correlation between the snow depth and snow avalanche frequency in Bâlea glacial area (on the righ), in winter (on the top) and in winter (on the botom) 4. Discussion The perspective of snow avalanche-hazard zoning of Bâlea glacial area is very important, because, as might be expected, the human activity by tourist winter practices has increased in this mountain area. Therefore, considering the terrain factors, the starting and runout zones, comparing of snow avalanche paths, the climate variables and our experience (FREER & SCHAERER (1980) we classified the snow avalanche paths. From the altitudinal (fig. 11) and morphological point of vue and according to LUCKMAN (1977, 1978), MARTINELLI (1974) and MCCLUNG & SCHAERER (1993)) we identified 4 major categories. According to the geographical classification of VANNI (1966), there are two types of track and snow avalanches in the Bâlea glacial area: medium high-mountain and valley floor avalanches with a local character. In the first case, 6 snow avalanche paths that occur within the glacial cirque at altitudes higher than m and 14 snow avalanche paths in the upper part of the Bâlea glacial valley, at altitudes between (2300) m. In the second case, 9 snow avalanche paths occur in the lower part of the glacial valley under the altitude of 2000 m and at the contact between the alpine level where the tree line is delineated at m. On the other hand, according to statistical analysis of altitude of starting and runout zones (BOVIS & MEARS (1976), BUTLER (1979) and BUTLER et al. (1992), 17 snow avalanche paths are located in the alpine level, 7 snow avalanche paths are located in subalpine level and 5 snow avalanche paths, on the eastern slope of Bâlea glacial valley are located in the upper part of the forestry level.

11 Fig. 11. Bâlea glacial area - altitudinal classification of snow avalanche paths A first category of snow avalanche paths is represented by those which are characterized by a starting and a runout zones in alpine level, being dominated by a rocky ground and sometime by meadows. This is the case of the snow avalanche paths between number 22 and 24, located on the walls of the Bâlea glacial cirque. The leeside slopes, which accumulate snow by drifting, or cornice development, located especially above the Bâlea glacial cirque but also between the snow avalanche paths number 22 and 24. In this morphological instance, the snow avalanches become wider in the second sector of the versant (LUCKMAN 1978). This is the case of the snow avalanche path number 24, where, the huge snow avalanche, known in the Romanian Carpathians occurred on April 17 th 1977, killed a group of 23 skiers (VOICULESCU 2009b). Also 4 other peoples were killed in snow avalanche path number 22 (fig. 12). The second category is represented by the snow avalanche paths which have their starting zone in the alpine level and the runout zone in the subalpine level, such as the case of the snow avalanche paths between 19 and 21 and between 25 and 29, with an evident meadow - type grassy sector or a particular vegetation cover, represented by Salix reticulate, Salix herbacea, Pinus mugo, Juniperus nana, Vaccinium myrtillus and Rhododendron kotschyi. The cliff sites with small gullies, where the snow can accumulate, and with cornice development. Considering their general morphology as well as their relatively small length, the snow avalanche activity is represented by small avalanches (LUCKMAN 1978), but killed over time 8 peoples (6 in snow avalanche path number 21, one person in snow avalanche number 6 and one person in snow avalanche path number 39) and buried seven peoples in snow avalanche number 21 (fig. 13). The cliff sites with large gully in the upper part of the slope and open slope in the lower part of the slope, characteristic for the western slope of Bâlea glacial valley. In

12 this case, the snow avalanches are significant, due to the the length of the snow avalanche paths, compared to those on the eastern slope of the Bâlea glacial valley. Fig. 12. Snoew avalanche path number 24. One can see the directions of huge snow avalanche on April 17 th 1977 Fig. 13. Snow avalanche path number 21. One can see the deflecting dike (on the left), snowpack support (on the right) and the Trasfăgărăşan highway, affected each year by snow avalanches The third category is represented by the snow avalanche paths with both the starting zone and the runout zone in the subalpine level, such as the case of the snow avalanche paths between 6 and 18 with small gullies where the snow can accumulate, and with cornice development also. The fourth category of snow avalanche paths are those having their starting zone in the subalpine level and their runout zone in the forestry zone, whose vegetation is basically represented by the Picea alba, in the case of the snow avalanche paths between 1 and 5 (fig. 14). The cliff sites are characterized by small gullies, sometimes deep in the upper half, where the snow can accumulate, and with cornice development. All these snow avalanche paths stand for a particular feature of the area Bâlea glacial valley, that force shaping ecosystems in mountain forests (RIXEN et al. 2007), representing an important disturbance agent of tree line (WALSH et al. 1994, WALSH & WEISS 2004, WEIR 2002) and severely damaged vegetation along their tracks (DUBÉ et al. 2004). Based on the data collected in the field, the analyses of the topographical maps and the satellite images related to the different types of the starting, track and runout zones and according to BUTLER (1979), we completed a comparative analysis of the snow avalanche paths, according to FREER & SCHAERER (1980). We noticed the predominance of the bowl-shaped and coalescing types in the starting zones, the preponderance of the narrow and neck-shaped and rectilinear types in the track zones and the spatulate, tongue-shaped and digital-shaped types in the runout zones (table 4). From the run-off direction point of view, the snow avalanche paths from the Bâlea glacial area are confined (BURROW & BURROW 1976).

13 Fig. 14. Snow avalanche paths on the left slope of Bâlea glacial valley with starting zone in the subalpine level and their runout zone in the forestry zone Table 4 Types of starting zones, track zones and runout zones Type Starting zone type Number Track zone type Number Runout zone type Number Bowl-shapped and coalescing 21 Narrow, neck-shaped 14 Tongue-shaped 12 Rectilinear 8 Rectilinear 15 Spatulate 15 Digitate 2 We should also mention that we have found several unconfined snow avalanche paths (DE QUERVAIN 1966, QUINN & PHILLIPS 2000) or open slopes (LUCKMAN 1977, 1978; MCCLUNG & SCHAERER 1993; MARTINELLI 1974, SCHAERER 1972). Basically, these have a scarped slope, varying from 30 to 45, the vegetation is hardly present here, e.g. between the snow avalanche paths number 21 and 22, between number 23 and 24 and between number 24 and 25, located on the walls of the Bâlea glacial cirque. The general slope of the starting zone fluctuates between 35 o and 45 o and also between 45 o and 55 o, extensively triggering minor but frequent slab snow avalanches (MCCLUNG & SCHAERER 1993). This situation characterizes in particular the eastern slope of the Bâlea glacial valley. On the western slope, where the starting zone appeared as bowl-shaped with slope values varying from 25 o to 35 o, the slab snow avalanches are often wider and the frequency of the wet snow avalanches is quite diminished (MCCLUNG & SCHAERER 1993). The track zones are located both in the alpine level (snow avalanche paths between number 21-25) but also in the subalpine level (most of them). The snow avalanches paths between number 1 and 5 have their track zones fully located in the forest level, generating thus the altitudinal zoning of the forest (SUFFLING 1993). The track zones can be represented by the sequence of the restrained areas, such as the gulches or the channels in gullies (BUTLER & WALSH 1990) with slope areas, with the occurence of certain rocky bars or necks, as it is the case of the snow avalanche paths on the eastern slope. In the track zone, the snow avalanche reaches its maximum speed (MCCLUNG & SCHAERER 1993, SMITH 1995) generating a major erosion which explains the location of the erosion material, the debris as well as the vegetative material, such as the trees, bushes, trunks, and branches (MCCLUNG & SCHAERER 1993).

14 The runout zone has a variable width (BUTLER & Walsh 1990) being located either in the alpine level (snow avalanche paths between number 21 and 25), or in the subalpine level (snow avalanche paths between number 6 and 21 and 25 and 29) or even in the forest level (snow avalanche paths number 1 and 5). Basically, this zone is marked by a reduced value of its slope, between 8 o (10 o ) and 20 o and by the implicitly diminished speed of the avalanche. The runout zone has a concave longitudinal profile and a convex transversal profile (BUTLER &WALSH 1990), indicating either the snow avalanche sediment deposits (HUBER 1982) or the debris accumulation (MCCLUNG & SCHAERER 1993; SMITH 1995), a particularly underlined phenomenon especially for the flowing avalanches. At the same time, a mixture of trees, bushes, tree trunks, and branches, depending on the avalanche magnitude, is also deposited (WALSH et al. 2004) (fig. 15): Fig. 15. The mixture deposited in runout zone of snow avalanche path number 27 As for the dry avalanches, the deposit corresponds to a sedimentation of the snow cover (ANCEY 1998), spreading along much further against the place imposed by the slope, on the opposite slope. Considering the lithologic structure (BUTLER 1979), the snow avalanche paths which are quite friable in this case, the snow avalanche activity generates significant quantities of debris and rock debris which is transferred on steep slopes or at the base of the snow avalanche paths (BELL et al. 1990, LUCKMAN 1977, 1978, 1988, 1992, PEEV 1966). Making a correlation between the snow avalanche paths and the characteristics of the landforms, according to BUTLER (1979) and according to the shapes studied and described by several authors (GARDNER 1970, 1983, GRAY 1973, JOMELLI 1999, LUCKMAN 1970, 1971, 1977, 1978, 1988, 1992, 2007, LUCKMAN et al. 1994, MATTHEWS & MCCARROLL 1994, PEEV 1966, SMITH et al. 1994, STOFFEL et al. 2006) but also according to our observations, we discovered the following situation (table 5): Table 5 Landforms associated with snow avalanche paths in Bâlea glacial area Landforms Number of snow avalanche path Avalanche boulder tongue - Avalanche cone 2 (18 and 19) Alluvial fan at base of snow avalanche path - Slushflow levee - Lake at base of snow avalanche path 1 (24) Debris avalanche within confines of snow avalanche paths 10 (between number 12-21) Stream channel within confines of snow avalanche paths 19 (between number 6-19 and between number 25-29) Protalus rampart at base of avalanche chute 1 (24) Rock glacier associated with snow avalanche path 1 (22)

15 6. Conclusions By this study, we attempted to prove the snow avalanche potential in one of the most representative glacial area in the Făgăraş massif. Such potential is also helped by the inherited glacial morphology, the global tectonics and by certain specific climatic conditions. The surface of snow avalanche paths represents almost 23% of the overall surface of the Bâlea area. It is important to notice that, from the 29 snow avalanche paths with high risk of occurrence, 24 snow avalanche paths (82.7%), are spread among the highest altitudes of the area and the tree line, an aspect that amplifies its geomorphic importance for triggering the snow avalanches (GARDNER 1970, LUCKMAN 1977, 1978) in Bâlea glacial valley. The eastern slope with the 21 snow avalanche paths has the highest potential of the snow avalanche activity, although their morphogenetic characteristics are reduced compared to those of the snow avalanche paths on the western slope or from the glacial cirque. This aspect is underlined by a high number of fatalities that have occurred in this glacial valley, being also acknowledged by both the statistics of the Bâlea Lake Mountain Rescuers and by the evidences of the activity carried out for supervising the snow avalanches from the Bâlea glacial area (ADMINISTRAŢIA NAŢIONALĂ DE METEOROLOGIE , , , ). Therefore, at the base of the eastern slope where, the summer tourist markings is overlapped by both the winter tourist signs and the ski path, there were registered, through the years, 8 fatalities. In Bâlea glacial cirque, 27 fatalities were recorded, from which 23 were caused by snow avalanche path 24. Although, the snow avalanche paths on the western slope of the Bâlea glacial valley have morphometric characteristics considered to be superior compared to all the other snow avalanche paths, no fatalities were recorded here. The major landforms produced by snow avalanches have both an erosion-based and deposit-based source (GARDNER 1970, LUCKMAN 1971, 1977, 1978). In the first case, the snow avalanches act upon the snow avalanche paths, by modifying them and permanently generating the debris or rock debris. In the second case, the snow avalanches bring their contribution to the generation of particular snow avalanche landforms like debris-covered slopes, moraines, talus in lower part of alpine level and above tree line in subalpine level. These processes are also helped by the friable geological structure, by the severe climatic conditions, by the existence of 2 periods of freezing cycles and by the large snow quantities that last for over months per year. These have a large content of debris or rock debris which is transferred and redistributed at the same time with the snow mass (GARDNER 1970, LUCKMAN 1977, 1978), failing thus to generate clean snow avalanches (Rapp 1960). Because the snow avalanche activity is under the incidence of both the regional climatic conditions and the local factors, the effects of the snow avalanches must be reconsidered in certain safety projects to prevent and mitigate snow avalanche hazards (GERMAIN et al. 2009). And that is why we consider that a good understanding of the snow avalanche activity would represent an important step in zoning of snow avalanches (Freer and Schaerer, 1980), in hazard zoning (HÖLLER 2007, KEILEr 2004) and for the optimum management of the risks of snow avalanches (BRÜNDL et al. 2004, CIOLLI et al. 1998, FUCHS et al. 2004, JAMIESON & STETHEM 2002, LATERNSER & SCHNEEBELI 2002, STETHEM et al. 2003, WEIR 2002). We believe that the recent Bâlea Lake Programme of Nivometeorology will represent a step forward in the supervision of the snow avalanche activity being also the basis of the snow avalanche hazard zoning. Acknowledgements The present study data are part of the projects PNCD-I2 1066/2009, Creation of the data base and thematic maps for the ski areas in the Southern Carpathians using GIS. Analysis, up to date evaluation and prognosis within the global climatic changes perspective, supported by National University Research Council (CNCSIS), Romania and PNCD-I2 1075/2009, Methods for digital terrain analysis and automatic classification of the relief in the mountain area based on digital terrain models and remote-sensed data, supported by National University Research Council (CNCSIS), Romania.

16 The authors want to express our gratitude to Dr. David Butler (Department of Geography, Texas State University-San Marcos, Texas), to Dr. Brian Luckman (University of Western Ontario, Canada) and to Dr. Stephen Walsh (Department of Geography, University of North Carolina, Chapel Hill, North Carolina) for support, assistance and supply of very important articles, not available in Romania. We would like to thank to members of Bâlea Lake Programme of Nivometeorology and to the Mountain Rescuers for the information and observations on site. References ADMINISTRAŢIA NAŢIONALĂ DE METEOROLOGIE, : Bilanţul nivologic, sezonul nivologic, Secţia de Meteorologie Dinamică, Climatologie şi Agrometeorologie, Grupul de Verificare a Prognozelor şi Adaptare Statistică (GVPAS), Bucureşti, 147 pp. ADMINISTRAŢIA NAŢIONALĂ DE METEOROLOGIE, : Bilanţul nivologic, sezonul nivologic, Secţia de Meteorologie Dinamică, Climatologie şi Agrometeorologie, Grupul de Verificare a Prognozelor şi Adaptare Statistică, Colectivul de Nivometeorologie, 132 pp. ADMINISTRAŢIA NAŢIONALĂ DE METEOROLOGIE, : Bilanţul nivologic, al sezonului de iarnă, Laboratorul de Tehnici pentru Prognoza Fenomenelor Meteorologice Severe, Bucureşti, 120 pp. ADMINISTRAŢIA NAŢIONALĂ DE METEOROLOGIE, : Bilanţul nivologic, al sezonului de iarnă, Laboratorul de Tehnici pentru Prognoza Fenomenelor Meteorologice Severe, Bucureşti, Colectivul de Nivometeorologie, Centrul Meteorologic Regional Transilvania-Sud, Serviciul Regional de Prognoză a Vremii Sibiu, 178 pp. Alix, A. (1925): Les avalanches. Revue de Géographie Alpine, XIII: Ancey, C. & Charlier, C. (1996): Quelques réflexions autour d une classification des avalanches. Revue de Géographie Alpine, 1: Ancey, C. (1998): Guide, neige et avalanche. Connaissances, Practiques & Sécurité, 3 ème édition, École Polytechnique Fédérale de Lausanne, 281 pp. Ancey, C. (2001): Snow avalanches. - In: Balmforth, N. & Provenzalle, A. (eds.), Geomorphological Fluid Mechanics: selected topics in geological and geomorphological guid mechanics, Berlin: Springer Ancey, C., Gervasoni, C. Meunier, M. (2004): Computing extreme avalanches. Cold Regions Science and Technology, 39: Armstrong, R., Armstrong. B. (1987): Snow and avalanche climates in the western United States: a comparison of maritime, intermountain and continental conditions. Int Assoc Hydrol Sci Publ, 162: Armstrong, R.B., Williams, K., Armstrong, R.L. (1994): The avalanche book. Fulcrum Publishing, Rev & Updted Edition, 240 pp. Bader, H.P., Salm, B. (1990): On the mechanics of snow slab release. Cold Regions Science and Technology, 17: Besancenot, J.P. (1990): Climat et tourism, MASSON, Collection Géographie, 223 pp. Bell, I., Gardner, J., Scally, de F., 1990: An estimate of snow avalanche debris transport Kaghan Valley, Himalaya, Pakistan. Arctic and Alpine Research, 22 (3): Birkeland, W.K. (1997): Spatial land temporal variations in snow stability and snowpack conditions throughout the Bridger Mountains, Montana, A dissertation presented in partial Fulfillment of the Requirements for the Degree Doctor of Philosophy, Arizona State University, 206 pp. Birkeland, W.K., Mock, C.J. (2001): The Major Snow Avalanche Cycle of February 1986 in the Western United States. Natural Hazards, 24: Björnsson, H. (1980): Avalanche activity in Iceland, climatic conditions, and terrain features. Journal of Glaciology, 26 (94):

17 Bovis, M.J. & Mears, A.I. (1976): Statiscal prediction of snow avalanche runout from terrain variables in Colorado. Arctic and Alpine Research, 8: Bründl, M., Etter, J.-H., Steiniger, M., Klingler, Ch., Rhyner, J. & Ammann, J.W. (2004): IFKIS - a basis for managing avalanche risk in settlements and on roads in Switzerland. Natural Hazards and Earth System Sciences, 4: Burrows, C.J. and Burrows, V.L. (1976): Procedure for the study of snow avalanche chronology using growth layers of woody plants. University of Colorado, Institute of Arctic and Alpine, Research Occasional Paper, 23: 54. Butler, D.R. (1979): Snow avalanche path terrain and vegetation, Glacier National Park, Montana. Arctic and Alpine Research, 11 (1): Butler, D.R., Malanson, G.P. (1985a): A reconstitution of snow avalanche characteristics in Montana, U.S.A., using vegetative indicators. Journal of Glaciology, 31 (108): Butler, D.R., Malanson, G.P. (1985b): Effects of Terrain on Excessive Distance by Snow Avalanches. Northwest Science, 66 (2): Butler, D.R. (1986): Snow avalanche hazards in Glacier National Park, Montana: Meteorologic and climatologic aspects. Physical Geography, 7 (1): Butler, D.R. & Walsh, S.J. (1990): Lithologic, structural, and topographic controls of snow-avalanche path location, eastern Glacier National Park, Montana. Annals, Association of American Geographers, 80 (3): Butler, D.R. & Malanson, G.P. (1992): Effects of Terrain on Excessive Travel Distance by Snow Avalanches. Nothwest Science, 6 (2): Butler, D.R., Malanson, G.P. & Walsh, S.J. (1992): Snow-avalanche paths: conduits from the periglacial-alpine to the subalpine-depositional zone. In Dixon, J.C. and Abrahams, A.D. (editors), Periglacial geomorphology, London: Wiley, Butler, D.R. (2001): Geomorphic process-disturbance corridors: a variation on a principle of landscape ecology. Progress in Physical Geography, 25 (2): Campbell, C., Bakermans, L. Jamieson, B., Stethem, B. (2007): Current and future snow avalanche threats and mitigation measures in Canada, Prepared for: Public Safety Canada, Canadian Avalanche Centre, 109 pp. Capello, C.F. (1973): Il problema delle valanghe. Bolletino Societa Geografica Italiana, Suppl., II (10): Ciolli, M., Tabarelli, S., Zatelli, P. (1998): 3D Spatial data integration for avalanche risk management. ISPRS IV, Symposium on GIS-Between Visions and Applications, Stuttgart, Germany, 32 (4): Daffern, T. (1992): Avalanche safety for skiers and climbers. Rocky Mountain Books, Calgary, 192 pp. De Scally, F.A. (1992): Influence of snow avalanche snow transport on snowmelt runoff. Journal of Hydrology, 137: De Scally, F.A. (1996): Avalanche Snow Melting and Summer Streaflow Differences between Highelevation Basins, Cascade Mountains, British Columbia, Canada. Arctic and Alpine Research, 28 (1): Decaulne, A. (2007): Snow-avalanche and debris-flow hazards in the fjords of north-western Iceland, mitigation and prevention. Natural Hazards, 41: Decaulne, A., Sæmundsson, T., Petursson, O. (2007): Debris flow triggered by rapid snowmelt: a case study in the Gleiđarhjalli Area, Northwestern Iceland. Geogr. Ann., 87A (4): Dubé, S., Filion, L., Hétu, B. (2004): Tree-ring reconstruction of high-magnitude snow avalanches in the Northern Gaspé Peninsula, Québec, Canada. Arctic, Antarctic, and Alpine Research, 36 (4): Embleton, C. (1979): Nival processes. Process in Geomorphology, in Edward Arnold editor, London, Egli, L. (2008): Spatial variability of new snow amounts derived from a dense network of Alpine automatic stations. Annals of Glaciology, 49: