Réf. Schöner & al. 2009 - A

Référence bibliographique complète

SCHÖNER W., AUER, I., BÖHM, R. 2009. Long term trend of snow depth at Sonnblick (Austrian Alps) and its relation to climate change. Hydrological Processes, 23, 1052-1063.

Abstract: The extensive snow measurement network of the Sonnblick region (Hohe Tauern, Austrian Alps) is used to describe temporal trends of snow-depth as well as its relation to climate change for a high-elevated site of the European Alps (2400–3100 m.a.s.l.). Spatial representativeness of single snow stakes, with readings back to 1928, is derived for maximum snow-depth in May using a spatially dense snow depth probing from glacier mass balance measurements. Long-term trends of snow depth show a significant reduction in the contribution of snow accumulation from core-winter (1 December to 1 March) compared to early and late-winter periods. Largest values of snow-depth since 1928 were measured in the 1940s and 1950s. Comparison of monthly changes in snow-depth with precipitation measurements underlines the high influence of wind drift on snow-depth during winter season from 1 October to 30 April. Whereas inter-annual variability of maximum snow-depth is better explained by low elevation precipitation measurements than by local (high elevation) precipitation measurements, the longer-term mean of local precipitation measurements, however, fits well to the snow-depth measurements, if a mean snow-density of about 400 kg m−3 is assumed (which matches field observations). Both maximum snow-depth and winter season precipitation show a clear decreasing trend for inter-annual variability. A statistical relationship between air temperature and fraction of solid precipitation is used for estimation of temporal trends in the fraction of solid precipitation at measurement sites. For summer a decrease of about 1% of solid precipitation per decade was found for the lowest elevated sites whereas fraction of solid precipitation in winter remains stable. Relation between snow-depth and climate is investigated by means of local climate data of Sonnblick-Observatory (SBO) and by means of the North-Atlantic Oscillation Index (NAOI). Whereas winter air-temperature is significantly correlated with the NAOI, for winter precipitation and snow depth on 1 May no correlation was found with NAOI.


Snow - Climate variability - Trends - Climate change - Sonnblick - Alps - Snow precipitation - Alpine precipitation - Austria


Organismes / Contact

• Central Institute of Meteorology and Geodynamics, Hohe Warte, 38 A-1190-Vienna, Austria. E-mail: w.schoener@zamg.ac.at

The study has been partially performed as part of the ALP-IMP project supported by European Commission (EVK2-CT-2002-00148).


(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






Pays / Zone

Massif / Secteur

Site(s) d'étude



Période(s) d'observation

Austrian Alps

Hohe Tauern

Sonnblick-Observatory (SBO) region


2400–3100 m



(1) - Modifications des paramètres atmosphériques










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



(2) - Effets du changement climatique sur le milieu naturel





Snow depth belongs to the climate elements with most pronounced differences between high alpine and low elevation sites. Not only the amount of snow is different but also the annual course shows systematic shift from low elevation to high elevation sites. The annual course of snow depth is shown for all snow stakes at Goldbergkees and Kleinfleißkees. It appears from the figures that maximum snow depth is measured in May for Goldbergkees and May to June for Kleinfleißkees. Increase in snow accumulation is highest in spring with, however, much higher spatial variability for Kleinfleisskees compared to Goldbergkees. This is because of Kleinfleisskees is situated at the transition zone from a high level precipitation mountain area to an inner-alpine dry valley area in the south. After the build up of maximum snow cover in May snow ablation takes place very rapidly, especially in June, July and August.

The long term evolution of snow depth (monthly readings) in the Sonnblick region is shown for the example of three snow stakes. It appears quite clear that there exists a common trend and variability for all three snow stakes (this is also true for the additional two stakes not shown here), especially for observed maximum values. Largest snow depths were measured in the 1940s and 1950s. After the period of largest snow depth in 1940s and 1950s, snow depth on 1 May shows a very weak increasing trend since about 1970. Differences in the temporal trend of maximum snow depth are most likely due to snowdrift effects for the snow stakes in single years. The temporal trend of snow depth on 1 February is similar to the trend on 1 May. In comparison to snow depth on 1 May, summer snow depth (1 August) shows a clear decreasing trend for all the stakes. This finding fits in the pattern of mass balances of glaciers of the Alps which are negative since the beginning of the 1980s. However, the negative glacier mass balances are not reflected in the curve of snow depth on 1 October (remaining snow, Firn) as snow depth at that time is significantly determined from snow fall in September.

In more detail, a N-S difference of snow depth on 1 October for the period of positive glacier mass balances in the 1960s/1970s is also found. Whereas PG3 (which is north of the Alpine main divide) shows a maximum of snow depth on 1 October (remaining Firn) around 1960, PF4 (which is south of the Alpine main divide) shows the analogues maximum at the end of the 1970s. Interestingly, PG6 which is situated in-between PG3 and PF4 includes both maxima (not shown here).

Another important feature of snow depth climatology is related to the changes in the seasonal distribution of snow cover accumulation. This is shown by the temporal evolution of the seasonal share of snow cover accumulation for three sub-periods, early winter period (1 October to 1 December), core winter period (1 December to 1 March) and late winter period (1 March to 1 May), respectively. Again, the figure is restricted to three snow stakes which, however, describes the total sample of five snow stakes very well. Most pronounced trend in the seasonal distribution of snow cover accumulation can be seen for snow stake PF4 with a remarkable decrease in the contribution of core winter with respective increase of contribution in early winter and late winter period. Linear trend estimation shows a decrease in core winter by 2.7% per decade and an increase of 1.8% per decade for early winter with a 95% confidence interval. For other snow stakes the trends are somehow weaker but the decrease in the core-winter contribution appears as a general feature from all snow stakes.

Changes in the probability of specific snow depths are shown for stakes PG1, PG3 and PF4, respectively. A moving window technique (15-years window) is used to compute the probability of different snow depths for four different dates. It appears quite clearly that probability has undergone a remarkably change since the beginning of measurements. In particular, probability of snow depth of 400 cm on 1 May at PG1 reduced from about 70% to about 10% and probability of 50 cm on 1 August at PG1 reduced from about 100 to 50%. For stake PF4 probability of snow depth of 400 cm on 1 May reduced from about 80 to 20%. From all changes in probability of specific snow depths the trends are weakest for autumn.

Importance of winter accumulation for annual net balance of a glacier is quite often discussed in glaciology (e.g. Schöner et al., 2000). Similarly, we investigated the importance of maximum snow depth on 1 May for remaining snow on 1 October. It can be seen quite clearly that there is no significant correlation between snow depth on 1 May and that on 1 October. In addition, the years with the three largest snow depths on 1 May resulted in very low snow depth on 1 October. This proves the previous statement that mass balances of glaciers in general are determined from summer melt and winter accumulation role is quite low.

Extremes of snow depth in the Sonnblick region are as large as about 11 m of snow. Such high values were observed in 1944 and 1951. However, the accumulation process of snow in the maximum years 1944 and 1951 was quite different. In 1944 most important contribution came from March with more than 4 m of snow. For other months accumulation of snow cover was quite equally distributed. In 1951, the highest increase of snow cover was in January with 3–4 m of snow accumulation. Three metres of snow were accumulated within the period 17–21 January 1951 which resulted into an extremely dangerous avalanche situation in the Alps with several hundreds of persons killed by avalanches. Similar information about avalanche activity as of 1951 is not known from 1944. This is partly because of the more equally distributed snow accumulation in 1944 but also because of less availability of information as a result of World War II.

Important forcing factors of snow depth in the Alps, not treated here in detail, are wind speed and wind direction. From the monthly readings of snow depth the great influence of wind on snow depth can be seen quite well from the differences in snow depth from one month to the following one. For some of the readings differences of up to about 200 cm were measured (with precipitation sometimes for the same period). From these, high influence of wind speed and wind direction on snow depth can be assumed. Finally, the role of snow settlement has to be taken into account for investigating snow depth variability. Quantification of the influence of snow settlement on snow depth measurements for Sonnblick region is currently done by means of a physically based snow cover model including snowdrift (Mott et al., in preparation).


The NAO is known as a major driving factor of climate in Europe (e.g. Wanner et al., 1997). Scherrer et al. (2004) found that the NAO did not explain inter-annual variability of snow days for northern part of the Swiss Alps but explained a substantial part of inter-annual variability of snow days for southern part of the Swiss Alps. For decadal trends however the NAO significantly explains snow days’ variability via air temperature for northern part of the Swiss Alps and via air temperature and precipitation for southern part of Swiss Alps. The NAO is also known to be related to atmospheric blocking, with negative correlation for Atlantic blocking and positive correlation for blocking over central and Western Europe (Scherrer et al., 2006). Pinto et al. (2007) showed quite clearly that the extraordinary snow accumulation in central Europe in winter 2005/2006 was a result of high frequency of blocking west of central Europe, which caused below normal temperatures and therefore more precipitation as snow and reduced melting of snow cover. Similarly, Scherrer and Appenzeller (2006) found the high value of explained variance of a uniform new snow sum pattern in Switzerland explained by European blocking (43.5%), whereas a low–high elevation gradient pattern of new snow sum in Switzerland was well explained by the NAO (30.6%). From these studies a significant influence of NAO on Alpine snow cover is obvious.

For the region of Eastern Alps it is known that the NAO has significant influence on winter temperature (Auer et al., 2001), with temporal change in strength of influence. Such a feature can also be seen for relation between winter air temperature (period October to April) and the North-Atlantic Oscillation Index (NAOI) at Sonnblick. Whereas correlation between air temperature and the NAOI is high in the 1940s and 1950s (r = 0.8 derived from a 31 running window correlation) decreased correlation can be seen for the 1980s and the 1990s (r = 0.2–0.4). Especially the low NAOI in 1996 is not reflected in air temperature. In contrast to air temperature, winter precipitation (sum October to April) and snow depth on 1 May at Sonnblick show no general coincidence with NAOI. Whereas high snow depths in 1940s and 1950s are related to high NAOI values, similar high NAOI values in the 1980s and 1990s do not show increased snow depths. However, the year 1996 with a very low snow depth and low precipitation amount shows the lowest NAOI value in the series since 1928. Scherrer et al. (2006) showed a significant relation between NAOI and blocking intensity over Europe which further influences amount of winter precipitation. While increased Atlantic blocking is related to a negative NAOI mode, blocking in the region of the Alps is related to a positive NAOI. Similarly, Quadrelli et al. (2001) described increased west-Atlantic blocking activity (with negative NAOI) as a general pattern of increased winter precipitation over the entire Alps (because of the splitting of the jet stream). However, all these findings are related to winter period, December to February, which significantly differs from our definition of winter (October to April). Therefore, possible anti-cyclicity between NAOI and precipitation or snow depth may be levelled out by the choice of seasonal periods and cannot be seen in the figures shown.

Air temperature and precipitation are the major forcing factors for changes in amount of deposited snow. Air temperature plays a crucial role for estimation of fraction of snow during precipitation events as well as for estimation of snow melt. In this study we use a statistical model (tanh-fit) to compute temporal changes in the fraction of solid precipitation at the location of snow stakes from air temperature based on monthly mean values. The model is similar to the approach of Hantel et al., (2000) and is described in Schöner and Böhm (2006). As air temperature measurements were not available for the location of the snow stakes we used local lapse rates (details given by Schöner and Böhm, 2006) and the air temperature measurements at Sonnblick to derive temperature series at snow stake locations. Computed series of fractions of snow precipitation for all snow stakes are drawn as mean values for the winter period October–April and as mean values for the summer period May–September. It is evident that only lowest elevated stakes show a rather weak negative temporal trend in the fraction of snow precipitation in winter because of the high elevation (and therefore low temperatures) at the stake sites. In summer the decrease in fraction of solid precipitation is about 0.9% per 10 years (linear trend significant at the 95% level). Although this trend appears to be quite small its result on snow and ice melt during summer is important because of the albedo feedback mechanism on shortwave radiation balance.

Amount of precipitation is the most important predictor of snow depth as long as no melt occurs. For Sonnblick region changes in precipitation can be described by a dense totalizer network. Snow depth on 1 May is related to the precipitation amount (1 October to 30 April) from nearby totalizers for the example of two snow stakes PG3 and PF4, respectively. For both stakes a weak linear relation between snow depth and precipitation amount can be seen masking out the 3 years with largest snow depth. This outlier-like behaviour of the three largest snow depth measurements could raise some doubts on the reliability of these data. However, the spatial coherence of these maximum snow depths for all stakes and the simplicity of measurements raise some arguments against the reliability of precipitation measurements especially for such high amounts necessary to explain these large snow depths. The picture gets further clear if instead of the totalizer series a nearby grid point series of a gridded data set of Alpine precipitation (Efthymiadis et al., 2006) is used. However, the maximum snow depth in 1944 still appears as a significant outlier.

Whereas the inter-annual variability of snow depth at Sonnblick is better explained by low elevation precipitation measurements compared to the local precipitation measurements at Sonnblick, the quantity of the longer-term mean precipitation of the totalizer network appears to be highly plausible. This finding results from the computation of mean snow density from mean snow depth and mean precipitation (both for e.g. 10 year average for 1991–2000) which gives values close to 400 kg/m3 for sites with comparable location for both snow depth and precipitation measurements. Similar snow density values were also found from direct measurements at Goldbergkees.

In Figure 9 precipitation and air temperatures of individual years are also clustered to wet–cold (w–c), wet–warm (w–w), dry–cold (d–c) and dry–warm (d–w) using the 0.25 and 0.75 quantile, respectively. It appears quite clear that a clear clustering of extremes of snow depth with respect to a very cold/warm/dry/wet year cannot be found.

Whereas correlation between maximum snow depth and local precipitation measurements at Sonnblick is weak, inter-annual variability of maximum snow depth (1 May) can be well described by the inter-annual variability of winter precipitation (21-years moving window variability). For both snow depth and precipitation, variability was highest in the 1940s and 1950s and significantly decreased since then, which in turn meant that extremes of winter season precipitation and maximum snow height significantly lowered during the last approximate 80 years of measurements.






Sensibilité du milieu à des paramètres climatiques

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

(…) Various studies investigated the sensitivity of snow cover on climate change especially on air temperature (Hantel et al., 2000; Beniston et al., 2003a,b). From these references a strong height dependency of sensitivity of snow cover on air temperature can be concluded, which follows from the deviation of mean air temperature of respective altitudinal range from the melting point (0 °C).

The availability of satellite based snow records enabled scientists to investigate spatially large scale snow cover measures. In fact these series are especially useful to study temporal changes in snow cover extent. Unfortunately, the satellite snow records go back only a few decades. However, Alpine in situ observations of snow cover are available since the 19th century. In general snow depth was investigated for lower elevation sites (lower than 2000 m.a.s.l) of the Alps whereas information about snow cover variability and change for the period of instrumental measurements is sparse for the higher elevated sites of the Alps.

During the last few years much effort was also made to derive long term and homogeneous data sets of snow cover measures for instance the Swiss part of European Alps (Laternser and Schneebeli, 2003). On the basis of such a high quality data set Scherrer et al. (2004) and Scherrer and Appenzeller (2006) studied in detail the role of local- and large-scale climate variability on days with snow-cover, new snow sum and snow depth in Switzerland. Principal component analysis of new snow sum and snow depth in Switzerland yielded three major patterns which are a pattern of winter’s rich or poor of snow (54% of explained variance), a north–south gradient pattern (14% of explained variance) and an Alpine forelands-higher inner Alps gradient pattern (7% of explained variance), respectively. Moreover, they found a distinct elevation dependency of Empirical Orthogonal Function (EOF) loadings (correlation between original and principle component series). Similar studies as for Switzerland are not available for the other countries of the European Alps mainly because of the lack of respective digitized and homogenized data series.

For the Austrian part of the Alps some first analyses of snow cover changes were done by Fliri (1992a,b) and Mohnl (1994). Fliri (1992a,b) did not find a significant trend in the snow cover days over the last century for Austria despite the increase in winter temperature. On the other hand Mohnl showed a decrease of 10–30% in snow cover days in the last 100 years for most Austrian stations. In fact high quality data series of snow cover measures for Austria (with daily to monthly resolution) are still in the process of digitizing and data quality control (Jurkovic et al., 2005) with the exception of the regional snow measurement network in the region of the Sonnblick Observatory (SBO) (Hohe Tauern, Austrian Alps). The Sonnblick region constitutes an outstanding region of measurements of snow cover properties with respect to data quality, data series length (back to 1928), and elevation of measurements (between 2400 and 3100 m.a.s.l). From the network of several snow stakes data quality was stressed in detail and plausibility of data was significantly improved. Together with the long term climatological measurements at the SBO these snow measurements enable the investigation of long term changes of snow depth and snow precipitation at high elevated sites of the Alps and their relation to climate change. Initial work using the snow depth measurements of Sonnblick region was performed by Lauscher and Lauscher (1973), and by Böhm and Mohnl (1987) who did the first statistical analysis. Later Auer (1996) discussed the importance of this snow measurement network for investigation of glacier-climate relation.

Air temperature and precipitation are the two major determinants of snow precipitation and snow melt. Whereas air temperature is covered by reliable measurements at high elevated Alpine sites, precipitation is still influenced by various measurement errors due to strong winds and catching efficiency of precipitation gauges for snow precipitation. This results in a significant underestimation of precipitation at higher elevation sites (e.g. Sevruk, 1985). Because of simple and secure measurement technique snow depth measurements allow to better understand the reliability of precipitation measurements at high elevated sites, and therefore can contribute to this still unsolved problem in meteorology and hydrology.

This paper addresses the issue of trends and variability of snow depth at a high elevated site of the European Alps and its relation to climate change. In particular the representativity of measurements is treated by spatial correlation analyses and temporal trends and variability are investigated from a reduced series network by statistical analysis of measurements. The relation between changes in snow depth and climate benefits from the high quality and long term climate series from the nearby SBO. On a larger spatial scale, the relation between snow depth series and the North Atlantic Oscillation (NAO) is shown by the NAO index.

Sonnblick Mountain is situated at the main divide of the Austrian Eastern Alps (12°550E, 47°050 N, 3105 m.a.s.l). (…) The snow network involves measurements which are part of SBO programme as well as measurements which are part of a glacier monitoring programme (Schöner et al., 1999). Snow measurements as part of the SBO date back to 1892 for precipitation measurements (amount of precipitation, fraction of solid precipitation which is quantified by the observer from precipitation measurements according to classification with 0%, 50% or 100% solid/liquid precipitation), but more detailed measurements started in 1928 with monthly depth readings for five snow stakes (monthly reading). At the same time a precipitation network with six shielded totalizers (monthly reading) was installed. Continuous measurement of snow depth by means of ultrasonic sensor is available from one site (Fleißscharte) which was installed in 1991. Snow measurements from the glacier monitoring at Goldbergkees (1.5 km²), Kleinfleißkees (1.0 km²) and Wurtenkees (1.0 km²) started in 1982, which included measurements of snow depth by snow probing (100 m inter-point distance but without full coverage of glaciers because of safety reasons), snow density measurements as well as snow stratigraphy (snow temperature, snow crystal type, snow crystal size, snow layer hardness, and snow layer water content).

Snow depth measurements with monthly readings are performed by means of wooden stakes at the three glaciers. All measurements are relative to an initial snow depth of 0 cm on 1 October (which matches the definition of the hydrological year) when adjustments of snow stakes are done each year. In case of snow remains on 1 October it is defined as Firn of the past year. The monthly readings are recorded approximately on the first day of each month by field inspections of the observers from SBO. In case of hard weather, readings are recorded as close as possible to the first day of the month and interpolated to first day by means of data at SBO. All data of the SBO programme as well as of the mass balance monitoring programme were subjected to careful data quality control which is described in Auer et al. (2002).

As high alpine precipitation measurements are known to be influenced by measurement errors the gridded HISTALP precipitation data set were also used, which was computed by matching the high resolution precipitation climatology of the Alps of Frei and Schär (1998) with monthly anomaly grids from low elevation precipitation series only (details are provided by Efthymiadis et al., 2006). The HISTALP data set constitutes the best available data source with respect to data series length and data homogeneity and covers the Greater Alpine Region.

For the investigation of the relation between snow depth and larger scale atmospheric circulation the NAO Index (Hurrell, 1995; Jones et al., 1997) is used. Scherrer et al. (2006) showed that this index derived from climate station measurements fits very well to a NAO-pattern derived from an unrotated Principle Component Analysis of gridded pressure fields.

The dense network of snow depth measurements for estimation of winter mass balance (snow probing on 1 May) of the three glaciers can be used for investigation of spatial representativeness of single snow stake measurements. (…) In this study, the authors used a sample of the five snow stakes as a reduced network of the 80–100 data points of the snow probing for the time of maximum snow depth on 1 May for each year of the period 1991–2005. As snow depths from probing were interpolated to the entire glacier area (by means of a spatial interpolation algorithm in a Geographic Information System software (GIS)), these snow depth fields can be used to compute spatial representativeness of individual stakes for subregions of glacier (e.g. altitudinal ranges) as well as the entire glacier area by means of a correlation analysis. (…)


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










Paramètre de l'aléa

Sensibilité des paramètres de l'aléa à des paramètres climatiques

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





(4) - Remarques générales



(5) - Syntèses et préconisations


This study has shown that the snow stake network in Sonnblick region is suitable to describe long term changes of snow depth at high elevated sites of the Alps. Comparison with a high resolution snow probing shows that the stake network at Sonnblick is spatially representative. Temporal changes in maximum snow depth are remarkably high from about 2 m to more than 10 m. Highest values were observed during the 1940s and 1950s, when influence of the NAO was high on air temperature. The correlation between snow depth from stakes and local precipitation measurements is weak and seems to be affected from measurement errors and snow redistribution processes, although a longer term mean of local precipitation from totalizers fits well to a longer term mean snow height. The higher correlation between snow depth and a low elevation grid point series compared to high elevated totalizer measurements supports the finding of many investigators that reliable precipitation gauging at high elevated Alpine sites is still unsolved. Investigation of seasonality of snow cover accumulation show a clear decrease of contribution from core winter (1 December to 1 March) especially for the snow stake south of the Alpine main divide. Linear trend of change in contribution from core winter to total accumulation was estimated as –2.7% per decade for the southern snow stake.

A very clear climate signal can be found in the probability of specific snow depths which appears to be quite stable in autumn but decreases drastically in summer as well as at the time of maximum snow depth for the southern snow stake. This finding is also reflected in the negative mass balances of the glaciers in the Sonnblick region. Snow depth on 1 May does not show any correlation to the remaining snow on 1 October at the end of ablation season which coincides with known high sensitivity of mass balances of glaciers to summer air temperature and low sensitivity to winter accumulation.

Fraction of solid precipitation has reduced only rather slightly during winter (at lowest elevated sites for period October–May) but about 1% per decade for lowest elevated snow stake (2400 m.a.s.l.) during summer. Although this appears to be quite weak the effects on shortwave radiation balance via the albedo feedback are important.

From the correlation between high alpine snow height measurements at Sonnblick and local low elevation precipitation data from HISTALP coincidence of temporal trends of low elevation and high elevation winter season (October–May) precipitation can be derived. This finding is also evident from the fit in the inter-annual variability of both maximum snow height and low-elevation precipitation.

Work is currently performed to better understand the process of wind redistribution of snow in the Sonnblick region. This will be done by means of a physically based snow cover model linked to a prognostic wind field model. Such modelling approach will not only help to get more insight into the redistribution process of snow by the wind but also to better explain the formation of extreme snow depth and its relation to climate.

Références citées :

Auer I. 1996. Results of the precipitation and snow-stake network in the glacier region of Sonnblick. Zeitschrift f¨ur Gletscherkunde und Glazialgeologie 32: 185–192.


Auer I, Böhm R, Leymüller M, Schöner W. 2002. Das klima des sonnblicks—The climate of sonnblick. O¨ sterreichische Beiträge zu Meteorologie und Geophysik 28: 304.


Auer I, Böhm R, Schöner W. 2001. Long climatic time series from Austria. In History and Climate—Memories of the Future? Jones PD, Ogilvie AEJ, Davies TD, Briffa KR (eds). Kluver Academic/Plenum Publishers: New York; 125–152.


Barnett TP, Adam JC, Lettenmaier DP. 2005. Potential impacts of a warming climate on water availability in snow-dominated regions. Nature 438/17: 303–309, DOI:10.1038/nature04141.


Beniston M. 2000. Environmental Change in Mountains and Uplands. Oxford University Press, Arnold Publishers: London, New York; 172.


Beniston M, Keller F, Goyette S. 2003a. Snow pack in the Swiss Alps under changing climatic conditions: an empirical approach for climate impact studies. Theoretical and Applied Climatology 74: 19–31, DOI 10.1007/s00704-002-0709-1.


Beniston M, Keller F, Koffi B, Goyette S. 2003b. Estimates of snow accumulation and volume in the Swiss Alps under changing climatic conditions. Theoretical and Applied Climatology 76: 125–140, DOI 10.1007/s00704-003-0016-5.


Böhm R, Mohnl H. 1987. Statistische Analysen der Schneehöhen im Gebiet des Hohen Sonnblicks. Zeitschrift f¨ur Gletscherkunde und Glazialgeologie 23: 47–63. Vol.


Efthymiadis D, Jones PD, Briffa KR, Auer I, Böhm R, Schoener W, Frei C, Schmidli J. 2006. Construction of a 10-min-gridded precipitation data set for the Greater Alpine Region for 1800–2003. Journal of Geophysical Research 110: D01105, DOI:10.1029/2005JD006120D01105.


Fliri F. 1992a. Der Schnee in Nord- und Osttirol 1895-1991—Ein Graphik-Atlas, Band 1, Universit¨atsverlag Wagner, Innsbruck.


Fliri F. 1992b. Der Schnee in Nord- und Osttirol 1895-1991—Ein Graphik-Atlas, Band 2, Universit¨atsverlag Wagner, Innsbruck.


Frei C, Schär C. 1998. A precipitation climatology of the Alps from high-resolution rain-gauge observations. International Journal of Climatology 18: 873–900.


Haeberli W, Beniston M. 1998. Climate change and its impacts on glaciers and permafrost in the Alps. Ambio 27: 258–265.


Hantel M, Ehrendorfer M, Haslinger A. 2000. Climate sensitivity of snow cover duration in Austria. International Journal of Climatology 20: 615–640.


Hurrell JW. 1995. Decadal trends in the North Atlantic Oscillation and relationships to regional temperature and precipitation. Science 269: 676–679.


IPCC. 2001. Climate Change, The IPCC Third Assessment Report , vol. I (Scince), II (Impacts and Adaptation), III (Mitigation Strategies). Cambridge University Press: Cambridge, New York.


ones PD, Jónsson T, Wheeler D. 1997. Extension to the North Atlantic Oscillation using early instrumental pressure observations from Gibraltar and South-West Iceland. International Journal of Climatology 17: 1433–1450.


Jurkovic A, Auer I, Boehm R, Debit S, Orlik A, Schöner W. 2005. The new centennial snow initiative for the Greater Alpine Region, Status report and first results. In Croatian Meteorological Journal, 601–603.


Laternser M, Schneebeli M. 2003. Long-term snow climate trends of the Swiss Alps (1931–99). International Journal of Climatology 23: 733–750.


Lauscher A, Lauscher F. 1973. Der Aufbau und Abbau der Schneedecke auf dem Sonnblick im Wechselspiel der Wetterlagen, Jahresbericht des Sonnblick Vereins f¨ur die Jahre 1970–1971, 68–669, 3–30.


Mohnl H. 1994. Die Schwankungen der Neuschneehöhe und Schneedeckendauer in Oesterreich (Periode 1895–1992). 90.–91.Jahresbericht des Sonnblick-Vereins für die Jahre 1992–1993, 5–47.


Pinto JG, Br¨ucher T, Fink AH, Kr¨uger A. 2007. Extraordinary snow accumulations over parts of central Europe during the winter of 2005/06 and weather-related hazards. Weather 62/1: 16–21.


Quadrelli R, Lazzeri M, Cacciamani C, Tibaldi S. 2001. Observed winter Alpine precipitation variabilit`a and links with large-scale circulation patterns. Climate Research 17: 275–284.


Scherrer SC, Appenzeller C. 2006. Swiss Alpine snow pack variability: major patterns and links to local and large-scale flow. Climate Research 32: 187–199.


Scherrer SC, Appenzeller C, Laternser M. 2004. Trends in Swiss Alpine snow days: The role of local- and large-scale climate variability. Geophysical Research Letters 31: L13215, DOI:10.1029/2004GL020255.55.


Scherrer SC, Croci-Maspoli M, Schwierz C, Appenzeller C. 2006. Two-dimensional indices of atmospheric blocking and their statistical relationship with winter climate patterns in the Euro-Atlantic region. International Journal of Climatology 26(2): 233–249.


Schöner W, Auer I, Böhm R. 2000. Climate variability and glacier reaction in the Austrian eastern Alps. Annals of Glaciology 31: 31–38.


Schöner W, Auer I, Böhm R, Staudinger M. 1999. Gletschermonitoring im Sonnblickgebiet. O¨ sterreichische Beitra¨ge zu Meteorologie und Geophysik 21: 23–28.


Schöner W, Böhm R. 2007. A statistical mass-balance model for reconstruction of LIA ice mass for glaciers in the European Alps. Annals of Glaciology 46: 161–169.


Sevruk B. 1985. Systematischer Niederschlagsfehler in der Schweiz. Der Niederschlag der Schweiz. Beiträge zur Geologie der Schweiz— Hydrologie 31: 65–75.


Wanner H, Rickli R, Salvisberg E, Schmutz C, Schüepp M. 1997. Global climate change and variability and its influence on Alpine climate—concepts and observations. Theoretical and Applied Climatology 58: 221–243.