Numerical simulations using SNOWPACK, a snow-cover model, were carried out to confirm the model’s applicability to conditions in Hokkaido, Japan, where temperatures are fairly low and for 3 months the snow surface is usually dry except during occasional periods of rain or above-freezing temperatures. The simulations were conducted for Sapporo, Kitami and Niseko using meteorological data, and the results were compared with the observed snow profiles. In Sapporo, snow-profile observations were carried out every day for two winters. In Niseko, one of the most popular ski resorts in Japan, an avalanche accident occurred on 28 January 1998 and a snow pit was dug through the fracture line the next morning. The simulated snow profiles agreed fairly well with the observed ones. However, near the surface we observed depth hoar, which can be an important factor in avalanche release after successive snowfalls, that the model did not reproduce distinctly. Extending the model’s metamorphism laws with an expression of depth-hoar formation under a large temperature gradient, as formulated from an experiment by Fukuzawa and Akitaya (1993), the model reproduced the depth hoar adequately.

Mechanical properties of snow were investigated by means of a vibration response technique in a frequency range from 10Hz to 1MHz and a temperature range from –15° to –0.1°C with heating and cooling processes. The response signals were divided into two kinds of propagation, transverse and longitudinal waves through the snow sample. The temperature dependence of elastic-wave velocities showed a large decrease above –0.6°C. Poisson’s ratio and Young’s modulus of snow samples were derived from the longitudinal and transverse wave velocities. Poisson’s ratio of snow samples showed a value of 0.35 ± 0.01 below –0.6°C, and dropped to 0.29 or less at –0.1°C. Young’s modulus of snow samples at –0.1°C showed values seven-tenths as large as (25–34%less than) those below –0.6°C. These phenomena suggest weakening and slipping of boundaries between ice particles in snow samples near the melting temperature. The elastic-wave velocities and Young’s modulus change with the density of samples and with time and temperature cycling. These changes are related to the number and state of bonds between ice particles in snow samples.

Dense snow avalanches are regarded as dry granular flows. This paper presents experimental and numerical modelling of deposition processes occurring when a gravity-driven granular flow meets a fence. A specific experimental device was set up, and a numerical model based on shallow-water theory and including a deposition model was used. Both tools were used to quantify how the retained volume upstream of the fence is influenced by the channel inclination and the obstacle height. We identified two regimes depending on the slope angle. In the slope-angle range where a steady flow is possible, the retained volume has two contributions: deposition along the channel due to the roughness of the bed and deposition due to the fence. The retained volume results only from the fence effects for higher slopes. The effects of slope on the retained volume also showed these two regimes. For low slopes, the retained volume decreases strongly with increasing slope. For higher slopes, the retained volume decreases weakly with increasing slope. Comparison between the experiments and computed data showed good agreement concerning the effect of fence height on the retained volume.

Optoelectronic sensors using infrared light-emitting diodes and photo-transistors have been used for measuring velocities in snow avalanches for more than 10 years in America, Europe and Japan. Though they have been extensively used, how they should be designed and how the data should be processed has received little discussion. This paper discusses how these sensors can be applied to measure two-dimensional velocities. The effects of acceleration and structure in the underlying field of reflectance are carefully accounted for. An algorithm is proposed for calculating the continuous velocity vector of an avalanche, and a sketch of the mathematical analysis given. The paper concludes by suggesting design criteria for such sensors.

A constitutive law for snow derived from a complementary power potential is proposed. The total deformation of snow is divided into elastic and creep parts. A hereditary integral using Norton’s power law is employed to describe primary creep. The concept of effective stress, which takes compressibility of snow into account, is used to calculate creep deformation. The hereditary integral is approximated by a non-linear spring–dashpot model. Results from uniaxial compression experiments (stress range 15– 45 kPa) on sieved snow of density range 180–470 kgm-3 were used to determine the constants appearing in the constitutive equation. The response of snow to constant strain rate (7.4×10-6 s-1 to 2.2×10-5 s-1) under bilaterally confined conditions was found with an iterative scheme employing the proposed constitutive law. The simulated results agree well with the measured axial stresses and volumetric changes.

Many snow models have been developed for various applications such as hydrology, global atmospheric circulation models and avalanche forecasting. The degree of complexity of these models is highly variable, ranging from simple index methods to multi-layer models that simulate snow-cover stratigraphy and texture. In the framework of the Snow Model Intercomparison Project (SnowMIP), 23 models were compared using observed meteorological parameters from two mountainous alpine sites. The analysis here focuses on validation of snow energy-budget simulations. Albedo and snow surface temperature observations allow identification of the more realistic simulations and quantification of errors for two components of the energy budget: the net short- and longwave radiation. In particular, the different albedo parameterizations are evaluated for different snowpack states (in winter and spring). Analysis of results during the melting period allows an investigation of the different ways of partitioning the energy fluxes and reveals the complex feedbacks which occur when simulating the snow energy budget. Particular attention is paid to the impact of model complexity on the energy-budget components. The model complexity has a major role for the net longwave radiation calculation, whereas the albedo parameterization is the most significant factor explaining the accuracy of the net shortwave radiation simulation.

Avalanche hazard maps of high accuracy are difficult to produce. For land-use planning and management purposes, a good knowledge of extreme run-out zones and frequencies of avalanches is required. In the present work, vegetation recognition (especially focused on Pinus uncinata trees) and dendrochronological techniques are used to characterize avalanches that have occurred in historical times, helping to determine both the extent of large or extreme avalanches and their occurrence in time. Vegetation was studied at the Canal del Roc Roig (eastern Pyrenees, Spain) avalanche path. The avalanches descending this path affect the railway that reaches the Vall de Núria resort and the run-up to the opposite slope. During winter 1996, two important avalanches affecting this path were well documented. These are compared with the results of the vegetation study, consisting of an inventory of flora, the recording of vegetation damages along eight transverse profiles at different altitudes on the path and a dendrochronological sampling campaign. The data obtained contributed to a characterization of the predominant snow accumulation in the starting zone, the 1996 avalanches and the range of frequencies of large avalanches. Also, traces of avalanches that increase the path mapped in the avalanche paths map published by the Institut Cartogràfic de Catalunya in 2000 were identified, improving the initial existing information.

Nearest-neighbour models for avalanche forecasting have made little use of snowpack properties; however, slab thickness (H), slab load (Load) and a skier stability index (Sk38) have proven useful for regional avalanche forecasting in the Columbia Mountains, western Canada. This study explores 21 meteorological, snowpack and elaborated variables including Sk38, H and Load. A daily skier instability index (DSI) is developed as a response variable using skier-triggered avalanche activity on persistent weak layers and stability ratings at the end of the day. In rank correlation analysis, Sk38, Load, previous avalanche activity, H and some meteorological variables were highly ranked. The physical explanations are discussed. In classification-tree analysis, Sk38 was ranked as the most important variable and used in the development of the tree structure along with Load. Besides Sk38 and Load, snowpack thickness, the number of previously triggered avalanches and H have potential to predict DSI. Further we included once all 21 variables, and once all variables except Sk38, H and Load in nearest-neighbour models. Comparing the performance of these models shows that Sk38 along with Load and H have high potential to forecast the DSI on a regional scale.

Historical data about ancient avalanches are scarce in the Pyrenees. Dendrochronology can provide new data about past avalanches and their return period, but up to now little research has been carried out with this purpose. The Aludex project aims to obtain information about the frequency and extent of extreme avalanches, using a dendrochronological and a nivo-meteorological approach. In this paper, we present the results of a dendrochronological study of the Canal del Roc Roig avalanche path which was affected by two extreme avalanches in February 1996. This first dendrochronological study has permitted us to assess the tree-ring signals due to avalanche events in 53 out of 132 dated trees. The trees presented a variety of responses to the 1996 avalanche events. It is shown that the type of tree-ring signal depends on tree age. The methodology has proved successful in detecting the 1996 and 1972 documented avalanche events, and provided outstanding evidence of undocumented past events such as one in 1930.

A model is put forward which focuses on the dynamical evolution of the spatial distribution of snow water equivalent (SWE). We make use of the fact that when the accumulation and ablation process of the snow reservoir is modelled as a summation of a gamma-distributed variable, both skewed distributions, typical of alpine areas, and more normal distributions, typical of forested areas, can be accounted for. A particular problem is to represent fractional snow-covered area (SCA) within the distribution framework. The change in SCA as a response to a melting event is explicitly linked to the shape of the distribution of SWE and is estimated as the probability of non-exceedance of the melted amount from a scaled version of the spatial distribution of SWE. An extensive snow-measuring programme, where several snow courses have been measured repeatedly throughout the melting season, justifies the dynamical aspects of the snow distribution in the modelling approach. The modelling approach has been tested with the Swedish rainfall–runoff model, HBV, and estimated values of SWE and SCA are compared with results obtained using the statistical distribution (log-normal) traditionally used in the model.

Layers of faceted crystals adjacent to crusts form the failure layers for some unexpected dry-slab avalanches. This paper focuses on the case of facets that form when dry snow overlies wet snow. From a basic equation for heat flow in solids, the approximate freezing time of the wet layer is derived. Seven experiments are described in which a wet layer was placed between two dry-snowlayers in a cold laboratory. Measured freezing times are comparable to the freezing times from the approximate solution assuming that latent heat from the irreducible water content flowed up. In four of the experiments, evidence of faceting was observed at the base of the upper dry snow layer within 5 hours and before the wet layer froze. In all seven experiments faceting was observed in the upper dry layer after the wet layer froze. Simulations performed with the snow-cover model SNOWPACK yield freezing times that agree reasonably with the approximate solution and allow the influence of various parameters on the results to be explored. In addition, simulated temperatures and grain evolution are compared with observations, showing good agreement.

Spatial statistics of snow water equivalent (SWE) and melt rate were measured using spatially distributed, sequential ground surveys of depth and density in forested, shrub and alpine tundra environments over several seasons within a 185 km2 mountain catchment inYukonTerritory, Canada.When stratified by slope/aspect sub-units within landscape classes, SWE frequency distributions matched the log-normal, but multiclass surveys showed a more bimodal distribution. Within-class variability of winter SWE could be grouped into (i) windswept tundra and (ii) sheltered tund ra/forest regimes. During melt, there was little association between the standard deviation and mean of SWE. At small scales, a negative correlation developed between spatial distributions of pre-melt SWE and melt rate where shrubs were exposed above the snow. This was not evident in dense-forest, alpine-tundra or deep-snowdrift landscape classes. At medium scales, negative SWE and melt-rate correlations were also found between mean values from adjacent slope sub-units of the tundra landscape class. Themedium-scale correlation was likely due to slope effects on insolation and blowing-snow redistribution. At the catchment scale, the correlation between mean SWE and melt rate from various landscape classes reversed to a positive one, likely influenced by intercepted and blowing regimes, shrub exposure during melt and adiabatic cooling with elevation rise. Covariance at the catchment scale resulted in a 40% acceleration of snow depletion. These results suggest that the spatial variability and covariability of both SWE and melt rate are scale- and landscape-class-specific and need to be considered in a landscape-stratified manner at the appropriate scale when snow depletion is described and the snowmelt duration predicted.

The mechanisms leading to dry-snow slab release are influenced by the three-dimensional variability of the snow cover. We measured 113 profiles of penetration resistance with a snow micropenetrometer on an alpine snow slope. Seven distinct layers were visually identified in all snow micropenetrometer profiles. The penetration resistance of adjacent layers did not change abruptly, but gradually across layer boundaries that were typically 2 mm thick. In two layers, penetration resistance varied around 200% over the grid, possibly due to wind effects during or after layer deposition. Penetration resistance varied around 25%in five layers. Statistically significant slope-scale linear trends were found for all layers. The semivariogram was used to describe the spatial variation. Penetration resistance was autocorrelated, but the scale of variation was layer-specific. A buried layer of surface hoar was the most critical weak layer. It had little spatial variation. The layers in the slab above had higher spatial variation. The penetration resistance of each snow layer had distinct geostatistical properties, caused by the depositional processes.

Although scale issues have been defined to be one of the most crucial topics in geosciences, they have received only limited attention in avalanche research. A thorough understanding of these issues in avalanche forecasting, however, is fundamental for the development of useful prediction models. After the definition of relevant terms, the different aspects of scale issues are introduced in detail. We present hierarchy theory (Ahl and Allen, 1996) as a potential framework for the multi-scale characteristics of the avalanche phenomenon. We suggest a temporal hierarchy, where the main contributing factors are ordered into seven levels according to their temporal-scale characteristics with respect to avalanche forecasting. Within each level there are individual spatial hierarchies, which result in a two-dimensional structure. This process-oriented framework is compared to the data classification scheme of La Chapelle (1980), which is based on informational entropy. Scale issues in prediction modeling and the consequences of the new framework for modeling efforts are discussed in detail.

The combination of numerical weather prediction (NWP) and snow avalanche forecasting has been performed using the output of two weather models run at the University of British Columbia, Canada, and a local numerical avalanche-forecasting model developed for Kootenay Pass (McClung and Tweedy, 1994). The main motivations for this work are. (1) to extend the lead time of avalanche forecasts by using NWP forecasts of meteorological variables as input to statistical avalanche-threat models (instead of the traditional method of using current/past observed meteorological variables as input); and (2) to create another tool to help avalanche forecasters in their daily decision-making by making true forecasts instead of “nowcasts”. Therefore, verified weather-forecast model output was used as input for the local avalanche-forecasting model at Kootenay Pass. The resulting 24 hour avalanche forecast was compared to observed avalanche occurrences and to the 12 hour avalanche forecast with current weather observations. As a result, the avalanche-model output for the test runs with numerically predicted weather data is comparable in accuracy to the runs with observed weather data. The results also suggest that avalanches may be predicted statistically for 24 hours into the future when high-resolution NWP is used as input, weather- and avalanche-forecast errors taken into account during operational use.

We investigated a buried surface-hoar layer using the SnowMicroPen (SMP), an instrument designed to measure detailed snowpack profiles.We collected data from two adjacent parts of a slope 6 days apart. In addition, one manual snowpack profile was sampled each day, as well as 50 quantified loaded column tests (QLCTs) which provided an index of shear strength. For the SMP data, a 900 m2 area was sampled on both days in a grid with points 3 mapart, with some sub-areas of more closely spaced measurements. We collected 86 SMP profiles on the first day and 129 on the second day. Our analyses involved manually locating layer boundaries and calculating statistics for the force signal through the surface-hoar layer. The shear strength index increased by 40% between the two sampling days, but the SMP data show no statistical difference in layer thickness, and the mean, minimum, median, and a variety of percentile measures of the SMP force signal through the layer also do not change. Interestingly, the maximum hardness, and the variance and coefficient of variation of the SMP signal, increased. Since the small SMP tip might only break one or a couple of bonds as it passes through the weak layer, we interpret these changes as being indicative of increasing bond strength. Though we cannot specifically tie the increasing maximum hardness of the SMP signal to our QLCT results, our work suggests that the maximum SMP signal within buried surface-hoar layers may be useful for tracking increases in the shear strength of those layers.

The dataset of Northern Hemisphere EASE-Grid weekly snow cover and sea-ice extent (U.S. National Snow and Ice Data Center) for the period September 1972–August 2000 is analyzed to examine the possible influence of recent global warming on the seasonal change of snow cover in the Northern Hemisphere. It is found that the total snow-cover area in the 1980s and 1990s is diminished by 36106 km2, and the length of snow-cover season is reduced by 2 –3 weeks, as compared with the 1970s. In general, the contribution from earlier snowmelt is greater than that from delayed snow accumulation. In addition, the maximum snow-cover area during January–February has gradually decreased by about 36106 km2 within the two decades. Geographically, the rate of decrease of snow-cover duration is 50.1 week per year (wpy) in the high-latitude regions such as the Siberian Plains and Northwest Territories of Canada and 40.2 wpy in the high-elevation regions such as the Scandinavian Peninsula, Tibetan Plateau and Rocky Mountains. The earlier snowmelt in the high-elevation regions suggests that the snowfall amounts there are decreasing owing to global warming.

An assessment of possible snow changes in a changing climate for Finland is presented. The snowpack structure model SNOWPACK (developed at the Swiss Federal Institute for Snow and Avalanche Research) was used for calculating snow conditions at six different locations in Finland for the decades 1980–89 and 2080–89. Regional climate model (RCAO) data from the Rossby Centre, Sweden, were used as input to the SNOWPACK model. Ten years from the RCAO control run and scenario run were chosen, and the snow conditions for different snow zones were calculated for these winters. The snow-cover depth and duration decreased at all locations in the scenario run cases, and the snow-cover quality also changed between the control and scenario runs: grains were bigger, snow was warmer and denser, and the fraction of faceted snow decreased while the fraction of icy or melting snow increased, even in mid-winter. Finally, the variability between different global climate predictions was analyzed. Significant differences were found between different climate-model outputs. The inter-model variability is comparable to the interannual variability of a single model. The qualitative conclusions from the scenario run do not critically depend on the climate-model variability.

This study is part of research activities concentrating on the real-time application of the U.S. National Oceanic and Atmospheric Administration (NOAA) Advanced Very High Resolution Radiometer (AVHRR) sensor for snow-cover analysis of the European Alps. For mapping snow cover in heterogeneous terrain, we implement the widely used linear spectral mixture algorithm to estimate snow cover at sub-pixel scale. Principal component analysis, including the reflective part of AVHRR channel 3, is used to estimate fractions of “snow” and “not snow” within a pixel, using linear mixture modeling. The combination of these features leads to a fast, simple solution for operational and near-real-time processing. The presented algorithm is applied on the European Alps on 17 January 2003 and successfully maps snow at sub-pixel scale. The detailed snow-cover information makes it easy to recognize the complex topography of the Alps, more so than with either a classic binary map or a Moderate Resolution Imaging Spectroradiometer (MODIS) snow product. The sub-pixel algorithm reasonably identifies snow-cover fractions in regions and at altitudes where neither the classic binary map nor the MODIS algorithm detects any snow. Differences concerning the snow distribution are found in forested areas as well as in the lowest-elevation zones. The algorithm substantially improves snow mapping over complex topography for operational and near-realtime applications.

Local observations of snow layers are used as the basis for spatial extrapolation of snow properties and for establishing a time record of snow deposition, yet significant lateral variations in layer thickness, density and microstructure are well documented. Here we examine the nature of layer heterogeneity over distances of 10– 100 000 m using data from primarily flat locations in Alaska, Antarctica and Greenland. We find that at a scale of 10 m or less, perennial snow layers on glaciers and ice sheets are more uniform and laterally continuous than seasonal layers, which, in addition to heterogeneity introduced by wind and water percolation, are also affected by local topography and vegetation. At a scale of about 100 m, heterogeneity of seasonal and perennial snow layers converges and approaches a peak value. At larger scales (103–105m), local (order 100 m) forcing continues to produce most of the layer heterogeneity, with synoptic-scale variations adding small amounts. Cross-correlation at these larger scales is based on recognizing distinctive layer sequences or matching a few key layers of snow. Many layers cannot be correlated because they pinch out or change at scales (i.e. 100 m) smaller than the spacing between snow pits.