To investigate the internal structure and transformational processes from firn to ice of the Hisago snow patch on Daisetsu Mountain, Hokkaido, Japan, many drillings were made during the ablation period in 1986 and 1987. The ice cores were analyzed for stratigraphy, density and free water content. The internal structure of the snow patch was characterized by a wet firn layer with abundant ice layers and ice glands overlying a 4.3 m-thick ice body. The thickness of the firn layer varied from 12.7 m in June to 1.3 m in October. A water-saturated firn layer about 1 m thick existed just above the firn-ice transition. The annual layers were identified by dirt layers which showed that 1–2 m of ice was formed each year when the mass budget was positive. Although the densification of the water-saturated firn layer proceeded rapidly, the transformation from firn to ice could not be recognized during the ablation period. It was concluded that the formation mechanism of the ice body comprised three processes: the formation of superimposed ice, the densification of a water-saturated firn layer and the freezing of wet and/or water-saturated firn by cold wave penetration.

The choice of the best protection system against avalanches on a particular path requires an accurate description, or image, of these avalanches. In order to get this image, avalanche consultants can use several numerical models which are often difficult to handle. Moreover, these models deal only with a part of the phenomena involved in avalanches and ignore the others. As a result, the consultants must use their experience and knowledge to imagine the avalanches on any particular path.

This paper presents ELSA (Etude et Limites de Sites Avalancheux), a computer system dedicated to the modelling of the knowledge of avalanche experts and to the integration of the new symbolic computer models with the classical numerical models. The basic aim of integration is to build a unique computer system incorporating all these models.

After a description of the terrain representation, we present the different scenarios that ELSA takes into account. Then, the methods which deal with some phenomena occurring in avalanches are described. The problems involved in the integration of these methods close this paper.

Isotopic composition of solid and liquid portions of wet snow was investigated experimentally. The compositions changed with time, δ values of ice becoming heavier than those for water. A simple model was proposed to explain their temporal variation. It predicted, however, a more rapid change of δ values than the trend obtained in the experiments. This suggests the presence of a “diffusion layer” adjacent to growing snow particles, where isotope concentration has dropped at the ice-water interface because of the fractionation during grain coarsening. The slope in δD–δ18O diagram estimated by the model is compatible with the experimental data. It is considered, therefore, that the freezing fraction, the part of the liquid which refreezes to relatively large particles during grain coarsening, could be estimated by measuring the isotope concentration.

A three-dimensional, numerical simulation model for snow transport and drift formation is proposed in which saltation as well as suspension are considered as dynamic behavioral factors of moving snow particles. The procedure for simulation is as follows: (1) Air flow field is simulated solving the Reynolds equations and the continuity equation. (2) Using the result of the air field flow simulation, the blown-snow density field is simulated using the diffusion equations in which the fall velocity of blown snow particles is considered. In the boundary conditions, the particle movement of saltation is taken into consideration. (3) Finally, the snowdrift rate is computed based on the amount of snow particles not transported by saltation. This model was quantitatively tested for the phenomenon of snowdrift development. The computed results showed good agreement with observations.

The mechanisms of strong winds associated with snow clouds, and the relationship between strong winds and blowing/drifting snow, were investigated. A snowstorm occurred with a typical L-mode snow band which was generated and organized longitudinally during a continental cold-air outbreak over the Sea of Japan. Doppler radar observations revealed that the snow band consisted of small echo cells arranged along the direction of the snow band. When one of the echo cells passed, blowing/drifting snow was generated and intensified by a snow cloud-induced gust, and the horizontal visibility near the ground surface was significantly decreased. Doppler radar and radiosonde data showed that the gust was due to the cold air outflow (CAO) from the snow clouds. The leading edge of the CAO was about 9 km ahead of the center of the snow cloud and the depth of the CAO was about 600 m near the forward flank of the snow cloud. The CAO was caused by a downdraft at the center of the snow cloud, which was initiated at a height of about 1.3 km and with a velocity in excess of 1 ms−1. The observed CAO speed was explained by the theory of the gravity current.

The optical system of the snow-particle counter (SPC), which was developed by Schmidt in 1977, has been improved. A laser diode is used as a light source, achieving uniform sensitivity to a blowing snow particle regardless of the location of particle trajectory within a sampling volume. The light entering a slit, which may be affected by a blowing snow particle, is perfectly detected by use of a piano-cylindrical lens and a dual-type photodiode. A signal processor has been developed to get output voltage proportional to the mass flux of blowing snow.

From the estimates based on blowing snow characteristics and wind speed profile, the new SPC system can accurately detect all the particles of effective sizes at least at a height above 0.1 m when the wind speed at a height of 1 m is less than 15 m s−1.

Considering the Fraunhofer diffraction by both the wire and the particle, the relation between a particle diameter and sensor output of the new SPC system is derived from the calibration with spinning wires.

Mass flux obtained with the new SPC system was found to be close to that with a snow trap. The system was operated continuously for at least nine days using two 35 A h lead batteries.

This research investigates mechanisms of drifting snow accumulation, and of strong wind associated with snow clouds (developed cumulus). Detailed structure of snow drifting close to the ground was observed at several sites by use of snow particle counters (SPC), visibility meters and other meteorological instruments, simultaneously with observations of wind structure in the lower atmosphere using Doppler radar and radiosonde. Areal distributions of drifting snow were also observed as a function of time. Primary results include the following. (1) The intensity of drifting snow was found to have a strong correlation with fetch distance. Of obstacles to windward of the observation point, a river interfered with drifting snow more effectively than did a hamlet. (2) The areal distribution of drifting snow is explained fully. (3) Visibility, reflectivity and wind velocity at ground level are strongly related to drifting snow patterns.

Preliminary estimates of the thermal signature of ventilation in polar firn are obtained from two-dimensional numerical calculations. The simulations show that spatially varying surface pressure can induce airflow velocities of 10−5 m s−1 at 1.5 m depth in uniform firn, and higher velocities closer to the surface. The two-dimensional heat-transfer results generally agree with our earlier one-dimensional conclusions that the thermal effects of ventilation tend to decrease the temperature gradient in the top portions of the pack. Field observations of ventilation through temperature measurements are most likely to be observed when the firn temperature at depths on the order of 10 m is close to the air temperature, since steep temperature gradients can mask the thermal effects of ventilation. Preliminary indications are that, as long as surface-pressure amplitude is sufficient to move the air about in the top tens of centimeters in the snow, the resulting temperature profile during ventilation is fairly insensitive to the frequency of the surface-pressure forcing for pressure frequencies in the range 0.1–10.0 Hz.

Atmospheric circulation patterns associated with snowfall fluctuations in Japan are examined using a rotated empirical orthogonal function (EOF) analysis. We also compute correlation coefficients between the scores of EOF modes in the 500 hPa geopotential height field of the Northern Hemisphere (NH) and amounts of snowfall in Japan on annual, monthly and pentad time scales. It is found that recent variability of snowfall amount in Japan is closely related to the long-term variations of large-scale circulation patterns. It is suggested that the dominance of teleconnection patterns such as Pacific/North American (PNA) and Northern Asian (NA) are responsible for the increase of snowfall in the coastal regions of the Sea of Japan during the cold period for Japan (1977–86).

During two winter seasons, we observed avalanche winds and the air movement in the snow dust region. The wind velocity near the avalanche front was comparable to the averaged internal velocity of the front region. The avalanche wind blew down the valley. However, there was a rising air current just before the avalanche front. The wind velocity and direction showed a periodic change in the snow cloud region, from which the size of the ordered structure (vortex) was estimated to be 35–70 m. The density of the frontal part of one avalanche was estimated to be 1.9–3.8 kg m−3.

The average amounts of seasonal snow cover and snowfall in Japan were calculated as 7.9 × 1013kg and 1.2 × 1014kg, respectively. The mass of seasonal snow cover of a heavy-snowfall winter, 1980–81 (56-Gosetsu), was calculated as 1.3 × 1014kg. The amount of 7.9 × 1013kg was converted to water equivalent of 230 mm on the whole snow-covered area, including snow-prone area. A mean of 370 mm in snow water equivalent was calculated for the snow area where mean snow depth on the ground was more than 10 cm.

Basically there are three kinds of snow removal: mechanical snow removal (MSR), snow-thawing pipe systems (STP) and snow-conveying open channels (SCOC). This paper identifies inhabitants’ attitudes toward snow-removal options and quantitatively evaluates their preferences by applying the Analytical Hierarchy Process (ΑΗΡ) method. A hierarchy is specified separately for the road in front of a house and the site of a house. A case study carried out for citizens living in the city of Ojiya in Niigata reveals that the space available for walking is an important factor, and hence STP scores considerably higher than SCOC and MSR. For the evaluation of a house site, the frequency per season of removing snow from a roof is the most important factor. House designs for overcoming snow accumulation are also evaluated.

We compute the weight of the snow cover as a function of the daily quantity of precipitation and daily melting using only data from the Automated Meteorological Data Acquisition System (AMeDAS), which is used widely in Japan. The correlation between long-term measurements and meteorological data in AMeDAS factors was computed by statistical methods from the Forestry and Forest Product Research Institute, Tokamachi Experiment Station, in Niigata Prefecture, using data for 11 winter seasons (1977–87).

The daily quantity of melting is expressed with a three-day moving average of degree days. The coefficient of correlation between the daily groups of each value of the 1323 days during the 11 winter seasons was 0.986 with a standard deviation of ±590 Ν m−2. Thus, if air temperature and precipitation can be obtained for an area, the weight of the snow cover can be estimated with confidence.

One of the principal aims of avalanche warning is to prevent slab avalanches triggered by skiers. Other than explosives, the best practical tool for stability evaluation is the Rutschblock test. Whether the slab may be triggered by a skier depends on various slab characteristics. Important factors seem to be depth of the weak layer, slab hardness and sublayering of the slab. The stress distribution induced by a skier is calculated by the finite element method for typical snow-cover configurations. The additional shear stress is of the same order of magnitude as the shear strength of weak layers. Besides the critical weak layer — prerequisite for a slab avalanche — hard layers seem to be important. The analysis suggests that a shear failure is most probable at the transition from a hard to a soft layer. This corresponds well to observations of slab-avalanche profiles. The results may help to improve and quantify the analysis of snow profiles.

The transformation of snow to ice under compressive stress is a very slow process unless meltwater is present. A qualitative measure of permeability variation with depth was continuously determined for various shallow-ice cores from the dry-snow zones in both Greenland and Antarctica to investigate this transformation process. Results of the study provide insight into snow densification mechanisms and generally characterize the pore close-off phenomena. Measurements show that bulk densities may be expressed as a function of overburden; pore close-off takes place when the overburden is approximately 3.0–4.5 bar.

The initiation and evolution of ice sheets are investigated using a two-dimensional thermo-mechanical ice-sheet model. The importance of the amount of snow accumulation on the ice sheet initiation is summarised in the following three points: (1) an ice sheet can grow from an initial area of less than 50 km diameter with a positive but large-enough accumulation rate; (2) the pattern of multiple steady-state solution branches critically depends on the surface mass balance; and (3) snow accumulation strongly controls the growth rate of the ice mass, which is crucial for ice sheets evolving in the limited time available during Milankovitch cycles.

Continuous crystal-size measurements made on the G6 Antarctic ice core (100m deep) show enhanced growth rates above a depth of 30 m (Zone 1) and in the interval between 70 and 80 m (Zone 2). Crystal growth in Zone 1 most probably takes place by a process of sublimation and condensation. The higher growth rate in Zone 2 is most probably related to the pore close-off transformation process in which a non-uniform strain field is created to form air bubbles by plastic deformation and “cannibalization” of individual ice crystals.

A high-precision, self-operating storage precipitation gauge was designed and field-tested. It consists of a cylinder-shaped container that incorporates antifreeze solution to melt snow and a quartz-crystal pressure sensor to detect mass of the contained liquid. Field trials showed performance to be at least equivalent to the Japanese standard gauge. A novel feature of the developed instrument is that it also measures a parameter related to wind speed, which is the basis for a catch correction. The instrument is simple, robust and can be battery-operated, therefore making it useful for winter precipitation measurements at sites where electric power supply is unavailable.

A numerical simulation of simple two-dimensional shear of round uniform grains is used to investigate the dynamic friction characteristics of the layer of snow at the base of an avalanche. For steady, uniform flow on a uniform flat surface, the dynamic friction coefficient transmitted through the shear array is found as it varies with the shear speed and normal force applied to the top of the shear layer, and the properties of the particles in the shear layer.

For this simple model, the flow in the shear layer is found to be independent of the total number of layers in the shear flow. A slip plane is formed along which most of the shearing motion takes places, so that the shear is confined to just two layers of particles which slide over one another. In the absence of gravity this slip plane jumps up and down randomly within the shear layer, which is otherwise composed of agitated semi-dispersed particles.

A simple quasi one-dimensional model of flowing avalanches is presented. It is a further development of that used in the Swiss Guidelines for practitioners. It is shown that shearing in avalanche movement is concentrated near the ground and that, due to the geometrical roughness of the ground, a flow resistance proportional to the square of velocity must be taken into account in addition to dry friction. For the change of flow on changing slope angles it is demonstrated that under certain conditions for internal friction a “normal” flow on a flat lower part can no longer be attained; the avalanche behaves like a rigid body. The runout distance is in fair agreement with the Guidelines if a larger internal friction is used. The main differences lie in much smaller deposition depths and smaller velocities during runout.