Over 30% of Alaska’s 586 400 squares miles (1 518900 km2) is subject to snow-avalanche activity. For a state-wide avalanche hazard evaluation, Alaska has been divided into six major snow— avalanche regions on the basis of topography, climatological data, dominant snow—pack conditions, and typical avalanche activity. They are: Arctic Slope, Brooks Range, Western, Interior, South—central, and South—east.

Mountainous terrain was studied at scales of 1 : 250 000 and 1 : 1 584000; final compilation was at a scale of 1 : 2 500 000. Regional snow—pack and climatic conditions were cross—correlated with relief zonation of each avalanche region to produce a map of Alaska's provisional snow—avalanche potential.

Most of the mountainous areas in the South—central and South—east regions, because of their northern latitude, closeness to large masses of water, and large orographic and cylonic weather processes, are susceptible to major avalanche activity. For areas near population centers, the potential avalanche terrain has been identified from data on known and suspected avalanche activity through air photographs, terrain analysis, and documented snow—avalanche occurrences compiled at scales of 1 : 250 000 and 1 : 63 360.

The state—wide regional data compilation and study are initial steps toward avalanche zoning in Alaska. Local land—use planning and detailed investigations are needed to establish effective natural—hazard zoning in municipal areas as related to snow avalanche activity.

This paper reports on results from a series of experiments in a boundary-layer wind tunnel concerning snow-drift control adjacent to grade separations on interstate highways. A new application of similitude principles making use of theoretical relationships involving drifting particulate material is presented. The primary modeling parameters are considered to be a mass-rate parameter and an aerodynamic roughness parameter. They are derived by considerations of mass-transport rate of material in saltation and the equivalent roughness height of material in saltation. The parameters are combined empirically to correlate model snow-drift data successfully as well as to predict equivalent full-scale wind speeds and storm durations. The combined parameter has been used to compare and evaluate a variety of drift-control techniques adjacent to an interstate highway grade-separation crossing.

This paper presents results from studies of snow-drifts formed by vertical-slat “Canadian” and horizontal-slat “Wyoming” snow fences having 50% porosity and heights H from 0.8 to 3.8 m, on nearly level terrain. Characteristics of equilibrium lee drifts behind the Wyoming fence include cross-sectional area 19.3H 2, length 30H, maximum depth 1.20H. and water-equivalent volume 6.8H2.18. Up-wind drift dimensions include maximum depth 0.5H and length 12H. Drifts behind Canadian fences are about 25% smaller than for the Wyoming design. Polynomial regression equations are fitted to drift profiles for both fence types. Relationships are presented for the effects of wind orientation and curvature of drifts near fence ends, as well as empirical equations describing pre-equilibrium geometry relative to degree of saturation. Mean density ρ (kg/m3) of drifted snow having depth y (m), is given by ρ = 376+158 log y.

Drifts formed by snow fences appear to be scaled in proportion to fence height over a height range of nearly two orders of magnitude. This paper examines natural scaling in terms of commonly accepted similitude criteria. It is suggested that requirements for dynamic similarity of flow fields are satisfied over a period of time because surface roughness height z0 varies approximately as the square of shear velocity u⋆2 for saltating flows in air. This is demonstrated by intensive measurements of wind profiles over snow and ice surfaces. Satisfaction of similitude requirements for scaling particle motion is less certain, although natural sorting may effectively scale particles over a restricted range of wind speeds. Natural scaling of drifts suggests snow erosion and deposition can be studied with reduced-scale models on smooth surfaces outdoors. The feasibility of this technique is demonstrated by results from 1 : 30 scale models of 1.8 and 3.8 m snow fences on both level lake ice and irregular terrain constructed from compacted snow.

The mechanism by which blower fences can protect roads from snow-drifts is investigated, and the variation of the drift-free width behind a blower fence with the dimensions and inclination of the fence and with terrain features has been determined. The results are presented in the form of an equation and a nomogram to help in the design of blower fences.

An hypothesis relating the slip velocity for laminar flow over a porous surface to the specific permeability of the surface proposed by Beavers and Joseph (1967) is applied to the problem of estimating the surface shear stress on an extensive snow surface at low wind speeds. Vertical profiles of wind speed and the air permeability of the surface snow layer were recorded for five periods of neutrally stable, hydraulically smooth flow. The variation of the ratio of the slip velocity to the surface shear stress with the specific permability was examined. The data are compatible with the hypothesis, although they are too sparse to allow valid estimates of the coefficients in the hypothesis.

Cohesive forces are added to the analysis of forces on sand and soil particles to show that fluid drag, alone, often cannot initiate movement of a snow surface. The impact force of saltating ice spheres, however, can easily provide the force to break cohesive bonds, according to these calculations. The argument suggests a balance between the distribution of bond strengths of exposed surface particles and the distribution of saltation trajectory heights.

In order to gain more insight into the mountain snow-transport mechanisms wind and drift flux measurements have been executed on a ridge crest (mainly during snow-storms). Horizontal wind-speed profiles, measured between 0.3 and 6 m above snow surface, show a hump-shaped course especially for strong winds. Theoretical approximations substantiate that the Bernoullian pressure decrease on the crest may be the main cause for this type of wind profile. Roughness parameters (Z0, u⋆) are determined with the aid of the wind profiles and compared with those reported in the literature. Corresponding drift density profiles coincide with steady-state drift theories as long as wind speeds are low (u1≤ 7-10 m s-1), at greater wind speeds snow plumes of 1 to 1.5 m thickness develop immediately above snow surface. Areal measurements on snow mass-balance differences between windward and lee slopes are used to approximate the total transport over the ridge crest and to derive a quantitative relationship between crest winds and drift-snow deposition on lee slopes.

There are a number of published empirical formulae for drift-snow transport as a function of wind velocity. Comparing these formulae at the same wind velocity, however, results in considerable disagreement. It is hypothesized that the disparity arises from snow conditions and the various stages of development of drifting snow.

The horizontal distribution of drift flux was measured with snow traps along a transect parallel with the wind, beginning at an up-wind boundary that served as the starting point of drifting snow. Results indicate that drift-snow transport cannot be defined uniquely unless the drifting snow attains equilibrium (i.e. the snow profile is saturated).

Saltation of snow particles is thought to prevail near the snow surface. However, the vertical flux profile of saltating snow has never been measured. Vertical profiles of drift flux from the snow surface to a height of 30 cm were measured at nine levels, using snow traps composed of nine streamers (compartments). It appears that the saltation flux prevails up to a height of 7-9 cm above the surface, and the suspension flux gradually takes over as the drifting snow develops.

A description is given and an analysis is made of successful VTR recordings of the process of natural release of eight full-depth avalanches. The observations were made over 7 years on a slope, where avalanches took place. The VTR was capable of taking a visual daytime record of an avalanche in motion for a 12 h sequence per reel.

The sagas of Alaskan winter outdoor activities contain increasing accounts of sudden deaths from snow avalanches. Over 100 documented fatalities have occurred within Alaska since 1898 ; 22 people died in south-central Alaska in the past 8 years. With increases in population, development, and use of the mountainous areas, avalanche warning and public education have acquired a new and very important function in Alaska.

Some efforts towards establishing an avalanche warning system for south-central Alaska have been initiated by state and federal agencies. Snow-avalanche education programs presented by Alaska Division of Parks, local rescue groups, and outdoor educators have helped to increase the overall avalanche awareness in the Anchorage community. Formalization and expansion of current efforts are aimed to reduce future snow-avalanche accidents and fatalities.

The energy-line method results from the graphical representation of the energy law as applied in hydraulics. It makes it easier to understand the entire process of the movement of an avalanche from where it breaks away to where it is deposited. The technique, the energy lines, and the average slope of avalanches, as well as the theoretical energy lines of the one-coefficient and two-coefficient models of avalanche dynamics, are described. The method is explained by means of an example.

Internal snow–cover structure depicted by conventional snow profiles is commonly held to be important for avalanche forecasting but it is largely omitted from numerical forecasting schemes owing to quantification difficulties. A subjective test given to practising forecasters illustrated the complex roles of measured profile parameters in describing instability. A clustering technique of pattern recognition, after introducing a snow–profile training set, has demonstrated an objective ability to sort correctly 86% of the layer boundaries associated with slab–avalanche bed surfaces or lubricating layers. Further improvement in stability recognition comes from a newly devised shear test driven by a wedge which quickly locates failure planes within the snow cover. The cluster analysis is then able to sort with similar accuracy between stable and unstable (avalanching) failure planes so located.

The exceptional avalanches of December 1966 in the basins of the Gunes and Kashi Rivers caused much damage and blocked the highway for several months. The general avalanche conditions in this area and the exceptional weather which caused the exceptional avalanches are described.

The mountains of western and central T’ien-shan have extensive snow cover and consequent avalanches, however conditions in this area of central Asia are different from those in many other regions with avalanches. Snow cover is not very thick but it experiences very large temperature gradients which dominate its metamorphism, thus depth hoar forms extensively and becomes the principal snow type, so the density therefore remains low as does mechanical hardness. Avalanching activity and total volume vary enormously from year to year; 1968-69, with 211 avalanches of 147 000 m3, accounts for 53% of the avalanches and 75% of the volume of all the avalanches in the seven years 1967-74 in the area of the Gunes avalanche station. Although they can occur from November to April, the main months are January and March with a minimum in February. A snow depth of 50-60 cm is needed for avalanching. Below —10°C dry-snow avalanches occur, while above about —5°C wet-snow avalanches happen. These are often caused by melt water penetrating rapidly through the extensive depth hoar and initiating full-depth avalanches.

This paper reports the present stage in our research programme. We have used two very different models : (i) A numerical model treating the avalanche as a Newtonian liquid with a free surface; this model describes qualitatively the velocity fluctuations both of avalanche particles and of the front and gives some quantitative predictions, but much experimentation is needed to determine the large number of parameters involved. (ii) An analogue model in which powder avalanches have been simulated in an underground tunnel, taking account of the densimetric Froude number; this model reproduces the pressure rise which precedes the visible front, and also the pressure wave as the front passes which causes so much damage to engineering works—and human beings. These models have been applied to two cases—a bridge and a snow shed.

The interplay between models and experimentation is discussed and the need for improved data on the mass output of an avalanche, of its detailed velocity history, and of the force on large objects is emphasized. Methods being used to try to supply this information are discussed ; gamma-ray measurements for specific mass, high-speed stereophotogrammetry for velocity, and various pressure sensors for the force. The possibility of releasing real powder avalanches to enable such measurements to be made more reliably is being investigated.

This study describes the application of an incremental rheological law for snow in a finite-element model. The computation simulates the variation of strains and stresses in a layer of snow covering an inclined plane of variable slope. At the same time, a physical simulation was carried out in a cold room to verify the results obtained with the model. This work illustrates the value of a finite-element model for snow, but also the complications attendant on its use. Indeed, the integration of the law may lead to oscillations due to the rate of variation of the stresses, and this means that the time increment must be modulated during the computation.

Automatic measuring devices can greatly facilitate the difficult task of persons responsible for avalanche warnings or the closing of roads or ski runs. In favourable cases the traffic can be directly stopped by means of radar when an avalanche is released. The development of the necessary bases and technical realization are important tasks of research. Above all the snow height or the mass of snow (from a snow gauge) or the onset of movement of an avalanche need to be measured automatically. For research it is also interesting to determine the speed of the avalanche and its distribution in given coordinates—along and across the avalanche track. With automatic snow gauges two operating methods have proved successful : (1) ultrasonic sounding from a point above the snow; (2) absorption of cosmic radiation; apparatus for detecting this can also be built into the soil of a slope in the avalanche release zone. The moment of avalanche release can be identified by means of radar in the X-band with moving-target indicator and micro-computer evaluation.

Experimental evidence shows that snow is a pressure-sensitive, dilatant, strain-softening material in slow, constant-rate shear deformation. When strain-softening initiates in a weak layer underneath a snow slab, avalanche release is hypothesized to be possible with or without loading. Specifically, two cases are discussed : (i) a shear-crack-like disturbance can initiate by formation of a slip surface in the weak layer and traverse the layer by a self-propagating progressive failure with or perhaps without loading; (ii) a self- propagating shear instability can develop when a region of the weak layer is driven past peak shear strength by loading. These cases represent the extremes in weak-layer deformations under which strain-softening failures might precipitate avalanche release. For these cases the associated fracture sequences, fracture geometry, time scale of release, and temperature- related effects are consistent with the known facts of dry-slab avalanche release.

A model, relating spatial patterns of snow accumulation in the alpine belt of major mountain ranges to the angular relationship which exists among terrain features such as cirque basins, mean storm-path trajectories, and incident solar radiation, is presented together with a limited amount of relevant empirical data. It is suggested that an “orientation gradient” exists in both the temporal and spatial variation of accumulation and ablation in the alpine terrain, the importance of which varies as some function of the relative control exerted by wind redistribution of snow and of direct incident solar radiation on the accumulation and ablation processes. The data obtained to date suggest that the importance of the orientation gradient varies widely among the mountain ranges studied, being a good predictor of snow distribution patterns in the Front Range of the Colorado Rockies and the Snowy Range in southern Wyoming, both of which are relatively windswept, while in the much less windy Bridger Range, in south-western Montana, both orientation and elevation appear to contribute on an almost equal basis. An indirect method of estimating relative amounts of annual mass turnover, using eleven glaciers located in the Wind River of west central Wyoming, is presented.