This paper describes the calibration of sediment traps in melt-water intake structures of the Grande Dixence hydro-electricity scheme for the purpose of generating medium-term estimates of suspended sediment and bedload transport by pro-glacial streams. The frequency of emptying/purging of sediment from traps in 17 glacier basins in the Val d’Hérens is combined with their design sediment volume to provide first approximations of the mean annual volume of sediment purged. This is followed by more detailed consideration of 11 years of records from three of the basins, to investigate mean seasonal and inter-annual patterns of purging. The difficulties of relating sediment purging to sediment transport are considered. A sediment calibration strategy based on 2 years of turbidity records, hand sampling of suspended sediment concentration, and monitoring of the accumulation of sediment in sediment traps is proposed for one of the glacier basins, and is checked using hourly and 4 h suspended sediment observations from two further ablation seasons. The calibration is applied to 10 years of discharge and purge records to establish estimates of mean seasonal and inter-annual patterns of suspended sediment and bed-load transport.

This paper discusses the results of a sediment-monitoring programme carried out in connection with hydro-electric power development plans in the river basins surrounding the Jostedalsbreen ice cap in Norway. Whereas the highest suspended-sediment transport rates occur during years with several flash-flood events, the bed load is more dependent upon the duration of large magnitude flood events. Bed-load transport has been obtained from annual measurements of deltaic growth in small lakes at the front of glaciers. During the years 1968–86, the mean ratio of bed load to total load amounted to 0.30–0.50% of the total load, but in years with large magnitude floods this ratio decreased. The mean annual suspended sediment yield of Norwegian glaciers ranges from 100 tonnes km2 a−1 to 1300 tonnes km2 a−1. Valley glaciers cause the highest erosion rates, with the exception of the small cirque glacier, Trollbergdalsbreen, which is thought to be a soft bed glacier. The investigation programme undertaken involved monitoring the volume of suspended sediments, together with the size distribution, mineralogy and shape of grains. In general, the valley glaciers supply more sand than the smaller cirque glaciers, whereas particles in the fine sand and silt-size ranges in almost all of the glaciers are angular in shape. Methods of comparing the abrasive capacity of the sediment load at various intakes in a power plant are discussed. Long-term sediment supply from the glaciers was investigated by studies of varves and rythmites in sediment cores from glacier-fed lakes.

Simulations of run-off from the Greenland ice sheet were made as part of a feasibility study for provision of hydroelectric power for Ilulissat/Jakobshavn, West Greenland. The aims were to see if the available short series of run-off measurements are typical of those under present climatic conditions, and to assess possible changes in run-off likely to be caused by gross changes in drainage pattern on the ice sheet. Specific run-off was calculated from climatological data, whilst run-off volumes were calculated by integrating specific run-off over the area of the ice sheet. There have been substantial year-to-year variations in run-off, but the 6 year measurement period is reasonably representative of present climatic conditions. Run-off could be reduced by 21% as a result of changes in hydraulic conditions on the ice sheet without this having a significant effect on the economy of the planned hydro-electric power station.

Drifting snow creates problems which have practical effects in France in two situations. The first situation is when the result of wind action on snow is to cause accumulation of this snow on mountains, leeward of crests, and consequently to create the potential hazard of avalanche. The second is when the result is to pile up snow in flat regions wherever there are obstacles such as road cuttings and vegetation. This paper focuses on the second type of problem, which affects large areas of central and north-eastern France. We explain how we have chosen our test site, which initial experiments we have conducted in order to assess the general geometrical features which a snow fence must possess in order to function most effectively in collecting snow, and how subsequently we have assessed the properties and costs of different fencing systems.

Over 15 years of experience, and a present stock of nearly 140 devices, have made us familiar with the use of avalanche-blasting ropeways (CATEX) in artificial avalanche release. Two years ago, in consultation with all other users we started to establish a list of recommendations for the safe use of CATEX, so that efficiency and reliability could be improved considerably. These guidelines have been designed to achieve safety by adapting CATEX technology to avalanche problems concerned with mechanical equipment, explosives, staff and maintenance.

Up to the present time, quantitative investigations on wet-snow metamorphism have mostly been conducted on water-saturated snow, because of the difficulty in getting large enough wet-snow samples at a uniformly low liquid-water content. Using the dielectric properties of snow at a frequency in the range 20–100 kHz, a warming device has been developed which has enabled us to bring samples of 7 × 10−3 m3 snow to any desired liquid-water content. A maximum value of 8% by volume was reached within 2 h.

The warming device was used to reproduce natural wetness conditions in the laboratory in order to investigate wet snow metamorphism at low liquid-water content. Snow samples were brought to different liquid-water contents and held in that condition for about 2 weeks, during which grain-size was characterized using a picture-analysis system able to derive the mean radius of curvature of the cluster circumference. At any given liquid-water content value, the growth rate of the mean volume of the crystals building the clusters was constant, a pattern which has also been observed in water-saturated snow by previous investigators. This growth rate is well described by a power function of liquid-water content.

This paper presents the first step in the development of a new kind of computer-based tool which can help a specialist consultant in avalanche path analysis, a knowledge-based or expert system. In considering an avalanche path, the specialist often needs a simulation of possible avalanches on this path in order to obtain data for probable velocities and pressures. Use of classical numerical tools by an engineer requires a great deal of experience and knowledge. Such knowledge exists, but until the development of knowledge-based systems no computer-based tool was available.

After a short review of such systems, we describe a knowledge-based system currently being developed in CEMAGREF. Its problem-solving environment includes methods of analysis developed by a senior consultant, and also numerical and qualitative models. In the first step, which is presented here, it analyses the starting zone of an avalanche path. By means of a digital terrain model and a description of vegetation and main ridges (a model which still needs to be improved) this system can make use of the experience of specialists to produce an avalanche-starting simulation based on specified meteorological conditions. Quite simple rules are employed in order to take into account snow drift. First results are shown on accompanying maps of this qualitative modelling and these will need to be tested for validity.

An operational version of avalanche forecasting using the nearest neighbours method was given to the Parsenndienst, the safety service in the ski area of the Parsenn region. They have tested this NXD method through two winters and have now decided to acquire their own computer with a mass storage unit in order to run the programme. The forecasting is being run by the patrolmen themselves, including the input of weather and snow data, and they also update the file of avalanche and blasting activities with the details of operations, as required for ammunition accountancy. The nearest neighbours method provides the patrolmen with a list of the avalanches that have occurred on the 10 d in the past 20 years most similar to the day being considered. These data represent the nearest neighbours of the method's name. The men are able to interpret the list correctly without any need of instruction, and rely on getting information from it which will help them to decide when and where to use (or not to use) explosives. The method of interpretation of the information supplied is illustrated by way of four specific examples.

Agricultural irrigation is an important use of melt water in the Karakoram Range of Pakistan; indeed all crop cultivation relies on melt water from small glaciers and/or snow and ice patches in the zone above 3500 m. Inhabitants of Hopar settlement utilize run-off from a cirque-shaped basin covering about 11.5km2 between 3600 and 4900 m a.s.l. to irrigate 280 ha of cultivated terraces in the altitudes 2500–3000 m. The characteristics of discharge from this basin require specific adaptions to be made by irrigators. In particular, access, turbulent flow, water temperature and sediment load are problematic, although water allocation is not an important concern because supply exceeds demand for most of the local growing season. Supply characteristics are outlined in the paper and are related to features of the indigenous irrigation system.

Snow blocks were slid down natural snow slopes and filmed with a video camera. Friction coefficients were calculated from time-distance curves and the equation of motion. Dry-friction coefficients ranged from 0.57 to 0.84, and could be separated into Coulomb friction and a friction component proportional to the contact area of the blocks (adhesion). These values are greater than the values usually used in avalanche dynamics, but are consistent with previous coefficients obtained for snow blocks sliding over snow.

When uniform ploughing occurred and a shear layer developed along the track the apparent friction coefficients increased with velocity, and could be modelled by considering the kinematic viscosity of the snow. The values of kinematic viscosity ranged from 10−3 to 10−4 m2/s and agreed well with those values obtained by other researchers.

During the ablation seasons of 1983 and 1987, measurements of discharge and suspended sediment concentration of melt waters draining from Gornergletscher, Switzerland, were obtained at hourly intervals, permitting estimation of total daily sediment flux. Seasonal patterns of variation in sediment flux are interpreted in terms of development of the subglacial drainage network. Variations in flux relate to contrasting temporal patterns of run-off, and the differing incidence of subglacial hydrological events in the 2 years. During such events, in which basal water pressure is raised, large areas of previously hydraulically isolated sub-sole are integrated with flow, releasing quantities of sediment from basal storage. Several types of event are identified, arising during periods of generally increasing discharge in the early ablation season, resulting from temporary blocking of subglacial passageways or from outbursts emptying a marginal, ice-dammed lake, and related to rain-induced floods. Flow spreads out over the glacier bed as pressure increases, suggesting that the basal drainage system consists of a diffuse network of many linked cavities rather than fewer major conduits, particularly at the start of the season. A distributed cavity system may be simplified to fewer conduits, dimensions of cavities may enlarge or the area of bed over which cavities are developing may be expanded to supply debris to melt waters during events. Different partial areas of sub-sole become progressively integrated with flow during sequences of hydrological events. Later in summer, melt waters are confined to basal areas within which only limited sediment remains available for acquisition.

During the twentieth century, glacial outburst floods have been the most destructive natural events on Mount Rainier, a stratovolcano in the Cascade Range in Washington State, U.S.A. In the period between 1930 and 1980 numerous floods were reported from five glaciers on the mountain, most of which flowed from Nisqually, Kautz, or South Tahoma Glaciers on its southern flank. Such floods threaten lives and property because they occur without warning and quickly mobilize the loose volcanic debris into debris flows.

A monitoring program was begun in 1987 which was designed to measure the dimensions and timing of outburst floods, but this has been unsuccessful because no floods have yet occurred on the monitored streams. Four floods did burst from South Tahoma Glacier that was unmonitored, but in spite of this they have been useful in providing evidence of flood storage and release mechanisms. All flood volumes were found to be of approximately similar orders of magnitude, of 1 × 105 m3 of water, indicating that all floods probably had similar mechanisms for storage and release of water. Hydraulic pressure considerations indicate that such a large volume of flood water would be stored at the bed of the glacier rather than in isolated englacial cavities. The stepped bedrock terrain provides an ideal setting for the formation of subglacial cavities capable of storing the volumes of flood water noted.

Distribution of snow-water equivalence (SWE) in the Emerald Lake watershed located in Sequoia National Park, California, U.S.A, was examined during the 1987 water year. Elevations at this site range from 2780 to 3416 m a.s.l., and the total watershed area is about 122 ha. A stratified sampling scheme was evaluated by identifying and mapping zones of similar snow properties, based on topographic parameters that account for variations in both accumulation and ablation of snow. Elevation, slope, and radiation values calculated from a digital elevation model were used to identify these zones. Field measurements of SWE were combined with characteristics of the sample locations and clustered to identify similar classes of SWE. The entire basin was then partitioned into zones for each set of survey data. The topographic parameters of the basin used in classification, namely slope and elevation, are constant in time and did not change between survey dates. The radiation data showed temporal variability providing a physically justified basis for changes in SWE distribution through time. Although results do not identify which of the classification attempts is superior to the others, net radiation is clearly of primary importance, and slope and elevation appear to be important to a lesser degree. The peak accumulation for the 1987 water year was 598 mm SWE, which is about half the 50 year mean.

Bias-corrected suspended sediment rating equations are developed for the data from six separate ablation seasons in the Glacier de Tsidjiore Nouve basin. These equations differ significantly from one another. Suspended loads as predicted by equations used beyond their year of origin fall in the range of 34-278% of the true load. Mean absolute-percentage errors (MAPE) based on forecasts computed from each year’s model applied to the other 5 years of data range from 35 to 81%, with a mean of 52%. The equivalent MAPE for a lumped multi-year model is 38%, for an autocorrelation-correcting generalized least-squares (GLS) regression model, is 15%, and for a transfer function (TF) developed from a single 25 d period, only 5%. Simple structure-transfer functions appear to be more robust to temporal transfer than the usual type of rating equation, and offer the possibility of greatly improving upon the poor performance of OLS rating equations in providing estimates of suspended loads for periods beyond frames of origin.

The porosity and water saturation of the firn of South Cascade Glacier were measured in order to determine both the volume of water stored in it and the significance of this water content for the water volume stored in the glacier. The distance to water below the firn surface was found never to be greater than 4 m, and the average thickness of the water-saturated layer was estimated to be 1.25 m. The average firn porosity was 0.15, the water saturation was 0.61, and the total volume of water stored in the firn was approximately 1.78 × 105 m3 representing about 12% of the total spring storage. The water table was found to exhibit a pattern of diurnal fluctuation which starts in late June or early July, indicating that melt water from the accumulation zone may pressurize the bed at diurnal frequencies. The depth-averaged permeability was found to be 1.5 × 10−5 m/s, a value which compares favorably with those from other studies.

In order to improve their hydro-electric power production in the Grimsel area, Kraftwerke Oberhasli (KWO) plan to construct a new reservoir with a storage level about 110 m higher than the existing Grimselsee. This paper deals with the expected changes of Unteraargletscher after periodical contact with the resulting water body. Upon initial flooding, the lowermost section of Unteraargletscher, about 800 m long, will float, drift away, and melt. A rough estimate of the heat balance shows that the energy input into the lake would be sufficient to melt this ice within 2–3 years, so that calving and melting will continue at a frontal ice cliff. The main effort of the study was aimed at forecasting this retreat. A pre-existing seismic survey was supplemented by new soundings by radar and seismic reflection, resulting in reliable cross-sections and information about the sub-bottom material. The forecast is based on the existing mass flux and an empirical calving rate relationship with water depth and predicts an equilibrium position of the terminus some 3–4 km further back than today, and a gain of water storage volume of 50 × 106 m3 after 10 years.

Results and characteristics of three models for estimating avalanche flow speeds, flow heights, and run-out distances are compared: (1) Voellmy–Salm equation used with the traditional release, track, and run-out segmentation method; (2) Voellmy–Salm differential equation solved numerically along longitudinal profiles of avalanche paths, combined with modified assumptions for the flow in the run-out zone; (3) a granular-flow model introduced by Salm and Gubler. Within the limits of the accuracy of the field observations, all models are able to predict run-out distances correctly, at least for large avalanches, but the Voellmy–Salm type models significantly underestimate flow speeds. Modelling different flow regimes (sliding and partial fluidization) increases the range of avalanche sizes for which correct run-out modelling is possible without recalibration of model parameters.

At strain-rates below 1 × 10−4 s−1, low-density weak seasonal snow fails in a ductile manner. In inclined natural snow covers, local strain-rates hardly ever exceed 1 × 10−4 s−1, therefore initial shear failure, which eventually starts slow and finally fast fracture propagation causing slab releases, must be ductile. Changes in initial stability during precipitation periods are simulated using a microscopic mechanical model for weak low-density snow combined with equations describing strength increase by settling and sintering as a function of snow temperature. The results indicate, in good correspondence with experience, a strong dependence of fracture height of direct-action avalanches on precipitation intensity, snow temperature, and slope angle.

Assessing risks from potential glacier hazards in relation to safety considerations for settlements and other fixed installations in high mountain areas requires the application of experience gained from previous events, combined with simple rules derived from basic glaciological theory. The general characteristics of steep, and usually unmeasured, glaciers can be estimated on the basis of a rough parameterization scheme. Variations in glacier length, ice avalanches, and glacier floods then have to be considered for time periods ranging from a few years up to a few decades. As a result of such systematic assessments, maps of potentially dangerous zones can be prepared. Although the inhabitants of many Alpine villages have always lived with the risk of glacier hazards, it now appears that modern construction work, especially that connected with the development of tourism, has started to infiltrate previously avoided high-risk zones more and more. In order to plan reasonable safety measures, risks from glacier hazards have to be compared with those from other natural hazards in mountain areas, such as snow avalanches, landslides, rock falls! or storm-induced floods. Decisions about the acceptable level of risk are difficult and subjective; they are also often influenced by political and economical considerations rather than by scientific reasoning.

Over 80% of the flow of the Upper Indus River is derived from less than 20% of its area: essentially from zones of heavy snowfall and glacierized basins above 3500 m elevation. The trans-Himalayan contribution comes largely from an area of some 20 000 km2 of glacierized basins, mostly along the axis of the Greater Karakoram range and especially from 20-30 of the largest glacier basins. Very few glaciological investigations have so far been undertaken in this the major glacierized region of Central Asia. Biafo Glacier, one of the largest of the Karakoram glaciers, drains south-eastwards from the central Karakoram crest. Its basin covers a total area of 853 km2, 628 km2 of which are permanent snow and ice, with 68% of the glacier area forming the accumulation zone. This paper describes investigations of snow accumulation, ablation, glacier movement, and glacier depth undertaken in the period 1985–87, set against a background of investigations carried out over the last 130 years. Biafo Glacier differs from most of the other Karakoram glaciers in being nourished mainly by direct snowfall rather than by avalanching; this has the advantage of allowing extensive investigation of accumulation over a broad range of altitude.

Snow-accumulation studies in the Biafo Glacier basin have indicated that annual accumulation varies from 0.9 to 1.9 m of water equivalent between 4650 and 5450 m a.s.l. This suggests an annual moisture input above the equilibrium line of approximately 0.6 km3. Monopulse radar measurements indicate the presence of ice thickness as great as 1400 m at the equilibrium line, although these results may not be completely reliable. Mean surface velocity during the summer of 0.8 m d−1 has been measured near to the equilibrium line. Calculations of annual ice flux through the vertical cross-profile at the equilibrium line indicate a throughput of 0.7 km3 a−1. Estimates from stake ablation measurements also suggest that ice loss on Biafo Glacier is about 0.7 km3 a−1. The close agreement between these three sets of measurements is reassuring, indicating that the ablation zone of Biafo Glacier, whose area covers 0.09% of the whole Upper Indus basin, produces approximately 0.9% of the total run-off. However, it should be mentioned that this estimate does not include water originating from seasonal snow melt, either above or below the equilibrium line, or from rainfall. Net annual ice losses due to wastage of the glacier since 1910 are probably of the order of 0.4-0.5 m a−1; this would represent between 12 and 15% of annual water yield from melting ice.