The origin and nature of radio-echo internal reflections is the subject of much debate. Calculations of single- and multiple-layer reflection coefficients for dielectric changes in the ice sheet indicate that the observed reflection strength may result from reflections from many thin layers but the zone of thin layers may best be approximated by replacing the zone with a single thicker layer with average dielectric properties. Calculation of reflections from density changes, dust and dirt bands and anisotropy of ϵ indicate density variations as the most likely cause of the internal layering in the uppermost kilometer of ice sheets. Comparison of the depth of internal reflections at “Byrd” station, Antarctica, with the physical properties of the ice measured in a deep core hole at that location support this conclusion.

Electrical resistivity measurements were made along two perpendicular profiles on the Ross Ice Shelf, Antarctica, in 1973–74. Apparent resistivities are generally well determined at electrode separations from 10 m out to 600 m, where the effect of the highly conducting sea-water beneath the shelf becomes strongly fell. Schlumberger and equatorial-dipole data are in excellent agreement on each profile; apparent resistivities on the two profiles, however, differ by about 12% at separations greater than about 30 m. This apparent anisotropy is attributed to a presumed inhomogeneity at a few tens of meters depth, rather than to true anisotropy in the bulk resistivity.

A computer program has been developed to calculate apparent resistivities on an ice shelf in which the density and temperature, and thus the resistivity, vary continuously with depth. Temperatures have been calculated according to the analysis of Crary (1961 [b]) for a steady-state ice shelf; densities have been calculated from seismic velocity data. Several different models of the dependence of resistivity on density have been tested—one appears to fit the observations very closely, but it must be accepted only with great caution because the assumptions on which it is based are violated in the ice shelf.

The activation energy and the rate of bottom melting or freezing upon which the temperature-depth variations depend have been treated as variable parameters in the modeling. The most satisfactory model corresponds to a melt/freeze rate close to zero, and an activation energy, 0.25 eV (24 kJ mol−1), in agreement with laboratory measurements on Antarctic ice samples, although less than that suggested by previous field measurements. However, since the actual temperatures in the ice shelf are unknown, models that combine a substantial melt rate with a higher activation energy, or a substantial freeze rate with a lower activation energy, cannot be ruled out at present. Future measurements in places where the temperature profile is known should resolve this uncertainty.

The actual resistivity in the solid ice at a depth of about 100 m (temperature about —23°C), lies within ±10% of 70000 Ω m, thus once again confirming the very low resistivities typical of polar glacial ice. The resistivity is, in fact, only about half that found near Roosevelt Island to the north and “Byrd” station to the east. That difference is believed to be real, but its cause is not known and probably will not be known until the basic cause for the generally low resistivity of polar ice is better understood.

An approximate, analytic solution is found for the profile of a water-free crevasse in a field of closely spaced crevasses. The depth of penetration of the crevasses into the glacier is found. If the fracture strength of ice is taken to be zero, the penetration depth is equal to the value found by Nye and is independent of the crevasse spacing. This conclusion is in disagreement with results reported recently by R.A. Smith. If the fracture strength of ice is taken to be finite, the penetration depth is reduced if the spacing between crevasses is reduced.

The results of the analysis can be applied to other crack problems. In particular, it can be applied to thermal, secondary cracking that it is hoped occurs when cooling fluid flows through the cracks created by hydraulic fracture for the purpose of extracting geothermal heat from hot, dry rock masses.

Analysis of wind-speed measurements made over a six-month period on an Antarctic glacier showed that conditions near the surface were dominated by gravity winds flowing downhill. In such conditions there is no satisfactory method of calculating the amount of energy exchanged between the glacier and the atmosphere. It is also difficult to extrapolate satisfactorily energy changes measured at a single point to the whole glacier. Moreover the loss of five days’ meteorological records may cause an error as large as the total change in energy content of the glacier during a year. In view of these difficulties it is more fruitful to measure the changes in energy content of the glacier directly. This can be done by accurate measurements of ice temperature and density near the surface. By defining the total energy content of a glacier as the heat required to melt it, fractional changes in the energy content and mass occurring over a year are equal and indicate the probable lifetime of the glacier. Estimates based on data from an Antarctic glacier suggest that the long-term change of energy of the glacier (≈ 1.5 W/m2) could be measured with an accuracy of 10% within a year.

Two climatic models are proposed for the fluctuation of Austrian Alpine glacier termini. The glacier record, based on annual observations by the Österreichische Alpenverein. is related to climatological data from the high-elevation Sonnblick-Observatorium. Monthly mean temperature and total precipitation during the ablation season for the concurrent plus preceding eleven years are the variables used in this analysis. The preferred model, mean temperature for June, July and August for the concurrent plus preceding seven years, is inversely related to glacier advance. This model was derived by combining those variables with the highest correlation coefficient relative to the glacier record. Mean lag time between climate and glacier termini response seems to be seven years. The second and less preferred model is derived by multiple regression analysis and includes seven variables temperature and precipitation. Variation in the glacier behavior record accounted for by the two models is 71% and 67% respectively.

The stress field in an isotropic elastic hump representing a typical bedrock feature is obtained for plane strain conditions. Gravity effects are included and the applied load is a normal pressure distribution deduced from an idealized model of glacier flow. A Coulomb failure criterion is applied, including the effective stress change due to pore-water pressure, and stresses on the predicted failure planes determined for different pressure amplitudes and relative gravity contributions. The latter make little difference to the maximum “failure stress" but influence the regions where such stress levels occur. Levels of cohesive stress required to inhibit Coulomb failure are obtained, and are low in general, implying that coherent rock in the adopted hump profile, subject to the model pressure, would not fail. That is, this profile is stable unless jointing introduces an easier failure mechanism.

Ice-marginal features in South Georgia include minor moraine ridges running approximately parallel to the ice margin and developed as superficial forms in various depths of till. It is suggested that these developed as ice-contact landforms, some being created by the interaction of the ice edge and abutting marginal snow bank, whilst the glacier is advancing. This advance is essentially short-term, being part of a seasonal variation in the position of the ice margin. “Snow-bank push” is the term proposed to describe this mechanism of ridge formation.

Meteorological and snow pack variables are measured on chronologically ordered sequences of avalanche and avalanche-free days. Discriminant analysis is used to define a subset of variables which produce an optimal separation of the two multivariate group means. Two seasons are identified in each of the years considered corresponding to periods of dry- and wet-snow avalanches, and form the basic stratification of avalanche days in the analysis. Days are stratified further within each season on the basis of magnitude and number or releases. Weather and snow parameters are integrated over variable time steps prior to each avalanche or avalanche-free day. This introduces a recursive element into the forecast method. Preliminary testing of the method points to its potential in real-time snow avalanche forecasting on a regional basis.

Isoglacihypses depicting the configuration of the glaciation threshold (= “glaciation limit”) in Washington broadly parallel the crest of the Cascade Range and curve around the west and south flanks of the Olympic Mountains. In both uplands the glaciation threshold rises inland (eastward) with a mean gradient of 10-12 m/km. However, the gradient in the Cascades is more variable (7-25 m/km) due to five east-trending troughs in the glaciation threshold surface that coincide with topographic depressions along the range crest and that apparently result from greater eastward penetration of moist maritime air.

Mean accumulation-season precipitation correlates strongly (r2 = 0.86) with altitude of the glaciation threshold in the North Cascade Range, but the correlation of glaciation threshold with altitude of the July freezing isotherm, determined from the calculated July lapse rate within the mountains, is much weaker (r2 = 0.40). Multiple regression analysis relating independent climatic variables that affect the height of the glaciation threshold indicates that 90.4% of variance is explained by accumulation-season precipitation and estimated mean annual temperature at the glaciation threshold.

The glaciation threshold during the greatest ice advance of the last (Fraser) glaciation in the southern North Cascade Range (c. 18 000-22 000 years B.P.) was 900 ± 100 m below that of the present. Depression of the glaciation threshold by this amount most likely resulted from a change in accumulation-season precipitation of no more than 30% from present values and a decrease in mean ablation-season temperature of 5-5 ± 1.5 deg.

Three different types of saline ice were cut into sections a few millimeters thick along planes parallel to the surface. The resulting melts were analyzed quantitatively for chloride, sulphate, sodium, potassium, magnesium, and calcium. Two of the specimens were artificial sea ice, grown in the laboratory at ambient temperatures of —30°C and - 15°C, respectively. A portion of the latter exhibited a clear, glassy, fresh-ice structure. The third was natural sea ire. The resulting ion concentration profiles and ion ratios are presented. For sections of widely varying salinity, the various ion ratios assumed values fairly close to those in natural sea-water.

Side-looking airborne radar (SLAR) imagery obtained in April-May 1974 from the North Slope of Alaska between Barrow and Harrison Bay indicates that tundra lakes can be separated into two classes based on the strength of the radar returns. Correlations between the areal patterns of the returns, limited ground observations on lake depths and water compositions, and information obtained from LANDSAT imagery strongly suggest that areas of fresh-water lakes giving weak returns are frozen completely to the bottom while areas giving strong returns are not. This is a reasonable interpretation inasmuch as the reflection coefficient associated with the high-dielectric-contrast ice-water interface would be roughly twelve times that associated with the low-contrast ice-soil interface. Brackish lakes also give weak returns even when they are not completely frozen. This is the result of the brine present in the lower portion of the ice cover limiting the penetration of the X-band radiation into the ice. The ability to separate tundra lakes rapidly and easily into these two classes via SLAR should be useful in understanding a wide variety of problems.

The study of subglacial movements taking place in regions of cavitation beneath the Glacier d’Argentière, has demonstrated the necessity for using several cavitometers. The very unusual conditions in which such apparatus is installed to measure simultaneously the velocity and the position of the ice vault have led us to design and produce a prototype of a portable cavitometer which records the data on magnetic tape (cassettes).

A diamond wire saw was modified for cutting thin sections of frozen soil and suitable operating conditions were determined experimentally. It was found that a lubricated wire, 0.34 mm in diameter, operaied at cutting velocities of 100-300 mm s-1 and cutting forces 0.02-0.1 kg produced smooth cut surfaces un thin sections 0.4-0.5 mm in thickness. The; temperature and wire size were not critical operating parameters and the wire tensions recommended by the manufacturer were satisfactory. A method of mounting the thin sections is also described.

It has long been known that a temperature gradient in the snowpack, when temperatures are less than 0°C, can metamorphose the snow to produce a fragile euhedral crystalline end product called “depth hoar" which is considered to be responsible for snowpack collapse and climax avalanches. The wave of metamorphism advancing through anhedral snow converts it, first to subhedral, and finally to euhedral depth hoar. Our field observations, which are at the moment supported by laboratory evidence, reveal that the zone of minimum snow strength in the gradient metamorphic layer is associated with the intermediate subhedral phase rather than with the euhedral end product.

Calcite (Mohs hardness 3), fluorite (hardness 4) and periclase (hardness 5 1/2) have been eroded by snow blown at moderate wind velocities, at temperatures between —10 and —25°C. Future interpretations involving ventifacts and apparently wind-blasted rock exposures must take this information into account.

Success has been achieved in relocating wire netting, which was subsequently covered by a year’s net accumulation of snow on Berendon Glacier. A signal produced by a wire loop tuned to respond to a 15 kHz electromagnetic output was readily detected by a hand-held battery-powered electromagnetic detector. Sextant readings on fixed landmarks around the glacier aided in locating the search area.

It is shown that when the elastic properties of sea-ice sheets are taken into account, wind-induced tilt is insufficient to cause cracking. These results are contrary to those of Weber and Erdelyi (1976) who considered the ice to be rigid.