The thermal conductivity and diffusivity of porous ice accreted on spherical and spheroidal hailstone models were measured over a density range of 620–915 kg m−3. By scanning the evolving surface temperature distributions during cooling in a cold airflow the thermal conductivity was varied in iterative fashion until the internal heat flux produced the correct surface temperature distribution. The results indicate a linear dependence of the thermal conductivity, ki, and diffusivity, αi, on density. For example, lowering the density by 10% lowers ki by 15%. Within the range of cloud conditions, the density variations affect the thermal parameters more than temperature does. The results also indicate a continuous decrease of the thermal conductivity from bulk ice via consolidated porous ice to loosely packed snow.

Annual net balance eight North Cascade glaciers during the 1984-94 period has been determined by measurement of total mass loss firn and ice melt and ice melt and, residual snow depth at the end of the Summer season. Overall spatial density of measurment points is 200 points km−2. Mean annual balance of North Clascade glaciers from 1984 to 1994 has been −0.38 ma−1. The resulting 4.2 m loss in water-equivalent thickness is significant, since North Cascade glaciers have an average thickness of 30–50 m.

Cross-correlation of annual net balance Ior eight glaciers ranges from 0.83 to 0.97. This indicates the mass balances of the eight glaciers have been responding similarly to elimate conditions despite their range of topographic and geographic characteristics. Annual net balance of individual glaciers was correlated with climate records. The highest ablation-season correlation coefficient is mean May–August temperature, ranging from 0.63 to 0.84. The highest accumulation-season correlation coefficient is total accumulation-season precipitation, ranging from 0.35 to 0.59.

The Vavilov ice cap was perforated in 1988 by a drilling which reached the underlying frozen sediments. In contrast to the overlying glacier ice, the basal ice is composed of different ice layers with a variable debris load. The stable-isotope composition of these layers shows δ values much lower than everywhere else in the core or in the Vavilov ice cap. This is most probably the signature of a remnant of Pleistocene ice which, for the first time, is shown to occur in the Russian Arctic.

We use the Wetherald and Manabe climate model to predict the response of mountain glaciers to a doubling of atmospheric carbon dioxide. The response is measured in terms of a change in the equilibrium-line altitude (ELA) and the glacier terminus altitude (GTA), net accumulation–ablation on these altitudes and the melt runoff for 12 mountain-glacier regions in southeastern Alaska, U.S.A. The methods we use involve extrapolating climate-model temperature fields to a glacier’s location, and empirical–statistical relationships between air temperature and percentage of solid precipitation, and between summer air temperature and ablation and melt runoff. Our study shows that, under global warming, glaciation in southeastern Alaska will not disappear, but mass exchange of glaciers will be more intensive and the ELA value will increase by 300–760 m, depending on the glacier’s distance from the ocean.

In temperate glacier ice, in situ, besides water veins, there are water lenses, on grain boundaries more or less perpendicular to the direction of maximum pressure p1 (at the grain scale). Geometry of veins is developed. Grains are modelled as equal tetrakaidecahedra. The stress and temperature fields around a vein at a smaller, microscopic scale are estimated and the water discharge by a Vein is calculated. The time-derivative of the cross-sectional area S of a vein is governed neither by energy dissipation in the water nor by plasticity, but by capillarity effects and salinity. A “vasodilator threshold” pd for water pressure pw in the veins is defined. Normally, Pw < Pd, then S has a stable value, the same for any orientation of the vein, and the microscopic temperature is uniform. The coefficient of permeability is proportional to (Pd-pw)−4, and thus a true Darcy law does not hold. As an application, the percolation of internal meltwater is studied; in an upper boundary layer about 2 m thick this meltwater flows upwards, because in the bulk of the glacier pw is very close to P1, whereas it is zero at the surface. When, exceptionally, pw > pd, S increases irreversibly. Whether it leads to the formation of “worm-holes” is discussed.

Detailed digital elevation models (DEMs) do not exist for much of the Greenland and Antartic ice sheets. Radar altimetry is at present the primary, in many cases the only, source of topographic data over the ice sheets, but the horizontal resolution of such data is coarse. Satellite-radar interferometry uses the phase difference between pairs of synthetic aperture radar (SAR) images to measure both ice-sheet topography and surface displacement. We have applied this technique using ERS-1 SAR data to make detailed (i.e. 80 m horizontal resolution) maps of surface topography in a 100 km by 300 km strip in West Greenland, extending northward from just above Jakobshavns Isbræ. Comparison with а 76 km long line of airborne laser-altimeter data shows that We have achieved a relative accuracy of 2.5 m along the profile. These observations provide a detailed view of dynamically Supported topography near the margin of an ice sheet. In the final section We compare our estimate of topography with phase contours due to motion, and confirm our earlier analysis concerning vertical ice-sheet motion and complexity in ERS-1 SAR interferograms.

For Glaciar Santa Rosa, in the Cordillera Raura, Peru, the extent and volume are determined for 1980 and 1961 from topographic mapping, and for two undated earlier stages LME and MAX from moraines. The regime of the glacier in terms of ice flow and mass balance is evaluated for the 1977–83 period. More than half of the water discharge from the glacier is not renewed by precipitation but supplied by the ice thinning. The rate of surface lowering of 2 m a−1 liquid-water equivalent corresponds to an energy amount for melting of 22 W m−2. Sensitivity analyses of Glaciar Yanamarey in the nearby Cordillera Blanca show that such a decrease of energy available for melting could be produced by a cloudiness increase of one-tenth, an air temperature decrease of more than 2°C, a decrease in specific humidity of 1 g kg−1, or some combination of heat-budget processes. Such changes in the atmospheric environment would be required to stabilize the glacier at its recent volume. As another indication of the recent imbalance, the maximum volume flux is found about 100 m below the equilibrium-line altitude. Under continuation of recent climatic conditions, the glacier may disappear within a few decades. These inferences coorroborate a recent assessment for the neighboring Cordillera Blanca.

Radio-echo soundings provide an effective tool for mapping the thermal regimes of polythermal glaciers on a regional scale. Radar signals of 320–370 MHz penetrate ice at sub-freezing temperatures but are reflected from the top of layers of ice which are at the melting point and contain water. Radar signals of 5–20 MHz, on the other hand, see through both the cold and the temperate ice down to the glacier bed. Radio-echo soundings at these frequencies have been used to investigate the thermal regimes of four polythermal glaciers in Svalbard: Kongsvegen, Uvérsbreen, Midre Lovénbreen and Austre Brøggerbreen. In the ablation area of Kongsvegen, a cold surface layer (50–160 m thick) was underlain by a warm basal layer which is advected from the temperate accumulation area. The surface ablation of this cold layer may be compensated by freezing at its lower cold-temperate interface. This requires that the free water content in the ice at the freezing interface is about 1 % of the volume. The cold surface layer is thicker beneath medial moraines and where cold-based hanging glaciers enter the main ice stream. On Uvérsbreen the thermal regime was similar to that of Kongsvegen. A temperate hole was found in the otherwise cold surface layer of the ablation area in a surface depression between Kongsvegen and Uvérsbreen where meltwater accumulates during the summer (near the subglacial lake Setevatnet, 250 m a.s.l.). Lovénbreen w as frozen to the bed at the snout and along all the mountain slopes but beneath the central part of the glacier a warm basal layer (up to 50 m thick) was fed by temperate ice from two cirques. On Austre Brøggerbreen, a temperate basal layer was not detected by radio-echo soundings but the basal ice was observed to be at the melting point in two boreholes.

Stresses and velocities at depth are calculated across Jutulstraumen, an ice stream in Dronning Maud Land, draining about 1% of the Antarctic ice sheet. The force-balance study is based on data from kinematic GPS measurements on three strain nets, each consisting of 3 × 3 stakes. The maximum measured velocity is 443 m a−1 and the velocity variation over short distances is large compared with studied ice streams in West Antarctica. The surface topography together with the measured velocities across the profile indicate that the bottom topography has a great influence on the flow direction, even where the ice thickness is more than 2000 m. The basal shear stresses are calculated as 180, 227 and 146 kPa in the three Strain nets, while the corresponding driving stresses are 180, 122 and 111 kPa (±5%). The heat produced by sliding and internal deformation is sufficient to keep the base at the pressure-melting point. The annual basal melting is estimated to be about 60 mm. Investigations on the effect of temperature softening show that the flow parameter’s influence on the effective strain rate is more important than the flow parameter’s direct softening in the flow low alone. The mass flow calculated by the force-balance method is between 87 and 96% of pure plug flow and total discharge is calculated to be 13.3 ± 10 km3a-1.

The effects on very low-frequency surface-impedence measurements of lateral variations commonly found in ice environments have been measured and modelled numerically using die quasi-static two-dimensional boundary-element method. Results indicate that surface-impedance measurements made in the vicinity of crevasses oriented perpendicular to the plane Of incidence, and those made in the vicinity of moraines and melt streams, can all show significant changes to the measured apparent resistivity. It is, therefore, misleading to use such measurements in the interpretation of ice depth.

Following construction of a glacial ice runway on the Ross Ice Shelf, Antarctica, and prior to flight operations, the runway was proof-rolled. The proof exercise was designed to simulate typical heavy aircraft. Initial testing produced numerous brittle surface failures in the runway ice. Thin sections of ice cores taken from the failed areas showed large crystals (с axis vertical) of clear, blue ice with long, vertical bubbles, indicative of ice formed directly from meltwater. Uniaxial unconfined compression tests on core samples were used to compare runway ice strength with published data for polycrystalline laboratory ice. Since the frequent failure of surface ice had not been expected, it was critical to understand the formation and mechanical properties of the weak ice to prevent its occurrence in the future and to strengthen the existing problem areas. We discuss the likely scenarios for development of weak ice on the airstrip and the physical properties of this type of ice. Also, the procedure used to repair successfully the runway surface is described, which culminated in test flights, followed by full flight operations.

Seismic-reflection methods were used to determine the ice thickness and basal topography of Jakobshavns Isbræ, a large, fast-moving ice stream/outlet glacier in West Greenland. A method of data analysis was developed which involves the pointwise migration of data from a linear seismic array and a single explosive source; the method yields the depth, horizontal position and slope of the basal reflector. A deep U-shaped subglacial trough was found beneath the entire length of the well-defined ice stream. The trough is incised up to 1500 m into bedrock, and its base lies 1200–1500 m below sea level for at least 70 km inland. Center-line ice thickness along most of the channel is about 2500 m, or about 2.5 times that of the surrounding ice sheet. This prominent bedrock trough was not apparent in existing radio-echo-sounding data. Reflection coefficients indicate that much of the basal interface is probably underlain by compacted, non-deforming sediment. The large ice thickness, coupled with relatively steep surface slopes, leads to high basal shear stresses (200–300 k Pa) along the ice stream. The large shear stresses and lack of a deformable bed imply that internal deformation plays a dominant role in the dynamics of Jakobshavns Isbræ.

A finite-difference numerical model is used to simulate the temperature profile at the center of an ice sheet throughout the course of a glaciation. The ice sheet is gradually built to a thickness of 3000 m over about 10 000 years, starting on permafrost. A geothermal heat flux is applied at large depth. For an initial surface temperature of –12.5°C, our model shows that basal melting occurs 72000 years after the onset of the glaciation. The important parameters determining the basal temperatures are the initial temperature of the ice and substrate, the rate of downward advection of cold ice and, to a lesser extent, the thickness of the ice sheet. The growth history of the ice sheet does not significantly influence the time at which basal melting occurs. Our results show the possibility that the central parts of the continental ice sheets were cold-based for a significant part of their existence. Heating due to the geothermal heat flux cannot account for basal melting during most or all of a glacial cycle. These results may help to explain the existence of preserved land forms under the ice sheets.

40Ar/39Ar ages of most single ice-ratted amphiboles from Heinrich layer 2 (H2) from a core in the Labrador Sea, a core in the eastern North Atlantic and a core in the western North Atlantic range from 1600 to 2000 Ma. This range is identical to that for K/Ar ages from the Churchill province of the Canadian Shield that outcrops at Hudson Strait and forms the basement of the northern part of Hudson Bay. The ambient glacial sediment includes some younger and older grains derived from Paleozoic, Mesoproterozoic and Archean sources, but still the majority of the amphiboles have ages in the 1600–2000 Ma interval. The Ca/K ratios of these 1600–2000 Ma old amphiboles, however, have a bimodal distribution in contrast with the uniformity of the Ca/K ratios of H2 amphiboles. This indicates that 1600–2000 Ma old amphiboles of the ambient sediment were derived from an additional Early Proterozoic source besides Churchill province. In H2, Churchill-derived grains constitute 20–40% of the ice-rafted debris (IRD). The fraction in the ambient glacial sediment is 65–80%. Results presented here are consistent with the hypothesis that Heinrich events were produced by a sudden intensification of the iceberg discharge through Hudson Strait that mixed, in the North Atlantic, with icebergs that continued to calve from other ice sheets. The shift from mixed sources in the background sediment to a large dominance of Churchill province grains in H2 indicates that, even if calving of other ice sheets intensified during the Heinrich episode, the increase in the iceberg discharge via Hudson Strait from the Hudson Bay drainage basin of the Laurentide ice sheet was by far the largest.

Measurements of change in length of a borehole, or displacements of poles and of subglacial Water pressure were combined with drainage tests and electrode tests in boreholes, in experiments near the northern margin of Gornergletscher, Valais, Switzerland, at the upstream side of an overdeepening. The measurements suggest that the subglacial drainage system consisted of discrete conduits at that location, presumably linked cavities on “clean” bedrock. Changes in subglacial water pressure were followed by variations in subglacial water storage, together with similar variations in elevation of a pole. The variations in subglacial water storage caused changes in cross-sections of subglacial passageways and thereby changed the frequencies of slug-test oscillations. Similar experiments conducted near the center line of the glacier revealed different subglacial conditions: impeded drainage through a sediment layer, and different depths of water levels in different boreholes. These results are discussed in relation to previous studies.

A theory of erosion, transport and deposition of unlithified sediments by glaciers is presented. It predicts the large-scale areal distribution of zones and rates of erosion and deposition in time and space through a complete glacial cycle, together with the resultant intensity of large-scale lineations (drumlins) which will be incised in the landscape. The theory also predicts the dispersal patterns of subglacial lithologies, together with the form of dispersal trains derived from distinctive sources and the vertical and horizontal distribution of lithologies within a till. It predicts major erosional discontinuities within tills and the formation of boulder pavements. It suggests that the dominant proportion of the lowland tills produced by Pleistocene mid-latitude ice sheets was generated by subglacial deformation and explains why they are predominantly fine-grained.

The theory is based on an analysis of glacier-dynamic processes and therefore can be used to infer the dynamic behaviour of former ice sheets from the distribution of tills and their lithologic composition.

A study of sea ice in the northern Weddell Sea was done, relating the ice motion, determined using an array of satellite-tracked drifters, deployed into ice floes, to parameters describing the nature of the ice cover, obtained from an analysis of Advanced Very High Resolution Radiometer (AVHRR) imagery. It was found that the ice motion was predominantly wind-driven, responding to the passage of low-pressure systems across the area. The correlation length of the strain field over the entire measurement period was around 200 km. At high concentrations the ice responded as a rigid body with coherent motion, but below a concentration of around 93%, differential motion occurred. The nature of the ice motion was found to depend upon the lead parameters, with low values of pure convergence and divergence and larger values of vorticity and deformation of the ice field. The vorticity was found to be well correlated with the atmospheric pressure, with a time lag of less than 3 h, implying an almost instantaneous response of the ice cover to meteorological forcing.

Two large glacier tongues, which extend substantially across the coastline of King George V Land in East Antarctica, have been studied by remote sensing (synthetic aperture radar, JERS-1). The tongue of Mertz Glacier is in a state of advance, while the Ninnis Glacier tongue is retreating. Some more specific points are:

The relations between mass balance and meltwater refreezing were examined on the basis of glaciological observations carried out in summer 1993 on Xiao Dongkemadi Glacier, Tanggula Mountains, central Tibetan Plateau. On this glacier, a part of meltwaler refreezes at the snow/ice interface as superimposed ice. The amount of superimposed ice formation was determined by both meltwater supply and temperature condition of the glacier. Snow-layer thickness on the glacier ice body is less than 2 m, even in the higher accumulation zone. About 60% of meltwaler generated in the accumulation zone for the period May–September was trapped at the snow/ice interface by refreezing, and was not discharged out of the glacier. About 26% of accumulated snow to the glacier surface was replaced on the snow/ice interface by refreezing in the accumulation zone. These facts indicate that superimposed ice formation is quite significant for water retention in glaciers under low-precipitation conditions.

Intercept analysis of approximately bi-yearly vertical thin sections from the upper part of the GISP2 ice Core, central Greenland, shows that grain-size ranges increase with increasing age. This demonstrates that something in the ice affects grain-growth rates, and that grain-size cannot be used directly in paleothermometry as has been proposed. Correlation of grain-growth rates to chemical and isotopic data indicates slower growth in ice with higher impurity concentrations, and especially slow growth in “forest-fire” layers containing abundant ammonium; however, the impurity/grain-growth relations are quite noisy. Little correlation is found between growth rate and isotopic composition of ice.