Dynamic surficial changes and changes in the position of the firn line and the areal extent of Hofsjökull ice cap, Iceland, were studied through analysis of a time series (1973–98) of synthetic-aperture radar (SAR) and Landsat data. A digital elevation model of Hofsjökull, which was constructed using SAR interferometry, was used to plot the SAR backscatter coefficient (σ°) vs elevation and air temperature along transects across the ice cap. Seasonal and daily σ° patterns are caused by freezing or thawing of the ice-cap surface, and abrupt changes in σ° are noted when the air temperature ranges from ∼−5° to 0°C. Late-summer 1997 σ° (SAR) and reflectance (Landsat) boundaries agree and appear to be coincident with the firn line and a SAR σ° boundary that can be seen in the January 1998 SAR image. In January 1994 through 1998, the elevation of this σ° boundary on the ice cap was quite stable, ranging from 1000 to 1300 m, while the equilibrium-line altitude, as measured on the ground, varied considerably. Thus the equilibrium line may be obscured by firn from previous years. Techniques are established to measure long-term changes in the elevation of the firn line and changes in the position of the ice margin.

Previous models of ice–till deformation near subglacial channels or cavities neglect the fact that the motions of the two materials are coupled, and thus the interface between ice and till may not remain stationary. Here, we analyze in succession two models which address the effect of such coupling via specification of appropriate continuity conditions for stress and velocity across the interface. The modelled scenario is that of a shallow channel–cavity, with its long axis parallel to the principal ice-flow direction, overlying actively deforming till sediments. By applying asymptotic techniques, we investigate how the pattern and velocity of the creep flow depend generally on the ratio between the ice and till viscosities, and on the deforming-till thickness. A more sophisticated, non-linear rheology for till sediments is then introduced. It reveals that the two-way interaction between water percolation and deformation in the till will enhance the localization of sediment flow near the channel margins. The length scale over which transition of effective stress in the till takes place — from its relatively high, far-field value to the low, channel value — is found to depend critically on a dimensionless permeability parameter (Λ). In any case, coupled deformation causes sediment (and ice) flow towards the channel, subsidence of the ice–till interface just outside the channel, and extension of the area over which the ice is in contact with till. Apart from having direct implications for subglacial sediment transport, these results indicate that coupled deformation can contribute significantly to the spatial evolution of stress distribution under a glacier, and thus its incorporation into future sliding and drainage theories for a soft bed should be considered essential.

A lightweight, portable drilling system for coring up to 500 m depths has been developed and field-tested. The drilling system includes four major components: (1) an electromechanical (EM) dry-hole drill; (2) an ethanol thermal electric drill; (3) a drill set-up with a 500 m cable capacity; and (4) a controller unit. The system may be switched quickly from a dry-hole EM drill to an antifreeze thermal electric drill. This lightweight system makes ice-core drilling more cost-efficient, and creates a minimal environmental impact. The new EM drill, which recovers 100 mm diameter, 1 m long pieces of ice core, is 3.2 m long and weighs 35 kg. This drill and the drilling set-up were recently tested at the Raven (former Dye 2) site, southern Greenland, where a core was recovered to 122 m. The thermal drill is 2.9 m long and weighs 25 kg. It produces 100 mm diameter, 2.1 m long pieces of ice core, and was tested to 315 m in Franz Josef Land, Eurasian Arctic. The drilling set-up with a 250 m cable weighs about 100 kg (or 128 kg for 500 m of cable). After minor adjustments this drill system retrieved cores of better quality than those recovered by other drill systems under similar glaciological conditions. After adjustments to optimize its performance, the drill retrieved 5.25 m of core per hour over the depth range 0–21 m.

At Engabreen, Norway, an instrumented panel containing a decimetric obstacle was mounted flush with the bed surface beneath 210 m of ice. Simultaneous measurements of normal and shear stresses, ice velocity and temperature were obtained as dirty basal ice flowed past the obstacle. Our measurements were broadly consistent with ice thickness, flow conditions and bedrock topography near the site of the experiment. Ice speed 0.45 m above the bed was about 130 mm d−1, much less than the surface velocity of 800 mm d−1. Average normal stress on the panel was 1.0–1.6 MPa, smaller than the expected ice overburden pressure. Normal stress was larger and temperature was lower on the stoss side than on the lee side, in accord with flow dynamics and equilibrium thermodynamics. Annual differences in normal stresses were correlated with changes in sliding speed and ice-flow direction. These temporal variations may have been caused by changes in ice rheology associated with changes in sediment concentration, water content or both. Temperature and normal stress were generally correlated, except when clasts presumably collided with the panel. Temperature gradients in the obstacle indicated that regelation was negligible, consistent with the obstacle size. Melt rate was about 10% of the sliding speed. Despite high sliding speed, no significant ice/bed separation was observed in the lee of the obstacle. Frictional forces between sediment particles in the ice and the panel, estimated from Hallet’s (1981) model, indicated that friction accounted for about 5% of the measured bed-parallel force. This value is uncertain, as friction theories are largely untested. Some of these findings agree with sliding theories, others do not.

An ice-thickness distribution model based on physical ice classes is formulated. Pack ice is subdivided into open water, two different types of undeformed ice, and rafted, rubble and ridged ice. Evolution equations for each ice class are formulated and a redistribution between the ice classes is calculated according to a functional form depending on the ice compactness, thickness and velocity divergence. The ice-thickness distribution model has been included in a coupled ice–ocean model, and numerical experiments have been carried out for a simulation of the Baltic Sea ice season. The extended ice classification allows separation of thermally and mechanically produced ice. Inherent thermodynamic growth/melting rates of the ice classes can be introduced into the model, giving a more detailed seasonal evolution of the pack ice. In addition, the model provides more information about the surface properties of pack ice.

Numerical experiments for the Baltic Sea show that both the sub-basin and inter-basin ice characteristics were realistically simulated by the model. Deformed-ice production was related to storm activity. Most of the deformation was produced in the coastal zone, which is also an important region for thermodynamically produced ice because of the ice growth in leads. The modelled mechanical growth rates of ice were 0.5–3 cm d−1 on a basin scale, close to the thermodynamic ice-production rates. The deformed-ice fraction was 0.2 in mid-winter and increased to 0.5–1.0 during spring.

Five ice cores have been retrieved from a transect close to the terminus of Glacier de Tsanfleuron, Switzerland. The cores extend from the ice surface to the glacier bed, and are 3.5–44.8 m long. Stratigraphic logging based on bubble size and density reveals the presence of a highly metamorphosed basal ice layer, about 10 m thick, from which all traces of bubble-rich ice have been removed. This bubble-poor ice, which corresponds closely with clear-facies ice observed in cavities beneath numerous temperate-based glaciers, contrasts with the overlying bubble-rich or bubble-foliated englacial ice and the underlying debris-rich and bubble-free dispersed-facies basal ice.

Down-core patterns in major-ion composition, stable-isotope composition and total gas content and composition are generally consistent with formation of clear-facies ice by deformation-related metamorphism of bubbly, englacial ice. In addition, isotopic data suggest that storage of downward-percolating meltwaters occurs close to the upper surface of the clear-facies ice layer, perhaps reflecting a local variation in ice permeability across the transition from englacial to clear-facies ice. Enrichment in crustally derived ionic species is noted in the lowermost decimetres of the debris-free, clear-facies ice that immediately overlies debris-rich dispersed-facies basal ice. This ionic enrichment in debris-free ice is interpreted in terms of active inter-granular meltwater flow within some decimetres of the glacier bed.

An enhanced composite Advanced Very High Resolution Radiometer (AVHRR) image is used to map flow stripes and rifts across the Ross Ice Shelf, Antarctica. The patterns of these flow-related features reveal a history of discharge variations from the ice streams feeding the eastern part of the shelf. The most profound variations are visible in the track of rifts downstream of Crary Ice Rise, flow-stripe bends to the west of this ice rise and adjacent to Steershead ice rise, and changes in the northern margin of Ice Stream B. The track of rifts downstream of Crary Ice Rise indicates that the ice rise has existed for at least 700 years. The character of this track changes about 350 km downstream, indicating a rearrangement of flow patterns about 550 years ago. The large bulge in the flow stripes to the west of Crary Ice Rise is shown in detail, with bent flow stripes extending for several hundred kilometers along flow; this feature formed from the south, possibly due to a change in the discharge of Ice Stream A. The AVHRR image documents a complex history associated with the shutdown of Ice Stream C, with changes in the margins of Ice Stream C and the northern margin of Ice Stream B, and the grounding of Steershead ice rise with an associated bending and truncation of flow stripes. Landsat imagery shows a region that appears to be actively extending just downstream of the ice rise, as the shelf continues to respond to recent changes in ice-stream discharge. We present a four-stage flow history which accounts for the features preserved in the ice shelf.

A mechanical model with circular symmetry is examined to test the hypothesis that the Martian ice caps are composed of flowing water ice, together with some rock debris. In contrast with most or all previous models, no assumption of a steady state is made. Instead the accumulation and ablation is assumed to be insignificant, and it is suggested that after a sufficient time the profile would have settled down to a particular collapsing form calculated by Halfar (1983). Higher modes of flow would have decayed relatively quickly. To calculate the time constant, it is necessary to consider carefully the distribution of temperature with depth. The time constant is sensitive to the grain-size, which is assumed to be 1–10 mm and is a significant unknown, as is also the effect of preferred crystal orientation. Apart from this, the main uncertainty is the value of the upward heat flux. With a heat flux of 30 mW m−2, the water-ice hypothesis is consistent with an age of about 107 years for both the north and the south polar caps, the north cap being the younger by a factor of about 7.

A suite of subglacial water-pressure records from the 1996 summer field season at Trapridge Glacier, Yukon Territory, Canada, discloses a hydraulic event that cannot readily be explained by known forcings. We suggest that these records indicate covert failure of the pressure sensors caused by at least one large water-pressure pulse. The sign and magnitude of the pulse appears to have varied spatially and the pulse duration was less than the 2 min sampling interval of our data loggers. Laboratory experiments support this interpretation and indicate that the pulse magnitude exceeded 900 m of hydraulic head, roughly 15 times the ice-overburden pressure. Within glaciers, large water-pressure pulses can be generated when abrupt ice motion changes the volume of the subglacial hydraulic system.

Time-dependent ice-sheet modelling of a 176 000 km2 area in western Dronning Maud Land, East Antarctica, provided information on the ice sheet’s response to six climate-change scenarios. Another experiment was done to study changes in ice thickness, flow and basal temperature conditions between the present ice configuration and a simulated maximum palaeo-ice sheet. The input to the model included new datasets of bed and surface topography compiled for this study. The results of the six climate-change experiments, including a 0.5°C per century global-warming scenario, show that the ice sheet has a robust behaviour with respect to the different climate changes. The maximum change in ice volume was <5% of the initial volume in all climate runs. This is for only relatively short-term climate changes without major changes in global sea level, and also a simulated ice sheet without an ice shelf. The modelled long-term response time of the ice sheet, 20 kyr or more, indicates that the ice sheet may still be adjusting to the climate change that ended the Last Glacial Maximum. In the maximum palaeo-ice-sheet simulation, with a 5°C climate cooling and the grounding line located at the continental-shelf margin, ice thickness increased drastically downstream from the Heimefrontfjella mountain range but remained basically unaffected on the upstream polar plateau. Compared to present conditions, complex changes in basal temperatures were observed. The extent of areas with basal melting increased, for example in the deep trough of the Veststraumen ice stream. Areas at intermediate elevations in the landscape also experienced increased basal temperatures, with significant areas reaching the melting point. In contrast, high-altitude areas that today are clearly cold-based, such as around Heimefrontfjella and Vestfjella and the Högisen dome, experienced a 5–10°C decrease in basal temperatures in the palaeo-ice-sheet reconstruction. The results suggest that the alpine landscape within these mountain regions was formed by wet-based local glaciers and ice sheets prior to the late Cenozoic.

The depth of reflecting layers in Arctic ice sheets has been determined by electromagnetic echo sounding, using a varying distance between transmitter and receiver to determine the radar wave velocity. The depth of the radar reflecting layers is compared with a profile of electrical conductivity measurements (ECMs) from the Greenland Ice Core Project (GRIP) ice core, in order to determine the velocity of the radar waves in the ice cap. By using several reflecting layers, it is possible to isolate the firn correction of the wave velocity and to estimate the accuracy of the calculated electromagnetic wave velocity. The measured firn correction is compared with the correction calculated from the density profile, and a comparison between the depth profiles of ECM and radar based on the corrected electromagnetic wave velocity is presented. This profile shows that acid layers, which originate from major volcanic eruptions, show up as reflecting radar horizons.

The ability of synthetic-aperture radar (SAR) to record change in glaciers and icefields on seasonal to interannual time-scales is useful in maritime mountain regions where visible data are often obscured by clouds. A time series of RADARSAT and second European Remote-sensing Satellite (ERS-2) SAR images shows dramatic changes related to the onset and progression of glacier melting on the Juneau Icefield, southeast Alaska, U.S.A. Four “radar glacier zones” are interpreted from SAR images as snow that is dry during winter, an early snowmelt (M), a second phase of snowmelt (M2) and bare ice. These zones develop every year on the Juneau Icefield and other mid-latitude glaciers. Summer field observations on the Juneau Icefield during 1997 and 1998 constrain interpretations of the C-band radar glacier zones. Of the two zones that occur in melting snow (M, M2), M has low radar backscatter coefficients (σ° < −12), in contrast to the higher backscatter coefficients (σ° > −12) of the subsequent M2 zone. Snow moisture and surface roughness at the scale of the radar wavelength (5.6 cm) were measured to characterize the melt zones. Correlation length, wetness and grain-size in the two zones are not distinguishable in the late ablation season. Mean surface roughness, due to the presence of suncups, is higher in the M2 zone than in the M zone and probably causes the higher backscatter due to greater scattering.

It is of practical importance to have a description on time-scales of 1–100 years of the relationship between the mass imbalance of an ice sheet and its rate of change of thickness. In this paper, a linearized treatment of the relationship is described. Closed-form expressions are derived that relate the time-variant density in an isothermal firn layer to the fluctuations in accumulation rate and density that occur at the surface. These expressions are used to provide a spectral description of the contribution of surface accumulation and surface density fluctuations to the rate of change of thickness of an ice sheet. Using these, the contribution of firn densification to the variability of ice-sheet thickness is examined as a function of the time interval over which the ice sheet is observed. This contribution is illustrated for sites in Antarctica and Greenland. It is concluded that it is important to give greater attention than hitherto to the spatial scale of accumulation fluctuations if satellite observations of ice-sheet elevation change are to be used to estimate ice-sheet imbalance over short time intervals.

Surface flow in a 10 000 km2 expanse of the onset area of Ice Stream D, West Antarctica, was measured by repeat, precise global positioning system surveys over a 1 year interval. The pattern of velocity and strain rate shows the development of Ice Stream D, the major flow into which originates south of Byrd station and follows the course of a deep bed channel. Plotting of the driving stress vs the ratio of velocity and ice thickness identifies the onset of streaming flow (roughly 140 km downstream of Byrd station) as a transition between deformation flow and sliding flow. Along the kinematic center line of the developing ice stream, the ice rheology is linear at stresses below 0.6 bar, and appears temperate at the base well before the onset of streaming is reached. The onset corresponds to a maximum driving stress of 0.8 bar. It occurs downstream of a slight increase in longitudinal strain rate where stronger along-flow lineations are apparent in Landsat imagery, and after the ice has passed the center of an overdeepening in the bed channel. No current deviation from equilibrium is detected in this region, but a set of flow stripes misaligned with present flow indicates significant changes in flow have occurred in the past.