The former glacier at the type locality of the “Gschnitz Stadial” of the Alpine Late-glacial chronology is interpreted from a paleoglaciological and paleoclimatological point of view. The equilibrium-line altitude, ice flux through selected cross sections and mass-balance gradients are calculated from reconstructed glacier topography. They are used to determine total net ablation and accumulation and precipitation under the assumption of steady-state. The former temperature at the ELA and temperature change is estimated using various glacier—climate models. Precipitation was less than one-third of today’s values, and summer temperature was roughly 10°C lower than today. The climate during the Gschnitz Stadial appears to have been cold and continental, and was more similar to full glacial conditions than to the Younger Dryas climate in the Alps. This is further evidence for an older age of the Gschnitz Stadial.

Three cirques in the Ferwall group, western Tyrol, Austria, which are characterized by distinct Late-glacial moraines and rock glaciers, are discussed. The morphology of the moraines and the depression of the equilibrium-line altitude suggest they were deposited during the Egesen Stadial (Younger Dryas), which can be subdivided into three substages. Rock-glacier formation was initialized during or after the Egesen II substage. They became inactive at the Pleistocene—Holocene transition. ELA values are 290–320 m lower than the Little Ice Age ELA during the Egesen I substage, 190–230 m lower during the Egesen II substage and 120 —160 m lower during the Egesen III substage. The lowering of the rock-glacier belt (discontinuous permafrost) during and after the Egesen II substage is about 400 m, indicating a mean annual air-temperature depression in the order of 3 K. During the Egesen I (earlyYounger Dryas), the climate seems to have been rather cold and wet with precipitation similar to present-day values. During later phases (Egesen II and III), the climate remained cold and became increasingly drier. The rise of the ELA during the Egesen I—III substages seems to have been mainly caused by a decrease in precipitation.

Nearly 9000 individual lineations (drumlins and flutes) from two ice- flow systems have been mapped over a 22 500 km2 area of northern Sweden by interpretation of aerial photographs.The lineations were classified according to their size and orientation to allow a quantitative analysis of their distribution. The two glacial lineation systems, separated in time by at least 100 000 years and two individual overriding ice sheets, cross-cut at a high angle. The older system occurs over the entire study area and is interpreted to have formed by prolonged sheet flow during Marine Isotope Stage 5d, while the youngest system is spatially restricted, marking the presence of a laterally constrained zone of faster ice flow, developed near the end of the last deglaciation. The effect of the younger ice flow was erosion and elimination of smaller lineations in the older system, while the larger drumlinoid forms survived more or less intact, thus creating an apparent increase in average drumlin size of the older lineation system. One possible explanation for the unaltered shape of these larger drumlins is that the glacier bed of the younger ice sheet only reached the pressure melting point in the lower parts of the terrain, thus eroding the smaller forms, while the larger forms protruded into colder parts of the ice mass and were protected from erosion and deformation. There was no evidence of the deformation of older drumlins into barchanoid shapes or the development of lee-side tail features on the distal side of pre-existing lineations, as has previously been described for crossing drumlin systems. The lack of any such deformational features suggests that the younger lineation system was formed by net erosion, rather than deformation or deposition.

Along the northern part of the western coast of Sweden, the landscape is characterized by joint-aligned valleys with intervening rock hills. Precambrian bedrock is exposed over large areas and the Quaternary cover is mainly restricted to the bottom of narrow valleys. The well-exposed bedrock exhibits many glacial forms, but recently the major landforms have been interpreted as exhumed sub-Mesozoic etch forms. To gain insight into the relative roles of different processes in shaping the present forms, we studied a granite area where the rock hills contain fractures and narrow clefts with weathering forms and saprolite remnants. Marks of glacial erosion are found only in the upper parts of the clefts, where erosion by subglacial meltwater has exploited weathered preglacial sheeting structures to form furrows with concave cross profiles (p forms). On the summit surfaces, glacial erosion has been of greater importance, but the roche moutonnees found here are also structurally controlled. The main impact of glacial erosion is plucking on the order of tens of metres. Thus, landscapes labelled “areally scoured glacial terrain” can, at some locations, also be described as stripped and slightly incised etch surfaces.

In eastern Ireland, subglacial bedforms including drumlins and Rogen moraines were modified by headward erosion along two ice streams which had overlapping flow tracks. The ice streams, which had tidewater termini, are dated by geochronometric and morphostratigraphic methods to <15.014 C kyr BP (Castleblaney ice stream) and ~13.814C kyr BP (Armagh ice stream). Bedforms along ice-stream tracks show a morphological continuum which reflects a down-ice increase in the degree of modification by ice-stream activity (i.e. resulting in unmodified →remoulded/overprinted →crosscut →streamlined bedforms). The down-ice changes in bedform types are inferred to relate to changes in subglacial drainage and sediment-transport mechanisms. Bedform and sedimentary evidence suggest that discrete subglacial meltwater channels which developed up- ice changed in a down-ice direction to unchannelized flows which deepened towards the ice margin. Meltwater release from subglacial cavities, and produced by strain heating at sheared ice-stream margins, probably helped support ice-stream flow, which ended as the volume of subglacial meltwater discharge decreased. Dated millennial-scale cycles of ice activity may be related to instability at tidewater margins, followed by complex thermal and hydraulic responses within the ice mass.

Till lineations are widespread throughout the interior of Quebec—Labrador, Canada. We discuss a previously unrecognized lineation type, horned crag-and- tails, characterized by two till ridges at the lateral flanks of a bedrock crag. This lineation type appears to occur exclusively in the Ungava Bay lineation swarm and has not been reported from other glaciated shield areas. The morphology and spatial distribution of these peculiar landforms are analyzed in the context of Laurentide ice sheet dynamics in central Quebec—Labrador. We propose that a horned crag-and-tail ridge forms when the ice sheet is frozen to the summit of the crag, but basal melting occurs lower down on the crag. The lack of basal sliding at the summit of the crag inhibits till transport across the crag and prevents the accumulation of till in a direct lee-side position. The last phase of the formation is fossilization by basal freezing of the horned crag-and-tail system. This scenario requires that the ice sheet undergoes a transition from frozen-bed conditions to basal melting with frozen patches on topographical highs, and back to fully frozen-bed conditions. Horned crag-and-tails may thus indicate a unique sequence of basal thermal conditions over large areas of Quebec-Labrador.

Glaciogenic sediment up to 70 m thick is exposed in preglacial valley fills in south-central Alberta. The two youngest facies are comprised of glaciolacustrine sediments overlain by a subglacial till. Conformable relationships with the till and topographic relationships support a subglacial origin for the lake sediments. As these facies are restricted to the preglacial valley system, it is suggested that the preglacial valleys of south-central Alberta acted as natural cavities at the base of the Laurentide ice sheet. Sedimentology suggests these filled and drained often.

To determine the effectiveness of glacial erosion and the magnitude of cosmogenic nuclide inheritance from prior periods of cosmic-ray exposure, we measured the abundance of 10Be and 26Al in nine samples collected from bedrock, boulders and cobbles exposed by the retreat of Tumbling Glacier, Baffin Island, Canada. Most samples had nuclide concentrations so low that we were only able to set upper limits for nuclide abundance. Three boulders, two on a Neoglacial moraine of Tumbling Glacier that impounds Crater Lake and one on a roche moutonnée within the Neoglacial moraine loop, had nuclide abundances indicating no more than 900 yr of exposure at the surface. Three bedrock samples, striated by Tumbling Glacier and exposed by ice retreat within the last 20 yr, have similarly low nuclide abundances. One bedrock sample, covered by Tumbling Glacier ice for some part of the Holocene but not eroded, allows us to estimate crudely the duration of Neoglaciation at our sample site (about 5450 yr) and to provide a lower limit on the erosion rate of Tumbling Glacier (0.10 ± 0.03 mm a–1). We analyzed two cobbles collected from the tops of roches moutonnées at Crater Lake; one cobble had the equivalent of 3000 yr of exposure, the other < 900 yr.

In this paper, we explore geomorphological evidence allowing a first- order reconstruction of the extent and pattern of frozen-bed conditions under the last Fennoscandian ice sheet. We mapped relict landscapes, i.e. glacial landforms and subaerially developed ground surfaces predating the last ice sheet and marking sustained frozen- bed conditions, at four different spatial scales. At the ice-sheet scale, relict landscapes are most abundant between the Last Glacial Maximum ice divide and the elevation axis of the Scandinavian mountain range. The location of frozen-bed zones was mainly a function of dispersal centre location (low surface temperatures and small strain heating) and small ice thickness over the eastern flank of the mountain range. At the mesoscale (260 × 360 km map area), the pattern of relict surfaces is governed by inward-cutting ice-stream erosion.Topographical control was weak, but relation to flow pattern was strong, with the major frozen-bed zone located where ice flow was strongly divergent. At the regional scale (40 × 65 km map area) in hilly terrain, topographical control was strong with relict surfaces only appearing above a plane dipping in the up-ice direction. At the local scale (12 × 14 km map area), control by topography was likewise strong, but the detailed boundary pattern was irregular, with specific landforms occurring both up- and down-ice of frozen patches.

Modern terrestrial glaciers in the Canadian High Arctic range from polythermal to cold-based. Where polythermal glaciers override thick unconsolidated sediment, longitudinal compression and glaciotectonic thrusting produce thrust-block moraines. In contrast, the dominant geomorphic record of cold-based glaciers consists of lateral and proglacial meltwater channels. Geomorphic and sedimentary evidence indicates that late Quaternary fiord glaciers were also characterized by variations in basal thermal regime. Erratic dispersal trains and striated bedrock record the flow of warm- based ice during the Last Glacial Maximum. Emergent grounding-line fans and morainal banks, deposited during deglaciation, consist of heterogeneous glaciomarine deposits that record well-developed subglacial drainage and high sedimentation rates. However, in other fiords, subaqueous outwash and fine-grained glaciomarine deposits are absent and deglaciation is recorded by lateral meltwater channels graded to raised glaciomarine deltas, suggesting these glaciers were predominantly cold-based during retreat. Regionally, deglacial depocentres are located at pinning points within fiords and a prominent belt of glaciogenic landforms at fiord heads records stabilization of ice margins during early Holocene retreat, rather than the limit of late Quaternary glaciation. Collectively, these observations refute previous reconstructions which inferred a climatically controlled switch from cold- to warm-based thermal conditions in fiord glaciers during early Holocene deglaciation, and indicate that the dominant controls on thermal regime were glaciological.

The Ultevis plateau, northwestern Sweden, has a relief of less than 200 m, yet bears three different kinds of landscape, classified according to the degree of glacial erosion. The first type is restricted mainly to topographic highs and has almost entirely escaped erosion, despite complete and prolonged ice cover during the late Weichselian. The other two landscape types are distinguished depending on whether older land- forms have been completely erased or not. The latter two appear to have undergone erosion only briefly. The transitions between landscape zones are usually sharp, and specific boundary landforms occur. The Scandinavian ice sheet was cold-based in its central areas during its maximum. During the deglaciation, both the dry/wet-base boundary and ice margin migrated inwards, at different speeds. When the ice front retreated faster than the thermal transition zone, the wet-base marginal zone shrank and erosion was reduced or avoided. Where the wet-base zone was of limited longitudinal extent, as on the Ultevis plateau, conversion from a frozen to a thawed bed was incomplete, leaving a patchwork of preserved glacial and non-glacial morphologies.

Measurements made at John Evans Glacier, eastern Ellesmere Island, Nunavut, Canada in 1994 and 1996 provide new insight into the internal hydrology of polythermal glaciers. During the early part of each melt season, supraglacial waters enter the glacier via a crevasse field about 4 km from the terminus and are stored in a subglacial reservoir. Release of the water from the reservoir occurs initially via an artesian fountain on the glacier surface and by upwelling of waters through subglacial sediments at the terminus (event1). Channelization of the subglacial waters then occurs and water is discharged as an outburst flood (event 2), which releases a considerably larger volume of water than event 1. Thus, drainage of the subglacial reservoir follows a cyclical pattern in which discharge oscillations increase in amplitude over time. The cycle may end with complete reservoir drainage. The total volume of water released was much greater in 1994 than in 1996, primarily because the 1994 melt season was longer and warmer than the 1996 season. Interannual differences in the form of the outflow hydrographs, and in the extent and timing of connections between the subglacial reservoir and marginal melt streams, are linked to variations in the size and rate of growth of the subglacial reservoir. This hydrological behaviour may have important implications for the dynamics of polythermal glaciers.

The changing geometry and thermal structure of Scott Turnerbreen, a 3.3 km2 glacier located at 78° N in the Svalbard archipelago, is documented. A net mass balance of–0.58 m a–1w.e. is determined for the period 1936–93, by comparing a recent topographic survey with earlier maps. The thermal regime was investigated with multifrequency radar and borehole thermistors. Basal temperatures of–4.1° and –3.3°C were measured, and observed temperature gradients indicate that the entire bed is frozen. This interpretation is confirmed by continuous radar profiling, which demonstrates the absence of high-frequency scattering from temperate ice. However, with the reconstructed 1936 ice- thickness distribution, at least 2 km of the length of the glacier bed would be at the pressure-melting temperature. The 20th century mass-balance history of Scott Turnerbreen is likely to have been influenced by a surge occurring around 1930, which meant that the glacier was already in a state of disequilibrium before the abrupt climate perturbation marking the termination of the Little Ice Age. A significant loss of mass has been accompanied by a transition from inferred polythermal to entirely non-temperate thermal conditions. Current driving stress and velocity are very low, and the glacier has almost certainly fallen out of the surge cycle. Within 60 years, there has therefore been a wholesale transformation in the geometry, thermal structure and dynamics of Scott Turnerbreen.

The geomorphological effects of ice sliding over till, internal deformation of till and till sliding over bedrock are considered. Two questions are examined: (1) is the till-sheet flow unstable, i.e. is a layer of uniform thickness maintained or not, and (2) does the slip of till over bedrock cause amplification of relief of the bedrock? Such instabilities seem to be necessary to explain such features as drumlins and whaleback forms.

It is found that the answer to (1) and (2) depends on the position of the system in a parameter space, defined by the till rheology, and applied shear stress, the effective pressure at the ice—till interface, the thickness of ice and till and the wavelength of the instability. Two configurations are considered: one where the wavelength of the perturbation is much less than the the ice-thickness, which is related to the classical Nye—Kamb solution for flow over bumps; and one where the wavelength is much greater than the ice thickness, where the mechanics are described by the shallow-ice approximation. In both cases, substantial areas of parameter space, where till-sheet and bedrock modes are unstable, are found. The conceptually related Smalley—Unwin bifurcation is re-examined.

The physical mechanisms by which ice and till flows couple are examined. At very short wavelengths (~10 m), the ice is so rigid that it forces till waves to move at the ice velocity; while at long wavelengths (~1000 m), the flows become essentially uncoupled and till waves move at the kinematic velocity. At intermediate wavelengths (~100 m), high growth rates occur ; this is postulated to be the scale of drumlin seeding.

The occurrence of rocky and sedimentary glacier beds in the Swiss Alps was investigated using two approaches: (1) a detailed and systematic classification of glacier forefields in the completely revised Swiss glacier inventory; and (2) the application of an algorithm for estimating bed characteristics from glacier-inventory data. The algorithm is based on the assumption that the sediment balance of a glacierized catchment, i.e. the ratio between debris input from the surrounding rock walls and debris evacuation by the meltwater stream, determines the occurrence of rocky, sedimentary or mixed beds. The processes which govern the sediment balance within a glacierized catchment cannot be easily observed or measured in the field. However, the main factors reflecting them can be organized in the form of an erosion—sedimentation index. This index discriminates between glaciers eroding into bedrock and those building up thick sediment beds and, thus, allows for rough assessments concerning bed characteristics. Sedimentary and mixed sedimentary—rocky glacier beds clearly predominate in the Swiss Alps, but more detailed spatial analysis is required for better predictions.

This paper presents new data and interpretations of the structure, strength and behaviour of basal ice and the substrate of Suess Glacier, south Victoria Land, Antarctica, which is a small alpine dry-based glacier that has a basal temperature of—17°C. A tunnel excavated in the glacier revealed a substrate composed of frozen sand and gravel and a basal zone that was 3.8 m thick. From bottom to top, the basal zone was composed of 1.8 m of stratified, complexly deformed layers of ice and debris overlain by a 0.9 m thick layer of frozen sediment and a 0.8 m thick layer of discoloured ice lying immediately beneath clean glacier ice. Direct shear tests performed in the field on 36 samples show that the average peak shear strength of substrate samples was 2.53 MPa, which is almost twice as strong as the average value for basal ice (1.28 MPa) and the glacier ice samples (1.39 MPa). The direct shear tests suggest that the glacier substrate is unlikely to deform at the current temperature. However, observations of the structure and composition of the basal zone suggest that the substrate contains localized weaknesses that may reduce the peak strength sufficiently to permit bed deformation and entrainment. Two observed forms of weakness are ice lenses and layers in the permafrost and lenses of unsaturated permafrost, which may be inherited from the active zone of a former land surface or produced as pore ice sublimes into cavities which have formed at the glacier bed. Interpretation of data from this study suggests that glacier beds well below the freezing point can be deformed and eroded under certain conditions.

Fourteen subglacial debris samples have been recovered from the margins of, or beneath, Haut Glacier d’Arolla, Switzerland. The grain-size distributions of these sediments are presented and compared with each other as bivariate plots of percentage weight against (sieve-defined) particle size and log number of particles against log particle diameter. All of the samples recovered are composed of a broad range of clast sizes and approach self-similarity over the four orders of magnitude of grain-sizes analysed. Fractal dimensions range from 2.47 to 2.77. Sample intercomparison reveals the operation of at least two processes of textural evolution: the production of fines by in-situ weathering, interpreted in terms of abrasion associated with subglacial sediment deformation, and the loss of fines, interpreted in terms of eluviation by percolating subglacial meltwaters. These interpretations are supported and refined through comparison of the grain-size fractions gained (in the case of deformation) and lost (in the case of eluviation) with those fractions respectively generated in a laboratory-based simulation of sediment deformation and exiting the glacier suspended in the proglacial meltwater stream. While sediment deformation has the effect of increasing the fine fraction between 0 and 10ϕ and of raising the fractal dimension of undeformed sediments from 2.47 to 2.77, eluviation removes particles between 2 and 100, driving the fractal dimension of deformed sediments down from 2.77 to 2.54. These fractal dimensions are generally lower than those recorded at other comparable glaciers, consistent with the relatively low rates of sediment deformation inferred from other studies at Haut Glacier d’Arolla.

Over a two-year period, continuous measurements of temporal changes in tilt, conducted with a string of tilt meters in a borehole on Unteraargletscher, Bernese Alps, Switzerland, have been used to estimate the basal-motion component. This estimation is based on a comparison of the measurements with synthetic tilt curves, computed using a parameterization of a simplified flow field. The best agreement is found for a ratio of basal motion to forward motion due to ice deformation (slip ratio) equal to about 1.2. Measured tilt curves exhibit a number of different transient features. While an overall increase in tilt angle is observed at every tilt-meter location, two of the sensors recorded anomalous tilt behaviour. These anomalies are characterized by sudden and drastic variations in tilt. A particularly intriguing example of such short-term tilt variations was recorded with a tilt meter positioned 40 m above the bed during the 1997 summer melt season.

The subglacial drainage in a pronounced overdeepening of Grosser Aletschgletscher, Switzerland was investigated by borehole water-level observations and dye-tracer injections. In August 1990 and 1991, tracer was injected at the bottom of a borehole (depth 904 m and 710 m, respectively), and simultaneously in a nearby moulin. The moulin and borehole injections identified two different flow systems coexisting within the overdeepening. The moulin injection yielded short-lived, highly peaked break-through curves with high velocities, indicating a hydraulically efficient channelized system. The borehole-tracer return occurred in broad multiple peaks characterized by a striking diurnal periodicity and, in general, correlated inversely with discharge. Water level in the borehole experienced high diurnal variations in phase with discharge at the snout indicating “closed channel flow” over large parts of the glacier. We infer that the boreholes drained subglacially and were connected to a drainage system with significant long-term storage capacity. Release of labelled water was triggered by daily water-pressure cycles.

Several different types of laterally extensive debris bands occur along the western terminus region of the Matanuska Glacier, Alaska, U.S.A. An ice-bed process, which to our knowledge has not previously been recognized and described, forms the most common and most prominent type of debris band at Matanuska Glacier’s terminus. The debris bands are composed of one or several millimeter-thick laminations of silt-rich ice having much higher sediment content than that of the surrounding ice. Samples of these bands and their surrounding englacial ice have been analyzed for anthropogenic tritium (3H), oxygen-18 (δ18O), and deuterium (δD).We interpreted the laminated, silt-rich debris bands as basal fractures, along which silt-laden, glaciohydraulically supercooled and pressurized waters flowed, healing the fractures by ice growth. This process is analogous to the inward growth of hydrothermal quartz from the sides of an open fracture.