Glacial quarrying remains enigmatic despite being long recognized as a primary, perhaps the dominant, process by which glaciers erode bedrock. The rate-limiting process appears to be subglacial rock fracture due to ice-induced mechanical stresses.

To study this erosional process, a simple model of quarrying is developed for a glacier sliding over a periodic series of bedrock steps. Consideration of the balance of forces at the ice/rock interface and of the rate of cavity closure permits evaluation of ice-induced stresses on bedrock surfaces. The resulting stress distribution where ice loads are most concentrated near the corner of bedrock steps is evaluated using a simple elastic solution for the state of stress in a loaded quarter-plane. It is then used to determine whether fractures in the rock will grow, and to estimate the rate of progressive crack growth. Based on these crack-growth rates, an index of the quarrying rate is then calculated as a function of glaciological variables effective pressure and sliding velocity—and various bed parameteres. Considerable incentive exists for further analysis of quarrying, and for seeking field data to test the model.

Like many mountain ranges, the Coast Mountains of British Columbia, Canada, have undergone both local and ice-sheet glaciation. Effects of ice sheets are concentrated along major valleys and on adjacent spurs and passes which carried strong flows of diffluent ice. The major valleys are broad glacial troughs with frequent rock basins. Their slopes are broken into rounded, steep-sided bosses whalebacks abraded on all sides: they are of the order of 100 m to 1 km long, and 10 m high. In the southern Coast Mountains, the distribution of these whalebacks is consistent with a proposed pattern of former ice streams 1.0–2.1 km thick, within the Cordilleran ice sheet. They are best developed where geological structures parallel the valley and thus the former ice-flow direction, but they are found on a range of lithologies and some are transverse to structure. The whalebacks provide an impression of glacial streamlining, and occasionally grade into rock drumlins. Roches moutonnées are rare in the major troughs.

It is hypothesised that these whalebacks and rock drumlins develop under ice streams of Greenland or East Antarctic type, sliding rapidly over bedrock and exploiting rock weaknesses to produce streamlined features. Lee slopes are abraded when thick ice suppresses bed separation, even with rapid flow; basal ice of low viscosity would aid this suppression. Water pressures under the ice streams may have remained high, so that lee-side plucking was rare; such plucking is most likely where pressure fluctuates dramatically, and especially when lee cavities under active ice reach atmospheric pressure.

A glacier moving by deformation of subglacial sediments will tend to exhaust its own sediment supply, unless new sediment is generated subglacially. We explore the potential for deforming sediments to overcome this difficulty and replenish themselves by abrading their beds. We review abrasion experiments and theory for brittle materials and conclude that a theoretical calculation of abrasion is not possible yet. Instead, we use fault-gouge production data to estimate a likely upper bound to abrasion rates, and conclude that sufficient erosion to maintain a steady deforming-layer thickness is difficult to achieve, and will only be possible if the substrate is very soft and if there is a moderate rate of slip at the base of the deforming layer. Slow abrasion, which can leave a geologic signature, is possible under most deforming layers that are deforming at the sediment/bedrock interface.

This paper determines the provenance of solute in bulk meltwaters draining Haut Glacier d’Arolla, Valais, Switzerland, during the 1989 ablation season. Dissolved species are partitioned into components derived from sea salt, acid aerosol, dissolution of atmospheric CO2, and lithogenic sources, namely carbonates, sulphides and aluminosillicates. A major conclusion is that trace geochemically reactive minerals in the bedrock contribute the bulk of the solute found in runoff. Seasonal changes in solute provenance and in the dominant chemical weathering process are observed. Whereas the chemical weathering of aluminosillicate minerals by carbonation reactions remains relatively constant during the ablation season, the chemical erosion of carbonates shows distinct seasonal variations, reflecting changes in the nature of the subglacial drainage system. Subglacial drainage structure and bedrock type are key controls on the extent of subglacial chemical weathering.

The mechanical configuration of cavities associated with clasts is discussed, and how this affects the water pressure underneath the clast. The range of permissible water pressure is determined by flow parallel to the bed rather than by flow towards the bed. The actual pressure is controlled by hydraulic connections to areas away from the clast. The degree of cavitation provides an additional degree of freedom which adjusts to ensure that horizontal and vertical forces respect the Coulomb friction condition.

The observed ratio of removal of material by (i) scratches gouged by large clasts and (ii) polishing by debris suggests that the latter is the dominant form of bedrock erosion. We examine this with a multi-stage breakdown model of glacial debris, which includes active erosion by fines. These models suggest that a mother clast can, through the action of its daughter debris, erode several times its own volume of bedrock.

Till-sliding over bedrock is likely to be a potent source of scouring. By proposing a till/bedrock sliding law, we investigate the kinematics of sliding till bodies. Ice flow naturally tends to thin and extend till cover, even in the absence of longitudinal gradients in the applied stress. Thicker till cover has an increased effective pressure at its base, a lower sliding velocity and, for larger thicknesses, a decrease in sediment flux with thickness. This implies backward-moving kinematic waves and shocks. It is suggested that this is related to the blunt upstream faces of drumlins, and that drumlinization can be a consequence of debris sliding over bedrock.

Sediment yields and glacial erosion rates are evaluated for four Norwegian glaciers during the years 1989-93. Annual erosion rates were determined from measurements of sediment load and water discharge in glacial meltwater rivers. The mean sediment yield and the corresponding erosion rate of the valley glaciers Engabreen and Nigardsbreen were found to be 456 t km−2 year−1 (0.168 mm year−1) and 210 t km−2 year−1 (0.078 mm year−1), respectively. A small and slow-moving cirque glacier Øvre Beiarbre yielded a rate of 482 t km−2 year−1 (0.178 mm year−1), and the sub-polar Svalbard glacier Brøggerbreen yielded 613 t km−2 year−1 (0.226 mm year−1). The erosion rates are low compared to glaciers elsewhere. There are also considerable variations in sediment yields at each glacier from year to year. However, different factors are found to control the variability on each individual glacier. Analysis of the relationship between water discharge and sediment concentration in meltwater rivers suggests that changes in subglacial drainage systems cause variations in sediment availability and the way sediments are melted out from the ice. When water pressure drops, the drainage system in fast-moving, thick valley glaciers deforms at a more rapid rate than in thin, slow-moving ones. New volumes of debris-laden ice are thus more readily available for melting when water pressure next increases. Beneath the thin, slow-moving Øvre Beiarbre, single years with high transport rates and evacuation of sediment are followed by periods of low availability lasting for 2 years or longer. It is suggested that this pattern results from exhaustion of sediment in a stable drainage system, with more sediment becoming available when the position of the subglacial drainage system is changed.

The formation of ice-cored moraines and push moraines is discussed in the light of glacier thermal regime and glacier dynamics. Data from two Svalbard valley glaciers, Erikbreen and Usherbreen are presented. On Erikbreen, fossil forms were investigated, while on Usherbreen a surge ending in 1985 caused the formation of new push-moraine ridges. The push moraines are considered as a soil-mechanical problem. In a theoretical discussion the stress transmitted by the glaciers to the proglacial sediments is estimated. On Usherbreen, the compressive flow results in deformation both in old front ridges and in undisturbed frozen sediment layers in the front sandur. Thus, folding, thrust faulting and overriding all occur. Deformation of proglacial sediments seems to be highly dependent on the mechanical properties of the sediments. The sediments are strongly influenced by permafrost conditions. The unfrozen water content in the sediments governs the deformability, which in turn is partly determined by pore-water salinity. The distribution of push moraines in Svalbard is therefore restricted to areas below the Holocene marine limit, and they occur most frequently in areas of sedimentary bedrock. This study concludes that push moraines and ice-cored moraines require permafrost conditions. Push-moraine ridges are not formed in direct contact with the glacier, so they are geomorphologically not moraines, but deformed permafrost sediments. A model for glacier debris sedimentation and deformation is outlined for Svalbard glaciers ending on land.

Detailed changes in surface elevation of a recently deglaciated area have been mapped using a high-resolution photogrammetric method, with a view to estimating the contribution of debris from the proglacial area to the meltwater streams draining Storglaciären, northern Sweden, over the period 1980-90. The net contribution of sediments originating from the deglaciated area immediately in front of the glacier was of the order of 50% of the suspended silt load transported by meltwater at the flume Rännan downstream from the glacier, but at the same time, a similar amount of sediment accumulated along the streams. Though there is a significant exchange of mass, the net change is close to zero. Moreover, the survey provides detailed information about morphological changes in the landscape. Different processes, such as melting of permafrost, fluvial erosion and sedimentation, have been active.

Erosion and sedimentation rates were calculated from the difference between digital terrain models based on aerial photographs taken in 1980 and 1990. The result shows erosion in the central part of the proglacial area and accumulation of coarser sediments along the braided streams. In places, the ground is sinking, possibly due to melting of permafrost.

Where the ice is thinner, in the marginal zone, the thermal regime of ice in the tongue of Storglaciären corresponds well with the proglacial geomorphology. At present, the glacier has a 30-40 m thick cold surface layer which at the thinner marginal zone corresponds to a 100-200 m wide frozen rim. The temperature distribution within the ice was mapped using high-resolution radar.

Several dry-based alpine glaciers in the Dry Valleys of south Victoria Land, Antarctica, have prominent end moraines. Examination of their morphology, structure and sedimentology shows they consist of blocks of sand, gravel and organic silt within which sedimentary structures unrelated to entrainment and transportation by ice are well preserved. The nature and preservation of sedimentary structures, together with the presence of algae mats in the sediment, suggest formation by proglacial entrainment, transportation and deposition of frozen blocks of lacustrine sediment. Previous explanations of the formation of thrust-block moraines, including those that stress the importance of elevated pore-water pressure and Weertman’s ice-debris accretion hypothesis, depend on the presence of subglacial meltwater or the 0° C isotherm being situated close to the glacier bed. These models appear inappropriate for cold, dry-based glaciers because their basal temperatures are well below freezing point and they rest on deep permafrost. Three alternative models for the formation of thrust-block moraines at the margins of dry-based glaciers are examined in this paper: block entrainment of sediment associated with frozen-bed deformation; entrainment by overriding and accretion of marginal-ice and debris aprons; and transient wet-based conditions associated with glaciers flowing into ice-marginal lakes.

A theory of erosion and deposition as a consequence of subglacial sediment deformation over beds of unlithified sediment is reviewed and applied to large-scale till sequences formed on the southern flanks of the North American and British and European ice sheets during the last glacial cycle. The distribution of till thickness, till lithology in relation to source materials and intra-till erosion surfaces along a flowline in the Michigan lobe of the North American ice sheet are shown to be compatible with the deformational theory but not with other modes of till genesis. It is then demonstrated, in the case of the British ice sheet, how the assumption of a deformational origin for tills can be used to infer time-dependent patterns of ice-sheet dynamic behaviour. By reference to an example from the Netherlands, it is argued that many till sequences interpreted as melt-out tills are more likely to have formed by subglacial sediment deformation.

Glaciofluvial ridges, several hundred kilometres long, are commonly referred to as interlobate moraines because they appear to have formed at the convergence of two distinct ice lobes. Flow convergence is indicated by patterns of striations, streamlined forms and eskers. The so-called interlobate moraines are also thought to have formed asynchronously as the ice margin retreated. By contrast, we argue that the Harricana moraine of northern Quebec, Canada, formed following flow convergence in a regional-scale subglacial outburst flood. Flowlines constructed from streamlined bedforms mapped on the glacial map of Canada, reinterpretation of these streamlined forms as products of meltwater erosion, and field records of erosional marks (S-forms) in bedrock and glaciofluvial deposits to the lee of bedrock highs support this model. The effects of this flow convergence on the ice-sheet topography and drainage controlled the location of the broad conduit in which the Harricana moraine was deposited. Continued flow in this conduit and melting of the conduit walls explain the local patterns of striae, the supply of debris to the conduit, and the morphological and sedimentary characteristics of the moraine itself. From these characteristics, we conclude that the moraine was formed synchronously. This conclusion, if correct, is instructive regarding the deglacial hydrological organization of a large sector of the Laurentide ice sheet.

Detailed till-provenance studies of moraine complexes in the Wildhorse Canyon area, Idaho, U.S.A., allow inferences to be made regarding the flow paths and dynamics of Wildhorse and Fall Creek Glaciers, the two principal tributaries constituting a late-Pleistocene compound glacier. In particular, the distribution of statistically defined pebble and mineral assemblages within moraine complexes suggests that Wildhorse Glacier contributed a substantially greater volume of ice to the trunk glacier than did Fall Creek Glacier.

An initial group of glaciological reconstructions yields estimates for the equilibrium-line altitude (ELA) of the compound glacier that are consistent with those independently arrived at using other methods of ELA determination. Ice-flux calculations imply, however, that for each reconstruction the relative contributions of ice from Wildhorse and Fall Creek Glaciers were about equal, which is inconsistent with the inferences drawn from the till-provenance data. An alternative reconstruction incorporated possible orographic effects on accumulation and ablation by using different ELAs for the two tributary glaciers. Calculations for this reconstruction suggest that the ice flux of Wildhorse Glacier was about twice that of Fall Creek Glacier. This reconstruction is more consistent with the till-provenance data, and furthermore suggests that such data might be invaluable in choosing between seemingly equally viable glaciological reconstructions of paleoglaciers.

Passglaciären is a small cirque glacier in the Kebnekaise massif, northern Sweden. It is frozen to its bed over more than 70% of its area, and under present climatic conditions has little effect on cirque formation. More favourable conditions for cirque glacier erosion during the Holocene are of short duration. Assuming similar conditions during previous interglacials, it is suggested that forms such as the Passglaciären cirque developed mainly during the initial phases of glacials when they were part of networks of large valley glaciers or of a small warm-based mountain-centred ice sheet. Passglaciären has been examined in order to evaluate its erosive capacity and its association with the subglacial cirque morphology. The methods used are radar surveys and direct ice-temperature measurements. Erosion is restricted to a small section of the glacier bed, at present resulting in only partial deepening of the cirque and erosion of the backwall. In cold, arid regions with extensive permafrost, small cirque glaciers are largely frozen to the bed, and therefore cannot contribute significantly to cirque formation. In such regions glacial erosion by larger temperate glaciers is more likely to be the major cause of cirque excavation.

Glaciers on southwestern Bylot Island in the Canadian Arctic flow from an alpine setting in high-grade crystalline Archean terrane, on to coastal lowlands underlain by clastic sedimentary strata of Cretaceous to Tertiary age. We have used the contrasting mineralogy of the substrate as a tool to study subglacial entrainment and transport of debris in two large piedmont glaciers on Bylot Island. High chlorite/ and mica-illite/smectite ratios indicate that most basal debris is derived from crystalline rocks underlying the upper reaches of the glaciers. The subglacial accretion of Cretaceous–Tertiary sediments appears restricted to the lowermost part of the basal zone and is most noticeable near the glaciers’ termini. Ice associated with discrete silly–sandy debris layers is characterized by an isotopic signature indicative of refreezing of meltwater at the glacier sole. The compositional, textural and isotopic characteristics of basal sediment and ice lead us to conclude that subglacial entrainment through basal ice accretion occurs in both the upper and marginal parts of the glaciers.

Subglacial intakes in a tunnel system underneath Engabreen, northern Norway, provide access to the underside of a 200 m thick glacier. Detailed observations and measurements were made in several ice tunnels melted out along the glacier bed. A 20–200 cm thick basal sediment layer is overlain by clean glacier ice. Stratigraphy is complex, with alternating sediment-rich and sediment-free layers, and pervasive shearing. Throughout the basal ice are numerous spheroidal water pockets, which increase in both size and degree of elongation with distance from the bed. Ice cores were retrieved from ice-tunnel walls for sediment, cation and isotope analysis. Our observations and measurements provide evidence for both accretion in and water movement through the basal ice. This supports the modification to classical regelation theory proposed recently by Lliboutry in which water flow in the vein network is required to achieve net accretion of regelation layers.

It is known that regions of warm- and cold-based ice sheets modify and protect the Landscape, respectively. Investigations on a small plateau-top ice field, Øksfjordjøkelen (40 km2), in north Norway have indicated that this situation can exist at a small scale. Margins of the plateau, exposed by ice retreat since AD 1850, provide evidence of a complex basal thermal regime: in some localities blockfields with patterned ground and, in others, abraded and quarried bedrock forelands have been exposed. Exposed blockfields are interpreted as areas covered by cold-based, non-erosive ice. In areas of sliding ice, substantial quantities of erosion are evident. Locally, bedrock shows three joint sets intersecting which produce joint-bounded blocks. Removal of these blocks during the Little Ice Age has produced small rock steps about 5–10 m long and 1–3 m high. Present-day basal sliding velocities at the snout are low (15 m a–1) and ice thickness over the whole glacier is < 190 m. Simple modelling for block removal shows a direct relationship with glacier-sliding velocity and inverse relationship with ice thickness. Preglacial weathering is shown to influence the size of removable blocks.

The stratified-facies ice of the basal zone of Matanuska Glacier, Alaska. U.S.A., contains significant concentrations of anthropogenic tritium, whereas unaltered englacial-zone ice is devoid of tritium. Supercooled water flowing through subglacial conduits during the melt season likewise contains tritium, as does frazil and other platy ice that nucleates and grows within this subglacially flowing water. These initial results demonstrate net accretion of more than 1.4 m of stratified basal-zone ice since initiation of above-ground, thermonuclear bomb testing in 1952. Furthermore, these results support a theory of basal ice formation by ice accretion and debris entrainment from supercooled water within a distributed subglacial drainage system.

Successful core-drilling to bedrock of both the Greenland and Antarctic ice sheets offers unique opportunities for examining processes acting at the bed. At Byrd Station, Antarctica, penetration of the bed was accompanied by upwelling of glacial meltwater into the drillhole. The nature and disposition of sediment in the 4.83 m thick debris-rich basal ice, together with stable-isotope and gas analyses of the enclosing ice, confirm that incorporation of the debris occurred simultaneously with periodic “freeze-on” of basal meltwater. Currently, the presence of substantial meltwater at the ice/rock interface likely precludes any erosive activity at the bed. At GISP2, Greenland, basal silly ice, 13.1 m thick, is currently frozen to the bed at −9° C. Limited studies of the silty ice at GISP 2, together with results of more comprehensive investigations obtained by GRIP researchers on basal ice at a companion site at Summit indicate that the sediment-bearing basal ice likely formed in the absence of an ice sheet and was therefore unrelated to direct interaction of the present ice sheet with its bed. The fact that the basal ice at Summit is frozen to the bottom also precludes any likelihood of erosive activity at the bed.

A 20 km long, 2-5 km wide trench, which extends to 300 m below sea level, is believed to have been created by removal of sediment during the advance of Breiðamerkurjökull in the Little Ice Age. Between 1732 and 1890, the glacier advanced by 9 km, covered an area of 45 km2 and excavated a volume of 5 × 109 m3, equivalent to an average of 110 m over this area. Average sediment-removal rate during the 158 years was 32 × 106 m3 a−1 or 0.7 m a−1 km−2 averaged over the area covered by the advancing glacier. Calculated over the whole drainage area of the eastern branch of the glacier of 750 km2, the denudation rate would be 4 × 10−2 m a−1 km−2. Fluvial processes are estimated to have carried about 30 × 106 m3 a−1, and the sediment fluxes within the ice and by the deforming subglacial till are estimated to be 105 and 106 m3 a−1, respectively. The average sediment concentration in the glacial streams would have been about 10 kg m−3. Such concentrations have been measured in Icelandic rivers during jökulhlaups and surges. Several surging events took place during the advance of Breiðamerkurjökull, and jökulhlaups drain regularly beneath the glacier from ice-dammed marginal lakes. The present rate of transport, although considerable, seems to be about 10 × 106 m3 a−1, of which 30% is transported by the river to the sea and 70% is dumped into a proglacial lake.