Subglacial Abrasion Research Articles

Deep Electrical Resistivity Tomography for Detecting Gravitational Morpho-Structures in the Becca France Area (Aosta Valley, NW Italy)

Deep-seated gravitational slope deformations (DSGSDs) consist of gravity-induced, large-scale, gradual rock mass movements. In the Aosta Valley region (Valle d’Aosta NW Italy), DSGDs affect wide valley slopes and produce several interconnected morpho-structures that involve bedrock and Quaternary cover. Some DSGSD effects are not visible at the surface because of subglacial abrasion or burial by sediments and, therefore, are difficult to map with standard geomorphological surveys. This is the case for the Pointe Leysser DSGSD in the Aosta Valley, which is heavily influenced by the historical movements of the Verrogne-Clusellaz Glacier and its tributaries. We conducted a new geological investigation, integrated with deep electrical resistivity tomography geophysical surveys (ERTs). The ERT results were initially compared with geological/geomorphological evidence at the surface to define the correlation between the values and spatial distributions of electrical resistivity and the sediments, rocks, or morpho-structures. The resistivity values at various depths were subsequently analysed, interpreted, and discussed in conjunction with geological hypotheses. The geological and geophysical survey revealed three wide buried glacial valleys filled with glacial sediments and mapped the locations of gravitational morpho-structures at depth. These new data allowed us to draw a relationship between glacialism and gravitational evolution, distinguishing between pre-singlacial movements and postglacial movements.

Geochemical reactions initiated by subglacial abrasion of natural quartz and alkali feldspar

Geochemical reactions initiated by subglacial abrasion of natural quartz and alkali feldspar

Discovery of the world longest known intra-till cave (Jura mountains, France): Age and formation processes

We describe an atypical cave network in the Jura mountains (eastern France), currently the only one known worldwide to be entirely excavated within a subglacial deposits. Extending over 600 m of mapped galleries, this natural cavity developed entirely within a large impermeable drumlin shaped during last glacial stage MIS 2 (Wurm). By using lidar topography, petrophysical, sedimentological and geochemical measurements, as well as U-Th dating, this cave is mainly attributed to mechanical erosion processes (piping cave). The network's morphogenesis seems to be characterized by two main phases. Firstly, thin mini-pipes were probably dug along shear planes previously formed within the drumlin. The first mini-pipes may have appeared under the ice cover during the last glaciation in response to differential hydraulic pressures on either side of the drumlin. They may also have appeared later when the drumlin was free from its glacial cover, forming a natural earth dam across a thalweg and obstructing the flow of surface water. A second phase would correspond to the development of mega-pipes, resulting from sequential capture of flows from mini-pipes. Mega-pipes appeared >2500 years ago, as indicated by U-Th dating of one speleothem. In other regions of the world, intra-till caves have been documented, but are <20 m long. This limitation is mainly attributed to instability of the materials constituting subglacial tills, disfavoring a durable existence of galleries without collapses. In the Jura calcareous mountains the subglacial till contains a high proportion of very fine calcite matrix produced by subglacial abrasion, which was indurated by compaction under several hundred meters of ice. These particular petrographic features appear to be required for the creation and the preservation of this long intra-till cave. Similar caves may likely be found in other regions of the world displaying limestone bedrock and where subglacial till with abundant carbonate matrix was formed.

Reassessing a glacial landscape developed during terminal glaciation of the Late Paleozoic Ice Age in Uruguay

Whalebacks, roche moutonnées, and S-forms carved on Ediacaran granitoids near Cerro de las Cuentas, Uruguay, along with overlying diamictites, siltstones, and sandstones displaying soft-sediment grooved and striated surfaces in the Pennsylvanian San Gregorio Formation, record the glacial to post-glacial transition in the linked Norte, southern Paraná and Chaco-Paraná basins of Uruguay, Brazil, and Argentina respectively. Early authors reported these features resulted from subglacial abrasion and deposition as lodgement tills and glaciotectonites. Our re-examination reveals a nuanced setting with changing ice thicknesses, subglacial kinematics, and ice proximal glaciomarine dynamics associated with advance and retreat of an ice stream, or multiple advances of the Uruguayan Ice lobe, during glaciation of the Late Paleozoic Ice Age (LPIA) in these basins. The preserved landforms indicate temperate glacial conditions. Whalebacks formed under 1.6 to 2.5 km-thick ice and likely formed when the lobe extended across the Uruguayan and Rio Grande do Sul shields into the adjacent Paraná Basin. Previously unidentified m-scale roches moutonnées cut into one whaleback developed under thinner ice where reduced basal pressure allowed for the opening of air and water-filled cavities, thus facilitating quarrying on the lee side of basement bumps. S-forms provide additional evidence for the occurrence of subglacial waters, indicating that the basal ice was at or above its pressure melting point. The lower meter of the overlying strata consists of interstratified trace fossil-bearing, laminated siltstones; thin-bedded diamictites; and current-rippled sandstones. Trace fossils belonging to the Mermia ichnofacies within the basal siltstones, as well as acritarchs in the overlying siltstones, suggest that these sediments were deposited in ice-proximal subaqueous settings with contributions from meltwater discharge. Graded siltstone laminae suggest settling from suspension likely from meltwater plumes, while thin-bedded diamictites were deposited either as debris flows or as two-component sedimentation with fines settling from suspension and coarser particles introduce as iceberg-rafted dropstones. Current-rippled sandstones indicate the occurrence of underflow currents. Soft-sediment troughs, grooves, and striations cutting these sediments display curved and sinuous paths with some features oriented perpendicular, and one oriented opposite to the overall trend. They contain marginal and terminal berms typical of iceberg scour marks suggesting transit across the area by icebergs calving from a tidewater ice front located to the SE.

Modelling alpine glacier geometry and subglacial erosion patterns in response to contrasting climatic forcing

AbstractClimate exerts a primary control on glacier mass balance, driving changes in ice flux, which affects basal sliding and subglacial erosion. Although past glacier reconstructions have been widely used as paleo‐climate proxies, extracting quantitative temperature/precipitation data from paleo‐glaciers is still challenging. The iSOSIA ice model was used over a synthetic landscape to quantitatively investigate the non‐linear dependence of glacier geometry and ice dynamics on climate forcing, as well as to analyse how spatial patterns of subglacial erosion and erosion rates vary between the accumulation and ablation zones for different climatic settings. We performed 10 calibrated climatic scenarios with different combinations of temperature and precipitation rate in two separate sets of simulations [maintaining either same equilibrium line altitude (ELA) or same ice extent; SA and ST, respectively]. Our results reinforce the role of climate and the importance of ice flux in conditioning glacier geometry. In SA simulations, contrasting climatic scenarios showed a major influence on glacier length and thickness. In ST simulations, calibrated ELAs presented a variability of over 300 m in order to maintain similar ice extent, but showed reduced changes in glacier thickness. These results highlight the importance of ice flux to predict glacier geometry, when compared to the overall mass balance from an ELA estimate. Subglacial abrasion and quarrying showed notable differences between the accumulation and ablation zones for varying climatic forcing, with multimodal erosion distributions for increased precipitation/temperature, shifting erosion towards higher‐elevation tributaries and connecting erosion patches in the ablation zone. Such trends highlight the role of subglacial hydrology and ice‐bed contact (cavitation) in dictating patterns of subglacial erosion, while the depth of erosion relates closely to ice flux and climate inputs. Our results have potential implications for paleo‐climate reconstructions based on glacier geometry and provide insights on how subglacial erosion can help estimate paleo‐climatic conditions from paleo‐glacial records.

Glacial Erosion Driven by Variations in Meltwater Drainage

AbstractThe subglacial processes of abrasion and quarrying are thought to be primarily responsible for bedrock erosion by glaciers. While theory points to sliding speed as the dominant control on abrasion, rates of quarrying are likely scaled by a more complex combination of sliding speed, effective pressure, bed roughness, and short‐term water‐pressure fluctuations. Here we pair a model for quarrying based on statistical characterization of bedrock strength with a model for subglacial hydrology that describes the temporal evolution of cavities under the influence of variations in sliding speed and effective pressure. Using a finite element model, we simulate the evolution of the hydrological system at the base of a glacier and compute rates of abrasion and quarrying. Cavity lengths and channel cross sections evolve through time, causing temporal shifts in ice‐bed contact area, which in turn govern the differential stress that influences erosion over the course of a year. Our results demonstrate how variations in meltwater production amplify rates of subglacial erosion relative to the case of steady meltwater generation. The level of amplification depends on how the variations control the ice‐bed contact area. Seasonal variations are most effective in boosting mean rates of basal sliding and hence subglacial abrasion, whereas shorter‐term variations (monthly‐weekly) most strongly influence rates of subglacial quarrying through temporal amplification of differential bedrock stress around cavities. This influence of transient hydrology on subglacial erosion processes may explain why glaciers in temperate climates with strong variations in temperature and precipitation erode faster than similar‐type glaciers in polar environments.

The influence of basal-ice debris on patterns and rates of glacial erosion

Glaciers have played a key role for shaping much of Earth's high topography during the cold periods of the Late Cenozoic. However, despite of their distinct influence on landscapes, the mechanisms of glacial erosion, and the properties that determine their rate of operation, are still poorly understood. Theoretical models of subglacial erosion generally highlight the influence of basal sliding in setting the pace of erosion, but they also point to a strong influence of other subglacial properties, such as effective bed pressure and basal-ice debris concentration. The latter properties are, however, not easily measured in existing glaciers, and hence their influence cannot readily be confirmed by observations.In order to better connect theoretical models for erosion to measurable properties in glaciers, we used computational landscape evolution experiments to study the expected influence of basal-ice debris concentration for subglacial abrasion at the scale of glaciers. The computational experiments couple the two erosion processes of quarrying and abrasion, and furthermore integrate the flow of ice and transport of debris within the ice, thus allowing for the study of dynamic feedbacks between subglacial erosion and systematic glacier-scale variations in basal-ice debris concentration. The experiments explored several physics-based models for glacial erosion, in combination with different models for basal sliding to elucidate the relationship between sliding speed, erosion rate and basal-ice debris concentration. The results demonstrate how differences in debris concentration can explain large variations in measured rates. The experiments also provide a simple explanation for the observed dependence of glacier-averaged rate of erosion on glacier size: that large glacier uplands feed more debris into their lower-elevation parts, thereby strengthening their erosive power.

Subglacial abrasion rates at Goldbergkees, Hohe Tauern, Austria, determined from cosmogenic 10Be and 36Cl concentrations

AbstractWe report concentrations of cosmogenic 10Be and 36Cl used to determine erosion depths in the recently deglaciated bedrock at Goldbergkees in the Eastern Alps. The glacier covered the sampling sites during the Little Ice Age (LIA) until c. 1940. The youngest ages calculated from these concentrations match the known exposure time after the post‐LIA exposure of &lt;100 years. The apparent age (no cover, no erosion) of most samples, however, is significantly older. We show that the measured nuclide concentrations represent subglacial erosion depths, rather than exposure times. In particular, erosion depths calculated using 10Be and 36Cl concentrations of individual samples match well, whereas apparent 36Cl ages are consistently older than 10Be ages. The bedrock at the ‘youngest’ surfaces was deeply eroded (≥ 297 cm) by the Goldbergkees during the late Holocene. In contrast, bedrock at the margin of the LIA ice extent was eroded ≤35 cm. These values convert to subglacial erosion rates on the order of 0.1 mm/a to &gt;5 mm/a. While modeled erosion rates depend on the duration of glacial cover and erosion intrinsic to the different exposure scenarios used for calculation (700–3300 years), modeled total erosion depths are insensitive (5–20% change).Analysis of erosion depths on the transverse valley profile shows a general trend of greatest erosion part way up the valley side and less erosion under thin ice at the lateral margin. A second profile along the valley axis indicates depth of erosion is greatest where the ice abuts the foot of the investigated bedrock riegel and at its lee side just beyond the crest. Copyright © 2016 John Wiley &amp; Sons, Ltd.

Submarine geomorphology and glacial history of the Sea of the Hebrides, UK

The Sea of the Hebrides is an island-studded region of complex bathymetry on the UK continental shelf, west of the Scottish mainland. An extensive area (2200km2) of recently collected multibeam bathymetry data, combined with seismic reflection profiles reveal this part of the shelf to have been extensively modified by both glacial and modern processes. Our new geomorphological evidence strongly supports the contention that an ice stream drained ice from western Scotland and the Inner Hebrides towards the Barra Fan at the continental shelf break at the height of the last glaciation (Marine Isotope Stage 2–3). Convergent seabed glacial lineations and other subglacially streamlined features eroded in bedrock around the Islands of Canna and Rum preserve the direction of ice sheet movement, and strongly suggest the onset of ice streaming in a southwesterly direction on the continental shelf in the Sea of the Hebrides region. This fast-flow zone formed part of a larger convergent ice stream system draining much of western Scotland and the north of Ireland — the southern part of which has been postulated already. A number of rock basins, linear troughs and deeps west and south of Muck, kept sediment-free by the dominant modern tidal regime, are interpreted as being at least partly subglacial in origin. Similarly, a large complex of buried deeps between Eigg and Arisaig are interpreted as an overdeepened glacial drainage network. It is suggested that intense bedrock erosion by focused subglacial abrasion and subglacial meltwater discharge over multiple glacial advance and decay cycles led to significant modifications of the pre-Quaternary bedrock surface. Other features such as moraines are only found in shallower water (typically <50m), west of Canna, west of Rum and south of Skye. Their relative paucity indicates that the modern tidally dominated environment of the area may have removed much of the localised geomorphological evidence of offshore glacial limits within the Sea of the Hebrides, or alternatively that the ice sheet margin was floating and retreated predominantly by calving leaving little geomorphological evidence of its retreat.

Subglacial erosion forms in northwest Ireland

Subglacial erosional forms are commonly found on bedrock substrates inside the Late Weichselian ice margin in County Donegal, northwest Ireland, and can be used to provide detailed information on subglacial processes and environments. The erosional forms occur on spatial scales from whalebacks (tens of metres in scale), to asymmetric and channelized bedrock‐cut scours (tens of cm in scale) and striations (mm scale). Processes responsible for development of subglacial erosional forms occur along a continuum, from free meltwater existing as a laterally extensive sheet at the ice‐bed interface, to abrasion by basal ice. Channelized bedrock‐cut scours are particularly common in County Donegal, and show asymmetric and meandering thalwegs, U‐shaped cross‐profiles and steep lateral margins. Innermost parts of the scours are highly polished and have striations that follow thalweg direction. In places, bedrock surfaces are overlain by a delicate polish and thin calcite cement, and are buried beneath glacial till. Based on their morphology, the bedrock scours are interpreted as s‐forms caused by high‐pressure subglacial meltwater erosion. Striations within the scoured channels reflect periods of ice‐bed coupling and subglacial abrasion. The range of features observed here was used to consider relationships between subglacial topography, hydraulic processes and ice‐bed coupling. Precipitation of calcite cement took place in depressions on the bedrock surface by CO2 degassing. Infilling of depressions by glacial till formed a new type of ‘sticky spot’ related to spatial variations in subglacial water pressure. The temporal evolution of sticky spots reflects interactions within the subglacial environment between subglacial relief, hydraulic regime and ice‐bed coupling.

Megagrooves and streamlined bedrock in NW Scotland: The role of ice streams in landscape evolution

New multibeam bathymetry data, onshore high-resolution elevation data (NEXTMap) and fieldwork in the Ullapool area of NW Scotland reveal large-scale megagrooves and streamlined bedrock forms in a well-defined ∼ 20 km wide zone. The landsystem is typical of a coherent flow corridor within a grounded ice sheet on bedrock-dominated terrain. We describe the morphology of the large-scale features, discuss their likely formation, and consider the wider implications for ice-sheet dynamics. Based on the strongly convergent bedform distribution, the presence of megagrooves and highly elongate bedrock forms, we interpret the erosional landscape to be the signature of a fast-flowing tributary that once fed the The Minch palaeo-ice stream — a major artery of the last British–Irish ice sheet. The exact genesis of bedrock megagrooves remains uncertain, although focused subglacial abrasion is likely to have carved most of the shallow, strongly parallel, features; whilst glacial meltwater may have carved or modified others. Bedform morphometry is used to discriminate zones reflecting the degree of glacial streamlining (elongation ratios < 5:1 or > 5:1). We interpret these zones to represent the transition from potentially cold-based slow ice-sheet flow to warm-based fast flow. Based on these results, and the presence of ribbed moraines, we suggest a bedform continuum model for onset zones of palaeo-ice streams on rigid beds. Rapid spatial bedform evolution is suggested to reflect an increase in subglacial erosive power that may be diagnostic of palaeo-ice-sheet thermal boundaries (i.e. from cold- to warm-based), and is also consistent with the expected downstream increase in ice velocity within an ice-stream onset zone. Finally, this study speculates on the role played by basal meltwater in ice-stream initiation and the role of ice streams and their tributaries in landscape evolution.

Experimental rock‐on‐rock frictional wear: Application to subglacial abrasion

Rock‐on‐rock wear under known normal loads was measured experimentally using 62 mm diameter rotating discs machined from a range of silicic rock types: one granite, two metamorphic rocks, and three sandstones of porosities varying between 7% and 28%. Wear debris was removed continuously from the contacting surfaces by water. We obtained a wear law of the form dT/dx = Aσnϕm, where σ is contact normal stress (MPa), dT/dx is a dimensionless wear rate (meter wear (T) per meter shear displacement (x)), ϕ is porosity, and A, n, and m are empirical constants, 10−8.11, 8.33, and 4.50, respectively. Low‐porosity rocks displayed surface polishing, and abraded particle size was substantially smaller than the starting grain size. Whole grains were plucked from high‐porosity rocks, without any surface polishing, so that the wear product particle size remained the same as that of the starting material. The empirical wear law describes abrasion beneath a hard‐bedded temperate glacier, provided we can estimate both the in situ normal stress between entrained clasts and the bed and the horizontal velocity of the clasts in the flowing ice. Normal stress concentration is expected to develop at clast/bed contacts from drag against ice flowing toward the bed, where pressure melting occurs. These experimental data are consistent with wear rates inferred from silt outflow in subglacial streams, provided a tenfold stress concentration can occur relative to the ice overburden pressure. The rate of production of fine‐grained wear products under subglacial conditions is substantially less than by tectonic faulting.

Glacial comminution of quartz sand grains and the production of loessic silt: A simulation study

Traditional views have attributed the formation of loess-sized quartz silt particles to glacial processes. Glacial grinding was considered to be the only mechanism capable of producing loessic silt in the quantities necessary for the formation of loess deposits. This idea has been strengthened by the apparent close association between present or former glacierised regions and major global loess deposits. However, some doubts exist as to whether or not glacial grinding is capable of producing sufficient quantities of quartz silt. Because of the difficulties entailed in the observations and direct measurement of subglacial abrasion and crushing, this paper describes and discusses the results of a simulation study designed to examine abrasion and size reduction of quartz grains under “grinding” conditions analogous to those experienced in a subglacial environment. The findings from these experimental runs suggest that, although particle breakage and comminution does occur, little silt-sized material is actually produced. The amount of silt produced ranges from approximately 9% of the original sample in one experimental run to less than 1% for the other seven runs. It is suggested that the results obtained may show considerably less abrasion and loessic silt production than may be expected to occur within a natural subglacial environment because: (1) subglacial grains experience much greater overburden pressures for much longer periods of time; (2) the use of sand-sized grains may represent a stable component resistant to further size reduction through crushing or abrasion; and (3) the use of a monomineralic sample of relatively sound quartz grains. Although care must be taken when relating results obtained from these experimental runs to silt production within a glacial environment, findings from this work do question whether large amounts of quartz silts are produced within subglacial environments. The importance of glacial processes in the formation of loess deposits may lie in the reworking of pre-existing sediments by ice rather than the actual production of vast quantities of silt. This in turn suggests that care should be taken in the interpretation of the provenance of particles comprising loess deposits.

Influence of hydroxyl-bearing minerals on the isotopic composition of ice from the basal zone of an ice sheet

ONE of the important questions in ice-core studies is the extent to which isotope measurements made on ice loaded with bedrock debris can be trusted when interpreting the climate record from ice-core data. Here we show that the isotopic characteristics of basal ice containing debris from beneath the Greenland ice sheet can be understood only if isotope exchange between hydroxyl-bearing minerals and water is taken into account. Sub-glacial abrasion of clays and micas increases the number of broken bonds at the border of the mineral structure, promotes cleavage and so increases the rate at which these minerals exchange oxygen and hydrogen isotopes with water. One implication of our results is to call into question the recent suggestion1 that the basal ice of the Greenland ice sheet is 'superimposed ice', and that therefore extensive melting must have taken place during the last inter-glacial. The enrichment in 18O of the basal ice, thought to be acquired during the formation of the superimposed ice, may in fact be the result of isotope exchange with hydroxyl-bearing minerals.

Forms of Subglacial Erosion on Palaeozoic Sedimentary Rocks (Abstract only)

The relationship between bedrock erodibility and the characteristic bed form and roughness produced by subglacial abrasion is largely unknown. Less resistant bedrocks may be abraded down to a surface of low obstacle size and low roughness parameter. However, Hallet (1979) notes that bedforms are probably dampened at widely varying rates, and that certain bedforms may actually be amplified through time by rapid trough abrasion. The object of this paper is to examine the relationship between the mechanical strength of several bedrock types and the magnitude of subglacially streamlined bed forms found on them.Cambrian limestones, mudstones, dolomites, and chert, some with abundant quartz veins, are the bedrock types examined. They lie as a series of steeply-dipping beds up to 100 m thick with their strike across the floor of the Yoho valley, in the Rocky Mountains of British Columbia. They were covered by the ice of Yoho Glacier 30 years ago, and are now less than 0.5 km below the present ice terminus. The sites examined lie within an area less than 0.25 km2, up-glacier from a major riegel form. The Yoho River nearby flows in a deep gorge and has not affected this section of the former glacial valley floor.The bedrocks were analysed and tested in the laboratory for mechanical strength and resistance to abrasion.During summer 1979, streamlined bedforms and surface roughness were surveyed along the former glacier flowline, evidenced by striations, and plane table maps were completed across several beds. Small 1 m2 plots were surveyed in detail on each bedrock type. Analysis of surface forms has included 3-dimensional SYMVU plots. Preliminary indications are that the micro-roughness shows greater variability between bedrock types than do large wavelengths of relief. However, resistant bedrocks are areas of prominent positive relief while others are generally low-lying. Analysis of the results continues.

Energy Dissipation During Subglacial Abrasion at Nisqually Glacier, Washington, U.S.A.

Abstract This study examines the effect of subglacial abrasion on the basal sliding term of the gravitational energy balance of the dynamic, temperate Nisqually Glacier on Mount Rainier, Washington, U.S.A. Subglacial water flux is estimated as 3 × 107 m3 a–1 and suspended sediment flux as 3 × 107 kg a–1. Suspended-sediment flux is assumed to represent, within an order of magnitude, the annual mass eroded by subglacial abrasion. Subglacial abrasion involves both brittle fracture and plastic deformation. Field observations of bas-relief and grooved depression striations appear to have exact counterparts in rock mechanics experiments approximating subglacial velocities and normal stresses. Boulton's ([Cl974]) abrasion model and a new attritivity model proposed herein are shown to predict subglacial abrasion-rates within the limits of natural variability and the error range of measurements. The first crude gravitational energy balance for lower Nisqually Glacier (1.96 km2) is attempted and probably has only order-of-magnitude accuracy. The importance of subglacial abrasion in dissipating basal sliding energy at Nisqually Glacier is confirmed.

Energy Dissipation During Subglacial Abrasion at Nisqually Glacier, Washington, U.S.A.

AbstractThis study examines the effect of subglacial abrasion on the basal sliding term of the gravitational energy balance of the dynamic, temperate Nisqually Glacier on Mount Rainier, Washington, U.S.A. Subglacial water flux is estimated as 3 × 107 m3 a–1 and suspended sediment flux as 3 × 107 kg a–1. Suspended-sediment flux is assumed to represent, within an order of magnitude, the annual mass eroded by subglacial abrasion.Subglacial abrasion involves both brittle fracture and plastic deformation. Field observations of bas-relief and grooved depression striations appear to have exact counterparts in rock mechanics experiments approximating subglacial velocities and normal stresses. Boulton's ([Cl974]) abrasion model and a new attritivity model proposed herein are shown to predict subglacial abrasion-rates within the limits of natural variability and the error range of measurements. The first crude gravitational energy balance for lower Nisqually Glacier (1.96 km2) is attempted and probably has only order-of-magnitude accuracy. The importance of subglacial abrasion in dissipating basal sliding energy at Nisqually Glacier is confirmed.

The formation of the Labyrinth, Wright Valley, Antarctica

Astract Previous hypotheses to explain the origin of the Labyrinth have invoked floods, glacial erosion, and uniformitarian arguments of salt weathering and deflation. The landscape is one of reticulate troughs, deeper to the east than west, with irregular long profiles and enclosed depressions. The troughs are shown to belong to a hierarchy and display curvilinear form. Their origin is related to glacial plucking along joints and subsequent deepening by ice-marginal and proglacial meltwater systems. Periglacial alteration of the primary form, particularly asymmetric infill by solifluction and scree, is largely a result of debris supplied by freeze-thaw processes. Intertrough blocks initially formed upstanding masses between reticulate, joint-controlled depressions, and were subject to subglacial abrasion by a glacier sliding over its bed. The required wet-based ice conditions imply either extremely thick ice or a warmer climate than present. Landform is used to deduce a complex origin for the Labyrinth w...