Nicholas Eyles

Earth's glacial record and its tectonic setting☆

Glaciations have occurred episodically at different time intervals and for different durations in Earths history. Ice covers have formed in a wide range of plate tectonic and structural settings but the bulk of Earths glacial record can be shown to have been deposited and preserved in basins within extensional settings. In such basins, source area uplift and basin subsidence fulfill the tectonic preconditions for the initiation of glaciation and the accomodation and preservation of glaciclastic sediments. Tectonic setting, in particular subsidence rates, also dictates the type of glaciclastic facies and facies successions that are deposited. Many pre-Pleistocene glaciated basins commonly contain well-defined tectonostratigraphic successions recording the interplay of tectonics and sedimentation; traditional climatostratigraphic approaches involving interpretation in terms of either ice advance/retreat cycles or glacio-eustatic sea-level change require revision. The direct record of continental glaciation in Earth history, in the form of classically-recognised continental glacial landforms and “tillites”, is meagre; it is probable that more than 95% of the volume of preserved “glacial” strata are glacially-influenced marine deposits that record delivery of large amounts of glaciclastic sediment to offshore basins. This flux has been partially or completely reworked by “normal” sedimentary processes such that the record of glaciation and climate change is recorded in marine successions and is difficult to decipher. The dominant “glacial” facies in the rock record are subaqueous debris flow diamictites and turbidites recording the selective preservation of poorly-sorted glaciclastic sediment deposited in deep water basins by sediment gravity flows. However, these facies are also typical of many non-glacial settings, especially volcanically-influenced environments; numerous Archean and Proterozoic diamictites, described in the older literature as tillites, have no clearly established glacial parentage. The same remarks apply to many successions of laminated and thin-bedded facies interpreted as “varvites”. Despite suggestions of much lower values of solar luminosity (the weak young sun hypothesis), the stratigraphic record of Archean glaciations is not extensive and may be the result of non-preservation. However, the effects of very different Archean global tectonic regimes and much higher geothermal heat flows, combined with a Venus-like atmosphere warmed by elevated levels of CO2, cannot be ruled out. The oldest unambiguous glacial succession in Earth history appears to be the Early Proterozoic Gowganda Formation of the Huronian Supergroup in Ontario; the age of this event is not well-constrained but glaciation coincided with regional rifting, and may be causally related to, oxygenation of Earths atmosphere just after 2300 Ma. New evidence that oxygenation is tectonically, not biologically driven, stresses the intimate relationship between plate tectonics, evolution of the atmosphere and glaciation. Global geochemical controls, such as elevated atmospheric CO2 levels, may be responsible for a long mid-Proterozoic non-glacial interval after 2000 Ma that was terminated by the Late Proterozoic glaciations just after 800 Ma. A persistent theme in both Late Proterozoic and Phanerozoic glaciations is the adiabatic effect of tectonic uplift, either along collisional margins or as a result of passive margin uplifts in areas of extended crust, as the trigger for glaciation; the process is reinforced by global geochemical feedback, principally the drawdown of atmospheric CO2 and Milankovitch “astronomical” forcing but these are unlikely, by themselves, to inititiate glaciation. The same remarks apply to late Cenozoic glaciations. Late Proterozoic glacially-influenced strata occur on all seven continents and fall into two tectonostratigraphic types. In the first category are thick sucessions of turbidites and mass flows deposited along active, compressional plate margins recording a protracted and complex phase of supercontinent assembly between 800 and 550 Ma. Local cordilleran glaciations of volcanic peaks is indicated. Many deposits are preserved within mobile belts that record the subduction of interior oceans now preserved as “welds” between different cratons. Discrimination between glacially-influenced and non-glacial, volcaniclastic mass flow successions continues to be problematic. The second tectonostratigraphic category of Late Proterozoic glacial strata includes successions of glacially-influenced, mostly marine strata deposited along rifted, extensional plate margins. The oldest (Sturtian) glaciclastic sediments result from the break-out of Laurentia from the Late Proterozoic supercontinent starting around 750 Ma along its “palaeo-Pacific” margin with a later (Marinoan) phase of rifting at about 650 Ma. “Passive margin” uplifts and the generation of “adiabatic” ice covers on uplifted crustal blocks triggered widespread glaciation along the “palaeo-Pacific” margin of North America and in Australia. A major phase of rifting along the opposite (“palaeo-Atlantic”) margin of Laurentia occurred after 650 Ma and is similarly recorded by glaciclastic strata in basins preserved around the margins of the present day North Atlantic Ocean. Glaciation of the west African platform after 650 Ma is closely related to collision of the West African and Guyanan cratons and uplift of the orogenic belt; the same process, involving uplift around the northern and western margins of the Afro-Arabian platform subsequently triggered Late Ordovician glaciation at about 440 Ma when the south polar region lay over North Africa. Early Silurian glaciation in Bolivia and Brazil was followed by a non-glacial episode and renewed Late Devonian glaciation of northern Brazil and Bolivia. The latter event may have resulted from rotation of Gondwana under the South Pole combined with active orogenesis along the western margin of the supercontinent. Hercynian uplift along the western margin of South America caused by the collision and docking of “Chilinia” at about 350 Ma (Late Tournasian—Early Visean) was the starting point of a long Late Palaeozoic glacial record that terminated at about 255 Ma (Kungurian-Kazanian) in western Australia. The arrival of large landmasses at high latitude may have been an important precondition for ice growth. Strong Namurian uplift around virtually the entire palaeo-Pacific rim of Gondwana culminated in glaciation of the interior of the supercontinent during the latest Westphalian (c. 300 Ma). There is a clear picture of plate margin compression and propagation of “far field” stresses to the plate interior allowing preservation of glacially-influenced strata in newly-rifted intracratonic basins. Many basins show a “steers head” style of infill architecture recording successive phases of subsidence and overstepping of younger strata during basin subsidence and expansion. Exploration for oil and gas in Gondwanan glaciated basins is currently a major stimulus to understanding the relationship between tectonics and sedimentation. Warm Mesozoic palaeoclimates do not rule out the existence of restricted ice covers in the interiors of continental landmasses at high palaeolatitudes (e.g. Siberia, Antarctica) but there is as yet, no direct geological record of their existence. The most likely record of glaciers is contained in Late Jurassic and early Cretaceous strata. In any event, these ice masses are unlikely to have had any marked effect on global sea levels and alternative explanations should perhaps be sought for 4th order, so-called “glacio-eustatic” changes in sea level, inferred from Triassic, Jurassic and Cretaceous strata. The growth of extensive Northern Hemisphere ice sheets in Plio-Pleistocene time (c. 2.5 Ma) was the culmination of a long global climatic deterioration that began sometime after 60 Ma during the late Tertiary. Tectonic uplift of areas such as the Tibetan Plateau and plate tectonic reorganizations have been identified as first-order controls. Initiation of the East Antarctic ice sheet, at about 36 Ma, is the result of the progressive thermal isolation of the continent combined with uplift along the Transantarctic Mountains. In the Northern Hemisphere, the upwarping of extensive passive margin plateaux around the margins of the newly-rifted North Atlantic may have amplified global climatic changes and set the scene for the growth of continental ice sheets after 2.5 Ma. Ice sheet growth and decay was driven by complexly interrelated changes in ocean circulation, Milankovitch orbital forcing and global geochemical cycles. It is arguable whether continental glaciations of the Northern Hemisphere, and the evolution of hominids, would have occurred without the necessary precondition of tectonic uplift.

‘Zipper-rift’: a tectonic model for Neoproterozoic glaciations during the breakup of Rodinia after 750 Ma

The ‘Snowball Earth’ model of Hoffman et al. [Science 281 (1998) 1342] has stimulated renewed interest in the causes of glaciation in Earth history and the sedimentary, stratigraphic and geochemical response. The model invokes catastrophic global Neoproterozoic refrigerations when oceans froze, ice sheets covered the tropics and global temperatures plummeted to −50 °C. Each event is argued to be recorded by tillites and have lasted up to 10 million years. Planetary biological activity was arrested only to resume in the aftermath of abrupt and brutal volcanically generated ‘greenhouse’ deglaciations when global temperatures reached +50 °C. The ‘Cambrian explosion’ is regarded by some as a consequence of post-Snowball glacioeustatic flooding of continental shelves. We shall show by a systematic review of the model that it is based on many long standing assumptions of the character and origin of the Neoproterozoic glacial record, in particular, ‘tillites’, that are no longer valid. This paper focusses on the sedimentological and stratigraphic evidence for glaciation in the light of current knowledge of glacial depositional systems. By integrating this analysis with recent understanding of the tectonic setting of Neoproterozoic sedimentary basins, an alternative ‘Zipper-rift’ hypothesis for Neoproterozoic glaciations is developed. The ‘Zipper-rift’ model emphasises the strong linkage between the first-order reorganisation of the Earths surface created by diachronous rifting of the supercontinent Rodinia, the climatic effects of uplifted rift flanks and the resulting sedimentary record deposited in newly formed rift basins. Initial fragmentation of Rodinia commenced after about 750 Ma (when the paleo-Pacific Ocean started to form along the western margin of Laurentia) and culminated sometime after 610 Ma (with the opening of the paleo-Atlantic Ocean on the eastern margin of Laurentia). Breakup is recorded by well-defined ‘tectonostratigraphic’ successions that were deposited in marine rift basins. The base of each succession is characterised by coarse-grained synrift strata consisting of mass flow diamictites and conglomerates (many of the ‘tillites’ of the older literature). These facies are interbedded with large olistostromes and contain clastic carbonate debris derived from landsliding of fault scarps along rifted carbonate platforms. Diamictites and conglomerates are dominantly the product of subaqueous mass flow and mixing of coarse and fine sediment populations (the term mixtite has been used in the past). These facies are not uniquely glacial and are produced regardless of climate and latitude. Synrift deposits commonly pass up into thick slope turbidites recording enhanced subsidence and are capped by uppermost shallow marine strata that record a reduction in subsidence rates and overall basin shallowing. Tectonostratigraphic ‘cycles’ can attain thicknesses of several kilometres, but have been commonly misinterpreted as recording globally synchronous ‘glacioeustatic’, falls and rises in sea level. In fact, eustatic sea-level changes in rift basins are suppressed as a result of a strong tectonic control on relative water depths. The great length of newly formed rifted margins around the long perimeter of Laurentia (<20,000 km) ensured that deposition of tectonostratigraphic successions occurred diachronously as in the manner of a zipper between approximately 740 and 610 Ma. Some successions show a definite glacial influence on sedimentation as a consequence of latitude or strong tectonic uplift, but many do not. Regardless, all deposits record a fully functioning hydrological cycle entirely at odds with a supposed fully permafrozen planet. Mapping of those deposits where a definite glacial imprint is apparent indicates that Neoproterozoic glaciation(s) were likely regional or hemispheric in scope and latitudinally constrained. They were perhaps no more severe than other glaciations recorded in Earth history. The regional distribution of ice centres is argued to have been influenced by tectonic topography created by large mantle plumes and by rift shoulder uplift. Paleomagnetic data indicative of tropical glaciation are, in our view, ambiguous because of uncertainty as to when such paleomagnetic characteristics were acquired. A lower solar luminosity may have played a role in lowering snow line elevations and displacing glaciation into latitudes lower than those of Phanerozoic glaciations. Global tectonic and volcanic activity, especially the rapid burial or organic carbon in new rift basins, may explain extreme shifts in C-isotopic values evident in late Neoproterozoic strata.

The Late Devensian (<22,000 BP) Irish Sea Basin: The sedimentary record of a collapsed ice sheet margin

The Late Devensian (<20 ka BP) glacial geology of the Irish Sea Basin (4000 km2) is an event stratigraphy recording the entry of marine waters into a glacio-isostatically-depressed basin, and the rapid retreat of the Irish Sea Glacier as a tidewater ice margin. Marine limits occur up to 140 m O.D. Across much of the central basin, the ice margin was uncoupled from its bed exposing a subglacially-scoured topography to glaciomarine processes. The Irish Sea Glacier was a major drainage conduit of the last British Ice Sheet; calving of the marine ice margin resulted in fast flow (surging) of ice streams recorded by drumlin fields around the northern basin margin and tunnel valleys. Rapid evacuation of the basin may have stranded large areas of dead ice in peripheral zones (e.g. Cheshire/Shropshire Lowlands) and initiated the collapse of the ice sheet.Thick wedges of ice-contact glaciomarine sediments were deposited during ice retreat as morainal bank complexes by successive tidewater ice margins stabilized at pinning points around the Irish Sea coast. Where morainal banks occur on the seaward side of drumlin swarms there is a clear sequential relationship between rapid ice loss from calving ice margins, the development of fast flowing ice streams, drumlinization and the pumping of subglacial sediment to tidewater. Raised delta complexes are locally associated with marine limits along the high relief coastal margins of Wales, east central Ireland, and the Lake District. Associated valley infill complexes record downslope resedimentation of heterogenous sediments into the marine environment during ice retreat. Co-eval offshore deposits are represented by well-stratified glaciomarine complexes that infill a subglacially-scoured topography that shows networks of tunnel valleys. Glaciomarine mud drapes occur well to the south of the maximum limit of grounded ice in the basin (e.g. North Devon, Scilly Islands, Southern Ireland). The age of these distal sediments, previously mapped as pre-Devensian tills, is constrained by amino acid ratios.Basin rebound following deglaciation was rapid, with over 100 m recovery in 3 ka, and was followed by a low marine still stand. Peat, accumulating in offshore areas now as much as 55 m below sea level has been drowned by the postglacial eustatic rise in sea level.The glacio-sedimentary model identified in this paper, involving rapid ice retreat and related sedimentation triggered by rising relative sea level, suggests that isotatic downwarping is an important mechanism for deglaciating continental shelves.

Models of glaciomarine sedimentation and their application to the interpretation of ancient glacial sequences

Abstract This paper argues that glaciomarine environments can be regarded as special, glacially-influenced types of continental margin environments (e.g. continental shelf, slope, rise and basin plain). Knowledge of the stratigraphic architecture and typical sedimentary sequences of non-glacial margins is becoming well-known but remains limited for those that have been glacially modified. The principal influences on sedimentation in these environments relate to the glacial sediment input (controlled by relief of basin margin, glacier thermal regime and ice flow dynamics) and depositional environments (influence of traction currents, substrate relief and proximity to nearby ice margins). Typical ranges of sedimentation rates can be established for glacially-influenced continental margin environments and these may provide a framework for ancient sequences. Starvation of sediment supply to glacially-influenced continental margins is common. The nature of sub ice shelf sedimentation, a model that has been applied to many ancient glacial sequences is critically reviewed; the significance of such sedimentation in the rock record has probably been exaggerated because of oversimplistic interpretations of diamictite sequences. Existing process models of glaciomarine sedimentation derived from study of modern environments are sometimes difficult to employ in investigation of ancient sedimentary sequences because simple lithofacies criteria and typical vertical profiles are not available to aid in interpretation. In addition compositional data emphasized by many workers for distinguishing glaciomarine from continental glacial diamict(ite)s frequently fingerprint sediment source and not mode of deposition. The importance of facies analysis methods for isolating depositional environments is illustrated by three examples of ancient glaciomarine sequences. These are the Early Proterozoic Gowganda Formation (∼2.3 Ga) of northern Ontario, Canada; the Late Proterozoic Port Askaig Formation (∼670 Ma) of Scotland and Ireland, and the Late Cenozoic Yakataga Formation (∼20 Ma to recent) of the Gulf of Alaska. These examples illustrate the significance of detailed genetic studies of ancient glacial rocks in the interpretation of palaeogeographic and tectonic settings. Diamictite units in ancient glaciomarine sequences cannot be easily interpreted in terms of climatic or ice advance/retreat cycles, because of the varied controls on diamict accumulation and diamictite preservation in marine basins.

Timing of late Cenozoic tidewater glaciation in the far North Pacific

The onset of late Cenozoic glacial events in the far North Pacific Ocean is recorded by ice-rafted debris in the Yakataga Formation in the Gulf of Alaska. The dating of these events is controversial. Ages based on molluscs suggest that initial late Cenozoic tidewater glaciation occurred in the early middle Miocene (15-16 Ma). Previous work on planktic foraminifera indicates that this event is no older than late Miocene, probably latest Miocene (5-6 Ma). Resolution of this problem is important because the Yakataga Formation is the thickest, most complete, and best exposed repository of late Cenozoic glaciomarine rocks in the northern hemisphere. Investigation of this record has broader implications for global temperature gradients, paleoceanographic development of the far North Pacific, and northern hemisphere glacial history. New planktic foraminiferal data from onshore and offshore Yakataga Formation sections in the Gulf of Alaska are compared to a regional depositional and chronostratigraphic framework in order to test whether the Yakataga record is anomalous or consistent with the paleoclimatic record of the North Pacific. This comparison reconfirms that initial glaciomarine rocks of the Yakataga Formation are no older than late Miocene. New paleomagnetic data suggest that the base of the Yakataga Formation is within the lower Gilbert polarity chron, consistent with the planktic foraminiferal dating and K/Ar dates on glauconite. Correlation of the Yakataga record to key offshore sections in the far North Pacific (Deep Sea Drilling Project [DSDP] Sites 178, 183, 192) provides additional insights into the paleoclimatic significance of these rocks. Uplift of the Alaska coastal ranges, a necessary prelude to alpine glaciation, occurred during late Miocene time. Diatom biofacies (DSDP Sites 183, 192) indicate warm middle Miocene conditions, climatic deterioration during the late middle and early late Miocene time, and initial tidewater glaciation between 5.0 and 6.7 Ma. A relatively warm mid-Pliocene interval succeeded this initial glaciation, which was, in turn, followed by hemisphere-scale glaciation beginning at ∼2.5 Ma. This second phase of Yakataga glaciation may have begun as early as 3.0-3.5 Ma. Glaciomarine rocks in the basal Yakataga Formation mark an important climatic event in the North Pacific Ocean. They are latest Miocene in age, consistent with regional climatic information, and not a middle Miocene climatic anomaly.

Drumlins carved by deforming till streams below the Laurentide ice sheet

The Peterborough drumlin field (5000 km 2 ) is the largest in central Canada and was formed during readvance of the Laurentide ice sheet shortly after 13,000 B.P. Morphometric analysis of 998 drumlins, combined with study of 6921 borehole logs and downhole geophysical data, has established the distribution of drumlin types and their stratigraphy along a 70 km flow line through the drumlin field. In the north, drumlins are spindle-shaped, composed of till resting directly on bedrock, and were formed by deformation of till into low-pressure zones in the lee of bedrock scarps. Eskers record subglacial drainage by channeled meltwaters flowing on bedrock. To the south, drift thickness increases and eskers are absent; drumlins are larger and more ovoid and are composed of overridden proglacial and glaciolacustrine sediments overlain erosively by deformation till. Till is thin over drumlin crests and thick in interdrumlin swales and records subglacial cannibalization and mixing of overridden sediments; the drumlin form is the product of erosion by streams of deforming till flowing in swales separating cores composed of older sediments. Systematic down-field changes in drumlin shape and size can be related to decreased duration of subglacial deformation toward the limit of the readvance and increased thickness of preexisting sediments available for dissection by subglacial till streams. Absence of eskers in the central and southern parts of the drumlin field suggests that the ice base was drained by mass flow of the deforming substrate. The substantial subsurface data available from the Peterborough area indicate that drumlins are predominantly erosional in origin and are the geomorphological expression of a deforming subglacial bed below the margin of the Laurentide ice sheet.

Initiation and development of the Laurentide and Cordilleran Ice Sheets following the last interglaciation

Abstract Fossil records from sites overridden by or adjacent to the Laurentide Ice Sheet indicate that the climate of the last interglaciation (Oxygen-Isotope Substage 5e, ca. 130-116 ka) was warmer than today. Following the last interglaciation, the Laurentide Ice Sheet first developed during Stage 5 over Keewatin, Quebec and Baffin Island. Along its northern margin, the ice sheet reached its maximum extent of the last glaciation during Stage 5. The ice sheet advanced across Baffin Island onto the continental shelf early during Stage 5 (5d?), whereas the advance into the western Canadian Arctic occurred late during Stage 5 (5b?). The ice sheet also may have advanced into the St Lawrence Lowland during Substage 5b, although this event may be younger (Stage 4). The Hudson Bay lowland became ice-free during Substage 5a. Retreat of the ice sheet on Baffin Island occurred during late Stage 5, probably Substage 5a. The exact timing of retreat from the western Canadian Arctic is unknown, but it occurred before 48 ka. The southern sector, including the St Lawrence Lowland, was ice-free during late Stage 5. The Hudson Bay lowland may have remained ice free through Stage 4 and much of Stage 3. Because of conflicting chronologies, however, it is more likely that this area was glaciated throughout Stage 3 and perhaps Stage 4. Nevertheless, the data demonstrate that the lowland was ice-free during part of the last glaciation. An independent ice cap developed over the Appalachian Uplands and advanced across Nova Scotia during Stage 4, perhaps as far as the edge of the continental shelf. The ice cap remained active over Nova Scotia as a setellite to the Laurentide Ice Sheet throughout the remainder of the last glaciation. The ice sheet advanced into the St Lawrence Lowland during Stage 4 and subsequently overwhelmed the local ice cap in the Appalachian Uplands, advancing perhaps into northern New England, but not farther south. The Lowland remained covered by the ice sheet until late Stage 2. The ice sheet may also have advanced into the Lake Ontario basin during Stage 4. The position of the northern margin of the Laurentide Ice Sheet during Stage 4 is not known, but it remained an unknown distance behind its maximum position reached during Stage 5. Cores from Baffin Bay indicate a substantial decrease in high-latitude glaciation during Stage 4. Following retreat, the Keewatin sector of the ice sheet may have remained over much of northwestern Canada as a quasi-stable ice mass until it readvanced during Stage 2. Similarly, the Baffin Island sector of the ice sheet may have remained largely intact. The southern margin of the ice sheet may have advanced into the Lake Ontario basin and upper Mississippi Valley during the middle of Stage 3 (ca. 50 ka), reaching its maximum extent of the last glaciation during Stage 2 (ca. 18–21 ka). Advance of the northern margin was younger (ca. 8–13 ka) than that of the southern margin; this advance was less extensive than the penultimate advance (Stage 5). Paleoenvironmental records indicate that the last interglaciation in areas covered by and near the Cordilleran Ice Sheet was as warm as, or warmer than, present. The Cordilleran Ice Sheet appears to have developed during Stage 5 or 4. At that time, it advanced over southern British Columbia and into the northern Puget Lowland. There is no record of this event in northern areas that were later covered by the ice sheet. The ice sheet disappeared before 59 ka, at the beginning of a lengthy nonglacial interval. Paleoenvironmental records indicate that climate was similar to the present during part of this interval. The ice sheet was absent, and glaciers probably were confined to mountain areas, throughout Stage 3. Climatic deterioration marking the end of this nonglacial interval may have begun as early as 29 ka. By 14–15 ka, the ice sheet had achieved its maximum extent of the last glaciation. Because there are few suitable dating methods capable of resolving events beyond the radiocarbon limit and because sites that preserve a record of events from the last glaciation are spatially restricted, we consider this synthesis as tentative and subject to significant revision as dating methods improve. Nevertheless, this perspective of the North American ice sheets through the last glaciation demonstrates their complex and dynamic behavior and attendant rapid fluctuations in ice volume.

Architectural element analysis applied to glacial deposits: Internal geometry of a late Pleistocene till sheet, Ontario, Canada

Deposits left by continental ice sheets are characterized by sedimentological complexity and stratigraphic heterogeneity, but stratigraphic descriptions of such deposits, and resulting “first-generation” facies models, are still based primarily on one- or two-dimensional borehole or outcrop data. Reconstruction of depositional environments, hydrogeological investigations of Pleistocene glacial deposits, and hydrocarbon exploration in pre-Pleistocene glaciated basin fills require a more detailed understanding of the form and heterogeneity of lithofacies sequences in three dimensions. Architectural element analysis is used widely by sedimentologists for categorizing internal stratigraphic heterogeneity in sandstones, particularly those of fluvial origin. This paper demonstrates the first application of architectural element analysis to glacial deposits such as tills. Outcrop, borehole, and a broad range of subsurface geophysical data were collected from a thick (60 m) till sheet present across an 80 km 2 study area near Toronto, Canada. The till sheet is not homogeneous, but is composed of three distinct architectural elements and associated lithofacies, viz, diamict elements, interbeds of subglaciofluvial sediments, and glaciotectonically deformed zones. Application of architectural element analysis to these subglacial strata provides insights into the origin of drumlin bedforms and subglacial processes below the Laurentide Ice Sheet and creates a framework for understanding ground-water and contaminant movement in underlying aquifers.

Sedimentation in a large lake: A reinterpretation of the late Pleistocene stratigraphy at Scarborough Bluffs, Ontario, Canada

A lithofacies code developed for diamicts emphasizes facies variability in the classic upper Pleistocene glaciogenic sequence at Scarborough Bluffs, Ontario, Canada. Lithological breaks along the bluffs are traditionally assigned to incursions of grounded ice margins and to interstadial lakes formed during ice retreat, during the Early and Middle Wisconsin. Detailed sedimentological logging through three diamict units (previously formalized as Sunnybrook, Seminary, and Meadowcliffe Tills) and intervening sandy lithofacies shows absence of glaciotectonic structures and diamicts associated with grounded glacier ice, traction-current activity during diamict accumulation, postdepositional resedimentation of diamicts into topographic lows accompanied by turbidite activity, a subaqueous deltaic origin for intervening sandy lithofacies, and loaded, transitional, and interbedded contacts between sand and diamict. The Scarborough Bluffs sequence may be the preserved bottom stratigraphy of a large lake. The bottom stratigraphy results from repeated basinward progradation of deltaic sandy lithofacies over glaciolacustrine diamicts deposited below floating ice, whether ice shelf, ramp, bergs, or lake ice. A facies model is presented for glaciolacustrine diamict deposition on the floors of enlarged Pleistocene lakes trapping substantial volumes of fine-grained suspended sediment.

Constraints on the preservation of diamict facies (melt-out tills) at the margins of stagnant glaciers

Abstract This paper reviews the formation and preservation of diamict facies (melt-out till) generated in situ by the downwasting of stagnant ice at the margins of continental glaciers. Melt-out tills are commonly envisaged as being the product of the top (supraglacial) or bottom (subglacial) melt of stagnant debris-rich glacier ice but such deposits are relatively uncommon in modern glacier environments where they are thin, laterally discontinuous and have a low preservation potential. Theoretical considerations of the distribution of englacial debris in continental ice sheets show that only the glacier margin is likely to contain the necessary volumes of englacial debris for melt-out to be a significant diamict-forming process. However the application of thaw-consolidation theory to the formation of such deposits shows that substantial porewater pressures and self-sediment deformations can be induced during melt-out, especially in matrix-rich debris. The classic melt-out mechanism widely invoked in the literature since 1875 and involving passive in situ aggregation from thawing englacial debris can be seen to operate under restricted conditions. Diamict facies produced by melt-out are likely to form spatially disjunct and laterally discontinuous elements within complex and ice marginal stratigraphics; the melt-out process is not a mechanism whereby regionally extensive diamiet(ite) units can be deposited across glaciated basins.