P. Jay Fleisher

Tsivat Basin Conduit System persists through two surges, Bering Piedmont Glacier, Alaska

The 1993–1995 surge of Bering Glacier, Alaska, occurred in two distinct phases. Phase 1 of the surge began on the eastern sector in July, 1993 and ended in July, 1994 after a powerful outburst of subglacial meltwater into Tsivat Lake basin on the north side of Weeping Peat Island. Within days, jokulhlaup discharge built a 1.5 km 2 delta of ice blocks (25–30 m) buried in outwash. By late October 1994, discharge temporarily shifted to a vent on Weeping Peat Island, where a second smaller outburst dissected the island and built two new sandar. During phase 2, which began in spring 1995 and ended within five months, continuous discharge issued from several vents along the ice front on Weeping Peat Island before returning to the Tsivat Basin. Surge-related changes include a five- to six-fold increase in meltwater turbidity; the redirection of supercooled water in two ice-contact lakes; and an increase in the rate of glaciolacustrine sedimentation. U.S. Geological Survey aerial photos by Austin Post show large ice blocks in braided channels indicating excessive subglacial discharge in a similar position adjacent to Weeping Peat Island during the 1966–1967 surge. During the subsequent three decades of retreat, the location of ice-marginal, subglacial discharge vents remained aligned on a linear trend that describes the position of a persistent subglacial conduit system. The presence of a major conduit system, possibly stabilized by subglacial bedrock topography, is suggested by (1) high-level subglacial meltwater venting along the northern side of Weeping Peat Island during the 1966–1967 surge, (2) persistent low-level discharge between surges, and (3) the recurrence of localizing meltwater outbursts associated with both phases of the 1993–1995 surge.

The sedimentary architecture of outburst flood eskers: A comparison of ground-penetrating radar data from Bering Glacier, Alaska and Skeiðarárjökull, Iceland

We present ground-penetrating radar (GPR) profiles that reveal the sedimentary architecture of an esker deposited during a surge-associated outburst flood at the Bering Glacier, Alaska. The wide, up-flow end of the esker contains a transition from large backset beds to large foreset beds interpreted to reflect composite macroform development in an enlarged part of the conduit. By contrast, the narrow, down-flow portion of the esker is dominated by plane beds interpreted to have been deposited where the conduit was constricted and the flow was faster. A previously studied outburst esker at Skeidararjokull, Iceland, has a similar morphology and stratigraphic architecture. This suggests that outburst floods generate distinct depositional signatures in eskers, both in terms of morphology and sedimentary architecture. Identification of these distinct signatures in ancient eskers will help assess the paleohydraulic conditions under which ancient eskers formed and, by extension, the nature of meltwater drainage systems beneath the Laurentide and Eurasian ice sheets.

Tunnel channel formation during the November 1996 jökulhlaup, Skeiðarárjökull, Iceland

Abstract Despite the ubiquity of tunnel channels and valleys within formerly glaciated areas, their origin remains enigmatic. Few modern analogues exist for event-related subglacial erosion. This paper presents evidence of subglacial meltwater erosion and tunnel channel formation during the November 1996 jökulhlaup, Skeiðarárjökull, Iceland. The jökulhlaup reached a peak discharge of 45 000 to 50 000 m3 s–1, with flood outbursts emanating from multiple outlets across the entire 23 km wide glacier snout. Subsequent retreat of the southeast margin of Skeiðarárjökull has revealed a tunnel channel excavated into the surrounding moraine sediment and ascending 11.5m over a distance of 160 m from a larger trough to join the apex of an ice-contact fan formed in November 1996. The tunnel channel formed via hydro-mechanical erosion of 14 000m3 to 24 000 m3 of unconsolidated glacier substrate, evidenced by copious rip-up clasts within the ice-contact fan. Flow reconstruction provides peak discharge estimates of 680±140m3 s–1. The tunnel channel orientation, oblique to local ice flow direction and within a col, suggests that local jökulhlaup routing was controlled by (a) subglacial topography and (b) the presence of a nearby proglacial lake. We describe the first modern example of tunnel channel formation and illustrate the importance of pressurized subglacial jökulhlaup flow for tunnel channel formation.

Dead-ice sinks and moats: Environments of stagnant ice deposition

Glacier retreat from areas of moderate to high relief typically occurs along a lobate ice margin controlled by underlying topography, as was the situation on the Appalachian Plateau of central New York. Stagnation of entire valley ice lobes is attributed to restricted glacier flow across headward divides. Detachment of stagnant ice blocks is also related to limited flow through deeply incised valley meanders. Melting of a detached ice block, several thousand metres across and hundreds of metres thick, within a valley train creates a depression referred to as a dead-ice sink. Ice-lobe stagnation involving several closely spaced sinks results in a chain of depressions called a dead-ice moat. Topographic evidence of the dead-ice sink environment is an anomalously broad flood plain that spans most of the valley width and is bounded upvalley and downvalley by dissected valley-train outwash. Dead-ice sink sedimentation results in a stratigraphic sequence typical of an ice-contact environment and includes interstratified silt, sand, and gravel hundreds of metres thick. Juxtaposed water-well logs reveal stratigraphic units of highly variable texture and thickness that defy lateral correlation.

Structural controls on englacial esker sedimentation: Skeiðarárjökull, Iceland

Abstract We have used ground-penetrating radar (GPR) to observe englacial structural control upon the development of an esker formed during a high-magnitude outburst flood (jökulhlaup). The surge-type Skeiðarárjökull, an outlet glacier of the Vatnajökull ice cap, Iceland, is a frequent source of jökulhlaups. The rising-stage waters of the November 1996 jökulhlaup travelled through a dense network of interconnected fractures that perforated the margin of the glacier. Subsequent discharge focused upon a small number of conduit outlets. Recent ice-marginal retreat has exposed a large englacial esker associated with one of these outlets. We investigated structural controls on esker genesis in April 2006, by collecting >2.5km of GPR profiles on the glacier surface up-glacier of where the esker ridge has been exposed by meltout. In lines closest to the exposed esker ridge, we interpret areas of englacial horizons up to ~30m wide and ~10–15m high as an up-glacier continuation of the esker sediments. High-amplitude, dipping horizons define the base of esker materials across many lines. Similar dipping surfaces deeper in the profiles suggest that: (1) the dipping surfaces beneath the esker are englacial tephera bands; (2) floodwaters were initially discharged along structurally controlled englacial surfaces (tephra bands); (3) the rapid increase in discharge resulted in hydrofracturing; (4) establishment of preferential flow paths resulted in conduit development along the tephra bands due to localized excavation of surrounding glacier ice; and (5) sedimentation took place within the new accommodation space to form the englacial structure melting out to produce the esker.

A decade of sedimentation in ice-contact, proglacial lakes, Bering Glacier, AK

Abstract Bathymetric surveys during the 1991–2000 decade in two ice-contact, proglacial lakes on the eastern sector of Bering piedmont lobe captured the buildup effects of the 1993–1995 surge. Following ice-front advance of 1.0–1.5 km into Tsivat and Tsiu Lakes, the basins were significantly altered by surge-related sedimentation including the impact of a subglacial outburst into Tsivat Lake. The subsequent changes in basin shape, size, and morphology were monitored by six bathymetric surveys. Measured changes in water depth serve as a proxy for determining increments of sediment accumulation. Upwelling, ice-front vents fed by subglacial tunnels transported suspended fine sediment directly into the lake system. The rate of suspension settling within both lakes varied from 0.6 to 1.2 m year −1 prior to the surge. Suspended load during surge years increased sixfold from 1.7 to 13.9 g l −1 , accompanied by increased sediment accumulation of 2.2–3.1 m year −1 . Vent-related aggradation and subsequent filling of Tsivat Lake caused sediment bypassing to Tsiu Lake, where encroachment by delta growth contributed to a postsurge rate of bottomset accumulation of 3.0 m year −1 . The total sediment influx from subglacial sources is represented by the sum of bathymetrically determined accumulation, plus an estimated volume of sediment that remained suspended, thus passing through the lake system. Total sediment flux along the eastern Bering piedmont lobe from 1991 to 2000 is approximately 227 million cubic meters.

Pleistocene sediment sources, debris transport mechanisms, and depositional environments: a Bering Glacier model applied to northeastern Appalachian Plateau deglaciation, central New York

Abstract A modified ice-tongue model suggests that subglacial, saturated, fine sediment derived from local bedrock sources reduced basal shear strength and lowered the ice surface gradient sufficiently to produce ice tongues 20 km long in all major north-south oriented valleys on the northeastern Appalachian Plateau, while adjacent uplands were virtually ice-free. Associated environments of deposition produced two different landform assemblages, one representative of active ice retreat in through valleys and another that depicts widespread stagnation in non-through valleys. Pebble count data indicate that sediment transport by glacial flow was important to the moraine-building process, but the occurrence of isolated kame fields suggests an origin linked to inwash from major upland tributaries. All coarse valley fill (sand and gravel) is derived from two basic sources: (1) re-worked upland drift, and (2) resedimented debris from upvalley sources, including the glacier. Processes common to through valleys favor upvalley sources and active ice landforms, whereas inwash and stagnant ice sedimentation are typical of non-through valleys. Although extensive ice-free uplands served as a source of some fine sediment, a comparison of sediment volume to upland area indicates that inwash processes could not have yielded sufficient fines to account for the volume of fine sand and silt found within the valley fill. Meltwater flow via subglacial tunnels discharged saturated, fine sediment directly into proglacial lakes and served as the major source and transport mechanism for most sand and silt. The Laurentide deglacial environment throughout the upper Susquehanna region was characterized by proglacial lakes, detached remnant ice masses, dead-ice sedimentation and collapsed ice tongues. Stagnation and downwasting in ice-contact lakes peripheral to the eastern Bering Piedmont Glacier, Alaska, serve to depict analog conditions for retreat in central New York.

Laboratory Models of Glacier Dynamics

Excellent laboratory models of glaciers were produced by pouring a mixture of molding plaster and water down a prefashioned trough of nonabsorbent material. The plaster models simulate the dynamics of glacier flow and accurately display many surface and internal structures common to real glaciers, including crevasse and flow patterns, icefalls and ogives, shears, and folds. Internal and surface reference lines depict the nature of the deformation produced by flow. The technique is successful as a teaching aid and shows definite potential as a research tool.