No abstract

Borehole observations from the base of the West‐Antarctic Ice Sheet (WAIS) reveal the presence of a 10 to 15 m thick accretionary basal ice layer in the upstream area of Kamb Ice Stream (KIS). This ice layer has formed over a time of several thousand years by freeze‐on of subglacial water to the ice base and has recorded during this time basal conditions upstream of its current location. Analysis of samples and videos sequences from boreholes drilled to the bottom of KIS confirms that KIS‐stoppage was due to basal freeze‐on and that relubrication of the ice stream is well underway. These results further suggest that ice stream cyclicity may be shorter than expected (1000s of years) and that a restart of KIS may be imminent within decades to centuries.

A real-time video camera probe was deployed in a hot-water drilled borehole through the Amery Ice Shelf, East Antarctica, where a total ice thickness of 480 m included at least 200 m of basal marine ice. Down-looking and side-looking digital video footage showed a striking transition from white bubbly meteoric ice above to dark marine ice below, but the transition was neither microscopically sharp nor flat, indicating the uneven nature (at centimetre scale) of the ice-shelf base upstream where the marine ice first started to accrete. Marine ice features were imaged including platelet structures, cell inclusions, entrained particles, and the interface with sea water at the base. The cells are assumed to be entrained sea water, and were present throughout the lower 100-150 m of the marine ice column, becoming larger and more prevalent as the lower surface was approached until, near the base, they became channels large enough that the camera field of view could not contain them. Platelets in the marine ice at depth appeared to be as large as 1-2 cm in diameter. Particles were visible in the borehole meltwater; probably marine and mineral particles liberated by the drill, but their distribution varied with depth.

[1] We have constructed a numerical model of basal ice formation for glacier ice in contact with subglacial sediment. The model predicts four different ice facies whose formation is controlled by availability of subglacial water to satisfy the basal freeze-on rate. Clean (or clotted) facies may result from congelation (or frazil) ice growth occurring when (supercooled) meltwater separates ice base from substrate or if a groundwater source can supply water to the ice base at a velocity equal to the freezing rate. Laminated facies develops when the supply of subglacial water is sufficiently constrained for cryostatic suction to raise the subglacial effective stress above a threshold for intrusion of ice into sediment by regelation. Debris laminas ($1 mm) are entrained by short, periodic regelation events (of a few hours) separated by longer periods (days to months) of congelation. Further meltwater limitation produces a massive dirty ice facies due to stacking of debris laminas. The model predicts growth of solid dirty ice facies when the bed is meltwater-depleted with fast freezing (5 mm yr À1) causing enhanced erosion (30 mm yr À1 ). We find that basal ice facies and sediment entrainment are controlled mainly by the ratio of freezing rate to water supply rate. The predicted ice facies compare favorably with borehole camera imagery of the basal ice layer in Kamb Ice Stream, West Antarctica. Facies variability in this layer suggests complex hydrologic history for the West Antarctic Ice Sheet with significant changes occurring over a period of several thousand years.Citation: Christoffersen, P., S. Tulaczyk, F. D. Carsey, and A. E. Behar (2006), A quantitative framework for interpretation of basal ice facies formed by ice accretion over subglacial sediment,