Jonathan L. Bamber

The role of ice thickness and bed properties on the dynamics of the enhanced-flow tributaries of Bailey Ice Stream and Slessor Glacier, East Antarctica

Abstract Airborne radio-echo sounding investigations in the upper reaches of Bailey Ice Stream and Slessor Glacier, Coats Land, East Antarctica, have shown that enhanced-flow tributaries are associated with well-defined areas of relatively thicker ice, and are separated from each other by areas of relatively thinner ice. A numerical modelling study has revealed that while internal ice deformation might account for all the observed flow in inter-tributary areas and the majority in the Slessor tributaries, a significant proportion of the flow of Bailey tributary is attributable to basal motion. Further, investigations of depth-corrected basal reflection power indicate that the bed underlying both Bailey and Slessor enhanced-flow tributaries is significantly smoother than in the slower-moving inter-tributary areas. It is thus proposed that enhanced motion within Bailey tributary (and also perhaps Slessor) may be facilitated by a reduction in basal roughness, caused by the accumulation of water and/or sediments within subglacial valleys, or by the erosion and smoothing of bed obstacles.

The englacial stratigraphy of Wilkes Land, East Antarctica, as revealed by internal radio-echo sounding layering, and its relationship with balance velocities

Abstract The disruption of internal layering visible in radio-echo sounding (RES) data from East Antarctica has, to date, been attributed to ice flow around bedrock topography. However, observations of internal layer disruption in the Siple Coast ice streams of West Antarctica have led to the suggestion that increased strain at the margins of ice streams may also be responsible for the disruption of internal layers. Here we present a re-analysis of the extensive RES dataset collected between 1967 and 1979 over a large part of Wilkes Land, East Antarctica, and relate the location of continuous and disrupted internal layers to modelled balance velocities. We show that the mean balance velocity associated with all areas of disrupted layers is 2.5 times higher than that associated with areas of continuous layers. We also demonstrate that disrupted layers are associated not only with ice streams, but also with areas of enhanced ice flow, which penetrate inland from the grounding line up to several hundred kilometres into the interior of East Antarctica. Continuous layers always overlie disrupted layers, suggesting either a depth dependency in the process responsible for layer disruption, or subsequent deposition of continuous layers. In some cases, disrupted layers occur outside fast-flow features, and continuous layers occur within fast-flow features. Such regions are explained by short-term flow patterns, but might also be attributed to inaccuracies in the balance-velocity calculations.

Brain state dependent stimulus information in the auditory thalamocortical system

Activity in the absence of stimuli is ubiquitous across the thalamocortical system (TS), with patterns of spontaneous activity reflecting ongoing behavioural state. Under anaesthesia and during deep sleep the TS operates in an inactivated state (characterised by low frequency high amplitude oscillations in local field potential (LFP)) in which neurons collectively alternate between periods of local silence and high synaptic activity. During wakefulness and REM sleep, however, the TS operates in an activated state (characterised by high frequency low amplitude oscillations in LFP) in which neurons fire in a sustained desynchronised manner. Such brain states may be indicative of different modes of neural processing, but the effect of brain state on neural processing remains unclear. n nHere we analyse data recorded in the auditory TS of urethane anaesthetised rats subjected to single-click acoustic stimulation over a range of intensities, presented in both the inactivated state (natural under the anaesthesia) and the activated state (induced through electrical stimulation of the basal forebrain). Evoked spike trains were identified for single units in the auditory thalamus and across depths of the primary auditory cortex. Mutual information (MI) between stimulus and response was then computed for spike probability, counts and timing. n nEvoked spiking activity was observed to last around 300ms, consisting of distinct initial and secondary (rebound) activity. Analysis of responses of single units over a full 300ms window showed that spike count and timing measures gave little more MI than response probability, suggesting that stimulus intensity is primarily encoded probabilistically, at least at the level of the single unit. Moreover, whilst many single units are uninformative of stimulus intensity, the most informative thalamorecipient (TC) and infragranular (IF) single units show increased MI in the activated state. Additionally, upon temporally partitioning data according to initial or rebound activity, we see that TC units are more informative in the activated state during initial activity whereas IF units are more informative in the activated state during rebound activity, suggesting that information loss in the inactivated state may be cumulative in time as sensory signals propagate through neural circuits.