Search

B.J.D was funded in the form of a PhD studentship provided by the Scottish Association for Geosciences, Environment and Society (SAGES) and the University of St Andrews, UK. J.M.L is supported by a UKRI Future Leaders Fellowship (Grant No. MR/S017232/1). D.F would like to acknowledge the support of this work through the EPSRC and ESRC Centre for Doctoral Training on Quantification and Management of Risk and Uncertainty in Complex Systems Environments Grant No. (EP/L015927/1). Surface‐derived meltwater can access the bed of the Greenland Ice Sheet, causing seasonal velocity variations. The magnitude, timing and net impact on annual average ice flow of these seasonal perturbations depends on the hydraulic efficiency of the subglacial drainage system. We examine the relationships between drainage system efficiency and ice velocity, at three contrasting tidewater glaciers in southwest Greenland during 2014‐2019, using high‐resolution remotely sensed ice velocities, modelled surface melting, subglacial discharge at the terminus and results from buoyant plume modelling. All glaciers underwent a seasonal speed‐up, which usually coincided with surface melt‐onset, and subsequent slow‐down, which usually followed inferred subglacial channelisation. The amplitude and timing of these speed variations differed between glaciers, with the speed‐up being larger and more prolonged at our fastest study glacier. At all glaciers, however, the seasonal variations in ice flow are consistent with inferred changes in hydraulic efficiency of the subglacial drainage system, and qualitatively indicative of a flow regime in which annually‐averaged ice velocity is relatively insensitive to inter‐annual variations in meltwater supply – so‐called ‘ice flow self‐regulation’. These findings suggest that subglacial channel formation may exert a strong control on seasonal ice flow variations, even at fast‐flowing tidewater glaciers. Publisher PDF

Calving glaciers contribute substantially to sea level rise, but they are challenging to represent in models. Fine resolution is required for continental-scale models to accurately resolve calving dynamics, and in many cases glacier geometry is too complicated to be adequately reflected by more simplified models. Flowline models are able to resolve flow along the main branch of a glacier, but many of those in current use either ignore tributaries entirely or parameterize their effect using a measure of “equivalent width.” Here we present a simple method to simulate terminus advance and retreat for an interacting network of glacier branches, based on a model extending Nye's (1953) perfect plastic flow approximation to calving glaciers. We apply the method to case studies of four marine-terminating glaciers: Jakobshavn Isbræ and Helheim Glacier of Greenland, and Columbia and Hubbard Glaciers of Alaska. Given bed topography and upstream elevation history, our method reproduces observed patterns of terminus advance and retreat in all cases, as well as centerline profiles for all branches.

Since 1990, Yahtse Glacier in southern Alaska has advanced at an average rate of ~100 m year−1 despite a negative mass balance, widespread thinning in its accumulation area, and a low accumulation-area ratio. To better understand the interannual and seasonal changes at Yahtse and the processes driving these changes, we construct velocity and ice surface elevation time series spanning the years 1985–2016 and 2000–2014, respectively, using satellite optical and synthetic aperture radar (SAR) observations. We find contrasting seasonal dynamics above and below a steep (up to 35% slope) icefall located approximately 6 km from the terminus. Above the icefall, velocities peak in May and reach their minima in October synchronous with the development of a small embayment at the calving terminus. The up-glacier minimum speeds, embayment, and plume of turbid water that emerges from the embayment are consistent with an efficient, channelized subglacial drainage system that lowers basal water pressures and leads to focused submarine melt in the calving embayment. However, velocities near the terminus are fastest in the winter, following terminus retreat, possibly off of a terminal moraine that results in decreased backstress. Between 1996 and 2016 the terminus decelerated by ~40% at an average rate of ~0.4 m day−1 year−1, transitioned from tensile to compressive longitudinal strain rates, and dynamically thickened at rates of 1-6 m year−1, which we hypothesize is in response to the development and advance of a terminal moraine. The described interannual changes decay significantly upstream of the icefall, indicating that the icefall may inhibit the upstream transmission of stress perturbations. We suggest that diminished stress transmission across the icefall could allow moraine-enabled terminus advance despite mass loss in Yahtse's upper basin. Our work highlights the importance of glacier geometry in controlling tidewater glacier re-advance, particularly in a climate favoring increasing equilibrium line altitudes.