Self‐organization in the thermomechanical flow of ice sheets

Abstract

Abstract. This paper uses a three-dimensional, thermomechanical ice sheet model to investigate the initiation and evolution of ice streams within ice sheets. Ice streams are distinguished within general ice sheet flow by their relatively fast (of the order of kilometers per year) velocity and are thought to be crucial in determining the response of ice sheets to climatic change. We show that streaming can arise solely as a consequence of internal feedback between ice sheet flow and temperature. Inhornogeneities in topography and/or geology are not therefore necessary conditions for ice stream development. This spatial patterning replaces the large-scale surging reported from previous modeling work in one horizontal dimension. We suggest that surging is restricted to situations where ice flow is ch•nnelized (e.g., v•lley glaciers) •nd that it is replaced by streaming where horizontal ice flow is unconstrained (ice sheets). The ice streams in some of our experiments show switching behavior reminiscent of that known from the Siple Coast ice streams, West Antarctica. The cause appears to be the competition for ice discharge and drainage basin area between adjacent ice streams. This may lead to the gradual reduction of a stream's discharge below the threshold necessary to generate enough frictional heat to maintain warm, viscous ice at the ice stream bed. In this case the stream will stagnate abruptly, which then prompts localized ice thickening and the initiation of a new stream in a different location.

On the reconstruction of palaeo-ice sheets: Recent advances and future challenges

“…As a result of new satellite and digital elevation products, there has been a huge increase in the number of inferred ice streams ( Figure 5), such that we now have a good knowledge of their location in most palaeo-10 ice sheets Livingstone et al, 2012a;Margold et al, 2015). Moreover, recent work has revealed abrupt changes in the trajectory of palaeo-ice streams, known as 'flow switching' (Dowdeswell et al, 2006;Winsborrow et al, 2012), which has also been reported from observations of present-day ice sheets (Conway et al, 2002) and numerical modeling (Payne and Dongelmans, 1997). Thus, reconstructions of palaeo-ice stream activity provide new insights into the long-term behaviour of ice streams (and their potential forcing) and often help reconcile complex cross-cutting flow-set patterns (discussed in Section 2.1) (Dyke and Morris, 1988;Stokes et al, 2009).…”

The Dubawnt Lake palaeo-ice stream: evidence for dynamic ice sheet behaviour on the Canadian Shield and insights regarding the controls on ice-stream location and vigour

“…Some ice streams can arise purely by glaciological controls (Payne & Dongelmans 1997). In this case, the ice stream acted as a release valve on ice-sheet accumulation.…”

“…Bird 1953;Craig 1964) may have been an important seed point for ice-stream initiation. Increased velocity at the margin may have led to greater basal friction, increased basal meltwater for lubrication and warmer less viscous ice, leading to a further increase in basal sliding velocity (see Payne & Dongelmans 1997). These thermomechanical feedback mechanisms would have propagated warm-based conditions inland.…”

Small-scale topographically-controlled glacier flow switching in an expanding proglacial lake at Breiðamerkurjökull, SE Iceland

“…Flow switching is associated with changes in ice dynamics (e.g. Payne and Dongelmans, 1997;Conway and others, 2002) and could lead to either increases or decreases in velocity. It is therefore of importance to future mass-balance predictions, although few examples of flow switches taking place in modern glaciers have been documented.…”

“…(1) accumulation of sediment causing accommodation space to be filled (Dowdeswell and others, 2006); (2) variations in bathymetry causing asynchronous response of ice flow (Stokes and others, 2009); (3) variations in basal thermal regime (Ó Cofaigh and others, 2010); (4) variations in subglacial meltwater routing and sticky spot location (Anandakrishnan and Alley, 1997); and (5) large-scale changes in mass balance (Payne and Dongelmans, 1997). All research into flow switching to date has been carried out on flow switches that have occurred in the past, from scales ranging from the last 150 years (Retzlaff and Bentley, 1993) to several glacial cycles (Dowdeswell and others, 2006).…”