Ice sheet and glacier behaviour in Svalbard

Apparently deranged glacier behaviour on Ellesmere Island, with two big glaciers flowing in the same trough in opposite directions, and with five other smaller glaciers flowing into the trough from flanking tributary valleys.  What is the explanation?

Another glacier collision, this time in East Greenland.  The big glaciers are wasting away, but this is a very complex pattern of conflicting ice flows.

When we see images such as those above, the only explanation of the landforms and glaciers is that the landscape is "inherited" from one or more past episodes in which the big troughs were cut by ice streams that had little or no relationship with the glaciological conditions that currently prevail. 

Anyway, I came across this article a few years ago, and have taken another look at it in the realisation that it also says something about the pattern and history of glaciation in the Celtic Sea arena of the British Isles.  It's a very interesting and enjoyable piece of work.  

Landscape imprints of changing glacial regimes during ice-sheet build-up and decay: a conceptual model from Svalbard
Jon Y. Landvik, Helena Alexanderson, Mona Henriksen, Ólafur Ingólfsson
Quaternary Science Reviews 92 (2014), pp 258-268

https://www.academia.edu/18557412/Landscape_imprints_of_changing_glacial_regimes_during_ice_sheet_build_up_and_decay_a_conceptual_model_from_Svalbard?email_work_card=view-paper

Abstract

The behaviour of ice sheets and their geologic imprints in fjord regions are often multifaceted. Fjords, which were temporarily occupied by fast flowing ice-streams during major glaciations, and inter-fjord areas, which were covered by less active ice, show different signatures of past glaciations. The land and marine records of glaciations over the western Svalbard fjord region have been extensively studied during the last few decades. We have re-examined ice-flow records from stratigraphic and geomorphic settings, and propose a succession of ice-flow events that occurred repeatedly over glacial cycles: the maximum, the transitional, and the local flow style. The differently topographically constrained segments of the ice-sheet switched behaviour as glacial dynamics developed through each glacial cycle. These segments, as well as the different flow styles, are reflected differently in the offshore stratigraphic record. We propose that the glacial geomorphological signatures in the inter ice-stream areas mostly developed under warm-based conditions during a late phase of the glaciations, and that the overall glacial imprints in the landscape are strongly biased towards the youngest events.

Figure 3 from the article, showing ice flow directions of different ages at the selected sites.  The arrows are overlaid from oldest to youngest.  At first this might look like chaos, or "random" ice flows, but on closer examination you can seed that the oldest ice flows represent a phase (or phases) when the islands were overwhelmed by the ice of the Barents Sea ice sheet, flowing broadly westwards towards its terminus at the shelf edge; and that later phases show the assertion of local topographical controls, with local glaciers becoming increasingly important for the evacuation of glacier ice. 

The authors pull in a vast amount of evidence from submarine and land-based studies (much of it already published) to support the hypoythesis of three different glacial dynamic events for the Late Weichselian glaciation: the maximum glaciation phase, the tributary ice-stream phase, and the local ice caps and readvances phase. They also link these phases to “flow styles” as suggested by Landvik et al. (2013): maximumflow style, transitional flow style and local flow style based, on stratigraphic records as well as geomorphic data. The main criterion for the classification is the degree of topographic constraint on the flow.

1.  Maximum flow style existed when at least parts of the ice sheet experienced topographically unconstrained ice flow during “maximum glaciation” episodes -- of which there have been many. For the most extensive glaciation in the records (the Late Weichselian), the maximum flow style reconstruction is primarily based on seafloor megascale glacial lineations (MSGL) in the fjords and troughs, often associated with terminal moraines on the continental shelf reflecting largely topo- graphically unconstrained ice flow. In the fjords and troughs, the till deposition and MSGL formation suggest that ice-streams existed, whereas the inter-fjord areas experienced a less active ice flow towards the ice-sheet margin.

2.  Transitional flow style is topographically constrained, mainly in a fjord setting. It is found in “tributary ice- streams”, but may also be characteristic of topographically constrained non ice-stream behaviour. Similarly to the maximum flow style, the signatures of transitional flow include seafloor MSGL and drumlins, recessional moraines and grounding zone wedges, and the flow direction is easily identified when it differs distinctly from the preceding maximum flow. For the last glaciation, the transitional flow style is also documented by land records from sites along the fjords: glacial striae, till fabric and provenance of erratic boulders.

3.  Local flow style is also topographically constrained, predominantly in the terrestrial environment -- but here smaller tributary troughs and cirques have a powerful influence.. The most important signatures are terminal moraines and tills deposited by ice flowing from local ice caps, piedmont glaciers, or valley and cirque glaciers. Abundant geological signatures of local flow style from the last deglaciation overprint the older landscape and sediments. In the stratigraphic record, the local flow style is primarily represented by tills containing clasts of local provenance supported by preferred clast orientation and glaciotectonic structures, as well as proglacial meltwater deposits formed during periods of high relative sea level.

Quote:

Fjord/trough (ice-stream) and inter-fjord (inter ice-stream) areas exhibit different glacial landscape imprints, reflecting different flow styles and basal temperature conditions. Ice-streams in the fjords and troughs were probably dominated by thawed bed. The inter-fjord areas seems to have experienced thawed basal conditions only during a late phase of ice-sheet retreat, leaving the distinct glacial signatures of an active zone that migrated over the older landscape.

The composite western Svalbard glacial record shows that these ice-flow styles occurred in a preferred succession during both ice- sheet build-up and decay. Due to the higher preservation potential of the younger geological imprints, developed under warm-based conditions, most studies utilizing geomorphic data for ice-flow reconstructions have tended to be biased towards the latest ice-flow styles, usually the local or transitional flow style.

We suggest the succession of flow styles shown for western Svalbard occurred on a full ice-sheet scale and in particular that the transitional and local flow styles shifted time transgressively to- wards the palaeo centre of the ice sheet as a function of changing ice-sheet thickness and volume. This conceptual model should be valid also for similar palaeo ice-sheet settings in e.g. Greenland, western Norway, eastern Canadian Arctic and Iceland.

There is some interesting modelling in this paper, with the authors examining both the spatial and time dimensions.  They refer to "glacial cycles" and spend some time discussing the landscape inheritance of past glaciations and the powerful "skewing" effect of the LGM or  Late Weichselian glacial episode -- through the complete or partial obliteration of the field evidence (both stratigraphic and landform) from past episodes, some of which might have been more prolonged and more powerful.

 

Map showing the succession of flow styles during the LGM on the west coast of Svalbard.  Note that at the LGM several powerful ice streams occupy the main "glacier discharge routes" and that between these, in the inter-ice-stream areas, there is still enough ice flowing from the barents ice sheet interior to overwhelm the mountains and flow out to the ice edge on the shelf margin. Because the flow is topographically unconstrained, there are no "ice free enclaves" in this piedmont area; ice flowing sluggishly will spread laterally until it is confined or limited by more powerful ice associated with the main discharging ice streams.

The relevance of this for the Celtic Sea arena?  Well, it's not so different.  There was one large ice stream -- the Irish Sea ice stream (ISIS) or glacier, carrying ice from the ice sheet interior.  At some stages there was unconstrained flow across parts of Wales including Lleyn and Pembrokeshire; and at some stages the flow was more like the "transitional flow style" of Landvik and his colleagues, based upon the Welsh Ice Cap operating "autonomously" as an independent accumulation area with radiating ice flow.   The relationships between Irish Sea Ice and Welsh Ice are still being worked out -- and  the Celtic Sea piedmont musty also be worthy of study in its own right.  Again, as pointed out by James Scourse and many other researchers, sea-level rise and fall -- and isostatic adjustments -- will have exerted, at times, a powerful control on the position of the ice edge and the dynamics of the ice masses.

There has been much discussion on this blog about the "shape" and extent of the Celtic Sea ice lobe, as attested by assorted maps as reproduced below.

Glasser et al, 2018, showing what is in my view a highly unlikely eastern edge to the Irish Sea Ice Stream.  The Svalbard research shows that large ice streams in unconstrained situations flow laterally and engulf the lowlands -- meaning that the Bristol Channel is highly unlikely to have remained ice-free. 

An attempt by Prof David Evans to portray a Celtic Sea piedmont dominated by ice flowing from southern Ireland

My own attempt at showing what might have happened with unconstrained ice flow in the Celtic Sea arena during the LGM.  The extent of ice cover must have been even more extensive during earlier glacial episodes........