Enhanced erosion by ice sheets on obstructive hill masses

Ounastunturi in Finland -- a hill mass which was in the way of the streaming ice of the Scandinavian ice sheet as it flowed eastwards from its source area. The hills look very much like those of Preseli --  moulded by a long history of glaciations.

Written in 1975 by David Sugden and myself, this book remained in print for 20 years between 1976 and 1996.  That's an extraordinary length of time for a university text, given the speed at which science moves.........

I have always had an instinctive feeling that the northern flank of Preseli is a perfect place for enhanced erosion beneath an ice sheet or ice stream coming from the NW and heading off to the SE.   Actually that is more than just instinctive -- it's something that can be predicted in glaciological theory as well -- and it has a lot to do with ice thickness and thermal regime.  When David Sugden and I wrote our text called "Glaciers and Landscapes" we tried to work out why glaciers erode in some places and protect in others, and we formulated a series of models which served to encourage a lot of other researchers, in the years following publication, to delve deeper into the theory and to apply it to their own field observations in different parts of the world. Hundreds of papers have appeared since then, many of them touching on the pointe we raised -- and the science has moved on by leaps and bounds. I have mentioned many of these papers in this blog.......

Here is a page from the article I wrote with Lionel Jackson in 2009, explaining why we thought that under an active Irish Sea Ice Stream, erosion and block entrainment would be enhanced on the northern flank of Preseli:

I don't think we were right in all respects, especially on the matter of shearing and / or internal deformation -- but the principles stand.  I still think that if the ice was to come again, there would be more erosion on the north (upglacier) slope than on the summit of the ridge or on the southern (downglacier) slope.

I was looking today at a paper that was published five years ago and which I had not taken all that seriously, but on re-reading it, I realised it is making a rather interesting and important point.

Unequal ice-sheet erosional impacts across low-relief shield terrain in northern Fennoscandia

Karin Ebert, Adrian M. Hall, Johan Kleman, Jannike Andersson
Geomorphology 233, 2015, pp 64-74

ABSTRACT
Much previous work on Late Cenozoic glacial erosion patterns in bedrock has focussed on mountain areas. Here we identify varying impacts of ice sheet erosion on the low-relief bedrock surface of the Fennoscandian shield,and examine the geological, topographical and glaciological controls on these patterns. We combine GIS-mapping of topographical, hydrological and weathering data with field observations. We identify and investigate areas with similar geology and general low relief that show different degrees of ice sheet erosional impact, despite similar ice cover histories. On two transects with a total area of ~84 000 km across the northern Fennoscandian shield, we first establish patterns of glacial erosion and then examine why glacially streamlined areas exist adjacent to areas of negligible glacial erosion. The northern transect includes two areas of exceptional glacial preservation, the Parkajoki area in Sweden and the so-called ice divide zone in Finland, each of which preserve tors and deep saprolite covers. The southern transect, overlapping in the northern part with the first transect, includes areas of well developed glacial streamlining, with bedrock areas stripped of loose material and barely any weathering remnants. For both areas, we firstly present contrasting indicators for ice sheet erosional impact: streamlined and non-streamlined inselbergs; parallel and dendritic/rectangular drainage patterns; and the absence and presence of Neogene weathering remnants. This is followed by an investigation of factors that possibly influence ice sheet erosional impact: ice cover history, ice cover duration and thickness, bedrock type and structure, and topography. We find that the erosional impact of the Fennoscandian ice sheet has varied across the study area. Distinct zones of ice sheet erosion are identified in which indicators of either low or high erosion coexist in the same parts of the transects. No direct impact of rock type on glacial erosion patterns was found, but an indirect control appears clear. Bedrock geology and long-term differential weathering and tectonic evolution determined the topography of the pre-glacial landscape, and these topographic differences subsequently influenced ice sheet dynamics and thereby partly controlled patterns of ice sheet erosion. Ice cover duration and former ice thickness were not significant controls on glacial erosion patterns. Extensive preservation of pre-glacial relief through low glacial erosion is attributed to the maintenance throughout the Pleistocene of divergent flow and frozen-bed conditions in the Fennoscandian ice sheet. In contrast, glacial streamlining and strong glacial erosion were caused mainly by acceleration of flow around major obstacles and flow towards major depressions on the ice sheet bed. The relatively strong ice sheet erosion towards the Gulf of Bothnia is the result of a combination of favourable factors: bedrock structure and river valleys aligned sub-parallel to ice sheet flow and convergent ice flow towards the Baltic

Of course there were many differences between the various manifestations of the Irish Sea Ice Stream and its equivalents in northern Fennoscandia, but ice behaves in the same way wherever it is, and one of the most interesting points in this paper describes how an ice sheet behaves when it is flowing from its upland accumulation area out towards its peripheries maybe 1000 km away.  In the transects studied in northern Sweden and Finland, the authors noted differences in the long bed profile of the ice from source to snout, and the roughness or bed obstacles encountered.  To bring things to their simplest conclusion:  where there was a continuous gradient, as in profile A-B, velocity accelerates and erosion is enhanced in the outer part of the glacier.  On the other hand, where ground surface slope decreases, and the ice is flowing over flattish or undulating terrain in its outer parts, ice flow decelerates and land surfaces are protected under cold-based ice conditions.  But the most interesting thing is that where the ice encounters an obstacle -- in the form of a hill mass or ridge perpendicular to the ice flow direction, there is a sudden shift from low glacial erosion (shown yellow on the diagram) to ice deformation and enhanced erosion (coloured brown).  Streamlined erosional features and enhanced entrainment of a bedrock load are inevitable, and have been observed empirically by the authors of the article.

The conclusion?  My little model, based on glaciological theory, is seen through empirical observation in northern Scandinavia to be correct.  During a big glaciation, with ice covering the whole of the West Wales landscape, where would erosion and entrainment be most powerful?  Answer: on the north slope of Preseli,  in locations such as Rhosyfelin, Carn Goedog, Carnedd Meibion Owen, Ty Canol and Carn Alw.  

Who needs quarries when ice can do the job far more effectively?

We may also have an explanation here for the relatively few traces of glaciation in South Pembrokehire -- in the lee of the mountains.  I must give that some further thought.