The Characteristics of Ribbed Moraine and Assessment of Theories for Their Genesis

Ribbed (Rogen) moraines are large subglacially formed transverse ridges that cover extensive areas of the beds of the former Laurentide. Fennoscandian and Irish ice sheets. Since the flow speeds and stability of ice sheets are known to be sensitive to conditions operating at the bed. a full understanding of the processes of ribbed moraine genesis are critical if we are to appreciate their role in ice sheet dynamics. Numerous hypotheses exist that seek to explain how ribbed moraine are formed: by shearing and stacking of subglacial sediments (e.g. Aylsworth and Shilts. 1989: Bouchard. 1989): as a consequence of subglacial megafloods (Fisher and Shaw. 1992): by fracturing and extension of frozen till sheets (Hattestrand. 1997) or by sediment deformation (Boulton. 1987: Hindmarsh. 1998a.b. 1999). However, given that their formation has never been observed, advances in knowledge have to rely on abduction. That is. hypotheses, theories or models are generated that seek to best explain their characteristics (pattern, size, shape, internal composition etc). The most successful hypothesis will be the one that provides the most complete explanation of ribbed moraine characteristics. This approach is problematic given that ribbed moraine characteristics are poorly known. Scrutiny of the literature reveals that detailed observations are limited to small areas (normally < 1500 km2) and with small sample sizes (i.e. probably <50 ridges). Generalisations drawn from this base cannot be regarded as being representative and thus remain an inadequate data source for testing the various hypotheses. This paper addresses this deficit by producing the first representative data set on ribbed moraine characteristics.Various remote sensing and GIS techniques were used to record the morphological, morphometric and spatial characteristics of ribbed moraines in Ireland, Canada and Sweden, over a combined area of 81.000 km2. This database of 36.000 individually mapped landforms demonstrates that many published accounts are somewhat misleading and unrepresentative, and that ribbed moraine morphology is more complex than was hitherto reported. For example it is shown that ribbed moraines often form independent of topographic influences, are not always curved down-ice. do not have accordant summits, can have both steep proximal and distal sides, have undulating crests and resemble waves, are not always anastomosing and do not necessarily fit neatly together like a jigsaw. Our database of ribbed moraine ridge length, width, height and wavelength is also presented and reveals they exist over a much larger scale range than was previously thought. These new data were used to test the various ribbed moraine theories. This led to the rejection of the topographic model of shear and stack and undermined the credibility of all other shear and stack hypotheses, the two-step hypothesis, the megaflood hypothesis and the thermal fracturing model of formation. Given that ribbed formation is not due to localised factors such as topography and on the basis that their patterns are repetitive, organised into dominant wavelength and are widespread it is strongly argued they are most likely the product of an instability mechanism that operates in the subglacial environment.