Drumlins and mega-scale glacial lineations as a continuum of subglacial shear marks: A LiDAR based morphometric study of streamlined surfaces on the bed of a Canadian paleo-ice stream. (15th September 2022)

• Record Type: Journal Article
• Authors: Sookhan, Shane; Eyles, Nick; Bukhari, Syed
• Is Part Of: Quaternary science reviews. Volume 292(2022)
• Publication Date: 2022-09-15
• Subjects: Paleo-ice streams -- LiDAR mapping -- Drumlins -- Mega-scale glacial lineation continuum -- Streamlined surfaces -- Basal drag -- Subglacial traction -- Shear plane; Geology, Stratigraphic -- Quaternary -- Periodicals; Stratigraphie -- Quaternaire -- Périodiques; 551.79
• DOI: 10.1016/j.quascirev.2022.107679
• Languages: English
• ISSNs: 0277-3791

Abstract

Abstract: We employ and further develop quantitative analytical techniques used to map roughness on fault planes, to map and classify drumlins and mega-scale glacial lineations (MSGLs) on the bed of a terrestrially-terminating Late Wisconsin paleo ice stream lobe exposed near Peterborough in Southern Ontario, Canada. Geological data show these bedforms are erosional remnants of older sediments, principally coarse-grained proximal outwash, capped by a thin (<2 m) upper till. Our earlier work ascribed the formation of drumlins and MSGLs to differential erosion of antecedent sediment by a thin layer of deforming subglacial debris (the Erodent Layer Hypothesis) which we now test using new geostatistical and unsupervised machine learning techniques applied to large LiDAR-derived digital topographic datasets. Analysis of more than 1739 streamlined bedforms mapped by Curvature-Based Relief Separation, reveals a statistically-validated continuum of eight morphotypes (M1 to M8) ranging from drumlins to grooved multi-crested drumlins to MSGLs. Furthermore, cluster analysis of the entire morphotype population, reveals three statistically-validated groups that define three streamlined surface types (S1 to S3) which record progressive erosional lowering of the ice stream bed, as drumlins evolve into multi-crested morphotypes, and ultimately MSGLs. Calculation of surface roughness using a technique called Median Absolute Differences (MAD) shows the frictional drag of mega-scale glacially lineated surfaces (S3) is 60% less than drumlinized surfaces (S1). Our findings support earlier ideas that drumlins and MSGLs are 'longitudinal shear marks' akin to those cut by extended frictional wear on fault planes, to which can now also be added analogous streamlined surfaces cut by wear debris mobilized at the base of landslides, ejecta blankets and pyroclastic flows, and on engineered man-made surfaces. Highlights: Employs and develops quantitative analytical techniques used to map roughness on fault planes, to map and classify drumlins and mega-scale glacial lineations (MSGLs). New geostatistical and unsupervised machine learning techniques applied to large LiDAR-derived digital topographic datasets. 1739 streamlined bedforms mapped by Curvature-Based Relief Separation, reveals a statistically-validated continuum of eight morphotypes (M1 to M8) ranging from drumlins to MSGLs. Cluster analysis of the entire morphotype population, reveals three statistically-validated groups that define three streamlined surface types (S1 to S3) which record progressive erosional lowering of the ice stream bed. Calculation of surface roughness shows the frictional drag of mega-scale glacially lineated surfaces (S3) is 60% less than drumlinized surfaces (S1).