Collapse of sediment-filled crevasses associated with floods and mass flows in the proximal zone of the Pernunnummi sandurdelta, III Salpausselkä, SW Finland

Abstract

We aim to explain the complex geomorphology and sedimentology of the proximal zone of the Pernunnummi sandurdelta, which forms the most prominent portion of the III Salpausselkä ice-marginal complex of the former Scandinavian ice sheet in SW Finland. We focus on the sedimentology of distinct, diamicton-covered glaciofluvial and glaciolacustrine landforms (kames), making accurate interpretations of both proximal diamictons and related sedimentary facies: a crucial approach in the reconstruction of the depositional history of ice-marginal zones. We propose that the build-up of the sandurdelta was followed by the formation of a glaciofluvially-dominated supraglacial landsystem within a stagnant-ice zone, which fringed a glacial embayment between sub-lobes of different deglacial dynamics. The role of an integrated meltwater system including a major subglacial tunnel and water-filled crevasses was crucial in constructing the controlled topography of this supraglacial landsystem. The stagnant-ice zone contributed to the incremental retreat of the ice margin within a shallow-water environment, where growing end moraines and proglacial deposits led to the development of supra-aquatic conditions. Our results also evoke temporary, dammed water bodies, which were supported by ice-marginal frozen bed conditions within the Baltic Sea icestream. The collapse of crevasses with ice-contact glaciofluvial and glaciolacustrine sediments during the continued deglaciation led to outbursts of dammed meltwater, accompanied by extensive mass flows that formed the diamictons partially covering the ice-proximal zone. Because ice-marginal and esker-related, glaciofluvially-dominated landform complexes with disintegration topography are relatively common along the main ice-marginal deposits of the last great ice sheets, the interpreted sedimentary record must be more common than previously reported. Thus, extensive surficial and proximal diamictons of the ice-marginal complexes do not always indicate either a re-advance or a major oscillation of the ice front. This study promotes the understanding of deglacial icestream dynamics with no known modern analogues and hence, helps to improve both palaeogeographic reconstructions and the future compilation of even more accurate ice sheet models.