The Scandinavian Ice Sheet: from MIS 4 to the end of the Last Glacial Maximum KURT LAMBECK, ANTHONY PURCELL, JASON ZHAO AND NILS-OLOF SVENSSON BOREAS Lambeck, K., Purcell, A., Zhao, J. & Svensson, N-O. 2010 (April): The Scandinavian Ice Sheet: from MIS 4 to the end of the Last Glacial Maximum. Boreas, Vol. 39, pp. 410–435. 10.1111/j.1502-3885.2010.00140.x. ISSN 0300-9483. Glacial rebound modelling, to establish constraints on past ice sheets from the observational evidence of palaeo- shoreline elevations, is well established for the post- Last Glacial Maximum (post-LGM) period, for which the observational evidence is relatively abundant and well distributed spatially and in time. This is particularly the case for Scandinavia. For the earlier part of the glacial cycle this evidence becomes increasingly sparse and uncertain such that, with the exception of the Eemian period, there are very few, if any, direct sea-level indicators that con- strain any part of the Scandinavian Ice Sheet evolution before the LGM. Instead, we assume that ice-sheet basal conditions during Marine Isotope Stage 3 (MIS 3) are the same as those for the LGM, focus on establishing these conditions from the rebound analysis for the LGM and Lateglacial period, and then extrapolate to the earlier period using observationally constrained locations of the ice margins. The glacial rebound modelling and inversion follow previously established formulations, with the exception that the effects of water loading from proglacial lakes that form within the Baltic Basin and elsewhere have been included. The data set for the inversion of the sea- and lake-level data has been extended to include marine-limit data in order to extend the observational record further back in time. The result is a sequence of time slices for the Scandinavian Ice Sheet from the time of MIS 4 to the Lateglacial that are characterized by frozen basal conditions until late in the LGM interval when rapid thinning occurred in the eastern and southern sectors of the ice sheet. The primary function of these models is as an inter- polator between the fragmentary observational constraints and to produce quantitative models for the glaciation history with predictive capabilities, for example regarding the evolution of the Baltic Basin. Kurt Lambeck (e-mail: Kurt.Lambeck@anu.edu.au) and Anthony Purcell (e-mail: tony@rses.anu.edu.au), Re- search School of Earth Sciences, Australian National University, Canberra ACT 0200, Australia; Jason Zhao (e-mail: jason.zhao@ga.gov.au), Geoscience Australia, GPO Box 378, Canberra ACT 2601, Australia; Nils-Olof Svensson (e-mail: nils-olof.svensson@geol.lu.se), GeoBiosphere Science Centre, Lund University, 223 62 Lund, Sweden; received 17th April 2009, accepted 4th December 2009. Ice-sheet coverage over Europe since the last Last Glacial Maximum (LGM) is well established from sedimentolo- gical, stratigraphic and geomorphological data, from gla- ciological and climate forcing models, and from inversion of isostatic rebound sea-level data, or from a combination thereof (e.g. Boulton et al. 2001; Siegert et al. 2001; Ar- nold et al. 2002). Ice sheets for the earlier period of the last glacial cycle are less well constrained, in part because the older observational record has been overprinted by more recent glacial activity and because of the decreasing accuracy and reliability of the chronological control. As a result, ice-sheet models for earlier epochs become in- creasingly assumption-dependent and uncertain. The observations do point to major fluctuations in ice margins across arctic Eurasia during the earlier part of the glacial cycle as well as to periods of extensive ice- free conditions in otherwise cold-climate intervals. The evidence points to significantly higher-than-present sea levels during the Last Interglacial (Eemian) in the Gulf of Bothnia and to the simultaneous submergence of the North Siberian Plain. For other times, the field data are indicative of high-elevation pre-LGM marine limits. A principal reason for developing an ice model for the pre-LGM interval is to produce models for the glacia- tion history that have predictive capabilities. The re- sulting ice-sheet reconstructions would then provide boundary conditions for regional climate models with corresponding implications for human, floral and fau- nal distributions. Such results would be particularly important for the Marine Isotope Stage (MIS) 3 period, in which Homo sapiens first appeared in Europe (e.g. van Andel & Davies 2003) (see Table 1 for nominal ages and durations of the isotope stages). We have chosen to explore the glacial rebound ana- lysis approach for constraining past ice sheets. This has been successfully applied to the period extending from the late LGM to the present, an interval for which the geographical extent of the ice sheet is well constrained and for which there is an abundance of observational data to effect the inversion for ice thickness (Lambeck et al. 1998b, referred to as L98 hereafter). Inversion of glacial rebound data has also been attempted for the period from the end of the penultimate glacial max- imum (MIS 6) to MIS 4, for which there is a growing body of field data to constrain the ice margins and shoreline positions for some of the principal stadials and interstadials (Lambeck et al. 2006, referred to as L06 hereafter). To complete our model for the last glacial cycle, we consider the period from the start of MIS 3 (c. 60 kyr BP) into the LGM, for which the field evidence is minimal. We focus on an analysis of the glacial rebound signal at the time of the LGM, using observational evidence constraining some of the principal ice-sheet advances and DOI 10.1111/j.1502-3885.2010.00140.x r 2010 The Authors, Journal compilation r 2010 The Boreas Collegium