The Late-Devensian proglacial Lake Humber: new evidence from littoral deposits at Ferrybridge, Yorkshire, England MARK D. BATEMAN, PAUL C. BUCKLAND, BRIAN CHASE, CHARLES D. FREDERICK AND GEOFF D. GAUNT BOREAS Bateman, M. D., Buckland, P. C., Chase, B., Frederick, C. D. & Gaunt, G. D. 2008 (May): The Late-Devensian proglacial Lake Humber: new evidence from littoral deposits at Ferrybridge, Yorkshire, England. Boreas, Vol. 37, pp. 195–210. 10.1111/j.1502-3885.2007.00013.x. ISSN 0300-9483. Proglacial Lake Humber is of UK national significance in terms of landscape drainage and development of the British Ice Sheet (BIS) during Marine Isotope Stage 2 (MIS 2), yet it is poorly understood in terms of its dynamics and timing. Sands and gravels exposed at Ferrybridge, West Yorkshire, UK, are interpreted as part of the Upper Littoral sands and gravels related to a high-level Lake Humber, which inundated the Humber Basin to 30 m OD during MIS 2. Excavations exposed well-rounded gravels of local origin extending downslope from the 27.5 m OD contour and interbedded sands and fine gravels, which are interpreted as the coarse littoral deposits and nearshore associated deposits. A sample from the distal sands returned an Optically Stimulated Luminescence age of 16.61.2 kyr, providing the first direct age for the high-level lake and for when North Sea Basin ice must have blocked the Humber Gap. An underlying sequence included a diamicton dated to after 23.3 1.5 kyr and before 20.51.2 kyr, indicating that the Late Devensian ice reached at least 15 km south of the Escrick Moraine prior to the high-level lake. Previous to both the high-level lake and this ice advance, loess found at the two sites investigated indicates a long period of loess deposition earlier in MIS 2. These new data for the history of Lake Humber are discussed in the context of ice-marginal oscillations in both the Vale of York and the North Sea Basin. Mark D. Bateman (e-mail: m.d.bateman@Sheffield.ac.uk), Sheffield Centre for International Drylands Research, Department of Geography, University of Sheffield, Winter Street, Sheffield S10 2TN, UK; Paul C. Buckland, Den Bank Close, Sheffield S10 5PA, UK; Brian Chase, Oxford University Centre for the Environment, University of Oxford, South Parks Road, Oxford OX1 3QY, UK; Charles D. Frederick, Department of Geography and the Environment, University of Texas, Austin, Texas, USA; Geoff D. Gaunt, Department of Archaeological Sciences, University of Bradford, Bradford BD7 1DP, UK; received 20th February 2007, accepted 11th September 2007. Presently, approximately one-quarter of the land sur- face of England drains through the Humber Gap in the Chalk escarpment, 15 km west of Hull (Fig. 1) (Versey 1940; Rees 2006). At its narrowest point downstream of North and South Ferriby, where the Humber Bridge now crosses the estuary, the gap is only 4 km wide. With the Chalk rising rapidly to over 60 m on either side, it is apparent that any impediment to flow would lead to the development of an extensive lake to the west in the Humberhead Levels and the Vales of Trent and York. The concept of a large proglacial lake, ponded back by ice in the North Sea Basin, appears to have been first formulated by Lamplugh (1887) and developed by Lewis (1894). Clark et al. (2004a) have emphasized that, with the exception of the British Ice Sheet (BIS), Lake Humber is the single most prominent feature on the map of Late Devensian Britain, covering at its max- imum in the region of 4500 km 2 . However, its age has remained poorly constrained. Edwards (1937) mapped locally derived gravels along the dipslope of the Mag- nesian Limestone outcrop west of York, which he thought defined a shoreline at approximately 30 m OD (Fig. 2). More recent detailed work, largely by Gaunt (1976a, b; 1981, 1987, 1994; Gaunt et al. 1992, 1994), has clarified the position of a later low-level lake with a shoreline at approximately 7 m OD, the laminated clay- silt deposits of which form the gently sloping plain of the present Vale of York and Humberhead Levels. Gaunt’s work also indicates glacio-isostatic depression of lake shorelines northwards towards the BIS, which extended down the Vale at least as far as the York– Escrick moraines, and shows laminated lake sediments both underlying and onlapping with the moraine sequences (Gaunt 1970). A short phase when this ice surged forward into the lake appears to be defined by a discontinuous sand and gravel ridge running across the southern part of the Humberhead Levels NE of Doncaster (Gaunt 1976a, b), although this interpreta- tion has been questioned (e.g. Straw 2002a, b). Sum- maries of the broader Quaternary sequence in the region are presented in Bateman et al. (2001a) and Gaunt et al. (2006). This article provides new sedi- mentological and chronometric evidence for the estab- lishment and demise of the high-level lake and also its relationship with the dynamic margins of the BIS, both in the Vale of York and in the eastern part of the North Sea Basin. Previous chronological evidence Despite the extensive exposures of laminated clay-silt (Hemingborough Formation of Bowen 1999) and other sediments associated with the lake, no evidence of DOI 10.1111/j.1502-3885.2007.00013.x r 2007 The Authors, Journal compilation r 2007 The Boreas Collegium