SIGNIFICANCE OF STAGNATION ON THE TIMING OF MASS WASTAGE OF THE LAURENTIDE ICE SHEET
Glacial-budget estimates depend on adequately constraining ice-marginal advance and retreat rates together with ice profiles. Three glacial retreat models are recognized. The Midwest model suggests persistence of an active ice front and a largely ice-free foreground (active-retreat end-member). In the New-England model, stagnant ice lingered in bedrock valleys as interfluves were deglaciated (stagnation end-member). The between-member, High-Plains model is characterized by persistent, broad, and often thick stagnant highland ice detached from active, broad lowland lobes. Glacial budgets are more likely to be underestimated for the latter two models, with implications for the rate and timing of sea-level rise with deglaciation.
Breakup of the ice on the Missouri Coteau in Saskatchewan and the Dakotas is a case in point. Nearly 3 kyrs were required for ice to abandon the Big Muddy spillway south of the Coteau (14 ka), provide closure for Lake Regina in a broad lowland, open the QuAppelle spillway, and ultimately drain Lake Regina (11 ka). During this nearly 113-km retreat of the ice front, from 9 to more than 60 m of ablation drift and associated landforms developed on the Coteau. Marl and charcoal were underlain by stagnant ice and not in final depositional position around 10.8-10.9 ka. These deposits are presently located 120 to 305 m above younger proglacial Regina lake beds that extend nearly 72 km to the northeast, in the direction of ice retreat. From 60 to 200 m of stagnant ice lingered over a 64-km wide and 193-km long belt within and near map-sheet 72 H alone, insulating roughly 7.41e2 to 2.47e3 km^3 of ice by 10.9 ka. A similar, 2 to 3 kyr period of insulation was reported to the southeast in North Dakota.