Laurentide Ice-Flow Patterns: A Historical Review, and Implications of the Dispersal of Belcher Islands Erratics

This paper deals with the evolution of ideas concerning the configuration of flow patterns of the great inland ice sheets east of the Cordillera. The interpretations of overall extent of Laurentide ice have changed little in a century (except in the Arctic) but the manner of growth, centres of outflow, and ice-flow patterns, remain somewhat controversial. Present geological data however, clearly favour the notion of multiple centres of ice flow. The first map of the extent of the North American ice cover was published in 1881. A multi-domed concept of the ice sheet was illustrated in an 1894 sketch-map of radial flow from dispersal areas east and west of Hudson Bay. The first large format glacial map of North America was published in 1913. The binary concept of the ice sheet was in vogue until 1943 when a single centre in Hudson Bay was proposed, based on the westward growth of ice from Labrador/Québec. This Hudson dome concept persisted but was not illustrated until 1977. By this time it was evident from dispersal studies that the single dome concept was not viable. Dispersal studies clearly indicate long-continued westward ice flow from Québec into and across southern Hudson Bay, as well as eastward flow from Keewatin into the northern part of the bay. Computer-type modelling of the Laurentide ice sheet(s) further indicates their complex nature. The distribution of two indicator erratics from the Proterozoicage Belcher Island Fold Belt Group help constrain ice flow models. These erratics have been dispersed widely to the west, southwest and south by the Labrador Sector of more than one Laurentide ice sheet. They are abundant across the Paleozoic terrain of the Hudson-James Bay lowland, but decrease in abundance across the adjoining Archean upland. Similar erratics are common in northern Manitoba in the zone of confluence between Labrador and Keewatin Sector ice. Scattered occurences across the Prairies occur within the realm of south-flowing Keewatin ice. As these erratics are not known, and presumably not present, in Keewatin, they indicate redirection and deposition by Keewatin ice following one or more older advances of Labrador ice. The distribution of indicator erratics thus test our concepts of ice sheet growth.