Sediment sources of northern Québec and Labrador glacial deposits and the northeastern sector of the Laurentide Ice Sheet during ice-rafting events of the last glacial cycle

Abstract

Provenance studies of anomalously high-flux layers of ice-rafted detritus (IRD) in North Atlantic sediments of the last glacial cycle show evidence for massive iceberg discharges coming from the Hudson Strait region of the Laurentide Ice Sheet (LIS). Although these so-called Heinrich events (H events) are commonly thought to be associated with abrupt drawdown of the LIS interior, uncertainties remain regarding the sector(s) of this multi-domed ice sheet that conveyed ice through Hudson Strait. In Northern Québec and Labrador (NQL), large-scale patterns of glacial lineations indicate massive ice flows towards Ungava Bay and Hudson Strait that could reflect the participation of the Labrador–Québec ice dome in H events. Here we evaluate this hypothesis by constraining the source of NQL glacial deposits, which provide an estimate of the provenance characteristics of IRD originating from this sector. Specifically, we use 40Ar/ 39Ar ages of detrital hornblende grains in 25 till samples distributed along a latitudinal transect (lat. 58°) extending east and west of Ungava Bay. The data show that tills located west and southwest of the Ungava Bay region are largely dominated by hornblende grains with Archean ages (>2.6 Ga), while tills located east of Ungava Bay are characterized by grains with early Paleoproterozoic ages (2.0–1.8 Ga), although most samples contain a few Archean-age grains. IRD derived from the NQL region should thus be characterized by a large proportion of Archean-age detrital grains, which contrasts significantly with the predominant Paleoproterozoic 40Ar/ 39Ar ages (1.8–1.6 Ga) typically reported for the dominant age population of hornblende grains in H layers. Comparisons with IRD through the last glacial cycle from a western North Atlantic core off Newfoundland do not show evidence for any prominent ice-rafted event with the provenance characteristics of NQL glacial deposits, thereby suggesting that significant ice-calving event(s) from the Labrador–Québec sector may have been limited throughout that interval. Although these results tend to point towards a relative stability of this ice dome during H events, our study also indicates that further provenance work is required on IRD proximal to the Hudson Strait mouth in order to constrain with a greater confidence the sector(s) of the LIS that fed ice into Hudson Strait during H events. Alternatively, these results and other paleogeographic considerations tend to support models suggesting that part of the Ungava Bay glacial lineations could be associated with a Late-Glacial ice flow across Hudson Strait.