Till Stratigraphy and Sediment Provenance of the Dessert Lake Drumlin Field Area, West of Great Slave Lake, NT: Implications for Regional Glacial Dynamics Models

Abstract

The integration of till stratigraphic and sedimentologic studies with landform-based ice flow analysis has considerable potential for advancing our understanding of subglacial processes, particularly those related to subglacial sediment transport and the generation of subglacial landforms such as drumlins. Currently, few drumlin studies incorporate till provenance analysis, which limits our ability to establish relationships with subglacial sediment transport, drumlin forming processes, and the regional glacial history. Considerable effort has been made to reconstruct the last glaciation, with a particular focus on the geometry of the Laurentide Ice Sheet (LIS) ice margins and their retreat pattern and chronology. However, to comprehend the long-term evolution of the LIS during a glaciation, it is also essential to include the dynamics of ice streams, the positions of ice divides and their migration phases, and related shifts in ice flow. The history of glacial dynamics remains limited in many regions covered by the LIS. This research focuses on the Dessert Lake drumlin field, west of Great Slave Lake (GSL) within National Topographic System (NTS) map 85K. This area presents a significant opportunity to investigate tills and related sedimentary processes within a drumlin field due to road access and the presence of road cuts extending across several drumlins along Highway 3. Additionally, the bedrock geology consists of sufficiently contrasting bedrock units extending approximately perpendicular to the long axis of the drumlins, which is particularly useful for till provenance analysis. The glacial geology record of western GSL bears evidence for four distinct ice-flow phases of the LIS. These phases are preserved in various forms such as outcrop-scale erosional features, subglacial streamlined landforms, and in the orientation of clasts in different tills (a.k.a. till fabric) preserved within the drumlins, as well as in the composition of the till, which is useful to establish bedrock source regions (provenance). The sequence of ice flows indicates an initial southward (180 to 190 deg.) ice flow phase, followed by a clockwise shift towards westward (260 to 270 deg.) ice flow, then a slight deviation towards southwestward (215 to 245 deg.) flow. The counterclockwise shift is supported by the overprint relationship of southwesterly oriented landforms on westerly oriented landforms. The average direction of the drumlin’s long axis is 249 degrees, which is close to the top till fabric of drumlin section 1 (246 deg.). Finally, there is evidence along the shore of western GSL for a late northwestward (320 to 325 deg.) flow, which may have been limited in lateral extent. The till provenance analysis suggests some of the surficial till samples show a dominantly local provenance, but there is surficial till with a more distal signature such as more abundant Canadian Shield derived clasts. These changes in ice flow corresponded with the overall eastward retreat of the ice margin and the increasing influence of topography due to the diminishing thickness of the ice. In conclusion, glacial dynamics shifted in the area, and the evidence is well-preserved in both the erosional and depositional records. Till stratigraphy and related ice flow shifts are preserved within the drumlin cores. Since erosional processes are thought to play an important role in the formation of drumlins, it is likely, in the context of stratified drumlins, that the surficial till in the Dessert Lake drumlin field, and perhaps in other similar settings, consist of multiple till sheets deposited under different subglacial transport regimes and pathways and may thus have contrasting bedrock source regions. It is thus critical for applications like mineral exploration and environmental protection to determine if the till in lower stratigraphic positions occurs at the surface in the drumlin field.