Ice Flow Shift, Drumlin, and Bedrock Topography Effects on Glacial Dispersal from Kimberlites in the Lac de Gras Region, Northwest Territories, Canada

Abstract

Glaciations of the Quaternary Period have had a profound impact on the Canadian landscape and surficial sediments. Understanding glacial landforms and sediments has thus helped develop fundamental knowledge of past glaciations and how they affected landscapes, sediments, and climates. Because Quaternary Period sediments and landforms are ubiquitous in many regions of Canada, studying and understanding them is also important for numerous applications such as mineral exploration. Successful diamond exploration in the Northwest Territories, Canada, for instance, requires detailed knowledge of the glacial history, including ice flow phases and the net effect of several glacial processes. This is because glacial sediments often cover bedrock of interest, and because they can be used for tracing certain minerals, referred to as indicator minerals, back to a buried source. Traditional and proven exploration techniques using glacial sediments evolved from tracing mineralized boulders back to their source to a greater diversity of methods using indicator minerals and geochemical pathfinders. Typically, regional scale programs involve collecting surface samples of glacial sediment (till) and analyzing compositional patterns of interest, sometimes in combination with geophysics to find the most promising bedrock targets. However, this analysis becomes a complex 3D problem in areas of variably thick till. There is also a temporal component often involving shifts in ice flows and thus of sediment transport directions. To address these issues, companies sometimes use a grid-style drilling strategy where till samples are collected vertically at different depths. However, this is a costly and time-consuming approach that may be applied with incomplete knowledge of the full glacial history and/or with limited understanding of bedrock topographic effects and changing subglacial conditions on 3D dispersion. To address this knowledge gap, this research uses a large reverse circulation (RC) drilling and hand-sampling dataset from the Harry Winston (HW) camp south of Lac de Gras in the Northwest Territories to improve our understanding of how till thickness and bedrock topography control or affect sediment dispersal from known kimberlites in areas that were affected by shifting ice flow phases during the last glaciation. Two study areas were selected to investigate two different problems. The first problem is about the effect of subglacial landform forming processes on till composition and 3D dispersal patterns. For example, recent models about drumlin formation (aka drumlinization) suggest they could be dominantly erosional, raising the possibility for the preservation of dispersal patterns inherited from ice flow phases that preceded drumlinization. The second problem is related to kimberlites that did not produce a clear dispersal train in glacial sediments. Here, the research investigates the specific case of a kimberlite (Big Blue) nested in a bedrock topographic high consisting of more resistant rocks (granitoids) than the surrounding lower terrain (metasediments). In both study areas, landform-scale and outcrop-scale ice flow indicators record a clockwise shift in ice flow from southwest to west, to northwest. Although the most dominant and youngest ice flow is to the northwest, there are still erosional evidence leftover (e.g., roches moutonnées) from the oldest and greater sediment-generating ice flow phases. RC drilling is not sensitive enough to determine the detailed vertical till stratigraphy. Nonetheless, useful insights can be obtained from compositional data about the possible occurrence of relict till preserved at the base of deeper boreholes, or increasingly inherited till at depth, such as at the core of drumlins or in lee-side deposits. The research problems were investigated by analysing and interpreting clast lithology and kimberlite indicator mineral (KIM) patterns at depth and at surface throughout the two selected study areas within the large HWRC dataset. Kimberlite indicator mineral patterns in the east study area demonstrate well-preserved isolated 3D sediment dispersal in a drumlin field that persists through the full thickness of the till column. Field observations and satellite imagery of landforms and ice flow indicator measurements constrain till and indicator mineral provenance, while multiple datasets, including KIM grain counts (peridotic and eclogitic pyrope, olivine, Cr-diopside, chromite, and picroilmenite), pathfinder geochemistry (Ni, Cr, Co Nb, Ba, Rb), and clast lithology (granitoid pebble counts %), indicate high inheritance in the thicker till within the drumlin field associated with the oldest SW ice flow phase. In addition, partial overprinting and till re-entrainment by the younger W and NW ice flow events reshaped the landscape thereby fragmenting and smearing older dispersal trains that are preserved in the subsurface sediments. Although the surficial till in the drumlin field could be a discontinuous till cap, erosional processes could have exposed relict till at the surface. Bedrock topography protected till at depth in the west study area, preserving a weak till dispersal train from a distal kimberlite at the base of a bedrock topographic high, whereas the lack of a dispersal train from the subcropping Big Blue kimberlite nested in a granitoid knob may be explained by the formation of a lee-side cavity, causing the glacier to pass over the kimberlite. Lee-side cavity filling and the formation of a crag-and-tail landform around the granitoid knob are considered the main mechanisms for explaining till thickness variations and vertical compositional changes identified from deeper boreholes. Further analysis of other settings with soft rock bodies nested in bedrock knobs would need do be conducted to determine whether the lack of surficial dispersal trains is common in this situation, or if it is due to specific or rare local conditions. Field observations and satellite imagery of glaciofluvial corridors that wrap around Big Blue as well as small channels that run overtop of the granitoid knob could have received sediment supply from the kimberlite, or alternatively the corridors could have re-entrained and mobilized till containing inherited indicator minerals sourced from distal kimberlites up the older up-ice ice flow direction (NE). Overall, this study further refines and constrains the local ice-flow history south of Lac de Gras, characterizes a well-preserved palimpsest dispersal train in a drumlin field, and provides a model for explaining the lack of a clear dispersal train from a known kimberlite nested in a bedrock topographic high. Till deposited during earlier ice flow phases appear to have been better preserved within drumlins and other areas of thicker till, such as down-ice bedrock steps. Steps in bedrock topography served to prevent erosion of pre-existing lee-side sediment, as well as of a nested kimberlite, to the extent of preventing a surficial dispersal train from forming down-ice of the kimberlite. This study thus provides a conceptual model for explaining why, in certain cases, kimberlites that intersect the bedrock surface may not produce a clear dispersal train at surface. Results from this research suggest there is an opportunity to re-examine subsurface till dispersal patterns in legacy datasets in the Lac de Gras region to better consider effects of ice flow shifts, drumlin formation, and bedrock topographic effects. This could potentially lead to new discoveries.