Outlet Glacier and Landscape Evolution of Victoria Land, Antarctica

<p>Terrestrial cosmogenic exposure studies are an established and rapidly evolving tool for landscapes in both polar and non-polar regions. This thesis takes a multifaceted approach to utilizing and enhancing terrestrial cosmogenic methods. The three main components of this work address method development, reconstructing surface-elevation-changes in two large Antarctic outlet glaciers, and evaluating bedrock erosion rates in Victoria Land, Antarctica. Each facet of this work is intended to enhance its respective field, as well as benefit the other sections of this thesis. Quartz purification is a necessary and critical step to producing robust and reproducible results in terrestrial cosmogenic nuclide studies. Previous quartz purification work has centred on relatively coarse sample material (1 mm-500 μm) and is effective down to 125 μm. However, sample material finer than that poses significant purification challenges and this material is usually discarded. The new purification procedure outlined in this thesis shows that very fine sand size material (125-63 μm) can be reliably cleaned for use in terrestrial cosmogenic nuclide studies. The results below show that 35% mass loss in very fine-grained quartz is sufficient to remove major elements (Al, Ti, Na, K, Fe, Mg, Ca, Mn,) and trace elements (9Be, and 10B) along with meteoric 10Be. Insufficient leaching is most detrimental to Al concentration, however errors up to 27% in exposure age and up to 29% in erosion rate are possible if meteoric 10Be is not fully removed from quartz during the HF leaching stages. Outlet glaciers have been well observed since the beginning of the satellite era, approximately 60 years ago. However, we do not currently know how these important glaciers, which drain a significant portion of the Antarctic Ice Sheet, have behaved on centennial to millennial timescales. Dating glacial erratics deposited by a thinning outlet glacier provides a window into the long-term outlet glacier and ice sheet response to climatic forcing. New results in this thesis constrain the thinning history of Mawson and Tucker glaciers over the last several thousand years. Mawson Glacier undergoes rapid thinning from at least ~6.5 kya to ~4.9 kya then transitions to slower thinning until ~1 kya, with a minimum of 250 m of ice-surface-lowering. While Tucker Glacier ~450 km north undergoes gradual thinning from ~19 kya to ~5 kya with ~300 m of ice-surface-lowering. The results of this work show that either the Tucker Glacier was not significantly affected by the Ross Ice Shelf grounding line, or that Antarctic mountain glaciers respond differently to the outlet glaciers connected to the Easty Antarctic Ice Sheet. The style, rate, magnitude, and duration of thinning is unique to each outlet glacier, even with similar climate forcing. The results of this work shed light on the style and duration of outlet glacier thinning and retreat that is possible following a climate perturbation. Antarctica’s average bedrock erosion rate is consistently lower than 4.5 m/Myr, the lowest bedrock erosion rates for any region on Earth. Therefore, many cosmogenic dating studies assume zero erosion when calculating exposure ages. However, previous erosion rate work in Antarctica is biased to arid high-elevation inland sites (~60% of work) and the hyperarid ice-free McMurdo Dry Valleys (~40% of work). These studies do not capture the effects of coastal maritime climates, where many outlet glacier studies are conducted, on the rate of bedrock erosion. New results presented in this thesis show that the Northern Victoria Land coast has the highest known erosion rate in Antarctica. Two sample sites were selected, one coastal and one in the interior. The coastal bedrock erosion rates are 8.86±0.78 m/Myr and 7.15±0.6 m/Myr while the interior bedrock erosion rates are 1.07±0.08 m/Myr and 0.42±0.03 m/Myr. The coastal erosion rates are average for non-polar cold climates while the inland sites are below average for polar erosion rates. The results suggest a strong gradient in the rate of erosion is present from the Antarctic coastline inland. If exposure ages are not calculated with an appropriate erosion rate the apparent age may under-estimate the actual age by as much as 12%, which is thousands of years for Holocene thinning histories like those found in this thesis.</p>

Outlet Glacier and Landscape Evolution of Victoria Land, Antarctica

<p>Terrestrial cosmogenic exposure studies are an established and rapidly evolving tool for landscapes in both polar and non-polar regions. This thesis takes a multifaceted approach to utilizing and enhancing terrestrial cosmogenic methods. The three main components of this work address method development, reconstructing surface-elevation-changes in two large Antarctic outlet glaciers, and evaluating bedrock erosion rates in Victoria Land, Antarctica. Each facet of this work is intended to enhance its respective field, as well as benefit the other sections of this thesis. Quartz purification is a necessary and critical step to producing robust and reproducible results in terrestrial cosmogenic nuclide studies. Previous quartz purification work has centred on relatively coarse sample material (1 mm-500 μm) and is effective down to 125 μm. However, sample material finer than that poses significant purification challenges and this material is usually discarded. The new purification procedure outlined in this thesis shows that very fine sand size material (125-63 μm) can be reliably cleaned for use in terrestrial cosmogenic nuclide studies. The results below show that 35% mass loss in very fine-grained quartz is sufficient to remove major elements (Al, Ti, Na, K, Fe, Mg, Ca, Mn,) and trace elements (9Be, and 10B) along with meteoric 10Be. Insufficient leaching is most detrimental to Al concentration, however errors up to 27% in exposure age and up to 29% in erosion rate are possible if meteoric 10Be is not fully removed from quartz during the HF leaching stages. Outlet glaciers have been well observed since the beginning of the satellite era, approximately 60 years ago. However, we do not currently know how these important glaciers, which drain a significant portion of the Antarctic Ice Sheet, have behaved on centennial to millennial timescales. Dating glacial erratics deposited by a thinning outlet glacier provides a window into the long-term outlet glacier and ice sheet response to climatic forcing. New results in this thesis constrain the thinning history of Mawson and Tucker glaciers over the last several thousand years. Mawson Glacier undergoes rapid thinning from at least ~6.5 kya to ~4.9 kya then transitions to slower thinning until ~1 kya, with a minimum of 250 m of ice-surface-lowering. While Tucker Glacier ~450 km north undergoes gradual thinning from ~19 kya to ~5 kya with ~300 m of ice-surface-lowering. The results of this work show that either the Tucker Glacier was not significantly affected by the Ross Ice Shelf grounding line, or that Antarctic mountain glaciers respond differently to the outlet glaciers connected to the Easty Antarctic Ice Sheet. The style, rate, magnitude, and duration of thinning is unique to each outlet glacier, even with similar climate forcing. The results of this work shed light on the style and duration of outlet glacier thinning and retreat that is possible following a climate perturbation. Antarctica’s average bedrock erosion rate is consistently lower than 4.5 m/Myr, the lowest bedrock erosion rates for any region on Earth. Therefore, many cosmogenic dating studies assume zero erosion when calculating exposure ages. However, previous erosion rate work in Antarctica is biased to arid high-elevation inland sites (~60% of work) and the hyperarid ice-free McMurdo Dry Valleys (~40% of work). These studies do not capture the effects of coastal maritime climates, where many outlet glacier studies are conducted, on the rate of bedrock erosion. New results presented in this thesis show that the Northern Victoria Land coast has the highest known erosion rate in Antarctica. Two sample sites were selected, one coastal and one in the interior. The coastal bedrock erosion rates are 8.86±0.78 m/Myr and 7.15±0.6 m/Myr while the interior bedrock erosion rates are 1.07±0.08 m/Myr and 0.42±0.03 m/Myr. The coastal erosion rates are average for non-polar cold climates while the inland sites are below average for polar erosion rates. The results suggest a strong gradient in the rate of erosion is present from the Antarctic coastline inland. If exposure ages are not calculated with an appropriate erosion rate the apparent age may under-estimate the actual age by as much as 12%, which is thousands of years for Holocene thinning histories like those found in this thesis.</p>

Bedrock erosion in subglacial channels

The Labyrinth in the McMurdo Dry Valleys of Antarctica is characterized by large bedrock channels emerging from beneath the margin of Wright Upper Glacier. To study the morphodynamics of large subglacial channels cut into bedrock, we develop herein a numerical model based on the classical theory of subglacial channels and recent results on bedrock abrasion by saltating bed load. Model results show that bedrock abrasion in subglacial channels with pressurized flow reaches a maximum at an intermediate distance up-ice from the glacier snout for a wide range of sediment grain sizes and sediment loads. Close to the snout, the velocity is too low and the sediment particles cannot be mobilized. Far from the snout, the flow accelerates and sediment is transported in suspension, thus limiting particle impacts at the channel bottom and reducing abrasion. This non-monotonic relationship between subglacial flow and bedrock abrasion produces concave up bottom profiles in subglacial channels and potential cross-section constrictions after channel confluences. Both landforms are present in the bedrock channels of the Labyrinth. We therefore conclude that these geomorphic features are a possible signature of bedrock abrasion, rather than glacial scour, and reflect the complex interplay between transport rate, sediment load, and transport capacity in subglacial channels.

Central Himalayan rivers record the topographic signature of erosion by glacial lake outburst floods

In steep landscapes, river incision sets the pace of landscape evolution. Transport of coarse sediment controls incision by evacuating material delivered to river channels by landslides. However, large landslide-derived boulders that impede bedrock erosion are immobile even in major runoff-driven floods. Glacial lake outburst floods (GLOFs) mobilize these boulders and drive incision, yet their role in regional-scale erosion is poorly understood, largely because of their rarity. Here, we find a topographic signature consistent with widespread GLOF erosion in the Nepal Himalaya. In rivers with glaciated headwaters that generate GLOFs, valleys stay narrow and relatively free of sediment, with bedrock often exposed to erosion. In turn, tributaries to these valleys are steep, allowing less efficient erosional regimes to keep pace with GLOF-driven incision. Where GLOFs are less frequent, valleys are more alluviated and incision stalls. Our results suggest the extent of headwater glaciation may play an important role in erosion of Himalayan river valleys and deserves more attention in future work.

Mass balance controls on sediment scour and bedrock erosion in waterfall plunge pools

Waterfall plunge pools experience cycles of sediment aggradation and scour that modulate bedrock erosion, habitat availability, and hazard potential. We calculate sediment flux divergence to evaluate the conditions under which pools deposit and scour sediment by comparing the sediment transport capacities of waterfall plunge pools (Qsc_pool) and their adjacent river reaches (Qsc_river). Results show that pools fill with sediment at low river discharge because the waterfall jet is not strong enough to transport the supplied sediment load out of the pool. As discharge increases, the waterfall jet strengthens, allowing pools to transport sediment at greater rates than in adjacent river reaches. This causes sediment scour from pools and bar building at the downstream pool boundary. While pools may be partially emptied of sediment at modest discharge, floods with recurrence intervals &gt;10 yr are typically required for pools to scour to bedrock. These results allow new constraints on paleodischarge estimates made from sediment deposited in plunge pool bars and suggest that bedrock erosion at waterfalls with plunge pools occurs during larger floods than in river reaches lacking waterfalls.

Late Holocene canyon-carving floods in northern Iceland were smaller than previously reported

Catastrophic floods have formed deep bedrock canyons on Earth, but the relationship between peak discharge and bedrock erosion is not clearly understood. This hinders efforts to use geological evidence of these cataclysmic events to constrain their magnitude – a prerequisite for impact assessments. Here, we combine proxy evidence from slackwater sediments with topographic models and hydraulic simulations to constrain the Late Holocene flood history of the Jökulsá á Fjöllum river in northern Iceland. We date floods to 3.5, 1.5 and 1.35 thousand years ago and confirm that flow peaks during these events were at most a third of previous estimates. Nevertheless, exposure ages suggests that nearby knickpoints retreated by more than 2 km during these floods. These findings support a growing consensus that the extent of bedrock erosion is not necessarily controlled by discharge and that canyon-carving floods may be smaller than typically assumed.

Supplemental Material: Mass balance controls on sediment scour and bedrock erosion in waterfall plunge pools

Methods, Tables S1 and S2 (data), and Figures S1–S4.<br>