Impact of Past Ice Streams on Antartic Sea Floor

How have past ice streams shaped the sea floor around Antarctica?

Introduction

Ice sheet reconstructions are of significant importance when researching interactions of ice sheets with the climate and ocean. By examining how ice sheets have responded to changes in the past, we are able to further understand how these ice sheets will react and change over future long-term timescales (Davies, 2018). An ice sheet can be defined as ‘a mass of glacial land extending more than 20,000 square miles’ (NSDIC, 2017), however the Antarctic ice sheet that we are studying extends over 5.4 million square miles (NSDIC, 2017). The Antarctic ice sheet is comprised of ice streams that are “a part of an inland ice sheet that flows rapidly through the surrounding ice” (Bentley, 1987). These ice streams are important to research as their behaviour and stability is essential to the overall dynamics of the Antarctica ice sheet and mass balance (Bennett, 2003). Once identified, ice streams can be used as an insight into past glacial behaviour and allow future prediction to be made of the response of contemporary ice sheets to future climate perturbations. (Stokes, 2001). Research has been undertaken to show that during the Last Glacial Maximum (LGM) the extent of the Antarctic ice sheet contributed 14m to the lowering of eustatic sea level (Denton, 2002). However, since the LGM, the Antarctica ice sheet has retreated, leaving a large range of geomorphological features uncovered. The aim of this report is to examine the submerged landscape surrounding the Antarctica and identify and examine the key geomorphological features formed by these ice streams and the dynamic processes that formed them. More specifically, this report will look at the Pine Island Trough (PIT) in the Amundsen Sea and the Ground-Zone Wedges (GZW) that were formed through the retreat of the Antarctica ice sheet.

Methods

Analysis in this report was mainly carried out through the use of ArcGIS. The multibeam echo-sounder dataset of the sea floor collected from the Antarctic continental shelf by the Swedish Icebreaker, RV Oden, in 2010 was used to construct topographic maps of the PIT floor. A hillshade raster for the bathymetric model was then created, made transparent and then placed over the PIT dataset. This allowed features on the sea floor terrain to be identified and examined such as the GZW. Through use of the 3D Analyst tool on ArcGIS, quantitative and qualitative data was produced examining the PIT area and the GZW within. This then allowed analysis to be undertaken on the subsequent results to help look at how ice streams formed the key features that were identified.

Results

Discussion:

Through the use of ArcGIS, subglacial landforms on the sea bed in the Pine Island Trough (PIT) have manged to be identified and analysed. By looking at these landforms we are able to obtain a great deal of information about past behaviour of ice streams in the Antarctic and how these have formed the landforms identified.

The most prominent feature identified and analysed through ArcGIS, was the Grounding Zone Wedges (GZW) which are widely assumed to have been formed as the ice sheet retreated after the Last Glacial Maximum 21,000 years ago. When the ice sheet retreated, sub glacial till was allowed to accumulate into 2 distinctive GZW located in Area A and Area B as shown in Figure 4.1. The two GZW grade upwards as you go away from the Antarctic coast, inferring that these features were formed through ice flow as the ice retreated. This is proven in Fig 4.2 where the GZW 5 is seen as deeper (has a greater amplitude) than GZW4. We can infer that GZW5 is older than GZW4 due to the ice retreat accepted as retreating towards the Antarctic coastline, and therefore as the GZW5 has a greater amplitude, it has had more time for sediment to accumulate. These GZW have formed due to several subglacial mechanisms that allowed sediment to build up when the ice became stationary creating these distinctive features. Jakobsson et al. (2012) used sediment fluxes to estimate that it takes over 1000 years for a wedge to form, indicating that the ice must have been stationary for a significant period of time. Subglacial till underlying the ice that assists ice flow becomes pushed out when the ice is stationary for a while (also known as ice stagnation). Ice stagnation can occur when certain subglacial processes taking place. The first of these is where changes in topographical gradient can cause the ice to stop retreating (Graham et al., 2010). This could be an increase in the gradient of the sea bed, resulting in the ice needing more energy to go up the gradient, causing the ice flow to stop until enough energy is provided to overcome this.  Another of these factors that causes a significant decrease in ice velocity is a decrease in water pressure causing the ice stream and sediment coincide causing stagnation of the ice stream (Bennet, 2003). The wedge geometry that is formed is due to the lack of accommodation space at the ground line for sediment to move out as the ice retreats. The retreat of the glacier also leads to erosion through mechanisms such as plucking and abrasion at the bed leading to an increase in sediment being available to be deposited in a GZW.

The effect these wedges have on the overlying ice is also significant. The deposition of the GZW elevates the grounding line off the sea floor and thickens the ice above it (Alley et al. ,2007). This then buffers the entire ice sheet against small rises in sea level which is significant in more recent times as sea levels begin to rise.

In both Area A and Area C, Mega Scale Glacial Lineation’s (MSGLs) were identified as the main subglacial landform. As shown in Fig 2.2 and Fig 2.4, they can be seen clearly on the high resolution hillshade raster as ‘ridge-groove features formed in parallel patterns’ (Jakobson et al., 2012). MSGLs can be defined as “linear forms 10,000 to 100,000 m in length, which are most easily observed from aerial photographs or satellite images” (Clark, 1993). These MSGLs are said to have been formed from drumlins, as they are much larger and more elongated, and they provide evidence for areas in which there was a fast ice stream flow. Several theories as to how MSGLs form have been put forward but not one solution has been widely accepted due to the complexity of ice dynamics. Shaw et al. present the idea that MSGLs have been formed through erosion by turbulent meltwater flow however modern observations show that ice-sheet bedforms were created without meltwater floods. Another theory is the “bed deformation model” by that shows that MSGLs were formed through weak sediments being deformed by shear stress imparted on the sediment by the overlying ice. But this theory cannot explain the pattern phenomena through which drumlins and MSGLs form. The instability theory which is the most widely accepted, even though it cannot explain all formation aspects, was presented by Hindmarsh in 1998. It states that the coupled flow of ice and subglacial sediment may be inherently unstable, such that the ice-till surface could become spontaneously wavy with growth at a preferred wavelength. It implies that layers of materials with varying physical properties are prone to the development of instabilities (instabilities being said to occur in a system when small irregularities spontaneously grow to produce regular patterns, in this case MSGLs). This theory has not been researched enough to conclude whether this is how MSGLs are formed. These theories, plus others, are heavily contested and so MSGL formation is still very much unknown. Due to the importance of the presence of MSGLs linking to fast flowing ice streams, improved knowledge of how they are formed would result in a major advancement of the basal processes that impact on ice velocity and flow (Stokes et al., 2013). This entails further research being carried out on MSGLs identified around the Antarctic, not just in the PIT, to facilitate this advancement in knowledge.

In Area B iceberg ploughmarks could be seen and identified across the entire Area. The form of these ploughmarks can vary significantly and depend on the type of sediment, the shape of the iceberg keel (usually V-shaped) and the motion of the iceberg (Shakesby, 1986). The presence of these ploughmarks in the PIT indicates a period of slower ice flow, as sediment has been allowed to accumulate, and then the icebergs have formed depressions as the ice retreated. These depressions can be characterised as more V-shaped, providing evidence they were formed from icebergs (not other large forms of materials), as seen in the profile graph of Fig. 3.6. Graham et al (2009) states that iceberg ploughmarks record the latest phase of shelf evolution and were subsequently formed when the ice retreated. These ploughmarks are evidence that after a period of fast flow (Area C) the ice retreat then slowed, allowing these subglacial landforms to be formed. They highlight the significance of ice streams and in particular the speed in which they flow, in constructing the geomorphology of the sea floor surrounding the Antarctica.

Conclusion

In conclusion, past ice streams have had a huge impact on the features formed on the sea floor in the Pine Island Trough around the Antarctica. The range of features produced demonstrates the different states of flow the ice streams went through, with the GMZ and iceberg ploughmarks showing periods of stagnation/slow flow and the MSGLs showing periods of fast flow. Looking at Dowdeswell et al. (2008) model of Antarctic ice-stream retreat we can conclude that the PIT area conforms most to the episodic retreat with periods of fast and slow flow causing superimposed grounding zone wedges to be formed. These findings can now be used to help predict future flow rate and subsequent landforms as the ice continues to retreat, and at a faster pace, as a result of increased anthropogenic activity.

References:

• Alley et al., (2007), Effect of sedimentation on Ice-SheetGrounding-Line stability, Science, 315(5820), 1838-1841
• Bennett, M.R. (2003), Ice streams as the arteries of an ice sheet: their mechanics, stability and significance, Earth-Science Reviews, 61(3-4): 309-339
• Bentley, C. R. (1987), Antarctic ice streams: A review, J. Geophys. Res., 92(B9): 8843–8858
• Davies. B (2018) Palaeo ice sheet reconstruction [online] (5^th November 2018) Available at: http://www.antarcticglaciers.org/glacial-geology/palaeo-ice-sheet-reconstruction/ [Accessed 10^th December 2018]
• Clark, C.D (1993), Mega-scale Glacial Lineations and Cross-cutting Ice-flow Landforms, Earth Surface Processes and Landforms, 18, 1-29
• Denton G.H, Hughes T.J (2002), Reconstructing the Antarctic ice sheet at the Last Glacial Maximum, Quaternary Science Reviews, 21(1-3): 193-202
• Dowdeswell, J. A., D. Ottesen, J. Evans, C. O. Cofaigh, and J. B. Anderson (2008), Submarine glacial landforms and rates of ice-stream collapse, Geology, 36(10), 819-822.
• Graham et al (2009), Bedform signature of a West Antarctica palaeo-ice stream reveals a multi-temporal record of flow and substrate control, Quaternary Science Reviews, 28 (25-26), 2774-2793
• Hindmarsh (1998), Drumlinization and drumlin forming instabilities; viscous till mechanisms, Journal of Glaciology, 44(147), 293-314
• Jakobsson, M., J. B. Anderson, F. O. Nitsche, R. Gyllencreutz, A. E. Kirshner, N. Kirchner, M. O’Regan, R. Mohammad, and B. Eriksson (2012), Ice sheet retreat dynamics inferred from glacial morphology of the central Pine Island Bay Trough, West Antarctica, Quaternary Science Reviews, 38, 1-10.
• NSIDC (2017) [online] Quick Facts on Ice Sheets (14th February 2018) Available at: https://nsidc.org/cryosphere/quickfacts/icesheets.html [Accessed 1st December 2018]
• Shakesby (1986) [online] Iceberg Plough Marks, Available at: http://sk.sagepub.com/reference/dictionaryenvirochange/n1986.i1.xml [Accessed 4th December 2018]
• Stokes C.R, Clark C.D (2001), Palaeo-ice streams, Quaternary Science Reviews, 20(13): 1437-1457
• Stokes et al (2003), Formation of mega-scale glacial lineations on the Dubawnt Lake Ice Stream bed: 1. size, shape and spacing from a large remote sensing dataset, Quaternary Science Reviews, 77: 190-209