Australia: The Land Where Time Began
Widespread increase in dynamic imbalance in the Getz region of
Antarctica from 1994 to 2018
Ice is being lost from the Getz region of West Antarctica at an increasingly rapid rate, however, the mechanism of forcing has remained unclear. In this paper satellite observations as well as an ice sheet mode have been used to measure the change in ice speed and mass balance of the drainage basin over the last 25 years. It is shown by their results that between 1994 and 2018 there was a mean increase in speed of 23.8%, with 3 glaciers accelerating by more than 44%. Across the Getz Basin the increase in speed has been linear, with increase in speed and thinning being correlated directly, which confirms the presence of dynamic imbalance. 315 Gt of ice has been lost since 1994 which contributed 0.9 ± 0.6 mm global mean sea level, with increased loss since 2010 which was caused by a reduction in snowfall. Overall, the dynamic imbalance accounts for ⅔ of the mass loss from this region of West Antarctica over the past 25 years, with a longer-term response to ocean forcing the likely driving mechanism.
The Antarctic Ice Sheet has contributed 7.6 ± 3.9 mm to global sea level over the last 25 years (Shepard et al., 2018). It has been shown by observations that ice loss from Antarctica is dominated by the low-lying, marine-based sectors of West Antarctica (Shepherd et al., 2018; Rignot et al., 2008; Joughin Smith & Holland, 2010; Shepherd, 2012), where the Amundsen Sea Sector glaciers have thinned, accelerated, and retreat of grounding lines since the 1940s (Mouginot, Rignot & 2014; Rignot et al., 2014; Smith et al., 2017; Shepherd et al., 2019). Elsewhere, on the Antarctic Peninsula ice shelves have retreated (Cook & Vaughn, 2010) and collapsed (Rott, Skvarca & Nagler, 1996; Rack & Rott, 2004). In West Antarctica dynamic balance is driven by incursions of warm, modified Circumpolar Deep Water (mCDW) melting the ice that is floating (Joughin, Smith & Holland, 2010; Jenkins et al., 2010), with the interannual and long-term variability of ocean atmospheres linked to atmospheric forcing that is associated with the El Nino-Southern Oscillation (ENSO) (Dutrieux et al., 2014; Jenkins et al., 2018) and anthropogenic forcing (Holland et al., 2019), respectively. The contribution of the ice sheet to the global sea level budget remains the greatest uncertainty in projections of future sea level rise (Church et al., 2013), which is partially driven by positive feedbacks such as the Marine Ice Sheet Instability (Favier et al., 2014; Joughin, Smith & Medley, 2014), and with scenarios that are the most extreme, more than 1 m by 2100 being possibly only through the onset of Marine Ice Cliff Instability (MICI) (DeConto & Pollard, 2016; Edwards et al., 2019). Antarctica has been shown by satellite data that the dynamic ice loss (6.3 ± 1.9 mm sea level equivalent (sle) is 86% greater than the modest reduction in surface mass (0.9 ± 1.1 mm sle) since the 1990s (Slater, Hogg & Mottram, 2020). Long-term and emerging new dynamic signals must, however, both be measured accurately in order to understand better how ice sheets will behave in the future. 88% of the speedup of the ice has been observed occurring on glaciers located in the Amundsen Sea, Getz and Marguerite Bay sectors (Gardner et al., 2018). The timing and pace of dynamic imbalance is characterised poorly in regions that have been observed less frequently, and uncertainty has remained about the physical mechanisms that are driving this change.
The Getz drainage basin, which covers 10.2% (177,625 km2) of the West Antarctic Ice Sheet, lies in the coastal margin of Marie Byrd Land (Shepherd et al., 2019). Ice flows from the ice sheet into the Getz Ice Shelf through 14 distinct glaciers that extend inland about 145 km, and flow at average speeds of more than 500 m per year at the grounding line. The ice shelf, which is 650 km long by 110 km wide, which characterises the region, is the largest in Antarctica providing buttressing support to the grounded ice. There are 8 large islands and 23 pinning points which stabilise the ice shelf calving front, which has resulted in relatively small area change over the last 3 decades (Jacobs et al., 2013; Swithinbank et al., 2003), though retreat has been observed since 2003 (Christie et al., 2018). Strong thinning of the ice shelf (-16.1 m/decade), in spite of the absence of significant area change, has been observed since the 1990s (Paolo, Fricker & Padman, 2015), producing one of the largest sources of freshwater input to the Southern Ocean (Assmann et al., 2019), more than double the input of the ice shelves of the neighbouring Amundsen Sea (Nakayama et al., 2014). Ocean currents are channelled beneath the sub-ice shelf cavity by the complex network of topographic rises at the ice front, which drives a highly localised spatial pattern of ice thinning, which has the strongest rates that are observed at the grounding line (Hogg et al., 2020). It is expected that there will be spatially variable ocean forcing along the Getz coastline due to its zonal extent, about half the margin of the West Antarctic Ice Sheet, and its location between the colder Ross and warmer Amundsen Seas (Jacobs et al., 2013).
The Getz drainage basin has lost ~410 Gt of ice mass over the last 30 years (Shepherd et al., 2019), and the rate has increased by more than 40% since 2010 (McMillan et al., 2014). Regions of high elevation bed topography at the ice divide has been provided by the Flood, McCuddin & Executive Committee mountain ranges, which produces a broadly prograde bed slope across the basin (Fretwell et al., 2013). Inland propagation of strong ice sheet thinning is prevented by this geometry and makes the region less susceptible to onset of MISI, compared with the retrograde sloping of the marine-based glaciers in the marine-based glaciers in the neighbouring Amundsen Sea Embayment (Favier et al., 2014; Joughin, Smith & Medley, 2014). The snowfall pattern is heterogeneous across the basin, with the highest rates being deposited on topography that is steeply sloping which is aligned orthogonal (i.e. West to East) to the prevalent direction of atmospheric moisture flux (Lenaerts et al., 2018). Interannual variability in Surface Mass Balance (SMB) is driven by the Amundsen Sea Low, which accounts for 40% of the surface mass and 21% of the surface melt variability, while summer melt is limited to the lower elevation ice shelf (Lenaerts et al., 2018). In Antarctica, the key to understanding the atmospheric and oceanic forcing mechanisms that have been driving recent change is partitioning of the influence of surface mass and ice dynamic signals (Hogg et al., 2017). It has been suggested by studies that thinning of glaciers that flow into the Getz Ice Shelf are greater than the difference in ice discharge alone that has been estimated, which indicates that surface processes are responsible, at least in part, for the thinning that has been observed (Shepherd et al., 2019; Chuter et al., 2017). A multi-decadal, continuous record of ice velocity is, however, required to perform a detailed assessment of the change in ice flux. Satellite data were used in this study to measure the change in the speed of the ice of glaciers in the Getz drainage basin from 1994-2028, in order to assess the localised pattern of dynamic imbalance in this large and complex sector of West Antarctica. It was shown by the results of this study that between 1994 and 2018, there was widespread speedup of the majority of glaciers in the Getz drainage basin of West Antarctica. It was observed that there was a mean increase of speed of 23.8%, with 3 of the glaciers accelerating by more than 44%. 315 Gt of ice has been lost since 1994, which contributed to 0.9 ± 0.6 mm to global sea levels. Since 2010, increased loss was driven by a reduction in snowfall, with dynamic imbalance likely to have been driven by a longer-term response to ocean forcing.
The change in speed of glaciers in the Getz drainage basin was measured in 2 previous studies, in which it was found that the largest ice flow increase was observed behind Simple Island and at the far West of the sector between 2007/2008 and 2014/2015 (Gardner et al., 2018; Chuter et al., 2017). According to Selley et al. extend and fill gaps in the spatial coverage, and show that the speedup of the ice flow that is the largest and most extensive is concentrated on flow units 5 and 6 in the centre of the Getz drainage basin. These 2 glaciers combined account for 41.1% of the total increase in the speed of the ice across the Getz sector since 1994. The zone in which the highest speed increase occurs coincides with the thickest and most rapidly thinning part of the Getz Ice Shelf behind Siple and Carney Islands (Paolo, Fricker & Padman; Gardner et al., 2018). It was confirmed by this study that the 5 fastest flowing glaciers in the in the far West of the Getz drainage basin (flow units 10 to 14) have also undergone a significant increase in the speed of ice flow since 1994, ~25% on average, which accounts for 39.7% of the total speedup in the region. It has been proposed that DeVicq Glacier (flow unit 10) is a possible route through which future instability may propagate in Marie Byrd Land, due to a deep bedrock trough that lies more than 300 m below sea level at the grounding line and extends inland for more than 200 km. It is indicated by model studies that in spite of the geometry of the glacier it is not susceptible to imbalance even with the presence of warmer ocean water in the Amundsen Sea (Holschuh et al., 2014). It is shown, in contrast, by the results of this study that this glacier has undergone the 4th largest speedup in the Getz region over the last 25 years with ice speed increasing at a rate of 9.1 m/year2, though it is limited to ~40 km inland of the grounding line.
A region of orographically driven snowfall of more than 3,000 mm w.e./year, is resolved by a high spatial resolution (5.5 km) regional climate models, corresponds to the Berry, Venzke and Land Glaciers (Lenaerts et al., 2016). The highest surface mass variability in Antarctica outside of the peninsula in located in the Getz drainage basin (Shepherd et al., 2019), and it is shown by the cumulative anomaly that snowfall into the basin since ~2008 has been significantly lower than the long term mean. It is shown by the multidecadal time-series of speed change that the increase in ice flow that has been observed on all glaciers in the Getz study region, has been relatively linear over the past 25 years (Supplementary Fig. 1d). The accelerated mass loss that has been observed by this study and others (Shepherd et al., 2019; Rignot et al., 2019) from the Getz sector since ~2010, has therefore been driven by the long-term gradual increase in dynamic imbalance, as well as combined with the effect of a short-term surface-mass deficit. It is shown by this that extreme snowfall years have a significant influence on mass balance of the entire drainage basin, which suggests that localised regions of high snowfall that are resolved by high-resolution models may impact the balance of individual glaciers.
Located in the transition zone between the cooler Ross Sea and the warmer Amundsen Sea, with potential warmer temperatures that are present in the West of the Getz study region, is the Getz drainage basin. It is suggested by observations over the last 3 decades in West Antarctica that temperatures have warmed offshore (Rignot et al., 2004), and there have been periodic incursions of warm mCDW onto the continental shelf which drives shorter-term sub-decadal to decadal variability (Dutrieux, 2014; Jenkins et al., 2018; Jacobs et al., 2013). Annual mean temperature depth profiles in the Getz region, collected from ship-based CTD sampling between 1994 and 2018 have shown annual variation in the depth of the thermocline which divides a cold and fresh upper layer from the mCDW. The thermocline was ~2,000 m deeper in 1994, 2000, 2012, and 2014, which led to cooler integrated heat content of the ocean, which is in contrast to the warmer years in 2007, 2009, 2016 and 2018 when the thermocline was shallower. Selley et al. suggest these observations are in line with previous ocean temperature studies, and have been linked to the ocean response to interannual variability of the atmosphere (Jacobs et al., 2013; Assmann et al., 2019). Changes in the depth of the thermocline are observed as a band of variability of temperature between ~500 and 800 m below sea level, and in the west of the Getz sector it is particularly strong. The ocean heat content is made particularly sensitive to atmospheric forcing at the continental break, even more than in the Eastern Amundsen Sea (Orsi, Whitworth & Nowlin, 1995; Dotto et al., 2019), by the relatively short distance between the continental shelf break and the calving front of the Getz Glacier, which is consistent with the large ocean heat content viability.
Warmer ocean water must be transported from the open ocean under the ice shelf at depth (below ~400 m), where it can come into contact with and melt deeply grounded ice, in order for the spatial and temporal variability to affect the rate of melting of the ice and the dynamic imbalance of the Getz drainage basin. The Coriolis force, which causes warm ocean water to enter ice cavities on the Eastern side, before being guided by the sea bed and the geometry of the base of the ice and exiting on the Western side, directs ocean circulation beneath the ice shelves of Antarctica. Coriolis driven circulation of the Getz Ice Shelf, brings warm ocean water from the Amundsen Sea between Wright and Duncan Islands and to a lesser extent between Dean and Siple Islands. Warmer water moves along the grounding line towards the West, in the process melting the ice and gaining buoyancy and upwelling with potential freshwater, as well as buoyancy and circulation input from subglacial meltwater channels. Under the Getz Ice Shelf the pattern of ocean circulation is visible through high meltwater fraction that exits the cavity at ~150-200 m ice draft on the eastern side of islands, most clearly Wright, Duncan and Grant. Selley et al. attributed the ice speedup that was observed on flow units 4-6 partially to the impact of warm ocean water reaching the grounding line in these locations, which drives high rates of ice speedup and melting of the ice shelf (Paolo, Fricker & Padman, 2015; Hogg et al., 2020). It appears that relatively warmer water reaches the Western Getz (Nakayama et al., 2014). Within the Getz cavities where tides seem to be relatively weak and melting seems to be relatively large, however, most of the ocean circulation is thought to be driven by the melt-induced upwelling and conducted by the evolving geometry, much as is the case in Pine Island and Thwaites cavities (Nakayama, 2019). In the case of the Getz, its many ice stream tributaries, and islands which act as pinning points for the ice flow make for a circulation pattern that is complex and expectedly sinuous.
Buoyancy gains at the grounding zone provided by subglacial drainage outflows can also enhance ocean melt. It is indicated by the results of this study that a number of glaciers in the Gets study region that have increased in speed coincide with regions of high subglacial water flux beneath the ice sheet, including flow units 4-6 and De Vicq Glacier (flow unit 10). It is known that ice melt that is driven by subglacial runoff or, in the case of the Amundsen sector, ocean heat, decreases with distance from the grounding line (Dutrieux et al., 2014; Le Brocq et al., 2013; Wei et al., 2020), as the entrainment of warm water at the ice-ocean interface increases with buoyancy and velocity, though it diminishes downstream. Contrasting with this, beneath the ice shelf meltwater channels can extend to the ice front, depending on whether the supply of warm ocean water is sustained (Gourmelen et al., 2017), or if the features are simply advected downstream. It is known that plumes of buoyant freshwater drive high melt rates on Getz Ice Shelf hydrological pathways that route subglacial freshwater have a significant role in determining the location of and rate of basal melt beneath the ice shelf (Wei et al., 2020). It is shown by the change in the speed of the ice that increases in the Getz study region are strongest locally near the grounding zone and do not extend to the ice front, which indicates that subglacial hydrology could also be responsible for some of the speedup and thinning that has been observed. If the volume of subglacial water flux changes over time, as is the case of glaciers in Greenland that are due to seasonal surface melt (Sundal et al., 2011; Nienow, Sole, Slater & Cowton, 2017), this may be a factor that has not been accounted for that drives changes in ice flow and thinning in the regions of Antarctica with high subglacial water flux. The presence of subglacial water has not been accounted for in most ice flow models and to the knowledge of Selley et al. about any change in subglacial flux over the last 25 years is limited regions that are not characterised by active subglacial lakes. According to Selley et al. the Getz drainage basin may therefore be a valuable test region of any studies that investigate the coupling between subglacial hydrology and the ocean, and the impact this may have on the localised ice sheet dynamics pastern in Antarctica.
It is shown by the results of Selley et al. that ice speeds have increased on the majority of glaciers in the Getz drainage basin, ice speeds have increased at a rate that is broadly linear. Given that the Getz basin, more so than its unstable neighbours in the Eastern Amundsen Sea, is thought to be relatively immune to positive feedback processes like MISI and MICI due to the prograde topography its bed, it is difficult to explain this trend as a runaway response to a step or oscillatory ocean forcing, at least during the past 25 years (Jenkins et al., 2018). It is suggested by this that the dynamic imbalance that was observed in the Getz may be primarily a response to ocean forcing on the longer-term which is suggested by Selley et al. to possibly be of anthropogenic origin (Holland et al., 2019). Research programs in the future that deliver continuous annual monitoring of the velocity of the ice and ocean temperatures across the study region at present locations will be critically important, preventing gaps in the record and enabling an assessment of the link between the localised pattern and short-term variability of the dynamics of the ice and the complex transport of ocean temperature variability that is more direct.
The record of the ice speed in this study over 25 years shows for the first time that widespread linear speedup has occurred on the majority of glaciers in the Getz drainage basin of west Antarctica since 1994. It is shown by the changes in ice flow that are concentrated zones of very high speed up (>44%) at the grounding line of 3 glaciers (flow units 6, 5 and Venzke Glacier (flow unit 12), and high speedup of more than >20% on an additional glaciers (flow units 3, 4 DeVicq Glacier (flow unit 10)) since 1994. 46.8% of this acceleration is accounted for by central regions of the Getz drainage basin, and contains the most spatially extensive areas of change. A localised response on individual glaciers is indicated by this pattern of ice speedup, which demonstrates the value of high resolution observations that resolve the detailed pattern of dynamic imbalance across the Getz drainage basin. The speed increase coincides with regions of high surface lowering on all glaciers, with a approximately 50% of speed up which corresponds to a reduction of ~5% of the thickness of the ice. It is shown that 315 Gt of ice has been lost from the Getz drainage basin since 1994, by the results of the optimised model of Selley et al., which contributed 0.9 to 0.6 mm to global se levels, and increased the rate of ice loss by 4 times in the 2010s compared to the 1990s. The topography of the prograde bed makes the Getz region inherently less susceptible to unstable geometry that drives feedbacks such as MISI and MICI compared with its neighbours in the Amundsen Sea, and indicates that in the long-term warming of the ocean may be driven by the dynamic imbalance in this region of Antarctica. Understanding of the dynamic imbalance of remote areas of Antarctica in the future will be helped by consistent and temporally extensive sampling of both ocean temperatures and ice speed.
Selley, H. L., et al. (2021). "Widespread increase in dynamic imbalance in the Getz region of Antarctica from 1994 to 2018." Nature Communications 12(1): 1133.
|Author: M.H.Monroe Email: email@example.com Sources & Further reading|