Antarctica Has a Huge Mantle Plume Beneath it, Which Might Explain its High Degree of Instability

Australia: The Land Where Time Began

Antarctica Has a Huge Mantle Plume Beneath it, Which Might Explain its High Degree of Instability¹

The continent beneath the Antarctic ice sheet is covered by rivers and lakes, the largest lake being the size of Lake Erie. The ice sheet melts and refreezes over the course of a regular year, which causes the lakes and rivers to fill and rapidly drain periodically from the melt water. The frozen surface of Antarctica is able to slide around more easily by this process, as well as rise by as much as 6 m (20 ft) in some places.

There may be a plume beneath Marie Byrd Land, according to a new study² by researchers from NASA’s JPL. Some of the melting that occurs beneath the ice sheet may be explained by the presence of this source of geothermal heat, and why it is unstable at present. It could also help to explain how the ice sheet rapidly collapsed during previous periods of climate change in the past.

The study² was published recently in the Journal of Geophysical Research: Solid Earth. The motion over time of Antarctica’s ice sheet has always been a source of interest to Earth scientists. It is possible to measure the rate at which the ice sheet rises and falls which means it can be estimated where melting is occurring and the amount of meltwater at the base. These measurements led to the first speculations about the presence of heat sources beneath the ice sheet

Wesley E. LeMasurier from the University of Colorado Denver proposed the presence of a mantle plume beneath Marie Byrd Land more than 30 years ago. It was suggested by the research carried out by LeMasurier that a possible explanation for the volcanic activity in the region and a topographic dome feature. More recently seismic surveys provided support for the presence of a mantle plume.

It is not currently possible, however, to measure directly the region beneath Marie Byrd Land. The JPL study relied on the Ice Sheet System Model (ISSM) to confirm the existence of the plume. This model is essentially a numerical depiction of the physics of the ice sheet that was developed at JPL and the University of California, Irvine.

Seroussi et al. drew on observations of altitude changes of the ice that had been made over the course of many years to ensure the model was realistic. These had been made by NASAs Ice, Clouds and Land Elevation satellite (ICESat) and their IceBridge campaign that was airborne. Very accurate 3-D elevation maps have resulted from these missions measuring the Antarctic ice sheet for years.

Seroussi et al. also enhanced the ISSM by including natural heating sources and transport of heat that result in the freezing, melting, liquid water, friction, as well as other processes. Powerful constraints are placed on allowable melt rates in Antarctica by this combined data, allowing the team to run dozens of simulations and test a wide range of locations for the mantle plumes.

They found that the heat flux from the mantle plume would be no more than 150 milliwatts per m². Regions with no volcanic activity, in comparison, experience a heat flux between 40 and 60 mW/m², whereas the geothermal hotspots, such as the one beneath Yellowstone National Park, experience an average of about 200 mW/m².

The melt rate was too high where they conducted simulations exceeding 150 mW per m². An area inland from the Ross Sea, which is known for intense water flows, is one exception. A heat flow of at least 150-180 mW/m² is required by this region to align with its observed melt rates.

Seismic imaging in this region has also shown that heating might reach the ice sheet through a rift in the mantle. This is also consistent with a mantle plume, which is believed to be in the form of narrow hot magma streams rising through the mantle and spreading out beneath the crust. The crust is then caused to bulge by ballooning of the viscous magma beneath it.

Where there is ice on top of the crust above the plumes heat is transferred by this process into the ice sheet which then triggers significant melting and runoff. Finally, Seroussi et al. provide compelling evidence, which was based on a combination of surface and seismic data, that there is a surface plume beneath the West Antarctica ice sheet. It was also estimated by Seroussi et al. that this mantle plume formed about 50-100 Ma, long before the initiation of the West Antarctic ice sheet.

When the last ice age ended about 11,000 BP, there followed a period of rapid, sustained loss of ice. Warm water was pushed closer to the ice sheet as weather patterns and rising sea levels began to change. The study by Seroussi et al. suggests that this kind of rapid loss at the present could be facilitated by the mantle plume, much as it did at the time of the onset of the last onset of an interglacial period.

It is important to understand the sources of ice sheet loss beneath West Antarctica, as far as estimating the rate of ice loss may be occurring there, which is essentially predicting the effects of climate change. It is essential in the development of accurate models that will predict how rapidly the polar ice will melt and sea levels will rise, given that the Earth is again going through temperature changes, though this time due to human activity.

Seroussi et al. also suggest it also informs understanding of how the history of the Earth and shifts in climate are linked, and the way in which these influenced geological evolution.

West Antarctica – Influence of a Mantle Plume on the Basal Conditions of the Ice Sheet²

It is possible a deep mantle plume manifested volcanism in the Pliocene and Quaternary and an elevated heat flux in Went Antarctica has been studied for more than 30 years. The plume hypothesis that a plume was the cause of volcanism and structure in Marie Byrd Land (MBL) was supported by recent seismic images. The geothermal heat flux above mantle plumes may increase by more than double that of nominal continental values. Consequently, Seroussi et al. examined a realistic distribution of heat flux associated with what is possibly a mantle plume from the Late Cainozoic in West Antarctica and explore the impact it has on thermal and melt conditions at the base of the ice sheet. In order to produce geothermal heat flux at the base of the ice sheet they used a simple mantle plume parametrisation. An enthalpy framework and full-Stokes stress balance is included in the 3-D ice flow model. As both the location and extent of the putative plume are not certain (Seroussi et al., 2017) performed experiments that were broadly scoped to characterise the impact the plume had on geothermal heat flux and basal conditions of the ice sheet. It was shown by the experiments that mantle plumes must have an important impact locally on the ice sheet, with basal melting rates of several centimetres per year directly above the hotspot. The upper bound on the geothermal heat flux that was plume-derived is 150 mW/m². Contrasting with this, in the lower part of the Whillans Ice stream the active lake system suggests an anomalous heat flux in the mantle that is linked to a rift source.

For sharpening theoretical and numerical estimates of the contribution of ice sheets to sea level rise in the future it is important to improve knowledge of basal geothermal heat flux, q_GHF, in Antarctica. However, only a few direct in situ measurements have been conducted at the bottom of deep boreholes (Engelhardt, 2004; Fisher et al., 2015) because of the thickness of the ice. Tectonic correlations to surface maps (Shapiro & Ritzwoller, 2004) satellite magnetic data (Maule et al., 2015), and interpretations of ice penetrating radar (Schroeder et al., 2014), provide other inferences of geothermal heat flux. It has long been known that the important connection between heat and water is critical to an understanding of processes beneath the Antarctic ice streams (e.g., Blankenship et al., 1986), now there is a growing interest in the basal hydrologic conditions beneath the West Antarctic Ice Sheet (WAIS) that has been motivated both by the discovery of extensive subglacial water activity (e.g. Fricker et al., 2007; B. E. Smith et al., 2009; Creyts & Schoof, 2009; Siegfried et al., 2004; Fricker et al., 2016) and by the recognition that basal conditions are very important to the proper formulation of numerical simulations of the evolution of ice sheets in a climate that is warming (e.g. Norwicki et al., 2013).

In order to properly assess the time scales and amplitudes of potential collapse it is essential to quantify the thermodynamic state of the WAIS. As an assessment of how ice flow in the future might change is related to the heat flux condition that is applied at the solid Earth-ice interface, scrutiny has now increased of the examination of the error and uncertainty that is caused by characterisation of the ice sheet thermodynamic and phase states (e.g. Rogozhina et al., 2012; Larour et al., 2012a). A zeroth-order problem is to attack the uncertainty in background heat flux at the bed.

A continuous replenishment of a thick thermally insulating layer is provided by snow falling on an ice sheet. The sensitivity to geothermal heat flux, q_GHF, to basal melt is revised by the theory of Budd et al. (1984) for 1-D flow lines. It emphasises the relation between basal temperature and geothermal flux and predicts the presence of meltwater beneath most of the Antarctic ice sheet for q_GHF ≥ 80 mW/m². It is assumed by most numerical models of the polar ice sheets that the geothermal flux is 42 ≤ q_GHF ≤ 65 mW/m² (Siegert & Dowdeswell, 1996; Llubes et al., 2006), as has been pointed out (Rogozhina et al., 2012). In West Antarctica, however, which experienced active Cainozoic volcanism and the formation of rifts, the question of higher q_GHF is especially important as geothermal heat flux is greater than 70 mW/m² in analogous regions, such as the continental US, west of the Rocky Mountain Ranges (Ramirez et al., 2016; Davies, 2013; Blackwell, 1989).

It has been shown (Pattyn, 2010) that more than half of the base of the ice sheet reaches the pressure melting point, with estimates of the total basal meltwater production rate of about 65 Gt/yr across the entire continent. Such a production rate is substantial, amounting to roughly 3% of the surface accumulation rate in Antarctica. The analysis (Pattyn, 2010), which used lakes as an indicator of the presence of basal meltwater, similar to what was done (Siegert & Dowdeswell, 1996), derived a modified geothermal heat flux map connected to the observations of basal water conditions in Antarctica. The map, however, lacks any solid Earth information that can now be derived from large scale broadband seismic stations that are currently imaging the mantle and crustal environment (e.g. Chaput et al., 2014; Emry et al., 2015; Heezel et al., 2016). New constraints of the tectonic conditions beneath the ice are provided by the 3-D seismic wave velocity structure that is derived from this new data. Any ice sheet observations confirming the presence of mantle plume conditions has profound implications for the properties of the mantle beneath the WAIS. The viscosity of the mantle is exponentially temperature dependent, and the viscous response time to loading and unloading by ice is also governed by an exponential dependency on the viscosity. Therefore a hotter mantle predicts time scales that are drastically reduced over which glacial isostatic adjustment stress relaxation occurs (e.g. Ivins et al., 2000).

The earliest hypothesis of a mantle plume with sufficient transport of heat to manifest volcanism dating to the Cretaceous and Holocene, and present day seismicity, in West Antarctica, was proposed in 1980. No seismic imaging provided support for a plume beneath Antarctica prior to surface wave mapping (Sieminski et al, 2003) and the mapping of a slow structure in the top of the lower mantle beneath the western Ross Sea (Montelli et al.. 2006).

A dome containing 18 high standing volcanoes of felsic and alkali basaltic chemistry are central to the 3 km uplift of Marie Byrd Land. Linear chains are formed by the majority of these volcanoes that age towards the centre of the province (Storey et al., 2013). Seroussi et al. suggest that these volcanoes, which date to the Late Cainozoic, 35-28 Ma, could possibly be associated with a single active plume (LeMasurier & Rex, 1989; LeMasurier & Landis, 1996). An alternative suggestion is that a mantle upwelling that is longer-lived and more broadly scaled, may have arrived in the lithospheric environment about 100 Ma, with uranium and lead isotopic ratios supporting separation from slab material stagnating at the top of the mantle beneath Gondwana ( Steinberger, 2000; Panter et al., 2006; Sutherland et al., 2010; Spasojevic et al., 2010).

The broad-scaled model that is longer-lived appears to be supported by global seismic tomography, as the method is inherently of lower resolution, though the more recent high resolution tomography resulting from the broadband seismic Antarctic Polar Earth Observing Network seems to support a single model that is younger and focused more spatially (e.g. Hansen et al., 2014; Accardo et al., 2014; An et al., 2015; Emry et al., 2015; Lloyd et al., 2015; Heezel et al., 2016). The seismic images reveal the pattern and anisotropy, as well as lateral variability in shear compressional wave velocity and, collectively, lend support for the hypothesis that the plume originated in the lower mantle, as the cause of the younger MBL volcanism and geophysical structure. Long period earthquakes have recently provided compelling evidence of ongoing movement of magma at lower depths in the crust (Lough et al., 2013), these events also indicting that the locus of activity has moved ⁓55 km southwards of the eruptions dating to the quaternary of Mt. Waesche in Executive Committee Range (WCR). When the continental plate has moved very little over the past 30 Myr, as is the case with Antarctica, the surface manifestations of continental plumes are approachable by the use of simple models (e.g. Gripp & Gordon, 2002). Consequently, it may be possible to test further the hypothesis of a single focused plume by modelling its interaction with the ice sheet base.

In this study Seroussi et al. used numerical modelling to simulate the interactions between the solid Earth mantle plume and the thermal structure of the ice sheet. It is also instructive to perform computations for an area around Subglacial Lake Williams (SLW) where there is abundant evidence of the presence of basal water, though the primary focus of this study was on the Marie Byrd Dome. In both regions observations suggest there was magmatic activity as well as high geothermal flux. The measurement of heat flux near mantle plumes suffer from their spatial sparsity, with difficulties being associated with the logistics in the deep ocean, though some measurements were obtained (e.g. Nyblade &Robinson 1994). It was shown (Harris & McNutt, 2007) that there are significant deficiencies when sampling the ocean floor heat flow in studies which have targeted hotspots. These result from the lack of enough spatial density in combination with a heterogeneous fluid convective transport through the shallow crust. Studies at the hotspots track of La Réunion (e.g. Bonneville et al., 1997) and Hawaii, where the sampling reaches its highest level, tend, however, to show that with respect to a background plate model, the anomaly is small, being about 10 mW/m². It was concluded (Jaupart et al., 2015) that the heat that is supplied by the mantle plume hotspot in the oceanic lithospheric environment is suppressed by it locking deeply to the base of the lithosphere. Where rising plumes meet the lithosphere that is rheologically weaker than beneath ocean plates (e.g. Kohlstedt et al., 1995), studies reveal anomalies that are an order of magnitude larger. The Yellowstone hotspot is the most studied and data rich example. Therefore Seroussi et al. used an analytical mantle plume model parameterisation that is able to produce spatial patterns that are realistic of geothermal heat flux where there are abundant observations (e.g. DeNosaquo et al., 2009).

The enthalpy framework, which is an energy conserving formulation that allows cold and temperate ice to be modelled, and includes thermal advection, fusion, deformational heat, and basal heat flux, is the basis for the thermal regime of the 3-D ice flow model (Aschwanden et al., 2012). The seismic data only roughly infer the location of the mantle plume beneath the WAIS.

Caveats

The thermal steady state assumption made to derive the thermal state of the ice sheet is a limitation on the simulations by Seroussi et al. There is a minor impact of the assumption on century-scale simulations of the evolution of the Greenland ice sheet (Seroussi et al., 2013). The thermal state of the ice sheet and the ice sheet basal conditions are, however, influenced by variations in surface temperature and velocities of the ice sheets. The aim of this study was to assess the range of basal conditions beneath the WAIS and to provide new bounds on the geothermal heat flux as the result of the seismic mapping of the mantle that are newly emerging. Therefore, quantitative mapping of the basal melting rates is beyond the scope of this study. There is no account taken of the water gained and lost by hydrodynamic melting and freezing. According to Seroussi et al. this water amount is likely limited to <1% of the water budget by the hydraulic gradients in Antarctica (Carter et al., 2009).

There are limitations to the parameterisations of the mantle plume, as it does not capture details of the growth of the mantle plume or head evolution. Tough it does produce realistic heat flux within the range estimated and allows sampling a variety of plume characteristics.

Conclusions

In this study Seroussi et al. have investigated the basal conditions of West Antarctica by the use of a 3-D full-stokes thermomechanical model of ice flow and geothermal heat flux that is generated by a mantle plume parameterisation. The suggestion that a mantle plume has ascended from below the 660 km seismic discontinuity beneath the crust of West Antarctica, possibly in 2 or more phases, has been supported by recent seismic imaging. Mapping from space over the past decade has identified clearly events that can only be interpreted as pulsations of water transport at the base of the WAIS (Fricker et al., 2007; Siegfried et al. 2014; Kim et al., 2016). Beneath the Ross ice streams a ubiquitous, somewhat sustained transport of water is observed (Siegfried et al., 2016). It is shown by Seroussi et al. that beneath the Willians Ice Stream and the Mercer Ice Streams intrinsic heating is quite limited and the geothermal heat flux above 150 mW/m², over a large region, plays an important role in the production of water. The heating likely originated in the mantle, and was potentially caused by a rift in the lithosphere.

It is possible that there is a plume source for the Marie Byrd Dome, though it should not raise the surface heat flux to values above 150 mW/m², as this would result in excessive basal water. The most logical solution of the apparent paradox of limited lake density in Marie Byrd Land and the combined seismic and petrological data that support the presence of a plume is therefore provided by the explanation of limited heat transport.

Sources & Further reading

1. Williams, Matt, 17/11/2017, Universe Today.

2. Seroussi, H., et al. (2017). Influence of a West Antarctic mantle plume on ice sheet basal conditions.