Developing an emulator to calculate present temperature field in the Antarctic Ice Sheet

<p><span>Ice temperature within the ice is a crucial characteristic to understand the Antarctic ice sheet evolution because temperature is coupled to ice flow. Since temperature is only measured at few locations in deep boreholes, we only rely on numerical modelling to assess ice sheet-wide temperature. However, the design of such models leads to a number of challenges. One important difficulty is that the temperature field strongly depends on the geothermal flux which is still poorly known (see White paper by Burton-Johnson and others,2020 </span><span></span><span>). Another point is that up to now there is no fully suitable model, especially for inverse approaches: i</span><span>)</span><span> analytical solutions are only valid in slowly flowing regions; ii</span><span>)</span><span> models solving only the heat equation by prescribing geometry and ice flow do not take into account the past changes in ice thickness and ice flow and </span><span>do not couple </span><span>ice flow and temperature. Conversely, 3D thermomechanical models that simulate the evolution of the ice sheet take into account all the relevant processes but they are too computationally expensive to be used in inverse approaches. Moreover, they do not provide a perfect fit between observed and simulated geometry </span><span>(ice thickness, surface elevation) </span><span>for the present-day ice sheets </span><span>and this affects the simulated temperature field</span><span>.</span></p><p><span>GRISLI (Quiquet et al. 2018), belongs to this family of thermomechanically coupled ice sheet models An emulator, based on deep neural network (DNN), has been developed in order to speed-up the simulation of present-day ice temperature. We use GRISLI outputs that come from 4 simulations, each covers 900000 years (8 glacial-interglacial cycles) to get rid of the initial configuration influence. The simulations differ by the geothermal flux map used as boundary condition. Finally a database is built where each ice column for each simulation is a sample used to train the DNN. For each sample, the input layer (precursor) is a vector of the present-day characteristics: ice thickness, surface temperature, geothermal flux, accumulation rate, surface velocity and surface slope. The predicted output (output layer) is the vertical profile of temperature. In the training, the weights of the network are optimized by comparison with the GRISLI temperature. </span></p><p><span>The first results are very encouraging with a RMSE of ~ 0.6 °C (calculated from the difference between the emulated temperatures and GRISLI temperatures over all the samples and all the depths). Once trained, the computational time of GRISLI-DNN for generating temperature field of whole Antarctica (16000 columns) is about 20 s.</span></p><p><span>The first application (in the framework of the ESA project 4D-Antarctica, see Leduc-Leballeur<span> presentation in this session</span>) will be to use this emulator associated with SMOS satellite observations to infer the 3D temperature field and improve our knowledge of geothermal flux. Indeed, it has been shown that SMOS data, coupled with glaciological and electromagnetic models, give an indication of temperature in the upper 1000 m of the ice sheet. Our emulator could also be used for initialization of computationally expensive ice sheet models.</span></p>

Permafrost and Gas Hydrate Stability Zone of the Glacial Part of the East-Siberian Shelf

By using thermal mathematical modeling for the time range of 200,000 years ago, the authors have been studying the role the glaciation, covered the De Long Islands and partly the Anjou Islands at the end of Middle Neopleistocene, played in the formation of permafrost and gas hydrates stability zone. For the modeling purpose, we used actual geological borehole cross-sections from the New Siberia Island. The modeling was conducted at geothermal flux densities of 50, 60, and 75 mW/m2 for glacial and extraglacial conditions. Based on the modeling results, the glaciated area is characterized by permafrost thickness of 150–200 m lower than under extraglacial conditions. The lower boundary of the gas hydrate stability zone in the glacial area at 50–60 mW/m2 is located 300 m higher than the same under extraglacial conditions. At 75 mW/m2 in the area of 20–40 m isobaths, open taliks are formed, and the gas hydrate stability zone was destroyed in the middle of the Holocene. The specified conditions and events were being formed in the course of the historical development of the glacial area with a predominance of the marine conditions peculiar to it from the middle of the Middle Neopleistocene.

Geophysical constraints on the properties of a subglacial lake in northwest Greenland

We report the first ground-based observations of a subglacial lake in Greenland, confirming previous work base on airborne radar data. Here, we perform an active source seismology and ground penetrating radar survey in northwest Greenland where Palmer et al. (2013) first proposed the presence of a subglacial lake. From reflections of both the lake top and lake bottom, we observe a subglacial lake underlying approximately 845 m of ice, and constrain its depth to be between 10–15 m. Additionally, using previously reported estimates of the lake's lateral extent, we estimate the total volume of liquid water to be 0.15 km3 (0.15 Gt of water). Thermal and hydropotential modeling both suggest that the lake should not exist unless it either sits over a localized geothermal flux high or has high salinity due to significant evaporite source in the bedrock. Our study indicates that this field site in northwestern Greenland is a good candidate for future investigations aimed at understanding lake properties and origins or for direct lake sampling via drilling.

Hard thermal turbulence in Antarctic Subglacial Lakes

<p>Over 250 stable and isolated subglacial lakes exist at and close to the ice-sheet center in Antarctica. The physical conditions within subglacial lakes, and the differences between distinct lake settings, are critical to evaluating how and where life may best exist. Here, we demonstrate that upward heating by Earth’s geothermal flux provides efficient stirring of Antarctic subglacial lakes’ water, in a variety of ways related to their water depth, ice overburden and ceiling slope. We show that most lakes are in a regime of hard convective turbulence, enabling efficient mixing of nutrient- and oxygen-enriched top melt-water, which is essential for biome formation. Lakes beneath a thin (about less than 3 km) ice cover and lakes with a thick (more than 3 km) ice cover experience similarly-large velocities, but the latter have significantly larger temperature fluctuations and have a stable layer up to several tens of meters thick adjacent to the ice. We discuss the implications of hydrological conditions on the concentration of particulates in the water column.</p>

Is there 1.5-million-year-old ice near Dome C, Antarctica?

Ice sheets provide exceptional archives of past changes in polar climate, regional environment and global atmospheric composition. The oldest dated deep ice core drilled in Antarctica has been retrieved at EPICA Dome C (EDC), reaching ∼ 800 000 years. Obtaining an older paleoclimatic record from Antarctica is one of the greatest challenges of the ice core community. Here, we use internal isochrones, identified from airborne radar coupled to ice-flow modelling to estimate the age of basal ice along transects in the Dome C area. Three glaciological properties are inferred from isochrones: surface accumulation rate, geothermal flux and the exponent of the Lliboutry velocity profile. We find that old ice (> 1.5 Myr, 1.5 million years) likely exists in two regions: one ∼ 40 km south-west of Dome C along the ice divide to Vostok, close to a secondary dome that we name Little Dome C (LDC), and a second region named North Patch (NP) located 10–30 km north-east of Dome C, in a region where the geothermal flux is apparently relatively low. Our work demonstrates the value of combining radar observations with ice flow modelling to accurately represent the true nature of ice flow, and understand the formation of ice-sheet architecture, in the centre of large ice sheets.