Ice-sheet models (ISM) were initially developed to reconstruct the extension of ice sheets during the Quaternary period. These first generation ISMs used low resolution (typically coarser than 20 km) and simplified physics (asymptotic equations, e.g. [6]). Although known to present some biases in coastal regions, these ISMs were well adapted to reproduce long and large-scale ice sheets variations. However, besides not being capable of explaining rapid changes during deglaciations [7,8], these ISMs have been further disproved by being unableto reproduce the contemporary state of Greenland [9,10] and Antarctica (e.g. [11]). The lack of representation of small-scale processes was clearly pointed out. Therefore, improving the physics included in ISMs is a priority before any reliable projection can be produced [4,5], and is one of the main objectives of EIS (Task 1).

The difficulty to represent grounding line (the boundary between the grounded and floating parts of the ice sheet) dynamics in the first generation of ice sheet models led to specific developments focusing on the dynamics of outlet glaciers [e.g., 12]. Subsequent improvements of models allowed to better capture rapid paleo events such as melt water pulses during deglaciation [13] or Heinrich Events [14]. Recently, attempts to calibrate future Antarctic discharge on periods presenting very high sea-level stand periods, lead [15] to propose hydro-fracturing and ice cliff collapse as potential mechanisms able to further enhance rapid retreat of marine based sectors of Antarctica. Although this latter result remains controversial within the ice sheet modelling community [16], all these past studies indicate that modelling paleo ice sheets is crucial in the identification and the quantification of the different processes governing their evolution and their rapid large scale retreat. Hence, paleo simulations help constraining physical parameters and therefore improve the reliability of future simulations[15]. EIS will model large and fast retreat of former ice sheets (T2) to better constrainprojections (T3).

It is now acknowledged that the most significant threats regarding SLR stands in the potential instability of marine based sectors in Antarctica (the so-called Marine Ice-Sheet Instability (MISI), [5]), the Amundsen Sea Embayment being most probably the most vulnerable region [17,18]. However, the rate of the forthcoming retreat of the glaciers feeding the Amundsen Sea Embayment remains highly uncertain [19] and the conditions for initializing and maintaining an unstable retreat remains to be evaluated [20]. EIS will improve estimations of the evolution of this specific region (T3). Moreover, part of the uncertainty in ice-sheets projections also comes from the absence of ocean/ice-sheet coupling, inducing a poor connection between ocean warming and changing ice shelf melt rates, as well as a poor representation of the feedbacks of changing iceberg calving and meltwater fluxes on the global climate system [21]. This requires to progressively integrate dynamical ice-sheet models into Earth-System Models and therefore optimize the ergonomy of ice flow models to extend their use from their initial developers to a wider community interested in climate sciences. EIS will contribute to boost the coupling of ISM with numerical representations of other components of the climate system (T3) and will offer to the national (and international) climate community the next generation of ISM, namely Elmer/Ice-Sheet, with a robust representation of fast dynamical changes, ready to be coupled to IPSL-CM and CNRM-CM global coupled climate models.

Task 1 – Development task - Elmer/Ice-Sheet package and capabilities

The aim of this task is to ensure timely, continuous and long-term development of Elmer/Ice-Sheet. First, we will extend the Elmer/Ice-Sheet package through a number of possible configurations, standardize Input/Output (I/O), perform benchmarking to evaluate and improve performance of standard simulations and efficiently distribute all the developed tools (manage versioning tools, wiki, Elmer/Ice-Sheet international courses). Secondly, as Elmer/Ice-Sheet is built upon Elmer/Ice and therefore benefit from all the existing and future Elmer/Ice capabilities, we will continue the implementation of new processes relevant for ice-sheet scale applications in Elmer/Ice and continuously optimize its numerical performance for standard ice-sheet wide simulations. Prompt (<9 months) implementation of well-known parameterizations of Glacial Isostatic Adjustment (GIA) [79] and calving [28] will be quickly initiated in order to rapidly extend Elmer/Ice-Sheet capacities to paleo-simulations. Processes implemented (GIA, calving, friction laws) and technical possibilities will be continuously improved to ensure cutting edge applications all along the project and beyond.

Task 2 - Scientific task - Revisiting past sea-level changes

Our long term objective is to reconcile time-dependent sea-level proxies with Elmer/Ice-Sheetsimulations of former ice sheets. The priorities identified and proposed in the EIS project are the simulation of the past collapse of a marine ice sheet during last deglaciation (ST2.1) and reevaluating the geometries of Greenland and Antarctica during the last interglacial, which is an analog of a warmer-than-todayperiod (ST2.2). The LGM Barents-Kara marine-based ice sheet (BKIS) constitutes one of the best paleo-analogues of the present-day West Antarctic Ice-Sheet. Modeling BKIS with Elmer/Ice-Sheetconstrained with paleo reconstructions [29,30] will allow to investigatethe potential mechanisms involved in the BKIS deglaciation and, ultimately, to examine whether these processes could explain the rapid sea-level rise episodes such the Meltwater Pulse 1A. This will also help to constrain the potential forthcoming collapse of the WAIS. During the Eemian (127-115 ka BP), sea level was 5 to 10 m higher than at present [31]. Evidence that Greenland only contributed for 2-3 m [32, 5] suggests that an additional contribution of at least 3 m may be due to dynamical iceloss in Antarctica. Relying on the capacities of Elmer/Ice-Sheetto cope with coastal glacier dynamics, our objective is to disentangle the respective Greenland and Antarctic contributions to SLR during this period, shedding light on their plausible respective contribution to SLR in our forthcoming warmer world.

Task 3 - Scientific task - Estimating the future contribution of Antarctica to SLR

As a long term objective, we aim to compute an ensemble of centennial to multi-millennial projections of Greenland and Antarctica mass balance, constrained by paleo evidences and recent observations. Our first priorities are to continue the development of state-of-the-art statistical methods to best interpret ISM simulations (ST3.1), investigate rapid ice shelves collapse to provide more reliable higher-end SLR projections (ST3.2), prepare the coupling with ESM by investigating the consequences of high-end scenarios on the climate system (ST3.3) and evaluate the importance of ice-sheet/ocean feedbacks in the Amundsen Sea Embayment (ST3.4). ST3.4 will be key to launching MISOMIP2, i.e. inviting the international community to contribute to a new ice sheet/ocean multi-model ensemble projections of the Amundsen region.The ST3.3 and 3.4 activities also prepare the coupling of Elmer/Ice-Sheetwith CNRM-CM and IPSL-CM. More generally, this also prepares the French community to the next CMIP initiative (CMIP7) where dynamical ice sheets will most probably be a standard

[1] Dutton et al., Science 349 (2015).
[2] Deschamps et al., Nature 483 (2012).
[3] Shepherd et al., Science 338 (2012).
[4] Solomon et al., WG1-AR4 IPCC (2007).
[5] Stocker et al., WG1-AR5 IPCC (2013).
[6] Ritz et al., JGR-Atm 106 (2001).
[7] Charbit et al., Global and Planetary Changes 48 (2005).
[8] Bonelli et al., Clim. Past 5 (2009).
[9] Roche et al., GMD 7 (2014).
[10] Grev eet al., Ann. Glaciol 52 (2011).
[11] Maris et al., Clim. Dynamics 45 (2015).
[12] Pattyn et al.,The Cryosphere, 6 (2012).
[13] Golledge et al., Nature Com. (2014).
[14] Alvarez-Solas et al., PNAS, 110 (2013).
[15] DeConto and Pollard, Nature, 531 (2016).
[16] Edwards et al., Nature, 566 (2019).
[17] Favier et al., Nature Clim. Change 4 (2014).
[18] Joughin et al.,Science, 344 (2014).
[19] Nias et al., J. Glaciol, 62, 223 (2016).
[20] Gudmundsson et al., The Cryosphere, 6 (2012).
[21] Asay-Davis et al.. Current Climate Change Reports, 3,4 (2017).

Updated on 24 December 2020