Leger et al. (2025) The Greenland-Ice-Sheet evolution over the last 24 000 years: insights from model simulations evaluated against ice-extent markers

Identification

• Journal: ˜The œcryosphere
• Year: 2025
• Date: 2025-11-14
• Authors: Tancrède Leger, Jeremy C. Ely, Chris D. Clark, Sarah Bradley, Rosie Archer, Jiang Zhu
• DOI: 10.5194/tc-19-5719-2025

Research Groups

• School of Geography and Planning, University of Sheffield, Sheffield, UK
• Institute of Earth Surface Dynamics, University of Lausanne, Lausanne, Switzerland
• Department of Geography and Environmental Sciences, Northumbria University, Newcastle, UK
• Climate and Global Dynamics Laboratory, NSF National Center for Atmospheric Research, Boulder, CO, USA

Short Summary

This study simulates the Greenland Ice Sheet's evolution over the last 24,000 years using an ensemble of 100 high-resolution ice-sheet models, quantitatively evaluating them against empirical ice-margin extent data. The findings reveal the dynamics, drivers, and spatial heterogeneities of the ice sheet's past evolution, indicating a larger Last Glacial Maximum extent and faster deglacial mass loss than previously estimated, while highlighting persistent regional model-data misfits.

Objective

• To produce data-consistent simulations of the Greenland Ice Sheet's evolution over the last 24,000 years, evaluating its past volume, extent, flux, internal flow dynamics, and thermal conditions against empirical ice-margin data to improve future projection initializations.

Study Configuration

• Spatial Scale: Greenland-wide, with a horizontal resolution of 5 km and 101 vertical ice layers.
• Temporal Scale: 24,000 years Before Present (BP) to 1850 AD (Pre-Industrial era).

Methodology and Data

• Models used:
  • Parallel Ice Sheet Model (PISM) version 2.0.5 (ice sheet model)
  • isotope-enabled Community Earth System Model (iCESM) (climate forcing)
  • Offline Glacial Isostatic Adjustment (GIA) model (for non-local sea level changes)
  • PISM's Lingle-Clark-type viscoelastic deformation model (for local GIA)
  • PISM's Positive-Degree-Day (PDD) model (for Surface Mass Balance)
  • Sub-shelf melt model (Hellmer et al., 1998; Holland and Jenkins, 1999)
  • Strain-rate-based eigen calving law (Albrecht and Levermann, 2014)
• Data sources:
  • PaleoGrIS 1.0 (Leger et al., 2024) (Greenland-wide reconstruction of former ice margin retreat and ice-extent markers)
  • BedMachine v4 (Morlighem et al., 2017) (present-day GrIS extent, ice thickness, bedrock topography)
  • ALOS World 3D 30 m Digital Elevation Model (DEM; Tadono et al., 2014) (terrestrial topography)
  • General Bathymetric Chart of the Oceans (GEBCO Bathymetric Compilation Group 2022, 2022) (ocean bathymetry)
  • Geothermal heat flux data from Martos et al. (2018)
  • iCESM (Brady et al., 2019) simulations (surface air temperature, precipitation, ocean salinity)
  • iTRACE experiment (He et al., 2021a, b) (monthly-resolution climate output)
  • Osman et al. (2021) (LGM-to-present ensemble-mean sea surface temperature (SST) reconstruction, surface air temperature and δ18O reconstructions)
  • Buizert et al. (2018) (Greenland-wide temperature and precipitation reconstruction)
  • Holocene thinning curves (Vinther et al., 2009; Lecavalier et al., 2013) (for validation)
  • Joughin et al. (2018) (present-day GrIS surface velocities)
  • Mankoff et al. (2020) (present-day GrIS ice discharge rate)
  • Simonsen et al. (2021) (present-day GrIS mass loss rate)

Main Results

• The maximum grounded Greenland Ice Sheet (GrIS) extent, the local Last Glacial Maximum (lLGM), likely occurred between 17.5 and 16 thousand years Before Present (kyr BP), covering an area of 2.9 to 3.1 million square kilometers. Grounded ice reached the continental shelf break along the entire Western, Southern, and Southeastern Greenland coasts.
• The GrIS contributed between 6 and 7.5 meters to global mean sea-level rise between the lLGM and today, a value higher than most previous estimates.
• During the lLGM, the ice sheet was up to approximately 1 kilometer thicker and 500 meters higher in elevation in Southern and Northwestern Greenland, leading to ice divide migrations. However, at central and northern GrIS summits, the ice sheet was not necessarily thicker than today.
• lLGM discharge rates reached 1500–1900 gigatonnes per year (Gt/a), 2.8–4.3 times higher than present-day estimates, indicating substantial iceberg and freshwater delivery to the North Atlantic.
• Rapid and extensive GrIS retreat occurred between 16 and 14 kyr BP (late Heinrich Stadial 1 and Bølling–Allerød warming), primarily driven by ocean warming and increased sub-shelf melt, with mass loss rates up to 7 times greater than present.
• GrIS margin re-advances during the Younger Dryas cooling event were limited and of low magnitude due to strong ice-sheet inertia.
• The GrIS reached a minimum ice extent between 6 and 5 kyr BP, lagging the cessation of Holocene Thermal Maximum warming by centuries to a millennium. Retreat behind present-day margins (up to ~100 km) was modeled only south of 68° N.
• Model-data misfits in grounded ice extent are spatially heterogeneous, with larger discrepancies in Northern, Northeastern, and Central-eastern Greenland, where both LGM advance and deglacial retreat magnitudes are underestimated.
• No single ensemble simulation consistently matched empirical data across the entire 24-0 kyr BP period and all GrIS regions, indicating trade-offs in parameterizations.

Contributions

• Presents the first high-resolution (5 km) perturbed-parameter ensemble (100 simulations) of the entire Greenland Ice Sheet evolution from 24 kyr BP to 1850 AD, quantitatively scored against Greenland-wide empirical ice-extent markers (PaleoGrIS 1.0).
• Provides novel insights into the dynamics, drivers, and spatial heterogeneities of GrIS evolution during the Last Glacial Maximum, Late-glacial, and Holocene.
• Highlights the significant role of ocean forcing in rapid deglaciation and the substantial inertia of the ice sheet.
• Quantifies a higher-than-previously-estimated sea-level contribution and discharge rates during the lLGM.
• Identifies regional differences in ice-sheet thickness changes and ice divide migrations at ice core sites, with implications for paleoclimate records.
• Demonstrates the importance of chronologically ordered sieving in model-data comparison to avoid selecting simulations that reproduce present-day states for unrealistic paleo-evolutionary reasons.
• Reveals parameter clusters (precipitation offset, air temperature offset, flow law enhancement factor) that influence model-data fit, suggesting potential biases in climate forcings or model parameterizations.

Funding

• European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 787263)
• U.S. National Science Foundation (NSF)
• NSF National Center for Atmospheric Research (Cooperative Agreement no. 1852977)
• Swiss National Science Foundation (RECONCILE: project number: 213077)
• NERC independent fellowship award (grant no. NE/R014574/1)

Citation

@article{Leger2025GreenlandIceSheet,
  author = {Leger, Tancrède and Ely, Jeremy C. and Clark, Chris D. and Bradley, Sarah and Archer, Rosie and Zhu, Jiang},
  title = {The Greenland-Ice-Sheet evolution over the last 24 000 years: insights from model simulations evaluated against ice-extent markers},
  journal = {˜The œcryosphere},
  year = {2025},
  doi = {10.5194/tc-19-5719-2025},
  url = {https://doi.org/10.5194/tc-19-5719-2025}
}