Henz et al. (2025) Alps-wide high-resolution 3D modelling reconstruction of glacier geometry and climatic conditions for the Little Ice Age

Identification

• Journal: ˜The œcryosphere
• Year: 2025
• Date: 2025-11-19
• Authors: Andreas Henz, Johannes Reinthaler, Samuel U. Nussbaumer, Tancrède Leger, Sarah Kamleitner, Guillaume Jouvet, Andreas Vieli
• DOI: 10.5194/tc-19-5913-2025

Research Groups

• Department of Geography, University of Zurich, Zurich, Switzerland
• Institute of Earth Surface Dynamics, University of Lausanne, Lausanne, Switzerland

Short Summary

This study presents the first Alps-wide, three-dimensional, model-derived reconstruction of glacier surfaces during the Little Ice Age (LIA) at 50 m resolution, using the Instructed Glacier Model (IGM) to match empirically mapped LIA glacier extents. It reveals a total ice volume of 283 ± 42 km³ and regional/local patterns of equilibrium line altitudes (ELAs) influenced by climatic and topographic factors.

Objective

• To generate the first Alps-wide, three-dimensional, model-derived reconstruction of glacier surfaces during the Little Ice Age (LIA) in the European Alps, consistent with ice flow physics and mass conservation.
• To simulate glaciers to match empirically mapped LIA maximum extents at a resolution of 50 m.
• To infer and analyze the corresponding spatial patterns of Equilibrium Line Altitudes (ELAs) across the Alps, examining their relationship with independent climate variables and topographic factors.

Study Configuration

• Spatial Scale: European Alps (Alps-wide), divided into 14 subregions, with a grid resolution of 50 meters (100 meters for sensitivity analysis).
• Temporal Scale: Little Ice Age (LIA) maximum extent (1260 to 1860 CE), representing a quasi steady-state configuration around 1850 CE.

Methodology and Data

• Models used:
  • Instructed Glacier Model (IGM) v2.2.2: A physics-informed neural network solving higher-order Blatter-Pattyn ice flow equations, capable of 3D ice flow modeling.
  • Simple elevation-dependent mass balance model: Uses an input ELA and two linear mass balance gradients (0.005 per year for accumulation, 0.009 per year for ablation) with a fixed maximum accumulation of 2 meters.
• Data sources:
  • Input surface topography: ALOS World 3D 30 meter (AW3D30) digital elevation model (DEM).
  • Subglacial topography: Derived by deducting present-day glacier thickness (Cook et al., 2023) from AW3D30 DEM using IGM inversion.
  • Target glacier outlines: Empirically reconstructed LIA glacier outlines of the Alps (Reinthaler and Paul, 2025).
  • Climate data for analysis (not model forcing): Gridded pre-industrial climate snapshot simulation at 2 km resolution (Russo et al., 2024) for temperature and precipitation.

Main Results

• The study produced the first Alps-wide, high-resolution (50 meters), 3D reconstruction of LIA glacier surfaces for 4094 individual glaciers, consistent with ice flow physics.
• The total Alps-wide glacier ice volume during the LIA was estimated at 283 ± 42 km³.
• The reconstruction revealed distinct regional and local patterns of Equilibrium Line Altitudes (ELAs) for each glacier:
  • Lower ELAs (average < 2700 meters above sea level) were found in the north-western, northern, and north-eastern Alps.
  • Higher ELAs (average > 2700 meters above sea level) were observed in the south-western, southern Alps, and inner-alpine valleys.
  • Regional median ELA values differed by up to 500 meters.
  • South-facing glaciers exhibited ELAs approximately 200–300 meters higher than north-facing glaciers.
• Spatial ELA patterns are significantly influenced by air temperature, precipitation, and shortwave radiation.
  • Lower mean annual air temperatures were correlated with higher ELAs.
  • Increased precipitation allowed for warmer ELA temperatures (resulting in lower ELAs).
  • Smaller solar incidence angles (more direct sun rays) generally led to higher ELAs.
• A sensitivity analysis indicated an uncertainty of up to 14 % in the total ice volume estimate, but glacier-wise ELAs remained largely stable across parameter perturbations.
• The power-law relationship between modelled ice volume and glacier area yielded an exponent of 1.311, slightly lower than the theoretical 1.375, particularly for smaller glaciers.

Contributions

• Presents the first Alps-wide, high-resolution (50 meters), three-dimensional reconstruction of Little Ice Age (LIA) glacier geometry that is fully consistent with the physics of ice flow and mass conservation, utilizing the Instructed Glacier Model (IGM).
• Provides a comprehensive dataset of LIA glacier surfaces and physically consistent Equilibrium Line Altitudes (ELAs) for over 4000 Alpine glaciers, addressing limitations of previous methods that do not explicitly account for ice flow dynamics.
• Offers novel insights into the spatial variability of LIA ELAs and their drivers, including the development of a new, computationally efficient approach for correcting ELA based on the median sun incidence angle.
• Validates the Instructed Glacier Model's capability for large-scale, high-resolution palaeo-glacier modeling, demonstrating its robustness for complex, high-relief glacier systems compared to traditional ratio-based ELA reconstruction methods.
• The generated Alps-wide ELA dataset for the LIA serves as a valuable resource for the palaeo-glaciological community, enabling more accurate contextualization of past and future glacier fluctuations and climate change studies.

Funding

• Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (grant no. 200020_213077)

Citation

@article{Henz2025Alpswide,
  author = {Henz, Andreas and Reinthaler, Johannes and Nussbaumer, Samuel U. and Leger, Tancrède and Kamleitner, Sarah and Jouvet, Guillaume and Vieli, Andreas},
  title = {Alps-wide high-resolution 3D modelling reconstruction of glacier geometry and climatic conditions for the Little Ice Age},
  journal = {˜The œcryosphere},
  year = {2025},
  doi = {10.5194/tc-19-5913-2025},
  url = {https://doi.org/10.5194/tc-19-5913-2025}
}