Seasonal climate impacts on LGM glaciers in the Vosges(France): Insights from GRISLI modeling and paleo-extent

Glaciers are key hydro-climatic indicators and markers of atmospheric changes in the past, making them essential tools for reconstructing glacial paleoenvironments and paleoclimates. As a climatically stable period that is drastically different from today, the Last Glacial Maximum (LGM, 26–19 ka BP) is widely used as a benchmark for evaluating climate sensitivity (i.e., a key parameter linking atmospheric CO₂ to temperature) and for comparing climate model simulations with continental reconstructions from multiple proxy archives.

Pollen assemblages are a commonly used proxy for reconstructing past temperature changes, as they offer broad spatial coverage across Europe. However, particularly in Europe, simulated LGM annual temperatures often show substantial disagreement with reconstructions and appear highly heterogeneous across models. Dated glacier extents provide an independent archive, helping to assess data–model comparisons.

Temperature is a critical variable to estimate the surface mass balance of glaciers (i.e., the difference between accumulation and ablation). Surface mass balance models (e.g., the positive degree day, PDD, model; [1]) provide the climatic conditions required to reproduce the extent of paleo-ice sheets (inverse approach), as constrained by geomorphological evidence.

PDD-based ice sheet models in central Europe ([2]; [3]) indicate stronger LGM cooling than pollen reconstructions (e.g., [4]), a mismatch likely linked to seasonal biases given the high sensitivity of glaciers to seasonal temperatures ([5]; [6]). Yet, seasonal LGM reconstructions remain scarce, and recent syntheses highlight marked inconsistencies in seasonality anomalies across European glaciated regions, including the Vosges ([7]) - which are too small to be captured by climate models (Global Circulation Models, GCMs).

Using a new compilation of 10Be cosmogenic exposure ages ([8]; [9]) in the Vosges Mountains (NE France) and the GRISLI ice sheet model ([10]), this study investigates the impact of LGM seasonal and precipitation anomalies on simulated glacier extents and on LGM data-model cooling agreement.

As results, we deduce a high variability of LGM climate conditions sufficient to reproduce the paleo-ice sheet extent in the Vosges, yet none of them match the pollen-based paleoclimatic reconstructions ([11]). However, some LGM climate models produce temperature conditions (annual and seasonal) similar to the GRISLI results, while producing lower precipitation in the Vosges (60% to 120% lower than GRISLI results). While the calibration of the GRISLI model has a minor effect on these results, one of the more feasible ways to minimize data–model discrepancies in climate spaces - considering paleoclimatic reconstructions - would be to substantially increase precipitation (+380%, i.e., ~5 times modern precipitation) in the restricted Vosges massif during the LGM.

[1] Reeh, 11-128 (erschienen, 1991)

[2] Allen +, https://cp.copernicus.org/articles/4/249/2008/

[3] Heyman, https://doi.org/10.1016/j.yqres.2012.09.005

[4] Davis +, https://cp.copernicus.org/articles/20/1939/2024/

[5] Oerlemans and Riechert, https://doi.org/10.3189/172756500781833269

[6] Huss and Hock, https://www.nature.com/articles/s41558-017-0049-x

[7 & 11] Fénisse +, in prep

[8] Harmand, https://doi.org/10.4000/rge.9703

[9] Blard +, in prep

[10] Quiquet +, https://doi.org/10.5194/gmd-11-5003-2018