The palaeoclimates of Europe since the Last Glacial Maximum: multi-proxy approaches and data-model comparison
Abstract:
In the context of anthropogenic climate change, increasing atmospheric CO₂ concentrations generate a positive radiative forcing responsible for an unprecedented warming at the Quaternary scale (since 2.6 Ma), the future evolution of which remains uncertain. This unprecedented situation highlights the importance of extending analyses beyond instrumental observations in order to investigate the mechanisms of past global climate changes during well-constrained and markedly different climatic periods. The Last Glacial Maximum (23–19 ka) represents a cold and relatively stable climatic state, substantially different from present-day conditions. In Europe, LGM climate reconstructions are spatially uneven (primarily biased toward low-elevation sites) and affected by various methodological uncertainties. Nevertheless, the diversity of available proxies provides an opportunity to assess their respective biases and refine estimates of past climatic conditions. In addition, climate model simulations show large inter-model spread in simulated annual cooling over Europe during the LGM. This thesis aims to reconstruct and compile LGM paleoclimate estimates using a wide range of proxy data and new methodological approaches, with a particular focus on pollen-based reconstructions. At the European scale, pollen-derived estimates converge toward consistent LGM temperatures, indicating an average annual cooling of approximately 6°C. Independent reconstructions based on noble gas concentrations dissolved in groundwater that recharged aquifers in the Grand Est region (Vosges) during the LGM yield similar cooling estimates, consistent with pollen-based results. However, these estimates are systematically warmer than those derived from glacial inversion approaches, which reconstruct climate from a revised LGM glacier extent constrained by recent exposure-age dating of glacial landforms in the Vosges. One possible explanation for this discrepancy lies in differences in climate seasonality between the LGM and the present day. Using the GRISLI ice-sheet model, we explore the sensitivity of inferred climatic conditions by simulating multiple ice cap configurations under different climatic forcings and comparing them to reconstructed glacier extents. A reduction in precipitation increases the required cooling (0.03 °C/%), while enhanced LGM seasonality inferred from pollen data further amplifies this cooling (−0.06 °C/%), thereby exacerbating the mismatch between glacial and other proxy-based reconstructions. Finally, climate model outputs from both General Circulation Models (GCM) and the intermediate-complexity model iLOVECLIM, including a newly developed downscaled European configuration, are analysed in terms of annual cooling and changes in seasonality relative to preindustrial conditions. These simulations reveal a wide range of LGM cooling over Europe, with pronounced differences in seasonality largely driven by winter temperature variations, particularly in and around glaciated regions (8–12 °C STD). Sensitivity experiments with iLOVECLIM further indicate that seasonality changes are primarily controlled by summer temperatures rather than independently forced. In addition, the simplified atmospheric representation in iLOVECLIM leads to seasonality responses that differ substantially from those produced by GCM.
Thesis direction :
Pierre-Henri BLARD – Directeur de recherche, Université de Lorraine, CNRS, Centre de Recherches Pétrographiques et Géochimiques, Vandœuvre-lès-Nancy, France
David Vincent BEKAERT – Professeur junior, Université de Lorraine, CNRS, Centre de Recherches Pétrographiques et Géochimiques, Vandœuvre-lès-Nancy, France
Aurélien QUIQUET – Chargé de recherche, Université Paris-Saclay, CNRS, CEA, UVSQ, Laboratoire des Sciences du Climat et de l’Environnement, Gif-sur-Yvette, France
Rapporteurs :
María Fernanda SÁNCHEZ GOÑI – Professeur/Directrice d’Études, Université Paris Sciences et Lettres, École Pratique des Hautes Études, Paris, France
Didier ROCHE – Directeur de recherche, Université Paris-Saclay, CNRS, CEA, UVSQ, Laboratoire des Science du Climat et de l’Environnement, Gif-sur-Yvette, France
Examineurs :
Fanny LHARDY – Maitresse de conférences, École normale supérieure de Lyon, Lyon, France
Charloke PRUD’HOMME – Chargée de recherche, Université de Lorraine, CNRS, Centre de Recherches Pétrographiques et Géochimiques, Vandœuvre-lès-Nancy, France
Pascale BRACONNOT – Directrice de recherche, Université Paris-Saclay, CNRS, CEA, UVSQ, Laboratoire des Sciences du Climat et de l’Environnement, Gif-sur-Yvette, France
Guest member :
Masa KAGEYAMA – Directrice de recherche, Université Paris-Saclay, CNRS, CEA, UVSQ, Laboratoire des Science du Climat et de l’Environnement, Gif-sur-Yvette, France