Abstract
The evolution of temperatures during the Holocene is controversial, especially for the early Holocene. The occurrence of the Holocene Thermal Maximum (HTM) during the early Holocene has recently been reconsidered and seasonal biases have been suggested in the paleoclimatic proxies. High regional variability and a low number of reliable and continuous quantitative reconstructions compared with the oceanic realm further complicate study of the Holocene climate in the continental realm. We analyzed branched glycerol dialkyl glycerol tetraethers (brGDGTs), an organic paleothermometer, and palyno-logical signals as part of a multiproxy analysis of the sedimentary record from Lake St Front, in the Massif Central (France). Identification of a shift in brGDGT sources through the Holocene required removing terrigenous influences from the temperature signal. BrGDGT-and pollen-inferred paleotemperature reconstructions (based on the Modern Analog Technique and the Weighted Averaging Partial Least Squares method) were compared. Both showed a thermal maximum during the early Holocene followed by a decrease of temperatures. We evaluated biases which could potentially influence the reconstructed signal. There was no evidence for a summer temperature bias either for brGDGT-derived temperatures or for pollen-derived temperatures. The Lake St Front data, in agreement with other regional records, confirm the occurrence of the HTM as a general warm period during the early Holocene followed by mid-Holocene cooling in Western Europe and suggest that seasonal biases are not the main explanation of the Holocene conundrum d the disagreement between model simulations and proxy-based temperature reconstructions for the northern hemisphere.