Chironomid-inferred summer temperature during the Last Glacial Maximum in the Southern Black Forest, Central Europe

The location of Bergsee (382 m a.s.l.), between the Black Forest and northern Alpine glaciers during their maximum extent of the Würm glaciation, makes the sediment record of this lake a unique palaeoenvironmental archive that probably recorded the entire Last Glacial Period. Here we present a chironomid record from Bergsee covering ca. 35 thousands of years (kyr) including the period corresponding to the Last Glacial Maximum in the northern Alpine area. The record is divided into 6 biostratigraphical zones. Between ca. 45.4–30.1 thousand calibrated ¹⁴C years (cal ka BP) taxa typical for the littoral of relatively warm lakes (Parakiefferiella bathophila-type and Paratanytarsus penicillatus-type) are dominating. Then (ca. 30.1–23.3 cal ka BP), Sergentia coracina-type, a profundal and cold indicative taxon, becomes dominant alongside Parakiefferiella bathophila-type. Low diversity and high abundances of Sergentia coracina-type suggest the most severe environmental conditions of the record. The third zone (ca. 23.3–20.6 cal ka BP) is largely dominated by Parakiefferiella bathophila-type and warmer indicative taxa replace Sergentia coracina-type. In the fourth zone (ca. 20.6–16.9 cal ka BP), Paratanytarsus penicillatus-type and Tanytarsus pallidicornis-type dominate and the presence of Chironomus anthracinus-type suggests slightly warmer conditions, more nutrients or organic matter and lower oxygen availability in the lake. In the next zone (ca. 16.9–10.7 cal ka BP), Corynocera ambigua, a taxon with uncertain distribution in respect to temperature, becomes the dominant chironomid. Changing abundances of Corynocera ambigua match the climate variability of the Lateglacial, with high abundances corresponding to the cold phases of the Oldest and Younger Dryas interrupted by the warmer Bølling/Allerød interstadial. However, the high abundances of Corynocera ambigua are not necessarily due to colder temperature but could also be explained by changes in the trophic conditions in the lake or other environmental changes. The last zone (ca. 10.7–8.4 cal ka BP, onset of the Holocene), shows the highest diversity and presence of warm indicative taxa such as Endochironomus tendens-type and Polypedilum nubeculosum-type. The high abundances of Corynocera ambigua make the temperature reconstruction challenging for the Lateglacial since this taxon is restricted to cold lakes in some European chironomid-temperature calibration datasets, whereas it is known that it can also be found in high abundances in warmer lakes in other calibration datasets. A chironomid-temperature transfer function based on a Swiss-Norwegian calibration dataset reconstructed exceptionally cold values in the interval with maximum abundances of Corynocera ambigua, which disagree with reconstructions from other palaeoenvironmental archives in the region. To address this problem, the Swiss-Norwegian model was also used excluding Corynocera ambigua and another calibration dataset (Finnish), having a more realistic temperature optimum for Corynocera ambigua, was applied to the Bergsee chironomid record to evaluate the results. The reconstructions with the Swiss-Norwegian calibration dataset excluding Corynocera ambigua and with the Finnish calibration dataset resulted in a more realistic temperature development for the Lateglacial at Bergsee. The reconstructions based on these datasets indicate temperatures of ca. 11.6 °C for the period 45–30.1 cal ka BP, the coldest phase between 30.1 and 23.3 cal ka BP with July air temperature of ca. 10.5 °C, an abrupt warming at the beginning of the Bølling interstadial to ca. 16 °C and a decrease to 12.2 °C during the Younger Dryas. Another warming of 3.7 °C is estimated after the Younger Dryas and reconstructed temperatures reached ca. 16.4 °C in the Early Holocene. However, inferred temperatures in this range may already have been affected by the edge effect that can lead to underestimated temperatures close to the maximum temperatures covered by calibration datasets. The timing of the coldest phase is consistent with reconstructions of the expansion and maximum extent of northern Alpine glaciers, the pollen record from Bergsee and other palaeoenvironmental records in the region.