Late Quaternary climate variability from western Mediterranean lake archives by multi-proxy data

Climate change is under global debate, due to increasing occurrence of extreme weather events affecting the present-day life. Predicting future climate is challenging, because of the complex interactions in the climate system. Only full understanding of past climate processes and forcing mechanisms allows us to understand and evaluate on-going changes. Some regions are more vulnerable than others. The Western Mediterranean, due to the transitional location between complex atmospheric and marine circulation systems, is highly sensitive to climatic changes. This work is affiliated to a Collaborative Research Centre (CRC 806 “Our Way to Europe”) dealing with history of human mankind and is in particular a contribution to the sub-project investigating the disappearance of Neanderthals in the Iberian Peninsula, which is supposed to be connected to high frequent climate variability. Three different sites, aligned on a SW-NE-axis from Southern Spain to Northern Italy, have been under investigation for Late Quaternary palaeoenvironmental reconstruction. Aiming on terrestrial records, which are scarce for the Western Mediterranean, sediment cores from three lakes revealed palaeoinformation: i) Laguna de Fuente de Piedra (LFP; SW-Spain), ii) Lake Banyoles (NE-Spain), and iii) Lake Como (N-Italy). LFP a shallow endorheic saline lake, whereas Lake Banyoles is a tectono-karstic open fresh-water lake, and Lake Como is a large 450 m-deep, semi-closed lake, all three are obviously of particularly difference in their characteristics. Multi-proxy data acquisition has been carried out and palaeoclimatic records at different temporal resolution with different temporal range have been achieved. Due to the different control mechanisms on sedimentation processes, the selection of climate-sensitive proxies for each site was challenging. For the hydro-sensitive LFP, optical description of sediment cores (max. 14 m long), elemental data (Ti, Ca, S, Sr, ratios, and time series analyses), mineralogical data including crystal habits and stable isotopic signatures of carbonates and gypsum have been acquired. This dataset leads for lithofacies-based palaeohydrological reconstruction. The obtained age control has been achieved by radiocarbon dating. But the low TOC content, bacterial sulfate reduction processes as well as saline groundwater circulation (with long mean residence times) increased the complexity of this site and probably affected datings. The identified lake level changes have been correlated with North Atlantic cold events for the past approximately 28,000 years. The Lake Banyoles sediment core (66 m long) characterization contains optical description, elemental data, and sediment-physical properties, isotopic data, and diatom analyses leading to different lithofacies. Robust age control has been achieved by U/Th dating, which has been re-assessed by stable isotope record correlation for the upper 30,000 years. Radiocarbon ages were used cautiously, due to an approx. 5,000 year reservoir effect. The high content of endogenic carbonates in Banyoles sediments makes this site highly sensitive to terrigenous influx, as observed in elemental data and their ratios, particularly K/Ca ratio. Potassium-rich material has been correlated to North Atlantic cold events (Heinrich Events 1-5), which are supposed to originate in clay minerals transported either fluvial or aeolian to the lake. The impact of North Atlantic circulation has also be proven by stable isotopic signatures, which rather show an Atlantic than Mediterranean signal.From a sedimentary record (65 m long) from Lake Como, different lithofacies have been revealed by optical description, elemental and sediment-physical properties (core logging). The radiocarbon chronology encompasses to upper 32 m and two age-depth models have been discussed for the entire record. Implications for palaeoclimate have been obtained from the younger chronology, assuming ca. 17,000 years at the base. This site also recorded effects from North Atlantic cold events, for the Holocene and have been further discussed for the Deglaciation. Two event layers composed of significantly coarser material are most probable related to climatic forcings as they coincide with warming phases, but seismotectonic triggers cannot be fully excluded.This work narrows the gap in terrestrial long-term records from the Western Mediterranean. It is shown that the sites are affected by North Atlantic Circulation, but all three show certain differences in their response to past North Atlantic Cold Events, which needs further work for detailed causative analyses