A 500.000 yrs paleoenvironmental and paleoclimatic record from the Balkans inferred from Lake Ohrid pollen data

Lake Ohrid is considered as one of the deepest and oldest tectonic lake in Europe. A drilling campaign was carried out in 2013 as part of the International Continental Scientific Drilling Program (ICDP), within the scope of the Scientific Collaboration on Past Speciation Conditions in Lake Ohrid (SCOPSCO) project. This lake is located at middle altitudes in a rift basin (693 m) surrounded by high mountain ranges and provides an excellent opportunity to study the impact of past climate changes on mid- to high-altitude forests. At the confluence of temperate and Mediterranean climate influences it is an exceptional site to record climate changes over several glacial-interglacial periods. First investigations show that the DEEP site sediment sequence covers the entire lake history and an age model was established for the upper 500 kyr. Thus, the last 12 marine isotope stages have been analysed at a millennial-scale resolution by a European team of palynologists. Our aim is now to propose a robust and precise quantitative estimates of the climate from the DEEP pollen record with three objectives: (1) to explore climate change on the Balkan Peninsula at the confluence of temperate and Mediterranean climate influences, (2) to identify the major patterns of climate change during the last 500 kyr in this region with a focus on the seasonality of the reconstructed climate parameters, and (3) to reconstruct the climatic change during the glacial and interglacial periods to better understand the variability of these glacial–interglacial cycles in south Europe. based on a multi- method approach (including the Modern Analogues Technique, and Weighted Averages regression). The originality of our approach is to better assess the error of the climate reconstruction by integrating all the errors calculated by each method in an “optimal” temperature and precipitation reconstruction. The so far existing DEEP pollen sequence records the major vegetation and climate changes. Results show that there is a general good correspondence between forested/non-forested periods and glacial– interglacial cycles of the marine isotope stratigraphy, suggesting a regionally relevant vegetation and climate record. We have reconstructed the annual precipitation and the annual temperature, summer precipitation and temperature, alpha index and GDD5 with the different methods. All the methods give roughly the same pattern, even if the MAT show more variability; the different glacials and interglacial periods are well reconstructed. Our reconstruction shows 2 distinct periods:-from 500 to 200 ka, colder and wetter than today conditions;-from 200 ka to now, more warm and dry conditions, particularly during glacials MIS6 and MIS2. The interglacial complex MIS5 is also dryer than interglacials 7, 9, 11. To improve these results, tests without pine and more methods are still in progress to better assess the uncertainties in the predictions and the error of reconstruction.