Palaeoclimate Reconstruction in the Levant and on the Balkans


For an understanding of the climate system’s variability, knowledge of the past climate is essential. Continuous observations are not available for longer than the last 100 years, a period that is insufficient to understand the variability and sensitivity of the climate system. Since information of both is necessary to build good climate models, the reconstruction of the past climate is fundamental. With palaeoclimate reconstructions it is possible to get information about the past climate and the climate changes in the period and region of interest. In this work these are the Levant region and the Balkans. The Levant region is situated around the Jordan Valley in Israel.
In this work the presence of pollen and macrofossils is used as a proxy. In detail the method bases on the assumption that the presence of a plant or more general of biometypes in a certain area is addicted to the climate. This connection between the occurrence of the plants and the climate is described by transfer functions. The nature of these transfer functions has to be probabilistic because the climate-biosphere system is a stochastical system. In the Collaborative Research Center (CRC) 806 project B3 “Our way to Europe” high-resolution lacustrine sediment cores were drilled in March 2010 by Thomas Litt and his working group at Lake Kinneret and Birkat-Ram. The sediment core situated on the Balkans was retrieved at Lake Prespa in November 2009 as part of the CRC project B2.
This study presents the results of local palaeoclimate reconstructions based on methods which are a statistical extension of the concept of biomisation, plant functional types and mutual climatic range (MAR). In more detail the Bayesian Biome Model (BBM) is applied for Lake Kinneret and Ein Gedi and the Bayesian Indicator Taxa Model (BITM) for Lake Prespa. For Birkat Ram the Bayesian Indicator Taxa and Biome Model (BITBM) is newly developed and applied. This method combines the BBM and BITM. Reconstructed are the near surface temperatures, middle troposphere temperatures (850 hPa level), the annual climatic water deficit CWDANN and the annual precipitation amount PANN.
All presented palaeoclimate temperature reconstructions except Lake Kinneret share that the
surface and the middle troposphere temperature reconstructions are in accordance. It is also shown that the CWDANN palaeoclimate reconstruction works and does not contradict PANN. The marginal distribution for CWDANN is for example for Lake Prespa a reconstructed palaeoclimate variable which allows more identifiable variation than in PANN. For Lake Prespa there are four, for Birkat Ram three, for Ein Gedi also three and for Lake Kinneret no identifiable time ranges with different climate in the marginal probability density function (pdf)s. In the case that there are time ranges with different climate they are clearly identifiable in the palaeoclimate reconstructions since the marginal distribution profiles before and after differ more or less. Some of these time ranges are compared by application of a Student’s t-test for a significance test.
Also presented is an interpolation of local reconstructions situated in the Levant or more precise the Jordan Valley which allows a better assessment of climate changes. The Jordan Valley climate field reconstruction (CFR) results is a dryer palaeoclimate than the modern climate for PANN and no climate change for the considered temperatures for all fossil sites and considered time slices. This result remains uncertain since there are some difficulties with the climate database.


url Accessed 5 times | Last updated 15.09.2017


Thoma, B. (2017): Palaeoclimate Reconstruction in the Levant and on the Balkans. University Bonn, Meteorological Institute

Authors Benno Thoma
Type phdthesis
Title Palaeoclimate Reconstruction in the Levant and on the Balkans
Year 2017
Pages XVI, 266
Month 05
Organization Meteorological Institute
School University Bonn
Export BibTeX

Additional Metadata

Spatial {"type":"Polygon","coordinates":[[[-6,46.63],[38.44,28.01],[38.44,28.01],[-6,28.01]]]}
Export RDF
Back to dataset list