Reconstruction of palaeoclimatic changes in Central Europe between 10 and 200 thousand years BP, based on analysis of growth frequency of speleothems

The present study is an attempt to utilise the uranium-thorium dates ofspeleothems as a source ofpalaeoclimatic data. The clue is that the changing climate influenced intensity of speleothem deposition, which is reflected in clustering of U-Series dates of speleothems in certain time intervals. This work discusses and improves various methods of combined presentation of dates, with a special attention to the presentation in form of growth frequency ('pdf) curves. Using the 'bootstrap' method the confidence intervals of the 'pdf' curves could be determined. Also the algorithm, originally developed to determine parameters of arbitrarily chosen maxima in the curve, has been modified. Due to that an assessment of number of maxima has been possible. This method enables objective distinction of phases of speleothem growth, which cannot be done 'by eye' when the 'pdf' curve is smooth. The statistical tests show that the reliable 'pdf' curve should contain more than 150 dates. Basing on 308 U-series dates of cave speleothems from southern Poland and other regions of central Europe, the growth frequency curves for the Carpathians and Uplands have been constructed. Comparison of phases of speleothem growth, distinguished by various authors for several regions of Europe, indicates that the climatic changes were synchronous over the whole region. However, different shapes of the 'pdf' curves reflect increasing continuity ofspeleothem growth in the N-S transect southwards. This may be connected with the N-S climatic gradient in Europe. Using the 'pdf' curves from caves of Tatra and Low Tatra Mountains the most probable timing of development phases ofmountain glaciers has been delimited.