THE EFFECT OF PDF PEAKS’ HEIGHT INCREASE DURING CALIBRATION OF RADIOCARBON DATE SETS ADAM MICHCZYÑSKI and DANUTA J. MICHCZYÑSKA Department of Radioisotopes, Institute of Physics, Silesian University of Technology, Krzywoustego 2, 44-100 Gliwice (e-mail: Adam.Michczynski@polsl.pl) Abstract: Abstract: Abstract: Abstract: Abstract: Large sets of dates are often used to construct frequency distributions to investi- gate variability of some events, which can follow an environmental change (eg. crystalliza- tion of speleothems depends on climatic conditions). Examples of such distributions are probability density functions (PDF) created for radiocarbon dates. In order to reach reli- able conclusions concerning environmental changes, we should know how to interpret these distributions. In this study, the authors discuss the problem of a possible correlation between the presence of some high, narrow peaks in the probability density function and the shape of the calibration curve. Key words ey words ey words ey words ey words: RADIOCARBON, CALIBRATION, LARGE SET OF RADIOCARBON DATES GEOCHRONOMETRIA Vol. 25, pp 1-4, 2006 – Journal on Methods and Applications of Absolute Chronology 1. INTRODUCTION The problem of using large sets of radiocarbon dates in environmental studies was discussed in 2003 by Michczyñska. A statistical analysis of large set of radio- carbon dates for reconstruction of paleoclimate was reported by Michczyñska and Pazdur (2004). In that paper, the authors assumed that the random character of dates is preserved in the case of large set of 14 C dates (large territory, different investigators interested in various sci- entific disciplines collecting samples). The authors constructed a Probability Density Function (PDF) by summing the probability distributions of individual 14 C dates after the calibration. However the shape of the con- structed PDF was unexpected, because there were present high, narrow peaks (see Fig. 1, the lower graph). Essen- tial changes in the shape of PDFs were expected at those time periods, where changes in the environment were important, but the appearance of such narrow peaks seemed strange. The coincidence of the localisation of high, narrow peaks of the PDF with the localisation of the border of the Holocene subdivision (Starkel, 1999; Pazdur, 2004) suggested that the presence of the men- tioned peaks might be a result of the methodology of sam- pling. According to paper Michczyñska and Pazdur (2004): “Samples for 14 C dating are frequently collected only from selected horizons that are of special interest from the point of view of the investigator. Because of economic reasons only limited number of organic layers can be dated. The general rule of taking samples from places of visible sedi- mentation changes (e.g. from the top and bottom of the peat layer) may be the reason why samples from the bor- der of the Holocene subdivisions are collected more fre- quently” 2. METHODS In this study we would like to look carefully at the prob- lem of the shape of PDF once more. We compared the shape of the PDF for 785 14 C dates of peat samples and the shape of calibration curve. We noticed that some of the peaks of the PDF (the high, narrow ones) are corre-