AGE MODEL ESTIMATION IN PALEOCLIMATIC RESEARCH - FUZZY REGRESSION ANDRADIOCARBON UNCERTAINTIES

A fuzzy linear regression-based method to relate the depth and age of sediment layers is described. Because algae, moss or local plants resp ond to climatic change, the age and depth of sediment layers are inter related variables but the link between them is often imprecise. In mos t cases, due to the limited number of layers in a core (i.e., small nu mber of data points), estimation of the slope and uncertainties in cla ssical regression analysis does not take into account the uncertaintie s in radiocarbon dating. Here, fuzzy linear regression, which may be a pplied even to a very small data set, is utilized. The method, illustr ated through a practical example, with eight data points, appears to b e a promising tool in stratigraphic studies to link sediment age to la yer depth and takes into account uncertainties from radiocarbon dating .