Summed radiocarbon probability density functions cannot prove solar forcing of Central European lake-level changes.

It has been argued that Central European lake levels were driven by solar activity. This interpretation rests on the comparison of a score record of radiocarbon dates with the 14C residual curve. This score record is a variant of a cumulative probability density function (CPF). In this paper it is argued that this method is not valid because the shape of the CPF is determined by the calibration curve. This hypothesis is tested by a set of null models. An alternative interpretation is given according to which only few episodes of Holocene lake-level changes were climatically driven, while most of the time non-climatic factors were dominating. On a more general level some methodological issues of CPFs are discussed and exemplified by the comparison of dendrochronological results from archaeological settlements and a CPF created using radiocarbon dates from the same settlements. [ABSTRACT FROM PUBLISHER]