Common trends in elements? Within- and between-tree variations of wood-chemistry measured by X-ray fluorescence - A dendrochemical study.

Element composition of annually resolved tree-rings constitutes a promising biological proxy for reconstructions of environmental conditions and pollution history. However, several methodological and physiological issues have to be addressed before sound conclusions can be drawn from dendrochemical time series. For example, radial and vertical translocation processes of elements in the wood might blur or obscure any dendrochemical signal. In this study, we tested the degree of synchronism of elemental time series within and between trees of one coniferous (Pinus sylvestris L.) and one broadleaf (Castanea sativa Mill.) species growing in conventionally managed forests without direct pollution sources in their surroundings. Micro X-ray fluorescence (μXRF) analysis was used to establish time series of relative concentrations of multiple elements (Mg, Al, P, Cl, K, Ca, Cr, Mn, Fe and Ni) for different stem heights and stem exposures. We found a common long-term (decadal) trend for most elements in both species, but only little coherence in the high frequency domain (inter-annual variations). Aligning the element curves by cambial age instead of year of ring formation reduced the standard deviations between the single measurements. This points at an influence of age on longer term trends and would require a detrending in order to extract any environmental signal from dendrochemical time series. The common signal was stronger for pine than for chestnut. In pine, many elements show a concentration gradient with higher values towards the tree crown. Mobility of elements in the stem leading to high within- and between-tree variability, as well as a potential age-trend apparently complicate the establishment of reliable dendrochemical chronologies. For future wood-chemical studies, we recommend to work with element ratios instead of single element time series, to consider potential age trends and to analyze more than one sample per tree to account for internal variability.