Towards a better understanding of long-termwood-chemistry variations in old-growth forests: A case study on ancient Pinus uncinata trees from the Pyrenees

Dendrochemical studies in old forests are still underdeveloped. Old trees growing in remote high-elevation areas far from direct human influence constitute a promising biological proxy for the long-term reconstructions of environmental changes using tree-rings. Furthermore, centennial-long chronologies of multi-elemental chemistry at inter-and intra-annual resolution are scarce. Here, we use a novel non-destructive method by applying Micro X-ray fluorescence (mu XRF) to wood samples of old Pinus uncinata trees from two Pyrenean high-elevation forests growing on acidic and basic soils. To disentangle ontogenetic (changes in tree age and diameter) from environmental influences (e.g., climate warming) we compared element patterns in sapwood (SW) and heartwood (HW) during the pre-industrial (1700-1849) and industrial (1850-2008) periods. We quantified tree-ring growth, wood density and relative element concentrations at annual (TRW, tree-ring) to seasonal resolution (EW, earlywood; LW, latewood) and related them to climate variables (temperature and precipitation) and volcanic eruptions in the 18th and 19th centuries. We detected differences for most studied elements between SW and HW along the stem and also between EW and LW within rings. Long-term positive and negative trends were observed for Ca and K, respectively. Cl, P and S showed positive trends during the industrial period. However, differences between sites were also notable. Higher values of Mg, Al, Si and the Ca/Mn ratio were observed at the site with acidic soil. Growing-season temperatures were positively related to growth, maximum wood density and to the concentration of most elements. Peaks in S, Fe, Cl, Zn and Ca were linked to major volcanic eruptions (e.g., Tambora in 1815). Our results reveal the potential of long-term wood-chemistry studies based on the mu XRF non-destructive technique to reconstruct environmental changes. (c) 2017 Elsevier B.V. All rights reserved.