Oxygen isotopes in tree rings of Abies alba : The climatic significance of interdecadal variations

We determined the δ 18 O variations in the latewood of tree rings from four silver firs ( Abies alba Mill.) for the period 1840–1997 at a mountain site in Switzerland, establishing the longest available tree ring record for δ 18 O in central Europe. The isotope ratios were determined on whole wood with a rapid continuous flow pyrolysis technique, thus avoiding cellulose extraction. We found significant correlations with δ 18 O tree ring records from the same region, although these involved different materials (cellulose extracted from whole rings rather than latewood) and different species. This indicates that physical factors are more important than biological influences as a determinant of δ 18 O in tree rings. The isotope tree ring chronology was highly correlated with the oxygen isotope variations in the June/July precipitation for the period 1972–1992 ( r =0.72), and δ 18 O in whole wood of tree rings is therefore well suited for the reconstruction of δ 18 O in precipitation. We found a slow, quasi‐periodic variation of the δ 18 O series with a periodicity of ∼24 years, which is correlated to variations in the July temperature. This could be caused by fluctuations in the large‐scale atmospheric circulation over Europe and the North Atlantic, which may result in a change in source and flow path of atmospheric moisture, affecting the isotope ratio of precipitation in Switzerland. Although a significant correlation with summer temperature was found ( p <0.01), the low correlation coefficient ( r = 0.31) indicates that the δ 18 O variations cannot be explained by temperature variations alone. However, even when considering that the factors influencing δ 18 O in precipitation are not yet fully understood, our study shows the potential of tree rings to provide long records of δ 18 O in precipitation for continental areas, which will improve our understanding of the causes of natural perturbations of the climate system.