Are pooled tree ring δ13C and δ18O series reliable climate archives? — A case study of Pinus nigra spp. laricio (Corsica/France)

Abstract

Stable carbon and oxygen isotopes in tree rings are considered as reliable climate archives as they provide past environmental information with high resolution. However, recent studies have shown that isotope chronologies may reveal a long-term age trend and be influenced by non-climatic factors as other tree-ring parameters as well. These trends can only be identified in chronologies measured on individual trees but not on pooled sample chronologies consisting of several trees. In order to test whether pooled chronologies from pine trees from the Island of Corsica (Mediterranean/France) can be used for climatic reconstructions, we compared calculated mean values from 5 individual trees with pooled chronologies from the same 5 individuals. Carbon and oxygen isotope chronologies for a 50-year interval with annual resolution and 400years with decadal resolution were analysed in order to document secular changes in inter-tree variability and to test for age related trends in the isotope ratios. Pooled carbon and oxygen isotope series correspond well to chronologies based on mean values calculated from analyses of individual trees. Inter-tree variability in oxygen isotope ratios is higher than in carbon isotope ratios but remains relatively constant over time. Similarities between the isotope series of individual trees are stronger over the 400-year time scale documenting a common long-term signal in the isotope values. No long-term age related trends are observed. Oxygen isotope values of the juvenile phase are characterised by a remarkable decrease over 40–50years, interpreted to be related to the less developed root system of the young trees. This age effect can be avoided by not considering the first 50years from the chronologies. Our results confirm that pooled carbon and oxygen isotope chronologies from Pinus nigra can be used for environmental reconstructions without statistical detrending.