Scrutinizing tree‐ring parameters for Holocene climate reconstructions

Independent evidence from Europe and Asia shows that tree‐ring stable isotopes can reveal persistent long‐term hydroclimate trends that are generally not captured by more traditional dendroclimatic studies using tree‐ring width or density. Since the recently observed long‐term discrepancy between flatter “growth‐dependent” and more varying “growth‐independent” climate proxy data is unrelated to possible biases of statistical age‐trend removal, I call for a conceptual rethinking of the predictive power of different tree‐ring parameters for reconstructing climate variability on interannual to multimillennial timescales. I describe why traditional “growth‐dependent” tree‐ring width and wood density measurements usually lack abiotic signals on ultra‐long timescales, whereas “growth‐independent” carbon and oxygen isotopic ratios from tree‐ring cellulose can capture environmental variation well beyond the segment length of individual tree‐ring samples. Caution is therefore advised when information from diverse tree‐ring parameters is combined in multiproxy reconstructions of Holocene climate that aim to reflect the full range of interannual to multimillennial variability. This Perspective not only emphasizes the paleoclimatic value that can be obtained from tree‐ring stable isotopes in living and relict wood. It also stresses the need for developing new high‐resolution isotopic datasets from different species and regions in both hemispheres to supplement the existing tree‐ring record.