Temporal instability of isotopes-climate statistical relationships - A study of black spruce trees in northeastern Canada.
Climate reconstructions using stable isotopes (δ¹⁸O and δ¹³C values) in tree rings are based on relationships between present climatic conditions and isotopic series. This widely used approach relies on the assumption that correlations between stable isotopes and climatic conditions are steady over...
| Published in: | Dendrochronologia |
|---|---|
| Main Authors: | , , , , , |
| Format: | Article in Journal/Newspaper |
| Language: | unknown |
| Published: |
2015
|
| Subjects: | |
| Online Access: | https://espace.inrs.ca/id/eprint/3892/ https://doi.org/10.1016/j.dendro.2015.04.001 |
| Summary: | Climate reconstructions using stable isotopes (δ¹⁸O and δ¹³C values) in tree rings are based on relationships between present climatic conditions and isotopic series. This widely used approach relies on the assumption that correlations between stable isotopes and climatic conditions are steady over time. In this paper, we evaluate the strength of the correlations between δ¹⁸O and δ¹³C series with several climatic parameters on fourteen black spruce trees coming from three different sites, in northeastern Canada. We applied a 21-year moving window on the r Pearson calculated between stable isotopes and March-May and June-August precipitation, June-August and April-June maximal temperatures. Our results indicate that despite the large distance and differences in stand conditions between the sites, the three sites responded in the same way over time. We show that because the climatic ambiance has changed during the 1980-1990 period due to a positive North Atlantic Oscillation index the δ¹³C values are not controlled anymore by spring precipitation or summer maximal temperature in the following two decades. As opposed to δ¹³C series, the relationship between summer maximal temperature and δ¹⁸O values was stable over time, and decreased only in the last decade. All these results attest of a “divergence problem” in the last decades which is most pronounced for δ¹³C series. We conclude that the spruce δ¹⁸O series appears to be the most appropriate indicator for reconstructing June-August maximal temperature in the studied area despite the divergence issue, given that the calibration-validation tests and reconstruction can exclude the divergent last decade. |
|---|