Tree-ring C-H-O isotope variability and sampling.

In light of the proliferation of tree-ring isotope studies, the magnitude and cause of variability of tree-ring δ(13)C, δ(18)O and δ(2)H within individual trees (circumferential) and among trees at a site is examined in reference to field and laboratory sampling requirements and strategies. Within this framework, this paper provides a state-of-knowledge summary of the influence of "juvenile" isotope effects, ageing effects, and genetic effects, as well as the interchangeability of species, choice of ring segment to analyze (whole ring, earlywood or latewood), and the option of sample pooling. The range of isotopic composition of the same ring among trees at a site is ca. 1-3‰ for δ(13)C, 1-4‰ δ(18)O, and 5-30‰ for δ(2)H, whereas the circumferential variability within a tree is lower. A standard prescription for sampling and analysis does not exist because of differences in field environmental circumstances and mixed findings represented in relevant published literature. Decisions in this regard will usually be tightly constrained by goals of the study and project resources. Sampling 4-6 trees at a site while avoiding juvenile effects in rings near the pith seems to be the most commonly used methodology, and although there are some reasoned arguments for analyzing only latewood and developing separate isotope records from each tree, the existence of some contradictory findings together with efforts to reduce cost and effort have prompted alternate strategies (e.g., most years pooled with occasional analysis of rings in the sequence separately for each tree) that have produced useful results in many studies.