Precision and accuracy of tree-ring-based death dates of mountain pines in the Swiss National Park

Key message

Mountain pines in the Swiss National Park show evidence of partial cambial mortality, which affects the precision of tree-ring-based death dates, followed by lagged crown mortality.

Abstract

The time of tree death is commonly reconstructed by dating the outermost ring of tree-ring series. However, due to the occurrence of partial cambial mortality, the date of the outermost tree ring may vary between different locations on the tree stem. Furthermore, a tree may continue to live following the formation of the most recent tree ring. In this study, we quantified precision and accuracy of tree-ring-based death dates from 229 dead mountain pines (Pinus montana) from a 28 km² study area in the Swiss National Park. For almost two-thirds of the trees, a maximum difference of just 0–4 years between the dates of cambial mortality from three increment cores was observed, however, for a few trees the difference reached 30–65 years. Higher maximum differences between the dates of cambial mortality are expected for trees on steep slopes, for old trees or for trees that died a long time ago. For 84 % of dead mountain pines, which were sampled in a permanent sample plot with 2-year remeasurement intervals, the difference between the date of observed crown mortality and the death date determined from three cores was 0–5 years. Sampling two or just one core per tree decreases the accuracy of tree-ring-based death dates. Based on the findings of our study, we recommend a prior assessment of the precision and accuracy of tree-ring-based death dates for any dendroecological study dealing with the reconstruction of tree mortality.