Tree-ring width and basal area increment
In late August 2018, two 5 mm diameter increment cores per tree were collected at breast height for each studied species using an increment borer, and tree diameter at breast height (DBH) of each sample tree was measured. A total of 272 tree cores were sampled from 136 trees, with one or two cores per tree and the number of cores for each species ranging from 36 to 40. Tree-ring width data were obtained according to the standard procedures described by Cook and Kairiukstis (1990). Samples were air-dried, glued and mounted on wooden staves, and then polished with 400 and 600 mesh sandpaper in sequence until the tree rings were clearly visible. Tree-ring measurements were crossdated by standard dendrochronological techniques (Schweingruber, 1988) and the quality of crossdated results was validated using the COFECHA program (Holmes, 1983). To remove the inherent effect on annual increments caused by tree aging and potential disturbance signals due to forest stand development, the crossdated tree-ring width measurements were detrended using a cubic smoothing spline with 50% frequency-response cutoff equal to two-thirds of the length of each series, which was performed using the ARSTAN program (Cook, 1985). All detrended series data for each species were separately averaged to obtain tree-ring width standard chronology using the bi-weight robust mean method (Cook & Kairiukstis, 1990).
The common statistical of tree-ring width standard chronologies including signal-to-noise ratio (SNR), expressed population signal (EPS) and all series correlation are shown in Table 1. The reliable periods of the chronologies were determined by the criterion of EPS surpassing 0.85 which were used for the subsequent climate-growth correlation analyses (Wigle, Briffa, & Jones, 1984). Mean sensitivity was calculated as a measure of growth sensitivity to climate based on the method by Speer (2010). The higher mean sensitivity value reflects greater variability in the chronologies and indicates a stronger response to inter-annual climate change (Fritts, 1976; Speer, 2010).
In order to measure the radial growth rate, the cumulative basal areas (CBA) at the same cambial age (Tognetti, Cherubini, & Innes, 2000) for each tree radius were calculated as follows:
\(\text{CBA}_{i}=\pi\times r_{i}^{2}\) (1)
where ri is the cumulative tree-ring width from the first year to the i th year. Since all species showed a relatively stable growth rate (Fig. 2), we calculated the CBA at 15-year-old (CBA15) of each species to compare radial growth rate at the same cambial age among the studied species.