Early summer hydroclimatic signals are captured well by tree-ring earlywood width in the eastern Qinling Mountains, central China

In the humid and semi-humid regions of China, tree-ring-width (TRW) chronologies offer limited moisture-related climatic information. To gather additional climatic information, it would be interesting to explore the potential of the intra-annul tree-ring-width indices (i.e., the earlywood width, EWW, and latewood width, LWW). To achieve this purpose, TRW, EWW, and LWW were measured from the tree-ring samples of Pinus tabuliformis originating from the semi-humid eastern Qinling Mountains, central China. Standard (STD) and signal-free (SSF) chronologies of all parameters were created using these detrending methods including (1) negative exponential functions combined with linear regression with negative (or zero) slope (NELR), (2) cubic smoothing splines with a 50% frequency cutoff at 67% of the series length (SP67), and (3) age-dependent splines with an initial stiffness of 50 years (SPA50). The results showed that EWW chronologies were significantly negatively correlated with temperature but positively correlated with precipitation and soil moisture conditions during the current early-growing season. By contrast, LWW and TRW chronologies had weaker relationships with these climatic factors. The strongest climatic signal was detected for the EWW STD chronology detrended with the NELR method, explaining 50% of the variance in the May-July self-calibrated Palmer Drought Severity Index (MJJ scPDSI) during the instrumental period 1953-2005. Based on this relationship, the MJJ scPDSI was reconstructed back to 1868 using a linear regression function. The reconstruction was validated by comparison with other hydroclimatic reconstructions and historical document records from adjacent regions. Our results highlight the potential of intra-annual tree-ring indices for reconstructing seasonal hydroclimatic variations in humid and semi-humid regions of China. Furthermore, our reconstruction exhibits a strong in-phase relationship with a newly proposed East Asian summer monsoon index (EASMI) before the 1940s on the decadal and longer timescales, which may be due to the positive response of the local precipitation to EASMI. Nonetheless, the cause for the weakened relationship after the 1940s is complex, and cannot be solely attributed to the changing impacts of precipitation and temperature.