Search for: All records

Delta blue intensity is a commonly used method to correct for the heartwood-sapwood color change in blue intensity (BI) measurements. It is based on the assumption that the heartwood-sapwood color change is similar in both earlywood and latewood. This assumption has not been supported physiologically. Furthermore, delta BI may confound the climate signals in earlywood and latewood BI as it is technically a linear combination of the other two. Here, instead of using delta BI, we used change point detection to identify the heartwood-sapwood transition, and corrected for the color change by rescaling the mean and variance of BI measurements after the transition to those immediately before. We tested three different change point detection methods and found that they agreed well with one another. Importantly, our approach preserves the climate signals in both earlywood and latewood BI data, while delta BI causes a total loss of climate signals in our test case. Therefore, we suggest that change point detection should be used instead of delta BI to account for the heartwood-sapwood color change. 

Abstract. We evaluate a range of blue intensity (BI) tree-ringparameters in eight conifer species (12 sites) from Tasmania and New Zealandfor their dendroclimatic potential, and as surrogate wood anatomicalproxies. Using a dataset of ca. 10–15 trees per site, we measured earlywoodmaximum blue intensity (EWB), latewood minimum blue intensity (LWB), and theassociated delta blue intensity (DB) parameter for dendrochronologicalanalysis. No resin extraction was performed, impacting low-frequency trends.Therefore, we focused only on the high-frequency signal by detrending alltree-ring and climate data using a 20-year cubic smoothing spline. All BIparameters express low relative variance and weak signal strength comparedto ring width. Correlation analysis and principal component regressionexperiments identified a weak and variable climate response for mostring-width chronologies. However, for most sites, the EWB data, despite weaksignal strength, expressed strong coherence with summer temperatures.Significant correlations for LWB were also noted, but the sign of therelationship for most species is opposite to that reported for all coniferspecies in the Northern Hemisphere. DB results were mixed but performedbetter for the Tasmanian sites when combined through principal componentregression methods than for New Zealand. Using the fullmulti-species/parameter network, excellent summer temperature calibrationwas identified for both Tasmania and New Zealand ranging from 52 % to78 % explained variance for split periods (1901–1950/1951–1995), withequally robust independent validation (coefficient of efficiency = 0.41 to0.77). Comparison of the Tasmanian BI reconstruction with a quantitativewood anatomical (QWA) reconstruction shows that these parameters recordessentially the same strong high-frequency summer temperature signal.Despite these excellent results, a substantial challenge exists with thecapture of potential secular-scale climate trends. Although DB, band-pass,and other signal processing methods may help with this issue, substantiallymore experimentation is needed in conjunction with comparative analysis withring density and QWA measurements.