Temperature sensitivity of blue intensity, maximum latewood density, and ring width data of living black spruce trees in the eastern Canadian taiga

The blue intensity (BI) technique provides opportunities to obtain surrogates to tree-ring density for reconstructing summer temperatures in high-latitude regions. In this study, we compare latewood BI (LBI) and delta BI (DBI), with the conventional X-ray maximum latewood density (MXD) and tree-ring width (TRW) data using 178 living trees of black spruce (Picea mariana (Mill.) B.S.P.), one of the most dominant species of conifers in the Northern Hemisphere, from 17 sites across the eastern Canadian taiga. The regional LBI and DBI chronologies are highly correlated to that of MXD (Pearson's r = 0.97 and 0.92, respectively), while DBI is also similar to TRW (Pearson's r = 0.67). Both LBI and DBI exhibit stronger responses to the May-August temperatures than TRW over larger time and spatial scales. However, only DBI is comparable to MXD data from inter-annual to decadal timescales. Low-frequency components of LBI data are likely distorted by color biases even if no obvious discoloration is present, as well as by the potentially low measurement resolution, leading to an overall weaker temperature sensitivity compared to the MXD data. Resampling experiments suggest that a minimum replication of 10 trees is needed to retain >= 90 % of the optimal temperature signal for MXD, LBI, and DBI data, and a minimum of 20 trees is required for TRW data.