Re-evaluating Divergence in Western Canada
The classic definition of the so-called “divergence problem” is a decoupling of temperature sensitive tree-ring chronologies from the instrumental record – expressed either as a loss or weakening in the inter-annual signal, or a divergence in trend in the recent period. Western Canada is one such area where divergence is ostensibly prevalent. However, recent observations using ring-width (RW) and latewood blue intensity (LWB) parameters challenge our views on divergence in this region: tree-ring parameter/climate response-change over time is much more complex than the traditional definition of this phenomenon suggests.
Tree-ring analyses using RW data from central (YUKcen) and south western Yukon (YUKsouth), and southern British Columbia (SBC) show that temporal instability in response to summer temperatures was found at lower frequencies, but it was not present at very high frequencies. However, the phenomenon is variable between regions. For unfiltered RW chronologies (detrended using age dependent splines), correlations with maximum temperatures are 0.22 (YUKcen), 0.35 (YUKsouth) and 0.35 (SBC). After first differencing, these correlations increase to 0.57, 0.67 and 0.55, respectively. For these regions using LWB chronologies, the unfiltered (1st differenced) correlations are 0.62 (0.70), 0.60 (0.70) and 0.62 (0.66). In our presentation, we will report time-varying climate response analyses which will detail that the problem with RW is not the inter-annual signal, but rather the low frequency signal, while LWB shows greater climate fidelity at all timescales. Ongoing research will explore these large scale coherent multidecadal and longer timescale trend differences in RW.