A methodological analysis of canopy disturbance reconstructions using Quercus alba

Abstract

Forest disturbance history reconstructions in the eastern US commonly rely on the analysis of a single tree-ring series per individual. However, this method can result in an underrepresentation of radial growth releases and canopy disturbance events. We analyzed paired tree-ring series from 884 Quercus alba L. individuals to quantify discrepant patterns of intratree release frequency, magnitude, and initiation years. We also developed a model for Q. alba that accounts for this underrepresentation of releases. Of the 884 trees analyzed, 216 exhibited radial growth releases. Only 13 of these 216 trees recorded the same canopy disturbance events in both series. Through analysis of a single growth-ring series per tree, a minimum of 39 and a maximum of 241 releases could be detected from the trees in the data set. Of the total number of release events, 238 (85%) occurred only in one of the paired tree-ring series. For stand development studies requiring the frequency of canopy disturbance alone, a multiplicative factor of 1.72 can provide the information necessary without deviating from the standard practice in the eastern US of collecting and analyzing a single increment core per tree. Studies requiring spatially and temporally explicit information regarding disturbance should extract and analyze two or more tree-ring series per individual.