A forward modeling approach to paleoclimatic interpretation of tree-ring data

We investigate the interpretation of tree-ring data using the Vaganov-Shashkin forward model of tree-ring formation. This model is derived from principles of conifer wood growth, and explicitly incorporates a nonlinear daily timescale model of the multivariate environmental controls on tree-ring growth. The model results are shown to be robust with respect to primary moisture and temperature parameter choices. When applied to the simulation of tree-ring widths from North America and Russia from the Mann et al. (1998) and Vaganov et al. (2006) data sets, the forward model produces skill on annual and decadal timescales which is about the same as that achieved using classical dendrochronological statistical modeling techniques. The forward model achieves this without site-by-site tuning as is performed in statistical modeling. The results support the interpretation of this broad-scale network of tree-ring width chronologies primarily as climate proxies for use in statistical paleoclimatic field reconstructions, and point to further applications in climate science.