Extended Data Fig. 2: Landscapes of relationships between the day of year on which 25% of annual growth is achieved (DOY25) and temperature in prior weeks for ring- and diffuse-porous trees at the Smithsonian Conservation Biology Institute (SCBI) and Harvard Forest.
From: Warm springs alter timing but not total growth of temperate deciduous trees
Shown are matrices of \(\beta \) coefficients from first-order linear regressions between mean maximum temperature (Tmax) and DOY25. Window Open and Window Close indicate number of weeks prior to DOY25 (listed in Extended Data Table 2). Yellow shading indicates neutral relationships, while orange or red shading indicates that DOY25 advances with increased Tmax over the given time window (negative \(\beta \)). Black circles indicate the critical temperature window selected based on minimization of ∆AICc, the difference in Akaike Information Criterion corrected for small sample size relative to a null model. Critical temperature windows are listed in Extended Data Table 2.
