Modeling jack pine branch characteristics in Eastern Canada

A total of 83 trees were sampled in three regions of Eastern Canada in order to model branch characteristics (number of branches per annual shoot, branch insertion angle and diameter) using linear mixed-effects models. Differences in branch characteristics according to branch type (pseudo-whorl at annual shoot apical end (PWA) versus pseudo-whorl between shoot apical ends (PWB)) were also studied. The number of both PWA and PWB branches are proportional to annual shoot length, whereas the number of PWB branches also decreases with tree age. Insertion angle was mainly driven by annual shoot number from apex (branch age). The diameter models showed the most complexity with branch vertical position and tree size (DBH and total height) among the statistically significant variables. Region and plot random effects were minimal compared with tree and annual shoot levels. Tree-level random effects were significant for every model and might be a symptom of genetic control over the number of branches and, to a small extent, branch diameter. Interaction between insertion angle and diameter is relatively strong because all the models using them as independent variables (except for the model of insertion angle for PWB branches) showed better fit statistics. These results lead us to believe that tree-, annual shoot- and branch-level variables should be further explored in order to better understand branch dynamics.

► We model branch characteristics (number of branches per annual shoot, branch insertion angle and diameter).
► Branch type (nodal versus internodal) has an influence on every characteristic.
► The number of branches per annual shoot is proportional to annual shoot length.
► Interaction between insertion angle and diameter is relatively strong.