Response of branch growth and mortality to silvicultural treatments in coastal Douglas-fir plantations: Implications for predicting tree growth
Static models of individual tree crown attributes such as height to crown base and maximum branch diameter profile have been developed for several commercially important species. Dynamic models of individual branch growth and mortality have received less attention, but have generally been developed retrospectively by dissecting felled trees; however, this approach is limited by the lack of historic stand data and the difficulty in determining the exact timing of branch death. This study monitored the development of individual branches on 103 stems located on a variety of silvicultural trials in the Pacific Northwest, USA. The results indicated that branch growth and mortality were significantly influenced by precommercial thinning (PCT), commercial thinning, fertilization, vegetation management, and a foliar disease known as Swiss needle cast [caused by Phaeocryptopus gaeumannii (T. Rohde) Petr.]. Models developed across these datasets accounted for treatment effects through variables such as tree basal area growth and the size of the crown. Insertion of the branch growth and mortality equations into an individual-tree modeling framework, significantly improved short-term predictions of crown recession on an independent series of silvicultural trials, which increased mean accuracy of diameter growth prediction (reduction in mean bias). However, the static height to crown base equation resulted in a lower mean square error for the tree diameter and height growth predictions. Overall, individual branches were found to be highly responsive to changes in stand conditions imposed by silvicultural treatments, and therefore represent an important mechanism explaining tree and stand growth responses.
Keywords: Douglas-fir; Intensive management; Crown dynamics; Branch modeling; Fertilization; Thinning; Swiss needle cast; Vegetation management; Branch mortality; Branch radial growth; Growth and yield modeling