A Bayesian approach for exact optimal measurement and cutting times for a Eucalyptus production forest

This paper addresses the design for supporting the optimal decision on tree cutting in a Portuguese eucalyptus production forest. Trees are usually cut at the biological rotation age, i.e. the age which maximizes the yearly volume production. Here we aim the maximization of the long-term yearly volume yield reduced by harvest costs. We consider different growth curves, with a known prior distribution, that can occur in each rotation. The optimal cutting time at each rotation depends both on the current growth curve and on the prior distribution. Different priors and strategies are compared with respect to the long-term production. Optimization of the cutting time allows an improvement of 16% of the long-term volume production. We conclude that the use of optimal designs can be beneficial for tree cutting in modern production forests.