Regeneration decision and land expectation value: Numerical results of decision model evaluation and optimization

摘要

Decision models for the planting density decision problem for Scots pine (Pinus sylvestris L.) plantations in Northern Sweden were evaluated. The management objective is to maximize the expected land expectation value. The survival rate and growth of planted seedlings in the beginning of a rotation and future timber prices are stochastic variables. Results show that the proper formulation of the decision model is dependent on the purpose of decision analysis. To determine only the optimal planting density, it is appropriate to use the deterministic decision model. Incorporating uncertainty into the decision model leads to a higher, but only slightly better optimal planting density. To find the optimal planting density and the expected land expectation value, random variations of the optimal final-harvesting age due to timber price uncertainty should be recognized in the decision model. Thinning as well as uncertainty in future stand states have little impacts on the optimal decision and, therefore, can be ignored. Such a decision model was used to determine optimal planting densities and harvest decision policies for different site indices. The obtained optimal decisions were then used to estimate the expected land expectation values. Effects of possible systematic prediction errors in the expected future stand states and stumpage prices on the optimal decision and on the expected land expectation value were examined using the deterministic decision model. 19 refs, 3 figs, 14 tabs