Adaptive optimisation of the management of Korean pine plantation

Abstract

Forest management planning faces uncertainties regarding future timber prices, tree growth, and survival. Future seed production is an additional source of uncertainty in Korean pine stands managed for the joint production of timber and edible seeds. Modern forest planning uses optimisation to determine the best possible cutting schedule. Optimisation can accommodate uncertainty by using decision rules for adaptive forest management instead of optimising cutting years and intensities. In this study, we optimised two adaptive decision rules for managing Korean pine plantations for the joint production of timber and pinecones when timber prices, tree growth, and seed production are stochastic. The first rule indicated the minimum price to sell timber, i.e., the reservation price, as a function of the mean tree diameter and stand basal area. The second adaptive rule expressed the mean tree diameter at which cutting is optimal as a function of timber price and stand basal area. Both decision rules resulted in nearly the same mean net present value when the optimised rule was applied to 100 stochastic scenarios for future timber prices, tree growth, and seed production. The net present values were over 20% higher than those for the deterministically optimised cutting schedules under the same scenarios. Therefore, the expected economic gain from switching from deterministic to adaptive stochastic optimisation was at least 20%. The cutting years of the adaptive optima were frequently later than those indicated by the deterministic optima, and optimal adaptive harvesting often involved waiting for high timber prices. The minimum price or minimum mean diameter to sell timber was higher when the income from seeds was considered in the optimisation. The cuttings were later, and the rotations were longer in the joint production of timber and pinecones than in timber production alone.