A dynamic programming algorithm for optimization of uneven-aged forest stands

Abstract

A deterministic dynamic programming formulation of the transition uneven-aged stand management problem is presented. Using a previously published northern hardwoods growth model, a forward recursive, discrete, two-state problem that maximizes the net present value of harvested trees at each stage is developed. State variables represent the total number of trees and the total basal area per acre. A neighborhood storage concept previously published is used to reduce the number of states considered at each stage. Two harvest allocation rules are used to assign the harvested basal area to individual diameter classes. Terminal end point conditions and stage to stage sustainability are not required. Results from four base runs of the model are presented and compared with previously published results. Each run produces significantly different optimal paths, with one showing a higher net present value than any previously published. Sensitivity runs illustrate the impact of changes in interest rates, width of neighborhood storage class, and initial conditions. Dynamic programming offers promise for analyzing uneven-aged stand management problems.