A mathematical model for the optimization of forest planning decisions: evaluation of the trade-offs between timber production and carbon sequestration

Resumen

A forest planning decision support system assists managers by providing a sequence of activities that each forest land area will receive through the whole planning horizon, as well as the schedule concerning all these activities are required to be performed. Each activity is a composition of silviculture treatments, and different combinations of activities are possible, which in turn depends on the type of trees. Also, the yield and operational costs are determined by the tree species, age, previous received treatments, among other features. In order to tackle the complexity of the optimal forest planning problem, a general mathematical framework based on a Generalized Disjunctive Programming (GDP) approach is proposed. The approach defines (i) the optimal combination of silvicultural treatments (a forest management plan), (ii) the proportion of land area to be harvested, and (iii) the volume of timber products produced at each harvest node. The maximization of the net present value and the minimization of carbon sequestration lost are used as a multi-objective function. The ϵ-constraint methods are used to evaluate the trade-offs between aforementioned objectives functions, which can be used with the aim of supporting forest planning management decisions. Data from a forestry industry in the northern province of Misiones (Argentina) are used to test the applicability of the proposed model.