Building Pareto Frontiers under tree-level forest planning using airborne laser scanning, growth models and spatial optimization

This article evaluates the conflicts between spatial, production and financial goals using tree-level decision-support built from stand dynamics equations, airborne laser scanning data and optimization based on mixed integer programming. The resulting Pareto Frontiers evaluated the role of spatial goals to create compact treatment areas along the 10-year forest plan composed of two periods. The weight of spatial goals was progressively increased to quantify the trade-offs towards (1) financial performance, (2) timber production, (3) clustering of harvesting decisions and (4) the Hart-Becking stand density index computed at tree-level using regions derived from ALS. The proposed framework was illustrated in pine forest located in Central Spain using tree-level inventory. The proposed tree selection method was feasible and provided results fast to support operational decision-making. Proved feasibility of the tree selection method optimization allowed the fast PF computation. The results showed the benefit of increasing the weight of spatial goals up to 20–40% to promote the clustering of tree harvests. The observed reduction of financial revenues from increasing priority to spatial clustering paid off, financially and operationally, considering the possible implementation of solutions, which were very dispersed when fully maximizing revenues or benefit. The assimilation of the Pareto Frontier supported with robust optimization contributes to improve forest management planning efficiency. This article turns tree-level decision making into contemporary by integrating multi-temporal decisions, using multi-objective scenarios to assess preferences, and making use of ALS technology as the vector to transform forest data into management decisions.