Tree-Level Harvest Optimization for Structure-Based Forest Management Based on the Species Mingling Index

This novel research investigated the use of a heuristic process to inform tree-level harvest decisions guided by the need to maximize the interspersion of tree species across a forest. In the heuristic process, a species mingling value for each tree was computed using both (1) neighbors that were simply of a different species than the reference tree and (2) neighbors that were uniquely different species from both the reference tree and other neighbors of the reference tree. The tree-level species mingling value was averaged for the stand, which was then subject to a maximization process. Constraints included residual tree density levels and minimum tree volume harvest levels. In two case studies, results suggest that the species mingling index at the stand level can be significantly increased over randomly allocated harvest decisions using the heuristic process described. In the case studies, we illustrate how this type of process can inform management decisions by suggesting the distance between residual trees of similar species given the initial stand structure and the objectives and constraints. The work represents a unique tree-level optimization approach that one day may be of value as new technologies are developed to map the location of individual trees in a timely and efficient manner.