Previously in my career, I worked on forest growth simulations where we looked at dozens of different harvesting strategies, distributed over the timber stands

Previously in my career, I worked on forest growth simulations where we looked at dozens of different harvesting strategies, distributed over the timber stands of a property over the course of 100+ years. 20 harvest strategies x 1000s of stands x 100 annual plans is too big to brute force (the underlying tree growth model was not fast) but we used simulated annealing to optimize for multiple objectives - wildlife habitat, water quality, harvest cost, and timber volume. We found a very interesting conclusion: there appears to be a combination of strategies that improves biological outcomes AND increases timber harvest volumes over the long term. The debate seems to always center on wilderness (no management) vs industrial timber farming (monoculture). But our models clearly showed a third path - it required a spatially-explicit, long-term plan with lots of labor-intensive interventions but it was absolutely possible to manage a forest for multiple purposes - timber, wildlife and human recreation - and do so without compromising any of the objectives. That these practices closely mirror the forest management of native people is not at all surprising.