Silviculture for Forest Diversity: Mixed Species Management to Meet Multiple Objectives

Abstract

Mixed species forest management is valued for the potential to increase productivity while generating other important co-benefits that may reduce risks associated with monocultures, such as varied species responses to drought, improved insect and disease resistance, enhanced wildlife habitat, and additional insurance against uncertain ecological change. However, diverse forest systems add a layer of complexity to management, including challenges with seedling regeneration and predicting varied forest development. This dissertation addresses three components of diverse forest management and research: (1) comparison of growth of five Pacific Northwest tree species in monoculture and mixed species forests, (2) browse deterrent strategies for regeneration of high-value timber species such as western redcedar (Thuja plicata Donn ex D. Don) and Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco), and (3) quantifying understory biomass response to forest management prescriptions using remote sensing methods. This work takes advantage of two long-term permanent plot datasets (20-year and 50-year) to explore mixed species and monoculture competition, six field trials to compare browse deterrent methods, and two remote sensing datasets to assess predictive model performance of understory biomass. Generalized linear mixed effect models were used to evaluate growth performance for species in mixtures versus monocultures and to investigate browse deterrent strategies alongside χ2 Tests of Independence; five different machine learning models allowed for comparison of remote sensing methods. Intraspecific competition is stronger than interspecific competition in mixed, mesic forested stands in the Pacific Northwest. The following pairings showed positive, small growth effects for both species under certain species basal area proportions: (1) Douglas-fir and bigleaf maple (Acer macrophyllum Pursh); (2) Douglas-fir and western redcedar; and (3) red alder (Alnus rubra Bong.) and western hemlock (Tsuga heterophylla (Raf.) Sarg.). Douglas-fir and western hemlock growth responses were also positive under some conditions but inconsistent across two datasets. Interspecific interactions were largely explained by competition or complementarity for limiting resources; positive growth pairings include shade-intolerant species paired with more shade-tolerant species. Browse deterrence for western redcedar showed effectiveness, ranked in terms of impacts on growth and tree form, from highest to lowest as fenced exclosures, Vexar® tubing, co-planting, Plantskydd and flexible fencing. For Douglas-fir, co-planting with Sitka spruce was more effective than Vexar tubing; fenced exclosures produced varying results, with the presence of invasive species and robust shrub cover associated with reduced Douglas-fir growth. Plantskydd® showed early promising results for western redcedar browse prevention, but challenges with application timing and need for consistent, sustained reapplication reduce utility. Reported costs for deterrent methods indicate large fenced exclosures are the most expensive method to implement, followed by chemical repellent, Vexar tubing, and Sitka spruce co-planting in that order. Lastly, low understory biomass values in an open forest system can be predicted using a stochastic gradient boosting model built with unmanned aerial vehicle (UAV) acquired LiDAR and multispectral data. Data acquisition and post-processing were more important than model selection in optimizing model performance and predictions. This study provides a framework for evaluating understory biomass efficiently in open canopy conditions, using limited ground measurements via calibrated photos and UAV remotely sensed data. Together, these results allow for streamlined decision-making for forest managers and researchers to identify complementary species pairings, increase regeneration success of high-value timber species, and produce wall-to-wall remotely sensed metrics for novel forest prescriptions with limited in-field data. In an era of ecological and societal change, alternative management options mitigate these risks, and this work equips managers with additional tools when monoculture forestry no longer meets objectives.