Tree-related microhabitats on live Populus tremula and Picea abies in relation to tree age, diameter, and stand factors in Estonia

Production forestry is known to reduce the naturally variable microhabitat pools in live trees and their biodiversity by tree removal and broad indirect effects of silviculture. However, the tree-scale processes are not known well enough for microhabitat management for the future. This study investigates how tree age affects the microhabitat occurrence in live trees, its effect modifiers, and explanatory value compared with tree diameter. We inventoried tree-related microhabitats on 879 live aspens (Populus tremula) and Norway spruces (Picea abies) of known age in 141 forests representing contrasting productive site conditions in Estonia. We analysed microhabitat incidence using logistic mixed models for significant tree-scale and stand-scale factors. Most microhabitat types appeared rare even in old trees, and the age effects had various patterns. Only 2% of trees bore five or more microhabitat types. Aspens and spruces had a similar microhabitat diversity before 80 years of age. Stand-level effects varied among microhabitat types but were less significant than tree-level effects; interactions were even rarer. Most effects found could be interpreted through known ecological processes; for example, humidity-dependent epiphytic growth; bark stripping by herbivores; pathogen effects in forests with distinct histories. In contrast, a reliable tree-scale prediction of microhabitat occurrence appears rarely possible; and depending on microhabitat type, either tree age or diameter can be a better predictor. We suggest that managing for tree-related microhabitats in production forests should combine facilitating ecological conditions for microhabitat formation, and early detection and retention of the trees with high microhabitat potential.