Management legacies and forest structure shape the root rot risk of coniferous forests in Finland

Forest management and land use can strongly alter the forest ecosystem with long-lasting legacy effects by shaping e.g. species composition and age structure in stand and landscape scales. These changes may lead to changes in the local disturbance regime. In addition, forest management may directly affect the dynamics of disturbance agents. Heterobasidion root rot is one of the most important diseases of conifers in the Northern hemisphere. The epidemiology of the fungus relies heavily on the availability of fresh wood material which the spores need to cause an infection. In managed forests, fresh stump surfaces provide massive amounts of perfect growth media.  Once the fungus has infected a stand, the disease remains in there over tree generations slowly deteriorating the timber quality, killing trees and predisposing trees to subsequent disturbance agents such as wind. Thus, host availability in landscape scale, stand conditions and the management history all are assumed to play an important role in the epidemiology. The aim of this study was to analyze the drivers of current distribution of Heterobasidion root rot in Finland in order to understand the disease dynamics better and to manage the disease in the future. Specifically, we asked how important the legacies of different past management and land-use methods are. The National Forest Inventory (NFI) in Finland has recorded root rot observations since 1995 covering in total over 348 000 sample plots over four inventories. We combined that database with 20 different explanatory variables with a hypothetical relation to the biology and epidemiology of the fungus. The variables were categorized to three categories; (i) management legacies, (ii) landscape structure, and (iii) site conditions. Management legacies included for example the historical locations of sawmills and the share of forest pastures. Landscape structure combined structural characteristics, such as Norway spruce and old forest (120+ years) shares from different time periods. Site conditions were described with e.g. temperature sum and Shannon index for tree species richness. By using Boosted Regression Tree and Generalized Liner Models, we found that variables from all the three categories contributed to the presence of Heterobasidion root rot. The distance from an NFI plot to the nearest sawmill operating in 1910 (historical intensity of logging) and the distance to a waterway (timber rafting as main transportation methods) were shown to be one of the most important variables together with temperature sum and current Norway spruce share in landscape scale. This indicates that the management legacies, especially the past management intensity, has a significant effect on the epidemiology of Heterobasidion root rot.