The relative importance of environmental drivers and their interactions on the growth of Norway spruce depends on soil unit classes: A case study from Saxony and Thuringia, Germany

Abstract

In forest management and science it is important to determine the drivers of tree growth and to quantify their relative importance with regard to forest site characteristics. The growth of individual trees depends on complex interactions of biotic and environmental drivers. Controlling the influence of biotic drivers (e.g. interspecific competition and bark beetles) is the main concern of forest management. However, large uncertainties emerge from environmental drivers and their impacts on tree growth. The aim of this study is to quantify the relative importance of environmental drivers (climate, soil, and terrain attributes) on the growth of Norway spruce trees (Picea abies (L.) Karst.). For that purpose, the relative basal area increment of individual spruce trees was modelled with a Boosted Regression Tree (BRT) approach. The approach is particularly suitable, since BRT quantify the relative predictor importance, taking nonlinearities and predictor variable interactions into consideration. We assume distinct differences in the growth responses to environmental drivers on three main soil unit classes (cambisol, podzol and waterlogged soils) in Saxony and Thuringia, Germany. The results of this study clearly demonstrate the importance of soil properties (available water capacity and sand content of the soil) on the growth of Norway spruce trees. Terrain attributes and water availability are crucial for Norway spruce growth on cambisol, podzol and waterlogged soils. Moreover, interactions among environmental drivers are more relevant on sites with cambisol as compared to podzol or waterlogged soils. Considering interactions between environmental drivers in the model led to significant differences in the identification of important environmental drivers. This observation was consistent among soil unit classes, especially for environmental drivers associated with water availability. Thus, the implementation of the results in growth models of high spatial resolution will support decision making in forest management, e.g. through identifying proper regions for spruce development and risk control.