Towards a framework for understanding the context dependence of impacts of non-native tree species

Biological invasions are a major driver of ecosystem change but causes of variation in their environmental impacts over space and time remain poorly understood. Most approaches used to quantify the impacts of non-native species assume there are interactions among per capita (i.e. individual level) effects, species abundance and the area occupied by the species. However, studies rarely evaluate these factors and their interactions and often fail to recognize that the magnitude of impacts can be highly context dependent. Understanding what drives the context dependence of non-native species impacts can improve our understanding and predictions of ecosystem change and better inform options for mitigation. Conifers, especially pines, are among the most problematic non-native plant species globally. We use Pinaceae to illustrate how context dependence in biodiversity and environmental impacts of non-native plant species can be generated by at least four processes: nonlinear density effects; intraspecific variation in functional traits; shifts in impacts over time; and persistence of impacts as biological or ecosystem legacies following non-native species removal. Using this understanding, we develop a framework to better quantify interactions of impacts along environmental gradients (e.g. soil fertility, climate, ecosystem age). We demonstrate how impacts of non-native species can occur at both low and high density, and that failing to account for intraspecific variation in effect traits can lead to significant errors in the prediction of impacts. By incorporating context dependence in regard to density and functional traits, we can measure how the interaction of this context dependence will shift along environmental gradients. Moreover, disentangling the roles of species and abundance along such gradients will provide new insights into the net effects of both the native and non-native components of communities. We use a working example of our framework that incorporates all four processes to demonstrate how to measure fire risk impacts of Pinus contorta. We show that ecosystem impacts of non-native tree species are not fixed but rather vary predictably along major environmental gradients. Moreover, removal of non-native species through management provides an important tool for revealing biological and ecosystem legacy effects. Although we focus here on relatively well-documented Pinaceae, the new insights into context dependence of impacts can be widely applied across species, environments and regions. A free Plain Language Summary can be found within the Supporting Information of this article.