Exploring Community Change of North American Stream Fish Communities Due to Species Introductions

The introduction of nonnative species has facilitated global changes in beta diversity, and these changes have been coined biotic homogenization (loss in beta diversity) and differentiation (gain in beta diversity). Homogenization represents a key conservation threat. Therefore it is imperative to understand the processes that allow nonnative species to establish and how these species contribute to beta diversity change. In this dissertation, I assess three key challenges exist in our ability to address these goals. One challenge is that we have inconsistent support among hypotheses of invasion drivers due to context dependency. Context dependency can arise due to differences in methodology and study design (i.e. apparent) or through differing ecological processes (i.e. mechanistic), accounting for both is critical to gain a general understanding of invasion processes. A second challenge is that many beta diversity change studies use taxonomic measures of diversity, which inconsistently relate to ecological processes. Beta diversity can be measured in multiple dimensions that better represent the complexity of community changes on ecosystems. Therefore, a multidimensional approach to beta diversity change is needed to understand ecological consequences of invasions. Finally, nonnative species do not universally cause homogenization to occur. Factors related to nonnative species origins and native community structure can determine trajectory of beta diversity change. Therefore, we need a better understanding of the factors that affect the trajectory of beta diversity change. I addressed these three challenges using fine resolution, continental extent stream fish community data for the United States. Specifically, I (1) demonstrated that both apparent and mechanistic context dependency can confound interpretation of invasion drivers, (2) classified communities into syndromes of multidimensional beta diversity change, demonstrating that taxonomic diversity is not sufficient to understand consequences of nonnative species, and (3) identified multidimensional beta diversity change drivers related to species origin and native community structure. From my findings, I present three recommendations to future invasion studies: make explicit and transparent methodological choices to reduce apparent context dependency, move away from one-dimensional diversity metrics and blanketed definitions of nonnative species, and consider the intersection of regional- and local-scale processes in biological invasions. By considering these recommendations, invasion ecology can benefit from a more general and mechanistic understanding of invasion processes.