The community ecology of invasive species: where are we and what's next?

Alien species are continually introduced in most regions of the world, but not all survive and coexist with the resident native species. Approaches analyzing the functional (or phylogenetic) similarity between invasive species and native communities are increasingly employed to infer the processes underlying successful invasions and to predict future invaders. The relatively simple conceptual foundations have made these approaches very appealing and therefore widely used, often leading to confusion and hampering generalizations. We undertook a comprehensive review and synthesis of the functional similarity approach in invasion community ecology to clarify its advantages and limitations, to summarize what has been learned thus far, and to suggest avenues for future improvement. We first present the methodological state of the art and provide general guidelines. Second, by organizing the published literature around seven key questions in invasion ecology we found cumulative evidence that: at large spatial scales phylogenetic relatedness of invasive and native species is a good predictor of invasion success, but a poor predictor of invasion impacts; at fine spatial scales, community resistance to invasion tends to increase with native species diversity and with similarity to the invaders, consistent with patterns emerging from biotic interactions. In general, the processes filtering invaders appear to vary across species' invasion stage and along environmental gradients. Nonetheless, we found conflicting evidence for differences in community assembly processes between invasive and native species, and between the invader's native and adventive ranges. Finally, we propose four important avenues for overcoming some of the identified methodological and conceptual limitations of the approach, in particular by integrating observational and experimental analyses and by explicitly considering species' demographic rates and complex biotic interactions. Although the functional similarity approach bears intrinsic limitations, it still offers many opportunities in invasion community ecology at large spatial scales.