The invasibility (susceptibility to the establishment of new species) of a community or habitat is influenced by a large number of factors, which operate at different spatial scales. Because most studies focus only on one factor or on one spatial scale, there is an urgent need to combine and integrate the findings of different studies to get a more complete picture of what drives invasibility. In this paper, we compile information on how invasibility studies differ from each other and examine at which spatial scales the most important invasibility-determining factors operate. Based on this information, and in combination with knowledge from biogeographical studies, we suggest a framework for facilitating the integration of results from different types of invasibility research. We discuss several possible applications of the framework and how it could be used in each of these cases. Further, we point out some limitations of the current framework and indicate other aspects of plant invasions, in addition to invasibility, that should be considered in order to improve our general understanding of biological invasions.
Our framework could be used to predict the invasibility of an environment towards invasion by a particular species, to improve our general understanding of mechanisms that determine invasibility, to identify research gaps, and to view results in a broader context. To this end, invasibility-determining factors should be classified according to the spatial scale at which they operate. To combine them, we suggest a hierarchical system in which factors operating at a larger spatial scale (e.g. climate, topography, land use) are dominant to those operating at smaller spatial scales (e.g. biotic interactions and micro-climate). This implies that in order to identify areas at risk for invasion by a particular species, first factors operating at a larger spatial scale should be considered and only if conditions at higher levels are satisfied, one should consider driving factors that are important at a lower spatial scale. However, the accuracy and spatial scale of prediction depend on the aim of the study, which in turn determines which data should be collected. Although it has limitations, we believe that this framework is a useful tool for improving our understanding of invasibility, for identifying areas at risk for invasions and for identifying research gaps.