When are species invasions useful for addressing fundamental questions in plant biology?

Species invasions have been championed as an ideal system for understanding key processes in ecology and evolution (Sax et al., 2002), including the impacts of climate change (Moran and Alexander, 2014). In exploring this idea, we have noticed a striking dichotomy: whether invasions present a useful model system depends on the degree to which the question is intrinsically about a system “at equilibrium” or a system changing directionally across time or space (Fig. 1). In the case of dynamic questions, such as those about rates of adaptation or range expansion, species invasions are ideal because they mimic the type of “natural experiment” necessary to test these questions. On the other hand, for questions where a stable equilibrium is implied, such as the number of species or traits an ecosystem can support, or the proximate causes of species range limits, the transient nature of species invasions can potentially lead to erroneous conclusions. Here we argue that as a field, we should recognize that invasions make powerful study systems for answering certain types of questions, but are potentially misleading for others. Invasions may be well suited as model systems for some questions but not others. In (A), the transient nature of invasion makes it a powerful test system for processes such as the influence of new abiotic or biotic conditions on the rate of adaptation through, for example, increased aboveground biomass. However, caution must be applied in (B), where the number of new species a community can support may initially increase, and then decrease, over a longer period of time. Similarly, in (C), since many invasives are still expanding in the introduced range, inferring the causes of range limits from species invasions must be done with caution (solid line indicates eventual species range limit in the introduced range; dashed lines indicate apparent range limits at earlier times). In most natural systems, change happens at temporal and spatial scales that are difficult or impossible to study. For several questions at the core of ecology and evolution, such as the rate and mode by which species adapt to new environments, invasive species have proved to be invaluable models because invasives, having left their home environment and coevolved ecological interactions behind, are expected to be under strong selection to adapt to new conditions. For example, Dlugosch et al. (2015) provided compelling evidence that since its arrival in North America, invasive yellow star thistle has rapidly evolved increased competitiveness through larger size and increased reproduction. This rapid evolution has allowed it to displace (nearly) all competitively similar natives at disturbed sites, even though it tends to be competitively excluded in intact grasslands. More generally, studies examining evolution in invasive species have contributed some of the central evidence that evolution can operate on ecological time scales (Carroll et al., 2007). Studying invasives can also help us to predict how fast species can spread across a landscape. One active area of research is determining how quickly native species can expand their ranges in response to climate change and which native species are most at risk for extinction through a failure to adapt or expand their range (Moran and Alexander, 2014). Since invasive species spreading in their new range will likely encounter novel biotic interactions, they also provide a model for how range expansion of native species will be impacted by changes in biotic interactions. For example, the enemy release hypothesis, originally conceived to explain how a loss of natural enemies could lead to some species becoming invasive upon introduction, may equally apply to native species, as they may also leave natural enemies behind as their range expands (reviewed by Stewart et al., 2015). The use of invasive species as models to explore how and whether species adapt to changing conditions is gaining interest, yet many questions remain unaddressed. Future directions include determining the factors that influence evolutionary change during spread, in particular those factors that influence spread itself. For example, adaptive evolution has been shown to contribute to the increased speed at which cane toads have expanded in their invasive range (Phillips et al., 2006). It is important that we understand how unique invasives are with respect to their evolutionary trajectories before using them to make predictions about natives. Invasives will be a poor model for many native species if they, for example, are more likely to exhibit genetic admixture from a range of source populations, or, conversely, are more likely to have experienced a bottleneck (reviewed by Moran and Alexander, 2014). Overall, when considering study systems to address questions about change or flux, species invasions provide opportunities to tackle questions that are otherwise difficult to address. While caution must be taken in making direct inferences from invasive to native species, we are optimistic that studying invasions in these contexts will have benefits beyond the field of invasion biology. How many species can an ecosystem support? What sets species’ range limits? Questions such as these imply “equilibrium” conditions (even if dynamic), and here we argue that the inherently recent nature of species invasions, in either time or space (which we treat separately below), can create problems. The question of how many species a region can support has been proposed as one that can be addressed through species invasions (Sax et al., 2007); it is also relevant to conservation because it addresses whether invasions tend to lead to local extinctions. Often invasions seemingly increase rather than reduce local biodiversity (Stohlgren et al., 1999). However, trajectories of biodiversity change may be misleading because they tend to be based on short-term data with spatial bias (as discussed by Gonzalez et al. (2016), although, see also Vellend et al. (2017)). In fact, current co-occurrence is not necessarily a reliable indicator of long-term coexistence, with many native species harboring “extinction debts”. Thus, the full impact of invasives on native biodiversity may not be seen for hundreds of years (Gilbert and Levine, 2013), and conclusions that native communities are “undersaturated” with species may be misleading until it becomes clearer when and where extinction debts or “colonization credits” are more likely (Jackson and Sax, 2010). Invasive species have quickly assembled latitudinal patterns that coarsely mirror those of native species, for example, increasing range sizes with latitude and strong correlations between latitudinal limits in native and exotic ranges (Sax, 2001; Guo et al., 2012). While invasions are the sort of large-scale biogeographic experiment that might help reveal mechanisms underlying these major biogeographic patterns, the challenge is that exotic ranges often increase with time since introduction (e.g., Dyer et al., 2016), suggesting that many invasive species have not yet reached stable limits in their new ranges. Thus, situations where patterns in native and exotic ranges match could be a transient phenomenon, not indicative of conservatism in the underlying determinants. On the other hand, differences in range limits between a species’ native and exotic ranges can suggest hypotheses for the causes of native range limits. For example, in climatic space, many invasive species show evidence of niche expansion or occupancy of novel climatic conditions in their exotic compared with their native range (Guisan et al., 2014). Subsequent investigation can reveal whether it is limits to adaptation, biotic interactions, or dispersal limitation that have been overcome in the exotic range to facilitate niche expansion. But, critically, these inferences about causes of native range limits are strongest when a species’ exotic range already exceeds its native range in at least some dimensions. Researchers must tread carefully when using invasive species to address questions that imply equilibrium. In particular, care must be used when determining what data are required to properly address the question, in addition to the temporal and/or spatial scale at which the dynamics might be operating. Biases in data collection, collecting data for too short of a time period, or collecting data at too small of a spatial scale could all lead to incorrect/premature conclusions. Invasive species continue to be an enticing study system, both in their own right, and for addressing fundamental questions in ecology and evolution. As our understanding of the factors driving invasions improves, so must our selectivity in terms of their suitability as models for the ecology and evolution of native populations and communities. The authors thank the members of the FLORUM discussion group at the Biodiversity Research Centre, University of British Columbia, for their suggestions for improving earlier versions of this essay. R.D.S., A.L.A. and J.L.W. were each supported by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.