Taking the long view on the ecological effects of plant invasions.

Abstract

Nonnative plant introductions, and our awareness of them, have risen considerably in recent decades. While most introduced species are not problematic, some escape and cause significant ecological and economic damage, including reduced biodiversity and altered fire regimes, nutrient cycling, carbon storage, and hydrology. Hundreds of studies demonstrate immediate effects of such plant invasions (for review, see Pyšek et al., 2012), but less is known about how ecological effects of invasions change over time. Because of the limited resources of landowners and management agencies, species must be prioritized for monitoring and treatment based on the magnitude and duration of invasion effects. If the effects of invasive plants decrease or increase over time, management efforts may need to be reprioritized to target the most problematic species. Long-term studies of the ecological interactions between native and nonnative species are particularly crucial for developing such management strategies. Here we call for concentrated efforts to quantify the ecological effects of plant invasions over time and the mechanisms that underlie shifting dynamics and impacts. Multiple community and ecosystem factors can determine whether the effects of invasive plants will increase, decrease, or be maintained over decades (Strayer et al., 2006). For example, although invasive species may initially benefit by escaping natural enemies from their home range, enemies may accumulate and reduce invasive plant populations and their effects on native species (Flory and Clay, 2013). Herbivores or pathogens could make an ecological jump from native species or undergo evolutionary change and affect invasive species. Enemies also could be introduced from the home range of the species accidentally or, in the case of biological control, intentionally to suppress invasive populations. Above- and belowground pathogens and negative plant–soil feedbacks, in which species alter soil in a way that eventually depresses their own population growth, may accumulate and reduce invader populations (Nijjer et al., 2007). Conversely, accumulating soil-borne pathogens could spill back to native species and exacerbate the effects of invasions (Eppinga et al., 2006). Establishment of beneficial mutualisms (Mitchell et al., 2006) and positive plant–soil feedbacks can enhance invasions and potentially promote persistent impacts (Suding et al., 2013), but Yelenik and D'Antonio (2013) suggested that positive plant–soil feedbacks might decline over time, thereby allowing colonization by native or other nonnative species. In addition, evolution of competing native species could also influence the outcomes of invasions. For example, Rowe and Leger (2011) showed that certain native grass populations have evolved greater competitive ability against the invasive grass Bromus tectorum in long-invaded sites, thus lessening invader impacts. A better understanding of evolutionary responses of native species to invasions could guide seed collection choices for restoration in the face of intractable invasive species. At the ecosystem level, positive responses of invaders to natural (e.g., fires, flooding) or anthropogenic (e.g., timber harvests, overgrazing) disturbances can maintain invasive plant populations and enhance colonization of new areas. For example, fire promotes invasions of introduced grasses and fire-invader interactions suppress native species and encourage a grass-fire cycle in many ecosystems (Mack and D'Antonio, 1998). In some settings without further fire, invader populations may decline as native species slowly return. In other systems, native species may benefit from natural disturbances and stressors (e.g., drought) if they are better adapted than the introduced species, as suggested by a meta-analysis of plant species responses to rainfall (Sorte et al., 2013). Restoration of natural disturbance regimes may reduce invader dominance and allow native species recovery, as suggested for Tamarix invasion in the western United States (Stromberg et al., 2007). The effects of invasions also may decline over time if ecological succession results in native species that alter abiotic conditions (e.g., light availability) that favor native species over invasive species. Many plant invasions are well known to promote and be sustained by repeated disturbance (Mack and D'Antonio, 1998), and invaded ecosystems that are altered beyond a particular threshold may not recover without extensive human intervention (Suding et al., 2004). To determine how invader impacts might change over the long term, we need more comprehensive and mechanistic evaluations of invasion dynamics and impacts. For example, the concept that herbivores and pathogens will eventually catch up to a broadly distributed species that grows at high densities is predicted by evolutionary theory, and expected in crop monocultures, but has received relatively little attention from invasion ecologists (Hawkes, 2007). Key questions that remain include: What traits of invasive species promote accumulation of enemies? Will enemies arise through evolutionary or ecological mechanisms? What is the time frame for such processes? Separately, disturbances are generally expected to promote plant invasions, and some invaders promote disturbances from which they benefit (Mack and D'Antonio, 1998). Yet long-term data sets that link invasive plant persistence and impacts to disturbance patterns and mechanistically predict when disturbance–invasion feedbacks will persist or subside are lacking. Additional long-term studies to evaluate how the distribution and abundance of invasive plants and their effects on native species may be altered by global climate and environmental change are needed (Bradley et al., 2010). Taking the long view on the effects of plant invasions will require carefully designed experiments, evaluation of long-term data, and patient and persistent research efforts. We encourage researchers to collect and report information on the age, density, and extent of invasive plant populations, which can be gathered from surveys, herbarium records, aerial photos, or land managers, and to deposit data in open source online repositories (e.g., datadryad.org) for later meta-analyses. Analyzing historical records, surveying chronosequences of invasive plant populations, or repeatedly surveying invasive populations over time can generate valuable information on longitudinal changes in invasive plant effects. However, studies on experimental invasions or removal in controlled settings conducted over years can be useful for revealing patterns not evident in observational studies and can improve interpretations of apparent invasion impacts. The cost of conducting experiments over many years can be high, but collaborations with land managers and agencies and building experiments into already planned restoration efforts can reduce costs. Most long-term evaluations of invader effects will be restricted to case studies of one or a few co-occurring species, but collectively they can inform predictive models of invader persistence and changing impacts and thus bring insight to management efforts. Extended time periods are necessary to determine whether invasions will decline or persist but reporting on the shorter-term effects of invasions can be important for informing immediate management priorities. With information on the long-term effects of particular invaders available in online repositories, land managers can target species with persistent impacts, while species that are predicted to decline can be designated as lower priority. Knowledge of the mechanisms underlying the effects of plant invasions, including traits of species and ecosystems susceptible to prolonged effects, will aid in predicting future problematic invasions and prioritizing management efforts.