Assessing Simple Versus Complex Restoration Strategies for Industrially Disturbed Forests

Assessing the community‐level consequences of ecological restoration treatments is essential to guide future restoration efforts. We compared the vegetation composition and species richness of restored sites that received a range of restoration treatments and those of unrestored sites that experienced varying levels of disturbance. Our study was conducted in the industrially degraded landscape surrounding Sudbury, Ontario, Canada. The Great Lakes–St. Lawrence Forest once present in this area was degraded through logging, mining, and smelting activities beginning in the late 1800s until restoration of the most visibly degraded areas began in 1974. Restoration treatments ranged from simple abiotic enhancements to complex, multistage revegetation treatments using native and non‐native species, which included fertilizing, spreading of ground dolomitic limestone, understory seeding, and tree planting. Canonical correspondence analysis was used to determine which restoration treatments explained differences in the community structure among sites. We found that native understory vascular species richness was similar in restored sites that received more complex restoration treatments and unrestored sites that were mildly disturbed; however, the role of planted trees and non‐native species in the restored communities remains unclear. Understory vascular seeding played a key role in determining community composition of vascular understory and overstory communities, but the time since restoration commenced was a more important factor for nonvascular communities because they received no direct biotic enhancements. The use of non‐native species in the vascular seed mix seems to be slowly encouraging the colonization of native species, but non‐natives continue to dominate restored sites 25 years after restoration began.     

Management intensity affects the relationship between non‐native and native species in subtropical wetlands

Question: Does management intensity affect the association between non‐native and native species and between non‐native species and soil nutrients in wetlands? Location: MacArthur Agro‐Ecology Research Center, Florida, USA. Methods: We evaluated native and non‐native plant richness and relative frequency in 15 1‐m² plots in 40 wetlands across two types of pastures, highly managed (fertilized, ditched, planted, heavily grazed by cattle) and semi‐natural (unfertilized, lightly seasonally grazed). Plant biomass was collected in five 0.25‐m² plots per wetland and sorted to species. Soil cores were collected to analyse soil total nitrogen (N) and phosphorus (P). An information‐theoretic approach was used to compare mixed effects models considering the association of non‐native richness, relative frequency, and biomass with native richness, relative frequency, biomass, C₃ grass relative frequency (a dominant native group), N, P and wetland‐type. Results: Non‐native richness was negatively correlated with native richness in semi‐natural wetlands, but there was no evidence of an association between these variables in highly managed wetlands. Non‐native richness increased with increasing soil N in semi‐natural wetlands, but not in the highly managed wetlands. Soil P was positively related to non‐native frequency in semi‐natural wetlands but negatively related in highly managed wetlands. Non‐native frequency and biomass were negatively related to relative frequency of C₃ grasses in both management types. Conclusions: Our results indicate that management intensity influences relationships between native and non‐native richness. Management intensity interacts with abiotic or biotic factors, such as soil nutrients and composition, in predicting where non‐native species will most likely need control.     

Opposite relationships between invasibility and native species richness at patch versus landscape scales

Using a nested plot design in oak forests in Minnesota, USA we measured the percent cover of young individuals of an exotic invading shrub, Rhamnus cathartica (common buckthorn), to assess the relationships at two scales between invasibility, propagule and light availability, and richness and cover of native species. Comparing patches (1 m²) within 17 Quercus -dominated stands (each 1 ha, within a 100 km by 150 km area), cover of young R. cathartica was negatively related to both species richness and cover of native species. In 1 m² patches, native cover was positively associated with native richness and thus cover-related competition was a likely mechanism by which richness influenced R. cathartica . At the landscape scale (comparing the aggregate stand-scale metrics among the 17 stands), native cover and richness were still positively related, but had opposite relationships with R. cathartica cover. R. cathartica cover was positively related to species richness and negatively related to native species cover. The observed switch at different scales from a positive to a negative relationship between R. cathartica cover and native richness supported the hypothesized scale dependence of these relations. Propagule pressure, which we estimated by measuring the size of nearby mature R. cathartica shrubs, had a large positive effect on R. cathartica seedling cover at the landscape scale. These results suggest that landscape patterns of invasion may be best understood in light of the combination of many factors including native diversity, native cover, and propagule pressure.     

Non‐native plants have greater impacts because of differing per‐capita effects and nonlinear abundance–impact curves

Invasive, non‐native species can have tremendous impacts on biotic communities, where they reduce the abundance and diversity of local species. However, it remains unclear whether impacts of non‐native species arise from their high abundance or whether each non‐native individual has a disproportionate impact – that is, a higher per‐capita effect – on co‐occurring species compared to impacts by native species. Using a long‐term study of wetlands, we asked how temporal variation in dominant native and non‐native plants impacted the abundance and richness of other plants in the recipient community. Non‐native plants reached higher abundances than natives and had greater per‐capita effects. The abundance–impact relationship between plant abundance and richness was nonlinear. Compared with increasing native abundance, increasing non‐native abundance was associated with steeper declines in richness because of greater per‐capita effects and nonlinearities in the abundance–impact relationship. Our study supports eco‐evolutionary novelty of non‐natives as a driver of their outsized impacts on communities.     

Invasion in space and time: non-native species richness and relative abundance respond to interannual variation in productivity and diversity

Ecologists have long sought to understand the relationships among species diversity, community productivity and invasion by non‐native species. Here, four long‐term observational datasets were analyzed using repeated measures statistics to determine how plant species richness and community resource capture (i.e. productivity) influenced invasion. Multiple factors influenced the results, including the metric used to quantify invasion, interannual variation and spatial scale. Native richness was positively correlated with non‐native richness, but was usually negatively correlated with non‐native abundance, and these patterns were stronger at the larger spatial scale. Logistic regressions indicated that the probability of invasion was reduced both within and following years with high productivity, except at the desert grassland site where high productivity was associated with increased invasion. Our analysis suggests that while non‐natives were most likely to establish in species rich communities, their success was diminished by high resource capture by the resident community.     

Incomplete recovery of plant diversity in restored prairie wetlands on agricultural landscapes

Restoration efforts are being implemented globally to mitigate the degradation and loss of wetland habitat; however, the rate and success of wetland vegetation recovery post‐restoration is highly variable across wetland classes and geographies. Here, we measured the recovery of plant diversity along a chronosequence of restored temporary and seasonal prairie wetlands ranging from 0 to 23 years since restoration, including drained and natural wetlands embedded in agricultural and natural reserve landscapes in central Alberta, Canada. We assessed plant diversity using the following structural indicators: percent cover of hydrophytes, native and non‐native species, species richness, and community composition. Our findings indicate that plant diversity recovered to resemble reference wetlands in agricultural landscapes within 3–5 years of restoration; however, restored wetlands maintained significantly lower species richness and a distinct community composition compared to reference wetlands located within natural reserves. Early establishment of non‐native species during recovery, dispersal limitation, and depauperated native seed bank were probable barriers to complete recovery. Determining the success of vegetation recovery provides important knowledge that can be used to improve restoration strategies, especially considering projected future changes in land use and climate.     

Landscape characteristics of non‐native pine plantations and invasions in Southern Chile

The spread of non‐native conifers into areas naturally dominated by other vegetation types is a growing problem in South America. This process results in a landscape transformation as the conifers suppress native vegetation leading to reduced biodiversity, lower water availability and altered nutrient dynamics. Previous research highlights the broad spatial extents of land cover change in parts of Chile. However, in Southern Chile, the extent of plantations and the landscape characteristics associated with plantations and ongoing pine invasions are poorly understood. Here, we characterised non‐native pine land cover within one Landsat scene (World Reference System 2 Path 232/Row 92; ~34 000 km²) in Southern Chile. We created training data based on historical high‐resolution imagery, derived land cover predictors from time series of Landsat observations and used a Random Forest classifier to map the distribution of non‐native pines. The overall classification accuracy was 88%, and the accuracy of the non‐native pine class exceeded 90%. Although 71% of non‐native pine patches were within 500 m of other non‐native pine patches, isolated non‐native pine patches were found to occur up to 55 km from the nearest neighbour. These distant plantations could exacerbate invasion risk by creating propagule sources for novel invasion fronts. In relation to landscape characteristics, non‐native pines were found to be more likely to occur in low slope and mid‐elevation areas. Because most of the study area is native forest, most non‐native pine patches border native forest. However, non‐native pine patches were almost three times more likely than random patches to border grass/agriculture. This suggests that grasslands and disturbed sites, which have low resistance to non‐native pine invasion, are disproportionately exposed to pine propagules. Our results indicate that non‐native pine plantations are extensive across Southern Chile, and well poised to cause future invasion.     

Restoring a Mediterranean‐climate shrub community with perennial species reduces future invasion

Successful restoration of an invaded landscape to a diverse, invasion‐resistant native plant community requires determining the optimal native species mix to add to the landscape. We manipulated native seed mix (annuals, perennials, or a combination of the two), while controlling the growth of non‐native species to test the hypothesis that altering native species composition can influence native establishment and subsequent non‐native invasion. Initial survival of native annuals and perennials was higher when seeded in separate mixes than when combined, and competition between the native perennials and annuals led to lower perennial cover in year 2 of mixed‐seeded plots. The plots with the highest perennial cover had the highest resistance to invasion by Brassica nigra. To clarify interactions among different functional groups of natives and B. nigra, we measured competitive interactions in pots. We grew one native annual, one native perennial, and B. nigra alone or with different competitors and measured biomass after 12 weeks. Brassica nigra was the strongest competitor, limiting the growth of all native species, and was not impacted by competition with native annuals or perennial seedlings. Results from the potted plant experiment demonstrated the strong negative influence of B. nigra on native seedlings. Older native perennials were the strongest competitors against invasive species in the field, yet perennial seedling survival was limited by competition with native annuals and B. nigra. Management action that maximizes perennial growth in early years may lead to a relatively more successful restoration and the establishment of an invasion‐resistant community.     

Restoring prairie pothole wetlands: does the species pool concept offer decision‐making guidance for re‐vegetation?

Question: Do regional species pools, landscape isolation or on‐site constraints cause plants from different guilds to vary in their ability to colonize restored wetlands? Location: Iowa, Minnesota, and South Dakota, USA. Methods: Floristic surveys of 41 restored wetlands were made three and 12 years after reflooding to determine changes in local species pools for eight plant guilds. The effect of landscape isolation on colonization efficiency was evaluated for each guild by plotting local species pools against distance to nearby natural wetlands, and the relative importance of dispersal vs. on‐site constraints in limiting colonization was explored by comparing the local species pools of restored and natural wetlands within the region. Results: Of the 517 wetland plant taxa occurring in the region, 50% have established within 12 years. The proportion of the regional species pool represented in local species pools differed among guilds, with sedge‐meadow perennials, emergent perennials and floating/submersed aquatics least represented (33‐36%) and annual guilds most represented (74‐94%). Colonization‐to‐extinction ratios suggest that floating/submersed aquatics have already reached a species equilibrium while sedge‐meadow and emergent perennials are still accumulating species. Increasing distance to nearest wetlands decreased the proportion of the regional species pool present in local pools for all guilds except native annuals and woody plants. The maximum proportion predicted, assuming no distance constraint, was comparable to the lowest‐diversity natural wetlands for most perennial guilds, and also lower than what was achieved in a planted, weeded restoration. Conclusions: A biotic constraints seem to limit the colonization of floating/submersed aquatics into natural or restored wetlands, whereas all other guilds are potentially constrained by dispersal or biotic factors (i.e. competition from invasive species). Using species pools to evaluate restoration progress revealed that immigration potential varies considerably among guilds, that local species richness does not necessarily correspond to immigration limitations, and that some guilds (e.g. sedge‐meadow perennials) will likely benefit more than others from being planted at restoration sites.     

Embracing variability: environmental dependence and plant community context in ecological restoration

In heterogeneous landscapes, one can expect a complexity of ecological restoration outcomes. The effectiveness of management often depends on environmental conditions (environmental context) and how management indirectly affects other components of the system (community context). Although managers appreciate this context dependency, it is difficult to translate it to decision‐making in restoration. We demonstrate one approach to improve this translation. We surveyed plant, soil, and landscape characteristics at 131 grassland and coastal sage sites that received herbicide treatments to remove non‐native plant species and/or propagule addition to increase native species. We used path analysis to describe how each management approach influenced target non‐native species and how interactions with environmental conditions and indirect effects of management influenced plant community composition. This approach enabled us to analyze a complex system with differing management histories to identify both direct and indirect effects of management. Management had the intended direct effects: the application of herbicide and propagule addition directly reduced non‐native species and increased native species, respectively. We found little evidence of environmental dependency: effects occurred largely independent of environmental conditions. However, management outcomes did depend on plant community context. Specifically, although herbicide reduced the cover of target, non‐native plant species, this reduction resulted in only slight increases in native species and instead led indirectly to increases in non‐targeted, non‐native species. We suggest that quantitative evaluation of variability in restoration outcomes allows management to be more adaptive and increase decision‐making efficacy in complex managed landscapes.     

Factors Affecting Revegetation of Oil Field Access Roads in Semiarid Grassland

We assessed vegetation recovery on access roads removed after well abandonment in an active oil‐producing region of northern Great Plains grasslands. We compared extant vegetation on 58 roads, restored 3–22 years previously, to records of species seeded on each and to adjacent, undisturbed prairie, to evaluate main differences between the restored and adjacent community and to explore patterns in the restored plant community over time. The restored plant community was dominated by low richness of seeded non‐native and native grasses and forbs, whereas adjacent prairie had numerous, abundant native graminoids and shrubs and higher richness of native forbs. Cover of seeded species on roads was double that of colonizing species. Disparity in cover of dominant native grasses between the adjacent community and relatively narrow restored roadway suggests that conditions for germination and survival in roadbeds are poor. This is at least partly due to persistence of seeded species. Differences in restored plant composition over time were best explained by changes in species seeded, from non‐natives to natives, and secondarily by successional shifts from ruderal to perennial non‐seeded species. Of the 30 species seeded at least once on these roads, only 10 were commonly used. The long‐term influence of seeding choices in grassland road restorations implies that improvements in these practices will be critical to reversing ecological impacts of roads.     

The distribution and habitat associations of non‐native plant species in urban riparian habitats

Questions: 1. What are the distribution and habitat associations of non‐native (neophyte) species in riparian zones? 2. Are there significant differences, in terms of plant species diversity, composition, habitat condition and species attributes, between plant communities where non‐natives are present or abundant and those where non‐natives are absent or infrequent? 3. Are the observed differences generic to non‐natives or do individual non‐native species differ in their vegetation associations? Location: West Midlands Conurbation (WMC), UK. Methods: 56 sites were located randomly on four rivers across the WMC. Ten 2 m × 2 m quadrats were placed within 15 m of the river to sample vegetation within the floodplain at each site. All vascular plants were recorded along with site information such as surrounding land use and habitat types. Results: Non‐native species were found in many vegetation types and on all rivers in the WMC. There were higher numbers of non‐natives on more degraded, human‐modified rivers. More non‐native species were found in woodland, scrub and tall herb habitats than in grasslands. We distinguish two types of communities with non‐natives. In communities colonized following disturbance, in comparison to quadrats containing no non‐native species, those with non‐natives had higher species diversity and more forbs, annuals and shortlived monocarpic perennials. Native species in quadrats containing non‐natives were characteristic of conditions of higher fertility and pH, had a larger specific leaf area and were less stress tolerant or competitive. In later successional communities dominated by particular non‐natives, native diversity declined with increasing cover of non‐natives. Associated native species were characteristic of low light conditions. Conclusions: Communities containing non‐natives can be associated with particular types of native species. Extrinsic factors (disturbance, eutrophication) affected both native and non‐native species. In disturbed riparian habitats the key determinant of diversity is dominance by competitive invasive species regardless of their native or non‐native origin.     

Diversity–invasibility relationships across multiple scales in disturbed forest understoreys

Non-native plant species richness may be either negatively or positively correlated with native species due to differences in resource availability, propagule pressure or the scale of vegetation sampling. We investigated the relationships between these factors and both native and non-native plant species at 12 mainland and island forested sites in southeastern Ontario, Canada. In general, the presence of non-native species was limited: <20% of all species at a site were non-native and non-native species cover was <4% m⁻² at 11 of the 12 sites. Non-native species were always positively correlated with native species, regardless of spatial scale and whether islands were sampled. Additionally, islands had a greater abundance of non-native species. Non-native species richness across mainland sites was significantly negatively correlated with mean shape index, a measure of the ratio of forest edge to area, and positively correlated with the mean distance to the nearest forest patch. Other factors associated with disturbance and propagule pressure in northeastern North America forests, including human land use, white-tailed deer populations, understorey light, and soil nitrogen, did not explain non-native richness nor cover better than the null models. Our results suggest that management strategies for controlling non-native plant invasions should aim to reduce the propagule pressure associated with human activities, and maximize the connectivity of forest habitats to benefit more poorly dispersed native species.     

Invasive species cover,soil type,and grazing interact to predict long‐term grassland restoration success

Grasslands are undergoing tremendous degradation as a result of climate change, land use, and invasion by non‐native plants. However, understanding of the factors responsible for driving reestablishment of grassland plant communities is largely derived from short‐term studies. In order to develop an understanding of the factors responsible for longer term restoration outcomes in California annual grasslands, we surveyed 12 fields in Davis, CA, U.S.A., in 2015 that were seeded with native species mixtures starting in 2004. Using field surveys, we investigated how invasive plant richness and cover, native plant richness and cover, aboveground biomass, grazing, soil type, and restoration species identity might provide utility for explaining patterns of restoration success. We found a negative relationship between invasive cover and restoration cover, which was attributed to the slow establishment of seeded species and subsequent dominance by weeds. The relationship between invasive cover and restoration cover was modified by grazing, likely due to a change in the dominance of exotic forbs, which have a more similar growing season to restoration species, and therefore compete more strongly for late season moisture. Finally, we found that soil type was responsible for differences in the identity and abundance of invasive plants, subsequently affecting restoration cover. This work highlights the value of focusing resources on reducing invasive species cover, limiting grazing to periods of adequate moisture, and considering soil type for successful long‐term restoration in California annual grasslands. Moreover, observations of long‐term restoration outcomes can provide insight into the way mechanisms driving restoration outcomes might differ through time.     

Using multi-scale species distribution data to infer drivers of biological invasion in riparian wetlands

Aim Biological invasion is a major conservation problem that is of interest to ecological science. Understanding mechanisms of invasion is a high priority, heightened by the management imperative of acting quickly after species introduction. While information about invading species’ ecology is often unavailable, species distribution data can be collected near the onset of invasion. By examining distribution patterns of exotic and native plant species at multiple spatial scales, we aim to identify the scale (of those studied) that accounts for most variability in exotic species abundance, and infer likely drivers of invasion. Location River Murray wetlands, south‐eastern Australia. Methods A nested, crossed survey design was used to determine the extent of variation in wetland plant abundance, grazing intensity and water depth at four spatial scales (reaches, wetland clumps, wetlands, wetland sections), and among three Depth‐strata. We examined responses of exotic and native species groups (grouped into terrestrial and amphibious taxa), native weeds and 10 individual species using hierarchical ANOVA. Results As a group dominated by terrestrial taxa, exotic species cover varied at reach‐, wetland‐ and section‐scales. This likely reflects differences in abiotic characteristics and propagule pressure at these scales. Groups based on native species did not vary at any scale examined. Cover of 10 species mostly varied among and within wetlands (patterns unrelated to species’ origin or functional group), but species’ responses differed, despite individual plants being similar in size. While flora mostly varied among wetlands, exotic cover varied most among reaches (26%), which was attributed to hydrological modification and human activities. Main conclusions Multi‐scale surveys can rapidly identify factors likely to affect species’ distributions and can indicate where future research should be directed. By highlighting disproportionate variation in exotic cover among reaches, this study suggests that flow regulation and human‐mediated dispersal facilitate exotic plant invasion in River Murray wetlands.     

Soil pH influences patterns of plant community composition after restoration with native‐based seed mixes

Reclamation of highly disturbed lands typically includes establishing fast‐growing, non‐native plants to achieve rapid ground cover for erosion control. Establishing native plant communities could achieve ecosystem functions beyond soil erosion, such as providing wildlife habitat. Pipelines, or other disturbed corridors through a landscape, present unique challenges for establishing native plant communities given the heterogeneity of soil environments and invasive plant propagule pressure. We created two structural equation models to address multiple related hypotheses about the influence of soil pH on plant community composition (current diversity and vegetative cover of the original restoration seed mix and background flora, and invasive plant density during mix establishment and current density) of a highly disturbed landscape corridor restored with native species. To test our hypotheses we conducted a plant survey on a gas pipeline crossing two state forests in the north‐central Appalachians that had been seeded with a native‐based mixture 8 years prior. Low soil pH was a strong predictor of density of the invasive annual plant, Microstegium vimineum, and had resulted in lower species diversity and cover of the seeded mix. Overall, our data provide evidence that native‐based grass and forb mixtures can establish and persist on a wide range of soil environments and thrive in competition with invasive plants in moderately acidic to neutral soils. Advancing knowledge on restoration methods using native species is essential to improving restoration practice norms to incorporate multifunctional ecological goals.     

Phenology in a warming world: differences between native and non‐native plant species

Phenology is a harbinger of climate change, with many species advancing flowering in response to rising temperatures. However, there is tremendous variation among species in phenological response to warming, and any phenological differences between native and non‐native species may influence invasion outcomes under global warming. We simulated global warming in the field and found that non‐native species flowered earlier and were more phenologically plastic to temperature than natives, which did not accelerate flowering in response to warming. Non‐native species' flowering also became more synchronous with other community members under warming. Earlier flowering was associated with greater geographic spread of non‐native species, implicating phenology as a potential trait associated with the successful establishment of non‐native species across large geographic regions. Such phenological differences in both timing and plasticity between native and non‐natives are hypothesised to promote invasion success and population persistence, potentially benefiting non‐native over native species under climate change.     

Early trajectories of forest understory development on reclamation sites: influence of forest floor placement and a cover crop

We tested whether direct placement of forest floor material (FFM: litter, fibric, humus layers and surface mineral horizons) and sowing of a cover crop (Melilotus officinalis) could facilitate the establishment of native forest understory species at a reclaimed coal mine in Alberta, Canada. FFM was salvaged at two depths (15 and 40 cm) from a recently harvested native aspen forest and immediately placed at the same depths on the reclamation site. Total richness (approximately 61 species in 96 subplots) was similar in each of 3 years post‐placement; total richness for all 3 years combined was 87 including 34 typical boreal forest understory species plus 30 other natives. The deeper treatment reduced cover of all species, native and non‐native species in year 1. In year 3, the deeper treatment still had lower cover of non‐native species but had higher cover of forest understory species in years 2 and 3. The deeper treatment also resulted in lower species richness per plot, but only in year 1. In year 2 (when the biennial clover was at its tall stage), the cover crop treatment was associated with lower cover of non‐native species but did not affect the cover of native forest understory species. Direct placement of FFM can help facilitate establishment of a diverse native boreal forest understory in a reclaimed landscape. Although richness and cover may be initially higher with shallower salvage and placement, deeper salvage may ultimately be better for encouraging establishment of native forest understory species.     

Does Seeding a Locally Adapted Native Mixture Inhibit Ingress by Exotic Plants?

Non‐native plant species often colonize retired agricultural lands, creating monocultures with low species diversity that provide poor wildlife habitat. We assessed whether sowing a mix of 29 locally adapted native species reduced invasion of non‐native plant species compared to allowing vegetation to colonize naturally following tillage. There was a sampling date × treatment interaction for canopy cover of perennial exotic plant species. Plots that were not sown to natives had two to six times greater canopy cover of exotic species than did plots with both preparation (woody vegetation removed, plowed, and disked) and control (no preparation or sowing) plots. Canopy cover of exotic plants was similar in prepared‐only and control treatments from October 2008 to June 2010, ranging from 8 to 40%. Percent absolute canopy cover of native vegetation was 10–20 times greater on prepared and planted plots than on prepared‐only plots during March 2009 to June 2010. Sowing a mix of locally adapted native species may inhibit encroachment by non‐native species for up to two years after sowing on retired agricultural land in the Lower Rio Grande Valley of Texas.     

Hotspots of plant invasion predicted by propagule pressure and ecosystem characteristics

Aim Biological invasions pose a major conservation threat and are occurring at an unprecedented rate. Disproportionate levels of invasion across the landscape indicate that propagule pressure and ecosystem characteristics can mediate invasion success. However, most invasion predictions relate to species’ characteristics (invasiveness) and habitat requirements. Given myriad invaders and the inability to generalize from single‐species studies, more general predictions about invasion are required. We present a simple new method for characterizing and predicting landscape susceptibility to invasion that is not species‐specific. Location Corangamite catchment (13,340 km²), south‐east Australia. Methods Using spatially referenced data on the locations of non‐native plant species, we modelled their expected proportional cover as a function of a site’s environmental conditions and geographic location. Models were built as boosted regression trees (BRTs). Results On average, the BRTs explained 38% of variation in occupancy and abundance of all exotic species and exotic forbs. Variables indicating propagule pressure, human impacts, abiotic and community characteristics were rated as the top four most influential variables in each model. Presumably reflecting higher propagule pressure and resource availability, invasion was highest near edges of vegetation fragments and areas of human activity. Sites with high vegetation cover had higher probability of occupancy but lower proportional cover of invaders, the latter trend suggesting a form of biotic resistance. Invasion patterns varied little in time despite the data spanning 34 years. Main conclusions To our knowledge, this is the first multispecies model based on occupancy and abundance data used to predict invasion risk at the landscape scale. Our approach is flexible and can be applied in different biomes, at multiple scales and for different taxonomic groups. Quantifying general patterns and processes of plant invasion will increase understanding of invasion and community ecology. Predicting invasion risk enables spatial prioritization of weed surveillance and control.