Invasive plants directly and indirectly disrupt the ecosystem functioning, of which indirect effects, for example, through trophic cascades, are particularly difficult to predict. It is frequently assumed that the impact of an invading species on the ecosystem is proportional (linearly related) to its density or abundance in a habitat, but this assumption has rarely been tested. We hypothesised that abundance and richness of plants and potentially pollinators of wet meadows change as a result of invasion of steeplebush Spiraea tomentosa and that these changes are proportional to the density of the shrub. We selected 27 sites amongst wet meadows habitats invaded by S. tomentosa with coverage ranging from 0% to 100% and examined the diversity of plants, as well as the abundance and diversity of flower visitors (bees, butterflies with moths and flies). Our results showed that the richness of plants, as well as the richness and number of individuals of flower visitors, decrease significantly and linearly with an increase of the S. tomentosa cover. This finding supports the hypothesis that the impact of an invasive species can be proportional to their population density, especially if this species is limiting the available resources without supplying others. Our study is the first to show such an unequivocal negative, linear effect of an invasive shrub on the abundance and richness of potential pollinators. It proves that the negative impact of S. tomentosa on the wetland ecosystem appears even with a minor coverage of the invader, which should be taken into account when planning activities aimed at controlling the population of this transformer species. The simultaneously detected linear dependence allows us to assume that the benefits of controlling secondary populations of the shrub can be proportional to the incurred effort.

bees, biodiversity, biological conservation, butterflies, flies, flower visitors, invasive plants, non-native species, wetlands

Freshwater wetlands are important refuges for hygrophilous and hygrobiont plants and animals and, as such, they support diverse and unique species assemblages (Burke and Gibbons 1995). Nevertheless, in spite of the ecosystem services they provide, these types of ecosystems have been lost, degraded or strongly modified by human activities worldwide (Moroń et al. 2008; Davidson 2014; Reis et al. 2017). One of the most important reasons for their degradation are biological invasions, to which these ecosystems seem to be especially vulnerable (Zedler and Kercher 2004) due to frequent, large-scale disturbances, influx of nutrients, runoff of propagules flowing down the ground and accumulating in land depressions, as well as their limited connection over the network of corridors (Zedler and Kercher 2004; Fletcher et. al. 2019). Invasive plants disrupt the functioning of wetland ecosystems at different levels of their organisation both directly, for example, through competition for resources or changes in habitat conditions and indirectly, for example, through the trophic levels (Dibble et al. 2013; Mazurczyk and Brooks 2021). By initiating food chains, these plants can affect trophic interactions (Pearson 2009), which may initiate a trophic cascade (López‐Núñez et al. 2017) manifested in population changes of consumers caused by direct and indirect relationships between them.

Amongst animals, insects are a group particularly sensitive to disturbances resulting from plant invasions (Van Hengstum et al. 2014; Schirmel et al. 2016), because they are highly dependent on plants as sources of food (Ehrlich and Raven 1964; Coley et al. 2006), shelter and breeding ground (Spafford et al. 2013). Therefore, the magnitude of response to changes in native flora composition caused by invasive plants varies extremely in different guilds of insects (Moragues and Traveset 2005; Fenesi et al. 2015). One of the crucial functions insects play in the ecosystem is pollination (Klein et al. 2007), which is the key in maintaining plant species diversity (Potts et al. 2010). However, data on the impact of invasive plants on pollinator populations are still scarce, whereas the effect of invasive plants on a pollinator community is equivocal (Bjerknes et al. 2007). Invasive plant species often compete with native ones for ecosystem services, such as pollination (Moragues and Traveset 2005; Larson et al. 2006; Bjerknes et al. 2007; Dietzsch et al. 2011; Williams et al. 2011). There are also known cases of pistil stigmas of native species being blocked by deposited foreign pollen, which may limit the reproductive success of native plants (Brown et al. 2002; Larson et al. 2006). Nevertheless, invasive plants, due to their higher level of generalisation with regard to pollinators when compared to native plants, may also play new, important topological roles. For example, they attract pollinators to visit invaded ecosystems (Albrecht et al. 2014), increase the frequency of visits of pollinators on flowers of plants found in the same habitat (Bartomeus et al. 2008), increase pollinator populations, flight frequency and range (Moragues and Traveset 2005) and fill the seasonal resource gap in the native flora (Tepedino et al. 2008).

The direction of the impact of invasive plants on pollinators is strongly dependent on the scale of the invasion (Powell et al. 2013; Rejmánek and Stohlgren 2015). It is often assumed that the impact of invasive plants is proportional to their population density (Yokomizo et al. 2009; Elgersma and Ehrenfeld 2011; Panetta and Gooden 2017), but, as noted by Elgersma and Ehrenfeld (2011), this assumption has been rarely tested and there is ample evidence that the effects of plant invasions on native pollinators are not linear (Moroń et al. 2019). Thus, the question arises whether the scenario of the non-linear effect of invaders on native plants and pollinators is a more general pattern. Therefore, in this study we assessed the impact of steeplebush on the species diversity of vascular plants, as well as the species number and diversity of pollinators in wet meadows.

The North American steeplebush (Spiraea tomentosa L., Rosaceae) is a highly invasive shrub found in central and northern Europe (Dajdok et al. 2011), with high potential for further spread (Wiatrowska et al. 2020). In the area of its natural occurrence, in the southern United States (USDA – The Plants Database 2022), S. tomentosa grows especially in wetland habitats and enters transitional and raised bogs (Reschke 1990; Schafale and Weakley 1990; Faber-Langendoen 2001). According to LaRosa et al. (2004), in its natural range, the shrub offers pollen and nectar as a reward for pollinators and, depending on the population density, it may have different (opposing) impacts on pollinators, such as bees (bumblebees) included in this study.

In Europe, S. tomentosa has been cultivated as an ornamental plant since the 18^th century (Symes 1983). Currently this shrub is found in nine countries: Sweden, Germany, Poland, Belgium, Denmark, Norway, Croatia, Romania and Great Britain (Dajdok et al. 2011; GBIF 2022), while it is considered as an invasive species in five countries: Belgium (Invasive species of Belgium 2022), Denmark (Bruus et al. 2007), Germany (Balkenhol et al. 2018), Sweden (Tyler et al. 2015) and Poland (Dajdok et al. 2011; Tokarska-Guzik et al. 2012). In the area of its secondary occurrence, this shrub forms monodominant plant stands completely changing the structure and plant species composition of wetland communities – wet grasslands, lowland peat bogs or marsh forests (Dajdok et al. 2011; Wiatrowska and Danielewicz 2016) (Fig. 1).

Figure 1.

Spiraea tomentosa in wet meadows located in the Lower Silesian Forests (photo by B. Wiatrowska).

However, still very little information is available on its effects on pollinators – flower visitors and the plant species composition. These data are urgently needed to communicate management priorities in times when invasions are a threat and challenge in nature conservation (Pyšek et al. 2020; Ricciardi et al. 2021; Moodley et al. 2022).

Spiraea tomentosa interacts with pollinators (LaRosa et al. 2004) and meets all four conditions (flower attraction, taxonomic affinity to native plants, formation of dense populations and generalised pollination strategy) that, according to Bjerknes et al. (2007), make alien plants strong interactors for pollination in the area of their secondary range. The shrub was brought to Europe (Symes 1983; Hardtke and Ihl 2000) because it has stately inflorescences (Dajdok et al. 2011) that can be attractive to pollinators (Chittka and Schürkens 2001; Brown et al. 2002). The plant has a taxonomic affinity to S. salicifolia L. (Mirek et al. 2002) native to Central Europe, as well as many other Spiraea spp. common in cultivation (e.g. S. douglasii Hook. or S. × pseudosalicifolia Silverside), which may facilitate attracting pollinators. Moreover, S. tomentosa develops denser and larger populations (Wiatrowska and Danielewicz 2016) than other species known as particularly attractive to pollinators (Ågren 1996; Mustajarvi et al. 2001; Thompson 2001). Probably, due to the relatively simple structure of flowers (S. tomentosa flowers are bisexual, 0.5–1.5 mm in diameter, with five free sepals, five petals orbiculate, free, usually light pink, five carpels and 15–20 stamens (Flora of North America 2022)), it also has a generalised pollination strategy, which means that it can attract many groups of pollinators.

Based on these premises, our main goal was to assess whether there was a relationship between the abundance of S. tomentosa and the diversity of plants, as well as the abundance and diversity of potential pollinators in wet meadow communities. We considered what influence the invader – non-linear or linear – has on plants and potential pollinators of indigenous communities. In the case of a non-linear relationship, the question arises whether there is a certain minimum point (tipping point), at which a pollinator population changes as the invasive plant becomes dominant.

We tested the hypotheses that: (1) the richness of plants and the abundance and richness of visiting pollinators change as a result of Spiraea tomentosa invasion and (2) changes are proportional (linearly) to the density of the shrub. The implementation of these studies was essential to provide insight into the density-dependent impact of the invasive alien species S. tomentosa on biodiversity, focusing on plants and pollinator species.

During overall 27 216 hours of the pan-trap deployment, we collected 3649 individuals of 191 potential pollinator species or families (in the case of flies). Flies (Diptera) formed the most abundant pollinator group, accounting for 75% of all trapped insects. Butterflies and moths (Lepidoptera) comprised 15% of the collected insects. Bees (Apoidea, Hymenoptera) were only a minor fraction of flower visitors (10% of the specimens). A total of 80 vascular plant species were recorded in the investigated study sites (Table 1).

A negative relationship was found between steeplebush cover and plant species richness (t = −7.15; Fig. 2), with a 60% decrease of plant species richness.

Figure 2.

The relationship between species richness of plants and site cover by Spiraea tomentosa. Points represent each of 27 sites. Point colours correspond to a meadow. The 95% CI are marked with polygons. Jittering was added to aid visualisation.

An increase in S. tomentosa cover correlated also with a decline in the number of bee, butterfly and moth and fly individuals by about 70%, 80% and 45% (Fig. 3A–C), respectively (bees: Z = −4.77; butterflies and moths: Z = −7.54; flies: Z = −2.81). Similarly, bee, butterfly (and moth) and fly species richness decreased across the range of S. tomentosa cover by approx. 70%, 70% and 30% (Fig. 3D–E), respectively (bees: Z = −4.35; butterflies and moths: Z = −6.22; flies: t = −2.46).

Figure 3.

The relationship between abundance of A bees C butterflies and moths and E flies, as well as species richness of B bees D butterflies and moths F flies and site cover by Spiraea tomentosa. Points represent each of 81 surveys. Point colours correspond to a meadow and point shapes correspond to a survey number. Legend as in Fig. 2.

The potential pollinator community in the habitats studied was significantly nested, indicating that species-poor samples (pan traps with a high rank) constituted subsets of species-rich samples (pan traps with a low rank) and that this pattern was not random. The nestedness rank significantly increased in proportion to S. tomentosa cover (t = 6.40; Fig. 4; see Appendix 2).

Figure 4.

Nestedness structure of pollinator communities in the study sites A maximally packed matrix and B single ranks (points) for each pan trap depending on Spiraea tomentosa cover. Legend as in Fig. 2 and Fig. 3.

The direction (negative vs. positive), the shape (linear vs. non-linear) and the strength of the relationship between the abundance of the invasive species and the diversity of native species determine which invaders pose the greatest threat to ecosystems (Bradley et al. 2019). It is often indicated that the impact of some invading species is proportional (linearly related) to its density or abundance in a habitat (Yokomizo et al. 2009; Elgersma and Ehrenfeld 2011; Panetta and Gooden 2017; Sofaer et al. 2018), but there is also much evidence that the impact of other invaders may elicit a non-linear response from native species (Elgersma and Ehrenfeld 2011; Panetta and Gooden 2017; Moroń et al. 2019).

In our research, we found a strong, negative, linear impact of the Spiraea tomentosa cover on vascular plant species richness, so this result positively validates the hypothesis. Our results showed that the diversity of plants decreased due to the increased invasive shrub coverage. These results correspond with a global meta-analysis that assessed the direction, shape and strength of the response of native communities to the increasing abundance of invasive species (Bradley et al. 2019). It was found that most often the impacts of invasive plants on native plant communities (at the same trophic level) are significantly negative and linear. However, the average impact of invasive plants on the diversity of native plants, estimated at 28% (Bradley et al. 2019), is much smaller than that shown for S. tomentosa in our research (60%).

In other studies on the impact of alien species on plant species richness, it was found that, amongst alien species entering wetlands in Central Europe, the invasion of Mimulus guttatus DC and Impatiens glandulifera Royle does not reduce the species richness of native plants. The invasion of Solidago gigantea Aiton and Rudbeckia laciniata L. decreases species richness by about 26% and 30%, while the invasion of Fallopia sachalinensis (F. Schmidt) Ronse Decraene, F. japonica (Houtt.) Ronse Decraene and F. × bohemica (Chrtek & Chrtková) JP Bailey contributes to the reduction of species richness by 86%, 73% and 66%, respectively (Hejda et al. 2009). Amongst the species entering wetlands, the impact of S. tomentosa on plants is, therefore, comparable to the impact of alien Fallopia sp. considered as an example of transformer species (Sukopp and Sukopp 1988; Cronk and Fuller 1995; Sukopp and Starfinger 1995), which changes the nature, condition and form or character of ecosystems (Richardson et al. 2000).

The direction, shape (linear vs. non-linear) and strength of the impact of an invasive plant species on insects, including potential pollinators, is more difficult to predict (Stout and Morales 2009; Lenda et al. 2010). When invasive species are at a lower trophic level, no consistent trend of impacts on native species or communities has been demonstrated (Bradley et al. 2019). This is due to the properties of environments that have different buffering potentials, for example, due to the condition of the recipient’s native ecosystem (e.g. Mason and French (2007); Pyšek et al. (2012)), different redundancy of the food web (Gilbert and Levine 2013), as well as properties of the invasive species themselves. The role of the invasive species impact on pollinator assemblages is ambivalent – some of them increase the base for pollinators (Moragues and Traveset 2005; Bartomeus et al. 2008; Hejda et al. 2009; Stout and Morales 2009), while others may limit it (Vanbergen et al. 2018). Reducing plant diversity is considered to be a factor that is directly manifested in reduced food resources for pollinating insects (Potts et al. 2003; Moroń et al. 2008; Senapathi et al. 2015, 2017), but, as noted by Vanbergen et al. (2018), surprisingly little research unequivocally indicates the negative impact of invasive species on the diversity or abundance of pollinators.

In our research, we found a strong, negative, linear influence of S. tomentosa on the abundance and diversity of flower visitors, which allows us to positively verify our hypothesis regarding the negative effect of the shrub cover on potential pollinators.

Butterflies and moths seem to be least resistant to S. tomentosa infestation, as in dense populations of this shrub, the number of individuals decreased by 80% and species richness was reduced by 70%. The strong response of this group of insects is understandable, because Lepidoptera species strongly depend on plants throughout their life cycle – they use them for breeding and as a source of food for larvae and adults (Altermatt and Pearse 2011), resulting in their limited resistance to disturbance. Butterflies and moths have been shown to suffer from invasion, because alien plant species can replace beneficial native partners (Lenda et al. 2013; Trigos‐Peral et al. 2018) and many butterfly species are recognised as food specialists, functionally constrained to exploiting a limited group of plants (Tallamy and Shropshire 2009), for which food source redundancy is relatively low (Moroń et al. 2019). Moreover, invasive plants due to their low nutritional value or toxicity may reduce their reproductive success and consequently lead to genetic bottlenecking (Davis and Cipollini 2014). Some of these plants can also affect the herbivores to become more attractive to predators (Bezemer et al. 2014). Although shoots and leaves of S. tomentosa can be used as a food source by a native moth species – Earias clorana larvae (Wiatrowska et al. 2018b), our results indicate that the buffering mechanism is not observed and butterflies, as well as moths, react quickly and very negatively to S. tomentosa invasion.

Other insects that have strong, often reciprocal, relationships with native species of flowering plants include bees. Pollinator bees are very sensitive to a particular diet source and combination of nutrients (Harmon-Threatt and Kremen 2015), such as the ratio of various essential amino acids (EAA) to carbohydrates and show poor growth and survival when reared on monotype or non-optimal nutrition diets (Vanbergen et al. 2018). It was found that the protein content of pollen varies with the plant species (Roulston et al. 2000), which influences the pollen-feeding behaviour of bees (Cook et al. 2003; Keller et al. 2005). The reduction or elimination of specialised partner plant populations is particularly important for certain solitary bee species that have specialised relationships with flowering plants (Mueller and Kuhlmann 2008). It is also known that even generalist bee species (which often benefit from abundant resources of mass-flowering invasive plants; Stout and Morales (2009)) require diverse sources of floral pollen, because pollen from different plant species varies in digestibility and nutrient content (Roulston et al. 2000). In our research, we found that bees are almost as sensitive to S. tomentosa invasion as butterflies. In dense populations of the shrub, a reduction in the number of bees by 60% was recorded and bee species richness decreased by 70%, which indicates that S. tomentosa, despite the large supply of pollen, which constitutes the majority of bee nutrition (Stout and Morales 2009), seems not to be attractive for this group of insects. It is worth noting that the response of bee populations to the presence of S. tomentosa in the ecosystem seems to be different in the area of the natural occurrence of the shrub and outside it. In North America, S. tomentosa offers both pollen and nectar as pollinator rewards and it is recognised as a species that attracts bumblebees (LaRosa et al. 2004). In Europe, the shrub provides only pollen, which may reduce its attraction to flower visitors looking for nectar. The differences between resources provided by S. tomentosa flowers in its native range and in invaded areas may be a result of the pollinator density differences between these areas (Muñoz and Cavieres 2008) or the lack of adaptation to obtaining flower reward (e.g. Liu and Pemberton 2009). However, the latter is unlikely in the case of S. tomentosa flowers, which are morphologically similar to flowers of other Spiraea species that naturally occur in Central Europe (Mirek et al. 2002).

Another important order amongst insects pollinating flowers around the world are flies from the Syrphidae, Bombyliidae and Muscoidea families (Larson et al. 2001), but also other erroneously overlooked Diptera pollinators (Ssymank et al. 2008; Orford et al. 2015). Amongst all the studied groups of pollinators, flies are the least specialised in flowers, because even though they use pollen and nectar, they also use other food sources (Brock 2015). Additionally, for this group of potential pollinators, a negative effect of S. tomentosa coverage on their number and diversity was revealed, but with the lowest impact. In dense populations of this shrub, a reduction in the number of individuals by 45% and reduction in species richness by 30% were found, which may result from the lowest dependence on a specific food source, but also from their greatest mobility amongst the studied groups of insects (Van Veen 2004). Unlike bees, whose females build nests for their offspring and are associated with them because they procure larval cells in the nest, expand it and sleep in it (Batra 1984), many flies have great power of dispersal spread and they are found far from their site of development (Ssymank 2001; Van Veen 2004).

For all the studied groups of potential pollinators, it was found that the influence of S. tomentosa is proportional to its density coverage. The number of individuals and richness of butterflies and moths, bees and flies significantly, linearly decreased with the increase in the steeplebush cover, which supports the thesis that the impact, at least of some invasive plants, is proportional to invader population density (Yokomizo et al. 2009; Elgersma and Ehrenfeld 2011; Panetta and Gooden 2017). Moreover, the potential pollinator community assemblages seem to be affected as S. tomentosa cover increases (Fig. 4). Changes in the community assemblages are a result of species loss rather than the constitution of new communities by species replacement. Thus, in areas highly impacted by invasive steeplebush, only a part of the original pool of species can persist. The next step should be to identify traits which make some species less vulnerable to biological invasions (Moroń et al. 2021).

Most studies indicate that the impact of invasive species on potential pollinators depends on whether the invasive species reduces resources, upon which the native species depends and also whether it acts as a novel resource for the native species (Bradley et al. 2019). In the case of the goldenrod (Solidago canadensis) invasion, it has been found that this plant reduces the resources, on which native pollinator species depend (it has a linear negative effect on plants). At the same time, it is also a food source for some butterflies and bees (Fenesi et al. 2015), which can cause a buffering effect and result in a non-linear plant impact on pollinators (Moroń et al. 2019). In the case of S. tomentosa, we found that this shrub is highly competitive with native plant species and displaces species of wet meadows, which reduces the resources available to native pollinators. Moreover, despite the large supply of pollen steeplebush is not attractive for most pollinators (Wiatrowska et al. 2018a), it does not have compensating properties that would buffer the displacement of native species, which is manifested in a strong, negative, linear response of potential pollinators to S. tomentosa invasion.

Effective nature conservation and management of invasive plant species should be based on a comprehensive understanding of the role they play in our ecosystems (Barney 2016). We found that S. tomentosa invasion in the wetland ecosystem has a very strong negative effect on the populations of plants and potential pollinators. In the areas invaded by this shrub, only a small part of the native species pool may persist, which implies that S. tomentosa should be considered as a transformer species.

The number and diversity of plants, butterflies and moths, bees and flies change at all points in the S. tomentosa invasion pathway (representing a linear response to invasion). Although it was assumed that invasive plant impacts are highly scale-dependent (Powell et al. 2013; Rejmánek and Stohlgren 2015), our study is the first to show such an unequivocal negative, linear effect of an invasive species impact on the abundance and diversity of potential pollinators.

Many studies showed that the management effort in the case of invasive species populations largely depends on the density–impact curve of the species and optimisation of management relies on minimising the sum of the costs of their impact and management (Yokomizo et al. 2009). For a linear effect of S. tomentosa on plants and potential pollinators, it can be expected that the benefits of controlling this plant will be proportional to the effort involved. As the strong negative impact of this shrub population is observed even at a small density of its population, the control of this plant population at the initial stage of invasion, before its indiscriminate spreading, seems to be the best and most promising approach.

This research was financially supported by the statutory activities of the Faculty of Forestry, the Poznań University of Life Sciences and partially financed by National Science Centre’s grant (UMO-2020/37/B/NZ8/01743) for Dawid Moroń.