Effects of Bothriochloa ischaemum on the Diversity of Pannonian Sandy Grasslands

Abstract

Changes in land use and agricultural practices have altered the resilience of plant communities and can lead to the emergence of invasive species. One of these is the perennial grass species Bothriochloa ischaemum (L.) Kleng., whose diversity-reducing effects are known from several studies. Our exploratory questions were as follows: How does the presence of B. ischaemum affect the diversity and ratio of the species of sandy grasslands? To what extent does this diversity change depend on site characteristics? The supporting studies were carried out in five low-lying sand dune slacks and six relatively higher areas in the upper-intermediate part of the dunes and on an abandoned old field located in the Hungarian Great Plain in the Carpathian Basin. The cover of vascular plant species was recorded in all sampling sites in twelve 2 by 2 m plots, and the dataset was analysed using agglomerative cluster analyses and a non-parametric Kruskal–Wallis test. Five significantly different groups were identified, separating the vegetation types of the sides of the sand dunes, the vegetation types of the dune slack and the old field, and a Stipa borysthenica Kolkov ex Prokudin-dominated vegetation type. Our results suggest that B. ischaemum is only present as small tussocks on the drier, more exposed sides of dunes, with 3.9–24.2% average coverage; is less able to outcompete Festuca vaginata Waldst. et Kit. ex Willd. and S. borysthenica; and is only able to form large tussocks mainly in the lower dune slacks, with 45.6–79.5% average coverage. Here, in the wetter areas, it achieves high cover with a considerable accumulation of litter, and it becomes a dominant species in this association. The diversity-reducing effect of B. ischaemum on old-field grasslands depends on the age of the site and on the stability of the vegetation.

1. Introduction

Invasive species can have multiple impacts on landscape diversity. Examples include disturbances to soil and ecosystem functioning, such as shifts in carbon and nitrogen cycling, which can lead to large-scale changes in ecosystem functioning [1]. They can alter soil nutrient dynamics, favouring their own growth while suppressing native plants [2]. However, this often leads to a loss of soil fertility and the alteration of microbial communities [3]. They can have a major impact on soil chemical parameters and microflora [4], and because they play a role in plant community composition, they also affect community succession, for example, by inhibiting the growth of native species [5].

Significant changes in the landscape ecosystem can be caused by the introduction of an invasive species with higher photosynthetic efficiency, such as Erigeron canadensis, which allows rapid colonisation and resource monopolisation [6], leading to the competitive exclusion of native species. This phenomenon is of particular concern in biodiversity hotspots, where the presence of invasive species can exacerbate pressures on native flora [7], leading to irreversible extinctions [8]. It can also mean that biotic homogenisation may increase local species richness (α-diversity) but decrease the overall diversity (β-diversity) of the wider landscape [9]. Nevertheless, the interaction between invasive species and native biodiversity is a complex phenomenon. Indeed, some studies have demonstrated that invasive species can increase local divergence while concomitantly reducing the overall ecological integrity of the landscape [10]. It is important to note, however, that although the ecological impacts of exotic plant invasions are typically negative, the effects are not uniform or unidirectional [11].

Healthy grasslands are resilient to invasion because of their native biodiversity, but this resilience is weakened by, for example, drought and degradation. Rising temperatures can help invasive plants dominate grasslands, while drought weakens the resilience of native biodiversity [12]. Low-diversity communities are more susceptible to invader dominance [12]. The present and predicted increasing frequency of droughts, heavy rainfall, and heat waves [13] due to climate change puts pressure on the habitats, with special emphasis on drylands [14,15,16]. The proliferation of Bothriochloa ischaemum (L.) Keng, often facilitated by disturbances that create openings in the existing vegetation, leads to the predominance of the species. Research indicates that as the cover of B. ischaemum increases, it can lead to substantial shifts in community composition, often resulting in the decline of native herbaceous species [17]. The negative relationship between native plant diversity and the invasiveness of B. ischaemum has been documented, suggesting that higher native diversity can inhibit the establishment of this invasive grass [18]. This finding underscores the importance of maintaining diverse native plant communities to mitigate the impacts of invasive species.

Grassland cover may be reduced by changes in land use, such as agricultural expansion, urbanisation, and land use change, which can also create favourable conditions for invasive species to proliferate, exacerbating the already negative impacts of climate change [19,20]. However, due to persistent droughts, many agricultural lands in the central sandy areas of the Carpathian Basin have also ceased to be used for agriculture in recent decades, and periodically, grassland grazing has been abandoned, which has had the further consequence of the emergence and increase of invasive species [21,22,23,24,25,26] in the plant communities in the region. Despite many adverse trends, however, nearly 40% of natural terrestrial vegetation is still grassland today, much of which is still semi-natural, thus being hotspots for biodiversity [27,28,29,30] and providing important ecosystem services [31,32]. Thus, there are international efforts to secure and maintain the many vital ecosystem services provided by these communities [33].

Although research on the distribution of plants is mainly focused on the threats posed by alien species, such as the perennial (Sporobolus cryptandrus (Torr.) A.Gray) [34,35,36,37] and annual (Cenchrus incertus M.A. Curtis) [22] grass weeds that are rapidly spreading in the central sandy areas of the Carpathian Basin, it should be remembered that some native species can also show invasive behaviour. Good examples in the region are Cleistogenes serotina (L.) Keng [17] and Calamagrostis epigejos (L.) Roth [38,39,40], which are becoming dangerous species in several habitats not only in this region but also across Europe [41,42,43]. B. ischaemum is a similarly widespread species that forms stable communities in steppe grasslands and even in coniferous plant communities in the Continental and Pannonian biogeographical regions [44]. It was introduced to the USA in the 1920s for erosion control and forage production. Since then, it has been planted in millions of hectares of unfavourable grasslands and roadsides [45,46], where it has spread into native grasslands and acts as an invasive species [47,48]. In China, it plays a significant role in erosion control [49], and despite not being as productive as other grass species in this region, it was planted earlier and has consequently occupied large areas [49,50]. Here, as well as in Romania and the Mediterranean [44,51], it is considered a co-colonial species, and it is conquering large areas due to its rapid reproduction, excellent regeneration and drought and disturbance resistance, and extreme adaptability [52,53]. It is also aided by inappropriate grassland management practices (such as overgrazing), compromising ecosystem health and the reliability of ecosystem services. Under-utilisation or abandonment following overgrazing is also highly conducive to B. ischaemum proliferation [54].

It can be seen from the above that the competitive advantage of B. ischaemum, due to its C₄ photosynthetic mechanism and morphological plasticity, mainly prevails in degraded and abiotically stressed sites, such as the open and closed sandy grasslands of the Great Hungarian Plain in the Carpathian Basin. Its success is also aided by its ability to quickly reach high densities, allowing it to colonise with greater success [55]. It then becomes an invasive species, displacing native species and becoming dominant in communities [56]. This competitive advantage can result in monocultures, which also reduce the habitat quality of native fauna and disrupt ecological interactions [57]. B. ischaemum dominance and high cover are interpreted as a result of the advanced degradation of natural communities [58,59], the intermediate state of degraded communities, and the process of regeneration of old fields [60]. This species is an excellent illustration of the importance of understanding ecological community dynamics to predict succession [61,62].

Concerning the study area, the vegetation of calcareous sandy soils in the central Carpathian Basin has undergone significant changes in the last centuries and has become mosaic-like [63,64,65,66]. According to Tölgyesi et al. [67], different vegetation patches have developed along environmental gradients such as soil moisture, soil structure, exposure, or temperature [68] in sandy areas between the Danube and the Tisza rivers. However, invasive species are now present in almost all of them and are causing a loss of diversity and habitat degradation due to the homogenisation of the flora [69]. The present study aims to investigate the effect of the presence of B. ischaemum on the diversity of different vegetation types in Pannonian calcareous sandy grasslands.

The following questions were formulated for this study:

• Q1: How does the presence and cover of B. ischaemum affect the diversity, naturalness, and ratio of the species of Pannonian sandy grasslands?
• Q2: To what extent does this diversity change depend on site characteristics?

2. Materials and Methods

2.1. Study Area

The sample area is located in the Carpathian Basin in the Great Hungarian Plain, which is on the border of the settlement of Fülöpháza (Kiskunság Sand Ridge) (Figure 1).

The average annual mean temperature is 10.5 °C, while the average mean temperature during the growing season is 17.4 °C. The average annual rainfall is 530 mm, of which 310 mm falls on average during the growing season, but the average rainfall was below 250 mm on two occasions in the last five years. The soil is calcareous quicksand of Ancient Danube origin. The diverse vegetation of the site was shaped by the continuous change in topography and the presence and movement of water [64,65]. These have led to the presence of a wide variety of vegetation types that are in close proximity to each other, interspersed with transitions.

The area, previously completely treeless, was slightly afforested in the 19th century. At that time, extensive grazing was predominant, but towards the end of the century, grazing around homesteads dominated. After the Second World War, the area continued to be somewhat overgrown with trees and shrubs (mainly Populus spp. and Robinia pseudo-acacia L.). Later, the area was continuously and spontaneously afforested (mainly Populus and Crataegus spp.) [70]. Until the 1950s, it was characterised by open sand surfaces, sparse vegetation, and mobile dunes. From the 1960s, the dune vegetation became more closed: firstly, because of the afforestation of the surrounding landscape and then because of the cessation of grazing. The last mobile dunes disappeared in the 1980s due to the lack of grazing. The central part has not been afforested in the last 60 years. Heavy, occasional overgrazing was last seen between 1979 and 1983. After that, the area was ungrazed for a few years and then grazed by a goat herd from the mid-1980s until 1994. It was then abandoned again. Subsequently, with the resumption of moderate grazing in the 2000s, some sand dunes started to move again. This process has continued since then [71]. Since the 1960s, areas under intensive agricultural cultivation have been largely abandoned, and it was therefore possible to analyse the vegetation of an old-field (65 years old) sample area.

2.2. Coenological Studies

Sampling was carried out in June 2023. The eleven vegetation plots (a–k) and distinct types were separated into the following 3 groups based on locality:

A: Dune slack vegetation “a–d” (Figure 2a) (semi-natural closed sandy grassland): The original vegetation is a sand steppe dominated by Festuca rupicola Heuff., Festuca wagneri (Degen, Thaisz & Flatt) Krajina, or Festuca pseudovina Hack. over Festuca vaginata Waldst. et Kit. ex Willd.

B: Dune-side vegetation types labelled “e–j”: Different sub-associations and facies of semi-natural open calcareous sandy grasslands that are mainly dominated by F. vaginata (Figure 2b)

C: Old-field vegetation type “k”: An old-field grassland formed on the humus-rich soil of a dune slack. The field was abandoned 65 years ago.

For each vegetation type, data were recorded for each species and their % cover in twelve 2 × 2 m plots. Species names were applied according to Király [72].

2.3. Data Analysis

The recorded data were not pre-treated, and statistical tests (cluster analysis, Kruskal–Wallis test) were performed directly on the recorded data. Cluster analysis was performed using a variety of clustering algorithms. The most commonly used clustering methods are hierarchical clustering, partitioning clustering, density-based clustering, and model-based clustering; however, each algorithm has its strengths and weaknesses in terms of data type and clustering objectives. In our work, we have chosen hierarchical clustering because similar objects are grouped into nested clusters. In this method, smaller clusters are iteratively merged into larger clusters based on their similarity or proximity. A dendrogram represents the relationships between objects in a dataset with a structure resembling a tree. The hierarchical clustering method is agglomerative, where objects are merged with their closest peers one after the other, allowing natural clustering to be identified in complex datasets. In our research, we considered it important to explore and assess the clustering relationships between vegetation types, for which a dendrogram of agglomerative hierarchical clustering (AHC) with the Euclidean distance (dissimilarity) model, a commonly used statistical method for such studies, is an ideal choice for visualising hierarchical relationships. To determine the cluster numbers, cluster analysis should use clustering indices, or to improve the reliability of the results, it is advisable to combine several clustering indices. In our work, we therefore chose three clustering indices based on very different computational approaches (Calinski & Harabasz index (adpt.), Hartigan index (adpt.), and Inertia), for which their results were in full consensus, so that the resulting 5 distinct cluster groups were confirmed by all three clustering indices.

For the analysis on the species coverage, diversity values, and conservation category of the studied vegetation types, the non-parametric Kruskal–Wallis test (α = 0.05) was used. In the case where the Kruskal–Wallis test was significant, the non-parametric Dunn’s test with Bonferroni corrections was performed for multiple pairwise comparisons. All statistical analyses were performed using the XLSTAT statistical and data analysis solution software version 2024.4.1 (Lumivero, New York, NY, USA) [73].

The conservation and habitat indications of the vegetation were assessed according to Simon’s [74] conservation value categories (CVCs) of the plant species, and a separate analysis was carried out on the functional group of weeds.

The dominant species of the communities were the primary focus of the study. In this case, the dominant species, the taxon F. vaginata, coincides with the characteristic species of the open sandy grassland association. In addition, the xerophytic species are highly adapted to arid climates and use different strategies to cope with drought; they are subdominant components or, occasionally, facies formers in the patches. These include the deep root system (S. borysthenica), succulent plants (Sedum spp.), narrow and folded leaves (F. vaginata), and the annual life form (Bromus tectorum L.).

Borhidi’s [75] relative water demand and relative nitrogen demand, as the driving environmental factors of open sand grasslands, were evaluated. For these methods, the value of each category was expressed as a group weight percentage.

3. Results

3.1. Characterisation of Each Vegetation Type

3.1.1. Vegetation Types in the Dune Slacks

a: Homogeneous, very dense vegetation consisting of large B. ischaemum tussocks was observed. Due to the very high cover of B. ischaemum, the cover of other grassland species is very low (0.1–1%). The tussocks are so dense that they are difficult or impossible to distinguish from each other. It is common for one large tussock to break up into several smaller ones. Nevertheless, the stand is locally species-rich. A total of 22 species were found in the plots, with an average of 7.42 species. The total cover was 86.6%, of which B. ischaemum accounted for 79.5%, with only Poa angustifolia achieving a higher average cover of 3.4%.

b: This vegetation type is an also closed B. ischaemum-dominated vegetation type with 40–50 cm diameter tussocks. A total of 20 species were recorded in the plots, with an average of 10.75 per plot. The total cover was 75.8%, of which B. ischaemum covered 63.9%. Only Stipa borysthenica Kolkov ex Prokudin (5.1%) and Thymus pannonicus All. (1.9%) achieved a cover above 1%. The coverage of F. vaginata was 1.6%.

c: This vegetation type is an also closed B. ischaemum-dominated vegetation type with smaller, separated B. ischaemum tussocks. Its physiognomy is characterised by prominent Stipa capillata L. individuals. A total of 16 species were found in the plots, with an average of 11 species per plot. The total cover was 68.5%, of which 53.1% was covered by B. ischaemum. In addition to Poa angustifolia L. (4.4%), F. vaginata (2.4%), and S. capillata (1.4%), Dianthus serotinus Waldst. et Kit. (3.4%) also achieved significant coverage.

d: The B. ischaemum tussocks are similar in size to the previous vegetation types but form a more open grassland. More species occur among the stems than in the previous vegetation types. A total of 31 species were found in the plots, with an average of 12.2 per plot, the highest of the vegetation types studied. The total cover was 58.5%, of which, despite the high number of species, 45.6% were covered by B. ischaemum, with only S. borysthenica (7.1%) and F. vaginata (1.7%) accounting for a further 1% or more.

3.1.2. Vegetation Types on Dune Sides

e: This vegetation type is an open vegetation type formed mainly by small B. ischaemum. In total, 26 species were found in the plots, with an average of 11.75 species per plot. The total cover was 38.4%, of which 24.2% was covered by B. ischaemum. S. borysthenica covered 45.2% and F. vaginata covered 3.3%. Three dicotyledonous species also reached an average cover of 1%: Teucrium chamaedrys L. (0.9%), D. serotinus (1%), and Eryngium campestre L. (1.7%).

f: This vegetation type is a mainly open vegetation type formed by medium-sized B. ischaemum tussocks. A total of 13 species were found in the plots, with an average of 11.58 species per plot. The total cover was 36.5%, of which 14.8% was covered by B. ischaemum. The sub-dominant species were S. borysthenica (8%), F. vaginata (4.9%), Koeleria glauca (Spreng.) DC., and Cynodon dactylon (Spreng.) DC. (2.0%). More common species were E. campestre (1.3%) and D. serotinus (0.9%).

g: B. ischaemum and F. vaginata are codominant in these open vegetation types, with few S. borysthenica individuals. A total of 15 species were found, with an average of 9.58 species per plot. The total cover was 36.0%, of which 13.3% was covered by B. ischaemum and 12.3% by F. vaginata such that the two species can be considered as codominant. S. borysthenica covered 4.2%, while Poa bulbosa L. covered 1.4%.

h: B. ischaemum and S. borysthenica are codominant in these open vegetation types. A total of 12 species were found, with an average of 9.25 species per plot. This vegetation type had the smallest number of species. The total cover was 31.7%, of which 13.8% was covered by B. ischaemum and 9.3% by S. borysthenica. Among the grasses, F. vaginata covered 1.8% and K. glauca exceeded 1.0%. Among the dicotyledonous species, Fumana procumbens (Dunal) Gren. et Godr. had the highest cover (1.5%). The size of B. ischaemum tussocks was similar to the previous (g) vegetation type.

i: Open vegetation type: The dominant species in the stand was S. borysthenica (8.7%), with 5% B. ischaemum cover and a total of 15 species, averaging 10.5 species per plot. The total cover was 19.3%. On average, the tussocks were 10–20 cm in diameter. The cover of grasses useful for grassland management was very low, with C. dactylon covering 1.8% and F. vaginata covering 0.8%. K. glauca can also be included here.

j: A more closed vegetation type than the previous one, with 49.1% total cover. It was dominated by 38.8% S. borysthenica. B. ischaemum covered 5%. A total of 18 species were found in the plots, with an average of 10.5 species per plot. The cover of other grassland species was very low here.

3.1.3. Vegetation of the Old-Field Area

k: This vegetation type comprises grassland formed on an abandoned field, with a large variation in the size of B. ischaemum tussocks. The total cover was 71.5%. Grasses were present here, with the highest cover of the 11 vegetation types. A total of 39 species were found, but only 12.7 species per plot were found on average. The average cover of B. ischaemum is 34.7%. The average cover of other grasses (26.25%) is well distinguished from the other vegetation types (ranging from 1.8 to 13.8%), especially the 18.9% cover of P. angustifolia and the 6.0% cover of C. dactylon. Elymus repens (L.) Gould (0.5%) and F. pseudovina (0.1%) were only present here. Among the weeds, grasses Secale cereale L. and Bromus hordaceus L. were present only in this vegetation type.

3.2. Cluster Analysis of Vegetation Types

The agglomerative hierarchical cluster analysis (AHC) combined clustering indices and all three clustering indices (Calinski & Harabasz index (adpt.), Hartigan index (adpt.), and Inertia) separated the same five clusters in two main groups (Figure 3). Clusters 1 and 2 include the vegetation types of the dune slacks. Cluster 1: a is the most closed dune-slack vegetation with a total cover of 86.6%; however, it is the poorest (8.4 species per plot) stand with the highest cover of B. ischaemum. Cluster 2: b, c, and d are the other vegetation types of dune slacks, with less but still significant (58.5–63.9%) B. ischaemum cover.

Cluster 3: e, f, g, h, and i are the open sandy grasslands located on the sides of the dunes, with low total vegetation cover between 31.7 and 38.4% and a lower cover of B. ischaemum (13.3–24.2%). Cluster 4: j had the highest total cover value of 49.1%, with a dominance of S. borysthenica and the lowest B. ischaemum cover (5%). Cluster 5: k is the vegetation of an old field with the highest species richness and 34.7% B. ischaemum cover. The classification analysis based on the 11 isolated vegetation types (Figure 3) clearly demonstrates that the vegetation types (a–d) in the dune slacks are separated from the other vegetation types by a significant distance. In contrast, the plots of the old-field area are closer to but also significantly different from the dune-side vegetation types.

3.2.1. Relationship Between B. ischaemum and Other Abundant Species

The cover values of B. ischaemum in the 11 vegetation types are shown in Figure 4. In vegetation types a–d, the highest B. ischaemum cover varied on average from 45.6 to 79.5%. In all four vegetation types, the total vegetation cover was above 50%, and other species covered only around 10%.

In the e–i vegetation types, the total vegetation cover varied between 20 and 40%, and B. ischaemum accounted for only about half of this. The cover for other species was more significant than in the dune slack vegetation types (a–d). However, in the case of vegetation type j, the total cover was the highest in the side position, and the lowest B. ischaemum cover was also present. In vegetation type k, the cover of B. ischaemum was again larger, in parallel with the more closed vegetation and dune slack position of the old field.

3.2.2. Relationship Between B. ischaemum and Diversity

In the 11 vegetation types, diversity values are inversely related to the cover values of B. ischaemum, almost mirroring the cover profile of the species (Figure 5). The cover of B. ischaemum is the highest in the vegetation types a–d, where diversity is the lowest. The higher the cover value of B. ischaemum, the lower the diversity values. This contradicts the assumption that one would expect to find higher species numbers and diversity in these habitats, with a less extreme water regime, compared to the dune sides.

The diversity values are highest in the e–j vegetation types, in which B. ischaemum’s cover values are the lowest. In vegetation type j, which is considered a more closed vegetation type, the diversity values are lower despite the low cover of B. ischaemum because the predominance of S. borysthenica slightly limits the other species.

In terms of vegetation, the old field (k vegetation type) is well separated from semi-natural grasslands. The highest number of species (41) was found here. However, the total cover of the plots (including B. ischaemum) and the number of species (8–16 species) vary widely. This large-scale spatial heterogeneity and the higher cover of weed species (Figure 6) confirm that the association was formed on arable land.

3.3. Results of Vegetation Types According to Conservation and Relative Ecological Indicators

In the 11 vegetation types, the distribution of conservation value categories without B. ischaemum was as follows (Figure 7). The cover of edificator species (E) and highly protected (KV) species varied in contrast to the cover of B. ischaemum. The latter category includes F. vaginata. The low degree of the degradation of vegetation type j is due to S. borysthenica, which is more resistant to invasion by B. ischaemum due to its excellent drought tolerance and strong competitive ability. In general, the dune sides (e–j) are more natural than the dune slacks (a–d). This correlation is also illustrated by the degree of degradation (Dʄ), i.e., disturbance species cover/natural species cover.

3.4. Water and Nitrogen Requirement of Vegetation Types

Based on the distribution of species with different relative water and nitrogen requirements, B. ischaemum was able to spread in vegetation types located in the better water- and nutrient-supplied dune slacks (Figure 8). The dry and nutrient-poor growing conditions of the dune sides and the good competitive ability of F. vaginata and S. borysthenica in these areas prevent this. Despite its high values in the old field (vegetation type k), the cover of B. ischaemum may be relatively low because it is at a stage of succession where the relatively good water and nutrient availability allows the reproduction of perennial grass species with good competitive abilities, such as E. repens, P. angustifolia, and C. dactylon.

4. Discussion

Plant invasion is highly prevalent in sandy grasslands, which is explained by the fact that many open sand grasslands are either of old-field origin or adjacent to old-field areas infested with invasive species; thus, they are either less resistant to invasion or have high propagule pressure in their vicinity. Invasive species have already invaded about 70% of local open sand grasslands and fescue grasslands and about 30% of dry and semi-arid grasslands [76] in the early 2000s, and this process has the potential to continue in the future.

In the central sandy area of the Carpathian Basin, the vegetation of the dune slacks has changed more in the last 250 years than that of the dune sides. One of the main reasons for this is the afforestation of the landscape, and another is climate change and the resulting decrease in groundwater levels, where relative water scarcity has become greater than on the dune sides [63]. As a result, the vegetation of the fen meadows and sand steppes that once were present in the slacks has degraded more than the open sand grasslands of the dune sides. One indicator of this could be the significant increase in B. ischaemum in these habitats, where it has sufficient soil moisture and nutrients to assert its competitive advantage [64,77]. Ploughlands, which were only established on small areas of soil with higher humus content in the landscape, have now been converted into old fields. In these areas, the inner regulating forces of association are still weak, and succession is slow. The spread of the species was also facilitated by occasional overgrazing in the landscape and subsequent abandonment [54]. Due to its better forage value and more stable crop distribution during the growing season, overgrazing was probably most pronounced in the dune slacks. Similarly, B. ischaemum’s resistance to grazing allows it to maintain a competitive advantage in overgrazed landscapes, where drought stress is often exacerbated during hot seasons [78].

On dune sides, S. borysthenica tolerates extreme drought better with its deep roots than F. vaginata [60,79], which is usually the dominant species in the association, and this is the reason why, in open vegetation types, B. ischaemum has not been able to significantly increase in open habitats, despite the fact that land use change and land abandonment often cause a loss of diversity in human-maintained habitats [80,81,82,83]. Grassland diversity is also significantly reduced in pastures, and the reduction in livestock is observed [84,85]. Both overused and abandoned habitats are subject to a number of risk factors. These include, for example, scrub encroachment and the rapid spread of invasive species [85]. Both processes significantly alter grassland structure and result in a loss of diversity, which is one of the most important conservation concerns [86]. Altering natural habitats through these activities not only promotes the spread of B. ischaemum but also upsets the delicate balance of local ecosystems, leading to a loss of species diversity [87].

The present studies confirm the findings that the increase in the abundance and dominance of B. ischaemum negatively affects the physiognomic structure of grasslands, causes species impoverishment, and reduces diversity [53,58,59,88,89,90]. In this study, as the target species cover increased, Simpson’s diversity in the intermound vegetation decreased from 0.39 to 0.16, while the values for the open sand grassland vegetation types overlying the mound sites varied from 0.58 to 0.74. A value of 0.37 for the vegetation type dominated by S. borysthenica indicates that this species also has good competitive abilities. The value of 0.64 for the old field is due to the fact that the dominant and subdominant grasses forming the association matrix are already established in the area due to the relatively good nutrient supply of the soil. Szentes et al. [54] have also shown the species’ beta-diversity-reducing effect at different spatial scales and thus may also reduce C₃-rich native species habitats at the landscape level. This decline could potentially threaten endangered rare species and reduce the number and density of native species [91,92,93]. In addition to its high morphological plasticity in dry grasslands under degraded abiotic conditions, its improved water utilisation due to its C₄-mediated photosynthetic mechanism [94] provides a competitive advantage over C₃ grassland species. Its spread is also aided by its allelopathic effect, which makes it difficult for other native species to recolonise once infested [93,95], as it releases chemicals into the environment that inhibit the growth of neighbouring plants [93]. This ability allows B. ischaemum to not only suppress native flora but also to create a more favourable environment for its continued dominance, leading to a significant reduction in species richness [96].

However, environmental stresses such as drought contribute to the decline of native plant species, and B. ischaemum has a negative impact on plant biodiversity through other ecological mechanisms as well. Although severe drought stress causes a significant reduction in total biomass even in the case of B. ischaemum, the root-to-shoot ratio and fine root biomass may increase significantly [94]. Decreased coarse root biomass indicates that severe drought stress leads to the non-uniform allocation of a carbon “sink” [94]. However, under mild drought stress, it can improve water use efficiency by improving water transport capacity (osmotic regulation), reducing water loss (reducing the transpiration rate) and changing the vertical distribution and morphology of roots to increase water uptake capacity [86]. It increases root density in the upper soil to absorb soil water and improves the water transfer ratio to the above-ground parts of plants [86]. As a result, it can maintain its growth even during drought partly due to osmotic adaptation and the increased production of proline and soluble sugars during water deficiency [97]. These mechanisms enable it to maintain its physiological functions and competitiveness even under dry conditions, especially in stressful environmental conditions where native species can barely survive due to limited water resources [98]. This ability makes it particularly suitable for invading disturbed areas [78,99]. Its drought tolerance is also closely related to fire ecology. Fire management strategies may inadvertently promote the species’ reproduction by creating post-fire environments that favour its growth [100,101], as it is often capable of flowering after fires [99].

Consequently, the control of invasive species is a complex task, involving a wide range of factors. Most of the time, it is too late, but arguably, the most effective way to manage invasive species is to prevent their introduction. Strict regulations on the introduction of potentially invasive species and public education are excellent tools for this [102], and involving local communities in invasive species management can increase the effectiveness of conservation efforts.

Even if introductions do occur, early detection and rapid response (EDRR) strategies are key to preventing new invasions [103]. Mechanical control, such as mowing at the right time and in the right way, has been shown to inhibit the invasion of certain species, such as several Solidago species, and can be an effective management strategy under certain conditions [104,105]. However, due to the annual dynamics and high morphological plasticity of B. ischaemum, this method is not suited for these conditions. Long-term mowing experiments concluded that sustained late-season mowing controlled the expansion of B. ischaemum on the tallgrass prairie of North America [106]. However, on sandy grasslands characterised by the low biomass grazing of small ruminants, sheep and goats could be a potential solution via appropriate late timing and low grazing pressure.

Chemical controls can be a solution if applied carefully, but especially in protected areas, they should only be used for spot treatment to minimise the impact on non-target species and the environment. Biological controls can reduce the population of the target species through the introduction of natural enemies of the invasive species (predators, parasites, or pathogens), but if not carefully managed, the introduced target species can also become an invasive species [107]. The introduction of native competitor species can help restore ecological balance and improve the overall biodiversity in areas affected by B. ischaemum.

Studies have shown that the inoculation of soils with native soil microbial inoculation improves the survival rate of native species and reduces the likelihood of re-invasion by B. ischaemum [108]. This approach takes advantage of the beneficial interactions between native plants and their associated soil microbiota, which can enhance nutrient uptake and suppress pathogens that favour invasive species [109].

Once invasive species have been controlled, the continued monitoring of restored areas is essential to assess the success of management and restoration efforts. Knowledge sharing between scientists, land managers, and policymakers can lead to more effective strategies and innovative solutions for invasive species management and ecosystem restoration [110]. The restoration of semi-natural grasslands can increase biodiversity and restore the hotspots of ecosystem services [111].

5. Conclusions

In both open and closed vegetation types in dry sand grasslands, the number of species in the vegetation types was also significantly affected by B. ischaemum cover, but this occurred primarily by narrowing the available habitat spaces, thus reducing the amount and cover of species. Based on the different sandy grassland vegetation types, B. ischaemum has undergone more significant changes in the more closed vegetation types in the dune slacks, where environmental factors were more favourable. B. ischaemum is less able to invade the open sand grassland vegetation of dune sides and transforms their vegetation to a lesser extent. This may be due to the more unfavourable abiotic factors, the competitive strength of F. vaginata, and the deeper-rooted S. borysthenica, which reaches a greater size in this environment than B. ischaemum.