What Are the Effects of Cattle Grazing on Conservation and Forage Value Across Grazing Pressure Gradients in Alkali Grasslands?

Abstract

Studying the effects of grazing pressure on species composition, beta diversity and yields is important for conservation purposes as well as for grassland management. The case study area in Hortobágy, which is one of the largest continuous grassland areas in Europe, has been managed for centuries by grazing of Hungarian grey cattle. The effect of grazing pressure was investigated in terms of distance from the livestock enclosure (50 m, 250 m, 500 m, 1000 m, and 1700 m) and in an ungrazed control area on dry and mesic alkaline grasslands in spring and autumn of 2024. In both types of grasslands at each distance, species composition and mean plant height were recorded in six 4 × 4 m plots. Overall, in both seasons the control areas were the poorest in terms of species richness. Among the grazed areas in both grassland types the ones at 1700 m distance had the lowest number of species. The species richness of mesic grassland decreased linearly with distance. The dry grassland showed a polynomial trend and was more species-rich at all distances than the mesic grassland. Green yield was the highest in the dry grassland at 250 m in spring and at 50 m in autumn, while in the mesic grassland it was highest at 1700 m in spring and between 500 and 1700 m in autumn. Forage quality in dry grassland was lowest at 50 m and highest between 500 and 1000 m. In mesic grassland, this parameter was equalized at all distances. The highest Simpson diversity was found at a distance of 500–1000 m from the livestock enclosure in both types. It is advisable to evaluate separately the spring and autumn characteristics of the alkaline grasslands, as there may be significant differences between them. Overall, it can be concluded that alkaline dry grasslands are particularly suitable for grazing because of their species composition and their good tolerance to grazing. Alkaline mesic grasslands are poorer in species and more sensitive to grazing; consequently, mowing or mixed utilization should be considered.

1. Introduction

Hungary’s first National Park was established in the Hortobágy in 1973, where grazing has a centuries-long tradition [1]. In 1999, the UNESCO World Heritage Committee inscribed the area on the World Heritage List. The designation was granted because the Hortobágy represents a cultural landscape shaped by pastoral communities, providing an outstanding example of the harmonious interaction between people and nature through traditional, sustainable land use practices that have persisted for millennia. In addition, it is the largest continuous natural grassland in Europe, with alkali grasslands forming its most characteristic habitats [1]. Globally, salt-affected soils cover approximately 1060.1 million hectares, and their extent is gradually increasing due to climate change. Despite this, the vegetation of the Hortobágy remains diverse and mosaic-like, a feature not resulting from deforestation but rather preserved through the practice of grazing [1,2]. This traditional land use remains evident to this day [2].

Livestock grazing is one of the main human activities shaping plant biomass and biodiversity in grassland ecosystems [3,4]. Grazing strongly influences the structure, composition, and functioning of grassland ecosystems [5,6,7,8]. It promotes spatial and temporal heterogeneity in vegetation biomass, structure, and species composition [9,10,11]. Such heterogeneity is further shaped by factors including landscape features, soil properties, and past grazing patterns [12]. Livestock graze selectively, preferring certain plant species or plant parts, which contributes to the patchy distribution of grazing pressure and thereby enhances diversity [13,14,15]. During selective grazing, cattle favour higher-nutrient, softer-stemmed plants, reducing the relative abundance of competitive, dominant perennial grasses and allowing space for more valuable forbs [16].

Grazing intensity is also of critical importance, as it can alter botanical composition and dominance patterns in grassland ecosystems. Increasing grazing intensity and stocking density may lead to more uniform forage use and reduced spatial heterogeneity [17,18]. By contrast, light grazing can enhance biomass and diversity by reducing competitive exclusion among plants and promoting compensatory growth in certain areas [19,20,21]. Under moderate grazing pressure—corresponding to the optimal stocking density of Hungarian Grey cattle—species richness generally increases, the excessive dominance of competitor species is reduced [22,23,24,25], and the mixing of plant species is promoted, which may enhance the overall resilience of the ecosystem [26,27].

In contrast to moderate grazing, at very low grazing intensity the vegetation is not fully utilized, while under high grazing intensity livestock tend to graze more uniformly [28]. Overgrazing suppresses grassland species of agronomic importance, as livestock preferentially consumes certain plants over others, leading to potential ecosystem degradation [27,29]. Biomass reduction and the dominance of highly drought-tolerant, lower-nutrient species over palatable herbs can further accelerate degradation [30,31,32,33,34]. Expansion of bare soil patches under overgrazing enhances desertification and vulnerability to soil compaction [29,34], exacerbating soil resource stress, which negatively affects both plant biomass and diversity [35,36,37,38]. Among the environmental impacts of grazing is the alteration of soil organic carbon (SOC) dynamics. Overgrazing reduces soil organic carbon stocks, potentially shifting grasslands from carbon sinks to sources of greenhouse gas emissions [39,40]. Determining appropriate grazing intensity is therefore critical; proper grazing techniques and stocking rates may help to maintain or even to increase SOC levels, thereby contributing to climate change mitigation efforts [41,42]. Grazing also affects nutrient cycling and water retention. Heavy grazing often results in nutrient loss and changes in soil structure, influencing hydrological processes and reducing the nutrient-holding capacity of the soil [29].

The extent of compositional changes can be well characterized by dissimilarity indices, which measure vegetation differences between zones exposed to different grazing pressure (e.g., heavily grazed vs. abandoned areas) [21]. Under grazing by Hungarian Grey cattle (Bos taurus primigenius var. hungaricus), the resulting mosaic landscape structure—formed by the alternation of continuously grazed, ungrazed, and regenerating patches—creates heterogeneous habitats that provide diverse microclimatic, edaphic, and community conditions, thereby enhancing spatial and compositional variability in grasslands [16,21]. Free-ranging grazing further contributes to the uneven spatial distribution of grazing pressure, as some areas are underutilized by livestock. This results in a heterogeneous distribution of vegetation, biomass, and species composition across the landscape [43,44,45,46].

The present study was carried out in Europe’s largest continuous grassland, a landscape with a millennia-long history of livestock grazing. Although several studies have examined the effects of grazing on alkaline grasslands [2,25,47,48], none of them have investigated grazing pressure as a function of distance from the fence. Our primary question was how grazing by Hungarian Grey cattle (Bos taurus primigenius var. hungaricus) alters the vegetation of the two fundamentally different, characteristic grassland types—dry and mesic alkali grasslands—in accordance with the changes in grazing pressure. We also examined whether there are differences between the spring and autumn aspects of vegetation. Furthermore, we asked whether the end of the growing season, after the summer drought, still provides vegetation of suitable composition and forage quality for cattle grazing. The study assesses the effects of grazing pressure on grassland species composition, and to identifies the level of grazing intensity that is most favourable both from the perspective of vegetation composition and grassland management. Our hypothesis was that in Hortobágy, the large continuous grassland expanse, together with its mosaic structure and the dual seasonal aspects of spring–early summer and autumn, provides the conditions necessary for sustainable cattle grazing, the maintenance of vegetation diversity, and the long-term usability of these grasslands for management purposes.

2. Materials and Methods

2.1. Sample Areas

The sample area is located in the Great Hungarian Plain in Hortobágy. The climate of the area is moderately warm and dry. Low rainfall is combined with abundant summer sunshine and high temperatures. The average annual mean temperature is 9.8–9.9 °C, with a breeding season of 17 °C. There are relatively many hot days. The main feature of the climate, which is extremely continental in the Carpathian Basin, is the high average annual temperature (around 24 °C), mainly due to the relatively harsh winters (average January temperature −3–−2 °C). The Hortobágy is a particularly arid area, with an annual rainfall of 520–550 mm [49].

The study sites were established near Tiszacsege, on the Nagy-Kecskés pasture (grazed area: N 47.649481, E 21.055003, control area: N 47.610803, E 21.044362) grazed by Hungarian Grey cattle (Bos taurus primigenius var. hungaricus) and managed by the Hortobágy Nature Conservation and Gene Preservation Non-profit Company. The herd consisted of 172 cows, 4 bulls, and 132 calves, grazing over an area of 523 ha (stocking rate: 0.46 Animal Units, AU) (Figure 1).

At each study site, we examined the vegetation of the two most extensive alkali habitat types: Artemisio santonici–Festucetum pseudovinae (Soó in Máthé 1933, corr. Borhidi 1996) (dry alkali grassland) (Figure 2) and Agrostio–Alopecuretum pratensis (Soó 1933, corr. Borhidi 2003) (mesic alkali grassland) (Figure 3).

2.2. Coenological Sampling

Since one of the main aims of the phytosociological surveys was to monitor changes driven by different grazing intensities along gradient, sampling was carried out in five groups of plots:

• K 50: plots located at a nominal distance of 50 m from the livestock enclosure
• K 250: plots located at a nominal distance of 250 m from the livestock enclosure
• K 500: plots located at a nominal distance of 500 m from the livestock enclosure
• K 1000: plots located at a nominal distance of 1000 m from the livestock enclosure
• K 1700: plots located at a nominal distance of 1700 m from the livestock enclosure

The suffix ‘A’ after the plot names mentioned above denotes mesic alkali grassland, while ‘F’ denotes dry alkali grassland.

In addition, in both vegetation types we designated one un-grazed site as a control area. In April 2024 within each sampling site, quadrats were randomly assigned for each grazing-intensity category. The corners of the quadrats were permanently marked with 20 cm long steel pegs sunk to ground level to withstand grazing disturbance. The first surveys were conducted in mid-May, corresponding to the spring aspect characterized by the flowering of the C3 photosynthetic grass Festuca pseudovina and the end of the flowering of Alopecurus pratensis. Following the summer drought period, autumn aspect surveys, representing the state of vegetation around cattle herding, were carried out in early November 2024.

At each study site, vegetation was recorded in 6 quadrats of 4 × 4 m in each habitat type for all grazing intensity. For each quadrat, we documented the percentage cover of plant species and their mean height, together with the percentage cover of cryptogams and litter, as well as the percentage cover of dung patches and bare soil surfaces. The average height of the shoots was calculated as the average of 10 repetitions, measuring the tip of the uppermost leaf on individuals. Species names follow the nomenclature of Király [50].

2.3. Yield and Quality Estimation

For forage yield estimation, we applied the method of Balázs [51,52,53,54], using data from phytosociological surveys and plant height measurements to evaluate the studied grasslands in terms of both forage yield and forage quality.

Green forage yield (P) was calculated as follows:

$$\mathrm{P}={\displaystyle \frac{\left(\mathrm{M}-\mathrm{s}\right)\times \mathrm{B}\mathrm{M}\times \mathrm{b}}{100}}$$

(1)

P: Forage yield [kg/ha]

M: Current average height of the grassland [cm]

s: Stubble height [cm]

BM: Biomass coefficient (400 kg/ha/cm)

b: Average cover of the grassland [%].

The forage quality of each grassland was determined using the following formula [51,52,53,54].

$$\mathrm{K}={\displaystyle \frac{{\sum }_{i=1}^n\left({\mathrm{k}}_i\times {\mathrm{b}}_i\times {\mathrm{m}}_i\right)}{{\sum }_{i=1}^n\left({\mathrm{b}}_i\times {\mathrm{m}}_i\right)}}$$

(2)

K: Forage value of the grassland

k_i: Forage value category of the i-th plant species

b_i: Cover of the i-th species (% or D_B)

m_i: Mean height of the i-th species (cm)

n: Number of species in the grassland.

2.4. Nature Conservational Value

The naturalness status of the study sites and its trends were evaluated using the nature conservation value categories (TVK) of Simon [55], according to which the state of the vegetation can be effectively characterized by the percentage distribution of the categories, which reflect deviations from the natural condition, degradation, or possible disturbance, while also taking local conditions into account. Species indicating natural state are: U: Unique species; KV: Strictly protected species; V: Protected species; E: Character species of the association; K: Companion species; TP: Pioneer species.

Species indicating degradation are: A: Adventive species; TZ: Disturbance-tolerant species; GY: Weedy species; G: Cultivated (economic) plants.

The degree of degradation/naturalness of the vegetation in the study sis was calculated using the following formula.

$$\mathrm{D}\mathrm{f}={\displaystyle \frac{\sum \mathrm{A}+\sum \mathrm{T}\mathrm{Z}+\sum \mathrm{G}\mathrm{Y}+\sum \mathrm{G}}{\sum \mathrm{U}+\sum \mathrm{K}\mathrm{V}+\sum \mathrm{V}+\sum \mathrm{E}+\sum \mathrm{K}+\sum \mathrm{T}\mathrm{P}}}$$

(3)

For diversity analyses, we applied Simpson’s Diversity Index (D).

Microsoft Excel 2509 and the R version 4.5.1. [56,57] were used to create the figures.

3. Results

3.1. Compositional Homogeneity, Species Numbers per Quadrat and per Habitat

In the dry grassland (Artemisio santonici–Festucetum pseudovinae), species richness was higher (Figure 4) than in the mesic alkali grassland dominated by Alopecurus pratensis (Agrostio stoloniferae–Alopecuretum pratensis) (Figure 5). In both vegetation types, the same tendency was observed: spring surveys were more species-rich than autumn surveys. Another common tendency in both habitat types and in both seasons was that the control sites had the lowest values for species richness at the site level and per quadrat too.

Regarding compositional homogeneity, a marked difference was observed between the two vegetation types: along the gradient of distance from the fence of the enclosure, i.e., under decreasing grazing pressure, species richness showed contrasting patterns. Homogeneity also varied both per quadrat and per site. In the dry grassland, species numbers showed no clear trend either in spring or in autumn. The highest species richness occurred at the 500 m and 1000 m distance categories in both seasons. In terms of species numbers per quadrat, a decrease with increasing distance was observed in spring, but no consistent trend line could be fitted.

In the mesic alkali grassland, however, species richness decreased with increasing distance both in spring and autumn, showing a negative linear trend both across distances and within quadrats.

3.2. Distribution of Species by Conservation Value Category

In evaluating the sites, not only species richness but also species composition proved to be highly important. To assess the degree of naturalness and degradation of the vegetation, we examined the distribution of Simon’s nature conservation categories [49] (Figure 6, Figure 7, Figure 8 and Figure 9). In the spring aspect of the dry grassland, the spatial variation of degradation with distance from the livestock enclosure could be described by a power function trend line (Figure 6). The highest values were recorded at 500 m, followed by a general decrease with increasing distance.

For weed species (GY), an exponential decrease was detected. Disturbance-tolerant species (TZ) showed a negative power function distribution. In the case of accompanying species (K), which are the best indicators of the naturalness of the association, a linear increase was observed. For the character species of the association (E), no consistent trend line could be fitted to the cover data.

In the autumn aspect, the trend of decreasing degradation level with distance disappears (Figure 9). A striking feature is the outstanding value of the heavily trampled K 50 F plots. For weed species (GY) and natural disturbance-tolerant species (TZ), the fitted trendline is polynomial, but with differing patterns. The cover values of weeds increased up to 1000 m, then decreased again in the most distant plots. Natural pioneer species disappeared from the quadrats of the study area. This category mainly included spring annuals such as Cerastium pumilum and Erophila verna. The trend of disturbance-tolerant species (TZ) differed: the high values at the K 50 F sampling site were linked to two species, Bromus hordeaceus and Trifolium striatum. A polynomial trendline could be fitted to the data. In the case of companion (K) species, a logarithmic increase with distance was observed. For the dominant (E) species, a change compared to the spring pattern occurred: the maximum shifted closer, with the highest proportion recorded in the 250 m zone.

In the spring aspect of the mesic alkali vegetation, the trend of the data based on the degree of degradation follows a polynomial pattern (Figure 10), as do the cover values of weed species (GY). The cover ratio of natural disturbance-tolerant species (TZ) was very low across all distance categories in this vegetation type. The edaphic species (E)—in this case primarily reflected by the presence of Alopecurus pratensis—also followed a polynomial trend line, reaching the highest value at the K 1700 A sampling site.

In the autumn aspect, the trend of the degree of degradation differs significantly from the one observed during the spring survey. From the K 50 A group to the K 250 A quadrat group, it gradually decreases, and beyond this distance it begins to increase again (Figure 11). The distribution of weed species (GY) and natural disturbance-tolerant species (TZ) also shows a polynomial trend. Natural disturbance-tolerant (TZ) species, similar to the spring condition, appear in the autumn surveys with only low cover values. The cover values of edaphic (E) species became highly variable, and no trendline could be fitted.

3.3. Changes in Yield and Forage Quality

From the perspective of grassland management, it is important to know how much biomass can be expected from the grassland within a given time window. In this respect, the dry alkali grassland and the mesic alkali grassland vegetation show different patterns (Figure 10 and Figure 11). Due to the unusual distribution of precipitation—with a dry spring and a rainy autumn—in both vegetation types, the yields of the spring and autumn aspects turned out to be nearly identical. In the dry alkali grassland, in spring the highest yield was recorded in the quadrats of the K 250 F sampling area, while in autumn the maximum shifted to the quadrats of K 50 F. Forage quality in spring increased progressively from the 50 m plots onwards, but at 1700 m—where grazing pressure was already hardly detectable—its value became very low. A similar pattern was observed in the autumn period as well.

In the mesic alkali grassland, yields were lowest in the quadrats at 50 and 250 m, then increased with distance. Forage quality showed better values in spring, but in autumn it decreased significantly with increasing distance.

3.4. Results of Diversity Analyses

The two vegetation types showed completely different patterns in terms of diversity values. The greatest difference was observed between the K 50 F and K 50 A sampling sites, which were subjected to the highest grazing pressure. The mesic alkali grassland (K 50 A) exhibited the lowest values among the grazed sites. Only the control site (TK A) showed lower values. In contrast, in the dry grassland, the K 50 F sampling site had the highest diversity values (Figure 12A,B). The lowest diversity values in this vegetation type were also found at the ungrazed control site (TK F). In the dry grassland, diversity values were balanced from 250 m onwards, showing a slight increase in spring and a minimal decrease in autumn from 250 m.

In the mesic alkali grassland, grazing had an opposite effect on diversity. At the K 50 A site, a plant community with lower diversity appeared compared to the other sampling sites, particularly in spring. In the mesic alkali grassland, diversity increased with distance up to 1000 m in spring, but at 1700 m a slight decrease was already observed (Figure 12C,D). Similarly, in this vegetation type, the lowest diversity values were recorded at the ungrazed control site (TK A) in both seasons.

By autumn, differences in both vegetation types shifted, and the primary factor limiting diversity became the thickness of accumulated litter from senescent and trampled vegetation. Accordingly, diversity values varied within a narrower range. A more pronounced decrease was observed at 1700 m (K 1700 F and K 1700 A) and at the control sites (TK F and TK A).

Based on these trends in diversity values, the dry grassland consistently showed higher values across all categories, with spring and autumn values following similar patterns. In the mesic alkali grassland, values were not only lower overall but also decreased across all distance categories in the autumn survey.

4. Discussion

The study area proved suitable for the parallel analysis of vegetation changes in dry and mesic alkali grasslands. This allowed us not only to confirm that grazing-based animal husbandry significantly affects grassland ecosystems, their plant biomass and biological diversity, but also to demonstrate that substantial differences in species composition of grasslands may develop even under the same grazing pressure [3,4]. This study also confirmed that grazing can alter resource competition among plants and the structural properties of soils through various mechanisms, thereby influencing both plant biomass and the diversity of grassland ecosystems [8,26,37,38,58,59]. High levels of functional diversity can enhance ecosystem functions such as soil formation, carbon sequestration, and water retention, thereby promoting resilience amid environmental variability 67–68.

The feeding preferences of grazing animals can directly modify plant community composition by selectively consuming certain species and reducing the standing biomass, which in turn may decrease competition for light among plant species [37,38,58,59,60,61]. In addition, it also affects the temporal asynchrony of individual species, so appropriate grazing pressure can increase the stability of the grassland against external influences [62]. ultimately altering both biomass and diversity over time. Trampling by animals also affects the habitat by compacting the soil, thereby reducing its water retention capacity and aeration [61,62,63,64]. Trampling may also result in bare soil patches, which can increase the germination rate of subordinate species and enhance the clonal spread of perennial plants [19,38,63].

In addition, the dung of herbivores increases nutrient input and stimulates soil microbial activity [64,65], which likely improves soil nutrient availability and promotes plant growth [26,58]. Manure and urine derived from grazing may also indirectly affect the spatial distribution of plants, since herbivores (e.g., cattle) avoid grazing near dung patches of conspecific individuals [4,37]. Moderate grazing by Hungarian Grey cattle is suitable for controlling dominant grass species while creating diverse microhabitats for less competitive species [2,25].

In the present case, this phenomenon was observed mainly in the dry grassland vegetation type, where besides Festuca pseudovina, other species such as Achillea collina, Achillea setacea, Bromus hordeaceus, Hordeum hystrix, Plantago lanceolata and Trifolium striatum achieved higher cover. This is consistent with the findings of Török et al. [2], who demonstrated in a mosaic alkali landscape that moderate stocking density of Hungarian Grey cattle (<1 livestock unit/ha) increased species richness and grass cover in dry grasslands [24,25].

A very important factor was that the survey was conducted in two aspects, which highlights why grazing could function on this area for centuries. One of the main reasons is that following the summer dry period, a second growth wave occurs in the vegetation due to autumn precipitation, which is linked to the within-year growth dynamics of C3 grasses [66]. An important finding of this study was that, in addition to parallel tendencies of the two vegetation types—such as the fact that in both vegetation types the autumn period showed a species-poorer vegetation compared to the spring aspect—there were also significant differences. The dry grassland was in every case more species-rich, and it responded differently to varying grazing pressure. Due to the dry spring and the rainy autumn, the yields of both vegetation types were nearly the same in spring and autumn. In the dry grassland (Artemisio santonici–Festucetum pseudovinae), the maximum yield was found at 250 m (K 250 F) in spring, which shifted to the 50 m quadrats in autumn. This seems to contradict the intermediate disturbance hypothesis [67] but it was caused by the fact that annual species grow more near the fence. Díaz and colleagues’ [68] results are not valid for our sample area either, because they found that grazing pressure does not affect annual species. There are several possible explanations for the high coverage of annual species near the enclosure. These include the selective feeding habits of cattle, the low coverage of perennial species [69], and nutrient-rich soil due to heavy fertilization [70], which favours the proliferation of ruderal species [71].

Forage quality in spring increased from the areas closest to the livestock farm (K 50 F) outwards, then declined at 1700 m (K 1700 F); this pattern was similar in autumn. In the mesic grassland (Agrostio stoloniferae–Alopecuretum pratensis), the yields were lowest in the quadrats at the K 50 A and K 250 A sites, then increased further away, suggesting that the grazing optimum lies beyond 250 m. These results can be effectively applied in planning annual grazing regimes.

In arid regions, the impact of cattle grazing on the resilience of grasslands is particularly significant. Drought conditions combined with high grazing pressure can lead to the depletion of vegetative resources, which negatively affects the overall functions of the ecosystem [16,72]. Nevertheless, cattle grazing can be incorporated into conservation strategies and may promote the development of a mosaic-like vegetation structure [73].

In the dry grassland, changes in species composition with distance from the livestock enclosure were less pronounced. However, in the un-grazed control area, species richness was significantly lower in both seasons compared to the grazed plots, demonstrating that abandoned areas deteriorate in every respect: litter accumulates, resulting in a decline in species richness [74,75]. Natural pioneer species disappeared from the quadrats. This is because this category primarily consists of spring annuals, such as Cerastium pumilum and Erophila verna, which cannot find suitable conditions due to the dominance of the main species and the accumulation of litter.

The uneven spatial distribution of grazing pressure can have considerable consequences for the structure, composition, and functioning of dry grassland ecosystems. It may lead to biomass and diversity loss, particularly in heavily grazed zones, and can promote the dominance of more drought-tolerant species [6,7,23,65,76,77], which was also observed in our study sites. In the most intensively grazed belt (K 50 F, 50 m), the dry grassland was dominated by the drought-indicating species Bromus hordeaceus and Hordeum hystrix.

In the mesic alkaline grassland (Agrostio stoloniferae–Alopecuretum pratensis), Alopecurus pratensis was the dominant species across all distance belts. The peak abundance of degradation-indicator species shifted from the K 50 A plots in spring to the K 250 A plots in autumn, indicating that continuous grazing exerts a strong degrading effect on this vegetation type. This is further supported by the fact that the cover of weeds increased with distance from the livestock farm in both seasons, particularly in autumn.

From the perspective of biological diversity, grazing can simultaneously damage and support grassland ecosystems. While excessive grazing pressure may lead to a substantial decline in native plant species, balanced grazing can promote favourable conditions for certain grassland birds and other wildlife [78]. The interactions between grazing and other ecological processes are complex. While moderate grazing can enhance plant species diversity, the absence of grazing may result in the encroachment of woody plants [41,79]. In contrast, in this study, intensive grazing pressure led to significant habitat destruction, promoting the dominance of less palatable species. In our case, however, plant biodiversity was not reduced, or was only reduced to a negligible extent [80], but the competitive dynamics within the community were significantly altered through the fragmentation of plant groups [81].

Cattle grazing in alkali grasslands, as in the present study, may have significant conservation implications, with differing ecological and socio-economic outcomes. The interaction between grazing practices and grassland health is complex, requiring a nuanced understanding of ecological dynamics. Proper management of cattle grazing is essential to mitigate negative impacts. Controlled grazing systems that limit livestock density and regulate grazing periods can prevent degradation while simultaneously meeting the economic needs of farmers. Such practices in grasslands enable the preservation of biological diversity and the maintenance of ecosystem services [82,83,84,85,86].

According to ecological assessments, the vegetation of the K 50 A plots was consistently the most degraded among the grazed sites in both seasons. In dry grasslands, the intensity of grazing pressure had a weaker effect on species composition, which reinforces the conclusion that, despite varying grazing pressures, vegetation change in this grassland type was less pronounced than in mesic grasslands.

We will continue to survey the plant population using this method to monitor inter-annual variability and the impact of climate change in the future. In addition to better separating the effects of grazing pressure and annual weather fluctuations on productivity, species composition and forage quality, a multi-year study would also provide information on the grazing pressure that develops the best adaptive capacity of the grassland to climate change.

5. Conclusions

Based on the results, it is advisable to assess both the spring and autumn aspects of alkali grasslands, as significant differences may occur between them. Moreover, this allows for predictions about how long the grazing season can be extended, which is of great economic importance. Overall, it can be concluded that for both vegetation types, the diversity of the ungrazed control plots was far lower than that of the grazed plots. The alkali dry grassland (Artemisio santonici–Festucetum pseudovinae) is particularly suitable for grazing, as it tolerates grazing well in terms of species composition, biomass, and diversity. The highest diversity was found, on average, at a grazing pressure of 0.46 LU/ha in the belt located 500–1000 m from the cattle farm.

The alkali mesic grassland (Agrostio stoloniferae–Alopecuretum pratensis) is poorer in species than the alkali dry grassland and responds more sensitively to grazing, showing significant degradation in all aspects examined. This also supports the earlier land use practice that the most suitable form of utilization for grasslands dominated by Alopecurus pratensis vegetation is gentle mowing or mixed utilization. The most diverse belt of this vegetation type was also found at 1000 m from the livestock enclosure.