Changes in Species Composition, Diversity, and Biomass of Secondary Dry Grasslands Following Long-Term Mowing: A Case Study in Hungary

: The focus of our study was the changes in the composition of semi-natural dry grass-lands in Hungary. Maintaining the favorable condition of grasslands is not only important from a theoretical nature conservation point of view, but it also has important economic implications. Since these valuable habitats were created with the help of humans, their preservation also requires active treatment. Our current experiment was aimed at investigating the suppression of tall grass, Calamagrostis epigejos L. Roth. In Hungary, in the Cserh á t Mountains, eight permanent plots were mown twice a year. We surveyed the vegetation twice a year between 2001 and 2011. The effects of treatment were studied with repeated measures analysis of variance (ANOVA). After 10 years, the C. epigejos cover of the mown plots decreased significantly, from the initial average of 62.38 to 7.50%. Surprisingly, we noticed a decrease in the control plots as well. While percentage cover of C. epigejos decreased in all plots, the decrease was significantly stronger in the mown plots. Regular treatment caused an increase in the number of species and diversity. Species richness increased continuously in both treatment types, which indicates the combined effect of vegetation succession and treatment. The biomass growth of other Poaceae and Fabaceae species, which are important from a grassland management perspective, was also facilitated by mowing. Our results allow us to conclude that long-term regular mowing is recommended for preservation from the perspective of the richness and variety of grassland management functional groups and the functioning of the ecosystem in semi-arid regenerating grasslands.


Introduction
Grasslands are some of the most versatile ecosystems.Climatic factors have played a role in the development of natural grasslands, while semi-natural grasslands have been created by human activity since the Neolithic Age.The traditional agricultural practices used for several centuries have helped the formation and survival of these grasslands [1,2].
For the maintenance of grasslands, traditional land use is needed, such as regular burning, grazing, or mowing and making hay [3,4].These human activities have contributed to species exchange between grasslands, balancing propagule availability [5,6] and they have created plant communities that are today the most important sites of European biodiversity [7].Maintaining the favorable condition of grasslands is not only Grasses 2024, 3 131 important from a theoretical nature conservation point of view, but it also has important economic implications.Many aspects of ecosystem services are associated with grasslands, e.g., the maintenance of the populations of pollinating insects, making the preservation of grasslands a global priority [8].The traditional management of abandoned vineyards and orchards has maintained the pattern and scale of the landscape, however, due to the intensification of agriculture, some areas are overused, while others are abandoned [9].
Land-use changes and land abandonment often cause a loss of diversity in man-made habitats [10][11][12][13].The diversity of grasslands is significantly reduced by the abandonment of pastures and the decline of livestock [14,15].Both overused and abandoned habitats are exposed to numerous risk factors.These include scrubbing, the advance of shrubs, and the rapid spread of invasive species [15].Both processes significantly change the grassland structure and result in a decrease in diversity, which is one of the most important nature conservation problems today [16].
The rapid spread of non-native, invader species can be observed; however, native species, especially tall grasses, are also capable of monotonically growing, in an invasionlike expansion.Deschampsia [17], Sesleria [18], Brachypodium [19,20], and Molinia [21] species are widespread dominant grass species in Europe, whose fast colonization leads to changes in nutrient availability, as well as in temporal and spatial niche divisions.This significantly reduces the germination, establishment, and growth success of other plant species, which ultimately results in the loss of species richness [22,23].
One of the indigenous but invader grasses is Calamagrostis epigejos (L.) Roth (see below C. epigejos), a strong, fast-spreading perennial grass species [24][25][26].It spreads successfully in areas where previous human activity has been abandoned [27][28][29].It has extremely high morphological and physiological plasticity [26] and resistance to various harmful environmental factors.C. epigejos is widespread in Europe, rarely occurring in undisturbed close-to-nature and natural grasslands; however, it can invade these habitats as well [30].It occurs in forests [31,32], river floodplains [33,34], disturbed sites, barren wastelands, and in reclaimed mines [35][36][37].It can be observed in many habitats [38], e.g., secondary habitats developing after deforestation or fallow lands and set-asides [39,40].According to Rothmaler [41], it occurs mainly in forest clearings and often indicates soil degradation.The nutritional value of the species is extremely low, due to its strong leaves and high level of dead matter accumulation.Its use as fodder is extremely rare, although there have been trials, mainly with goats [42].
The most important goal of our present study was to investigate the effect of multi-year mowing on the restoration of the botanical composition and productivity of the grassland.
In the tenth year from the start of the continuously conducted experiment, we evaluated the effect of mowing on the dominant C. epigejos cover and the occurrence of subordinated grassland species, as well as the changes in species richness.From the point of view of grassland regeneration, grazing would have a more favorable effect, but there are currently no livestock.At the same time, if it becomes possible to supplement the treatment with grazing in the future, it will be important to gather information about the change in the productivity of the grassland.
We examined the following questions: 1. How can the spread of dominant grass be controlled?Can mowing reduce C. epigejos coverage, and if so, how long does it take?2. How does the number of species and the Shannon diversity of the grassland change during the experiment?3. Does the grassland productivity value change during the treatment?How is the distribution of the main grassland utilization groups changing?

Study Area
The experimental area is located in the western part of the Cserhát Mountains, on the border of the villages of Vác, Rád, and Vácduka (Figure 1).It belongs to the territory of Duna-Ipoly National Park; the center coordinates are 47

Study Area
The experimental area is located in the western part of the Cserhát Mountains, on the border of the villages of Vác, Rád, and Vácduka (Figure 1).It belongs to the territory of Duna-Ipoly National Park; the center coordinates are 47°45′38.23″N, 19°12′47.53″E. The climate of the region is temperate.The annual precipitation is 520-590 mm and the average annual temperature is 8-10 °C [43].
The substrate is loess of various thicknesses deposited during the Tertiary period.Mountain ridges emerge from the loess layer, such as the Bükkös hill (190 m above sea level).The study sites are located on the west-facing slope (15 ha) of this hill The earlier natural grassland vegetation was Salvio nemorosae-Festucetum rupicolae Zólyomi ex Soó 1964 community [44], which corresponds to Natura 2000 habitat type 6240 "Sub-Pannonic steppic grasslands" [45].
In this mainly agricultural landscape, large areas were used for centuries as arable fields, vineyards, and orchards.These areas have been abandoned since the 1960s and 1970s for various economic reasons [46].
The present study is part of a larger series of experiments that are ongoing in three different areas ( [46,47]).This article summarizes the results of the experiment performed on the western slope of the Bükkös hill, which is the largest of the three areas.

Experimental Design and Data Collecting
In our present work, we focused on the fast-spreading, perennial rhizome grass, which strongly changes the composition of the plant community.With the mowing treatment, we aimed to reduce the vegetative and generative growth potential of C. epigejos.
Eight quadrats were designated, the corners of which were permanently marked.These permanent plots were 3 × 3 m in dimension, positioned randomly along the west slope.In addition to each mown plot, we also designated a control plot, also with permanent marking, where no treatment took place.We arranged mown and control plots in a split-plot design.Vegetation data were monitored in 2 × 2 m large permanent quadrats placed in the middle of each plot, i.e., there was a 2 m buffer zone between the paired (mown and control) quadrats; (Figure 2).This experimental layout was justified by two reasons: (1) avoidance of the forest edge effect and (2) selection of contiguous, homogeneous patches.The chosen small plot size allowed us to minimize topography, vegetation, and land-use heterogeneity, i.e., to ensure the most similar initial conditions.During our work, we performed stratified random sampling.We did not consider patches dominated by shrubs and Robinia pseudoacacia The climate of the region is temperate.The annual precipitation is 520-590 mm and the average annual temperature is 8-10 • C [43].
The substrate is loess of various thicknesses deposited during the Tertiary period.Mountain ridges emerge from the loess layer, such as the Bükkös hill (190 m above sea level).The study sites are located on the west-facing slope (15 ha) of this hill.
In this mainly agricultural landscape, large areas were used for centuries as arable fields, vineyards, and orchards.These areas have been abandoned since the 1960s and 1970s for various economic reasons [46].
The present study is part of a larger series of experiments that are ongoing in three different areas ( [46,47]).This article summarizes the results of the experiment performed on the western slope of the Bükkös hill, which is the largest of the three areas.

Experimental Design and Data Collecting
In our present work, we focused on the fast-spreading, perennial rhizome grass, which strongly changes the composition of the plant community.With the mowing treatment, we aimed to reduce the vegetative and generative growth potential of C. epigejos.
Eight quadrats were designated, the corners of which were permanently marked.These permanent plots were 3 × 3 m in dimension, positioned randomly along the west slope.In addition to each mown plot, we also designated a control plot, also with permanent marking, where no treatment took place.We arranged mown and control plots in a splitplot design.Vegetation data were monitored in 2 × 2 m large permanent quadrats placed in the middle of each plot, i.e., there was a 2 m buffer zone between the paired (mown and control) quadrats; (Figure 2).This experimental layout was justified by two reasons: (1) avoidance of the forest edge effect and (2) selection of contiguous, homogeneous patches.The chosen small plot size allowed us to minimize topography, vegetation, and land-use heterogeneity, i.e., to ensure the most similar initial conditions.During our work, we performed stratified random sampling.We did not consider patches dominated by shrubs and Robinia pseudoacacia and ignored patches with less than 60% C. epigejos cover.All treated plots were mown with hand clippers (CMI Art Num.234620, Emil Lux GmbH, Wermelskirchen, Germany), twice a year, in June and September.In each plot, we visually estimated the cover of the vascular plant species present, with an accuracy of 1 percent.After the plots were mown, the biomass was removed.The coenological survey and treatments were carried out every year between 2001 and 2011.random sampling.We did not consider patches dominated by shrubs and Robinia pseudoacacia and ignored patches with less than 60% C. epigejos cover.All treated plots were mown with hand clippers (CMI Art Num.234620, Emil Lux GmbH, Wermelskirchen, Germany), twice a year, in June and September.In each plot, we visually estimated the cover of the vascular plant species present, with an accuracy of 1 percent.After the plots were mown, the biomass was removed.The coenological survey and treatments were carried out every year between 2001 and 2011.The removed biomass was subjected to further examination in 2009.The cut plant material was collected, always using the same method.The biomass tests were carried out twice a year, in June and September.After the coenological surveys were completed, the biomass was sampled from the central 1 × 1 m grassland area with a hand-held cutter.We left 4-5 cm high stubble.In the case of the control quadrats, the biomass cutting was not collected from the place of the coenological recording, but outside it, from the area of the buffer zone, from the corner of the square.The biomass collected in this way was sorted into groups important for grazing and the forage value of main species according to Klapp [48].(Supplementary Materials Table S1).The values and designations of the grassland management categories were as follows: 1. Dominant grass species i.e.,: C. epigejos; 2. Subordinate grass species important for grassland management; 3. Fabaceae species important for grassland management; 4. Other Dicotyledonous species neutral for grassland management; 5. Thorny, prickly plants; 6. Litter, (standing dead biomass, and lying dead biomass).
All samples were dried to constant weight at 80 °C in tray drayer (Pink WertheimVSD-EX-650-650-140-4, Pink GmbH Thermosysteme, Barneveld, Netherland.The dry biomass measured with digital laboratory scale (Shanghai Ruishan Electronic Technology Co., Ltd.Shanghai, China.The biomass values of each grassland management category were given in grams.

Statistical Analyses
The effects of the mowing treatment on C. epigejos cover, number of species, and Shannon diversity were compared using analyses of variance (ANOVA) with treatments and years as fixed factors.
During the analysis of variance (ANOVA), we examine the differences between and within the groups, that is, we divide each sum of squares by the corresponding degrees of freedom.The result is actually a measure of variance.The F-ratio is then obtained by dividing MS (between) and MS (within).The p-value is the probability of obtaining an F- The removed biomass was subjected to further examination in 2009.The cut plant material was collected, always using the same method.The biomass tests were carried out twice a year, in June and September.After the coenological surveys were completed, the biomass was sampled from the central 1 × 1 m grassland area with a hand-held cutter.We left 4-5 cm high stubble.In the case of the control quadrats, the biomass cutting was not collected from the place of the coenological recording, but outside it, from the area of the buffer zone, from the corner of the square.The biomass collected in this way was sorted into groups important for grazing and the forage value of main species according to Klapp [48].(Supplementary Materials Table S1).The values and designations of the grassland management categories were as follows: 1.
Subordinate grass species important for grassland management; 3.
Fabaceae species important for grassland management; 4.
Other Dicotyledonous species neutral for grassland management; 5.
All samples were dried to constant weight at 80 • C in tray drayer (Pink WertheimVSD-EX-650-650-140-4, Pink GmbH Thermosysteme, Barneveld, Netherland).The dry biomass measured with digital laboratory scale (Shanghai Ruishan Electronic Technology Co., Ltd.Shanghai, China.The biomass values of each grassland management category were given in grams).

Statistical Analyses
The effects of the mowing treatment on C. epigejos cover, number of species, and Shannon diversity were compared using analyses of variance (ANOVA) with treatments and years as fixed factors.
During the analysis of variance (ANOVA), we examine the differences between and within the groups, that is, we divide each sum of squares by the corresponding degrees of freedom.The result is actually a measure of variance.The F-ratio is then obtained by dividing MS (between) and MS (within).The p-value is the probability of obtaining an F-ratio greater than the observed one, assuming that the null hypothesis that there is no difference between the group means is true [49].For post hoc test, the Tukey honestly significant difference (HSD) with corrections (adjusted p-values for the multiple tests) was used.
The diversity index is a mathematically expressed measure of the species diversity of a plant community.The Shannon diversity index (H) could be calculated from the species richness and number of species values of the individual coenological relevé.Diversity index reveal more information about the composition of the community than the number of species, as they also consider the relative abundance of different plant species.The diversity index theoretically depends not only on species richness, but also on how evenly each species is distributed in the same community [50].
The H index can be calculated as follows: where p i is the proportion of species i in the community.Most of the statistical analyses were performed with R statistical software (version 4.0.5, [51]).We used "vegan", "tidyverse", "ggpubr", and "rstatix" packages.For Shannon diversity, values were calculated with PAST 3.11 statistical software [52].

Effects of Mowing on the Cover of C. epigejos
C. epigejos was the most important plant species in all plots at the beginning of the investigation with similar average cover (62.38% in the plots that were assigned to be mowed and 69.38% in control plots).After two years of mowing treatment (in 2003), the difference became significant (p = 0.0019) (Figure 3).The significant difference that emerged in the third year of the study had remained throughout the experiment.If we compare it to the starting year as a control, we can conclude that the difference in the mown plots is also significant starting from the third year.At the same time, this is not the case with control plots.
Grasses 2024, 3, FOR PEER REVIEW difference between the group means is true.[49].For post hoc test, the Tukey honest significant difference (HSD) with corrections (adjusted p-values for the multiple tests) w used.
The diversity index is a mathematically expressed measure of the species diversity a plant community.The Shannon diversity index (H) could be calculated from the speci richness and number of species values of the individual coenological relevé.Diversity i dex reveal more information about the composition of the community than the number species, as they also consider the relative abundance of different plant species.The dive sity index theoretically depends not only on species richness, but also on how evenly ea species is distributed in the same community [50].
The H index can be calculated as follows: where pi is the proportion of species i in the community.Most of the statistical analyses were performed with R statistical software (versi 4.0.5, [51]).We used "vegan", " tidyverse"," ggpubr", and " rstatix" packages.For Sha non diversity, values were calculated with PAST 3.11 statistical software [52].

Effects of Mowing on the Cover of C. epigejos
C. epigejos was the most important plant species in all plots at the beginning of t investigation with similar average cover (62.38% in the plots that were assigned to mowed and 69.38% in control plots).After two years of mowing treatment (in 2003), t difference became significant (p = 0.0019) (Figure 3).The significant difference th emerged in the third year of the study had remained throughout the experiment.If w compare it to the starting year as a control, we can conclude that the difference in t mown plots is also significant starting from the third year.At the same time, this is not t case with control plots.After ten years of our study, in 2011, in the mown plots there was a considerab decline in the average cover of C. epigejos (from initial 62.38% to 7.50%).In the contr plots, the average cover also decreased to 56.88%.The total cover of all species had n substantially altered.(Table 1).After ten years of our study, in 2011, in the mown plots there was a considerable decline in the average cover of C. epigejos (from initial 62.38% to 7.50%).In the control plots, the average cover also decreased to 56.88%.The total cover of all species had not substantially altered.(Table 1).
Table 1.Change of different parameters in mown and control plots."Total cover absolute" = cover of all vascular plant species; "Cover of C. epigejos absolute" = the cover of Calamagrostis epigejos; "Cover of C. epigejos relative" = the cover of Calamagrostis epigejos relative to the overall coverage; "the cover of subordinate species absolute and relative" = cover of all species except Calamagrostis epigejos, "Number of species" = number of all vascular plant species.All coverage data are expressed as a percentage.For each variable, the same letter shows if the temporal difference between means is not statistically significant.In 2011, in the mown plots there was a considerable increase in the average cover of subordinate species (from the initial 33.16% to 104.41%).In the control plots, the average cover also increased from 38.71% to 54.85%.

Effects of Mowing on the Number of Species and Diversity
The total number of species increased during the study period.Comparing the data on the number of species on the study site, we found a significant difference between the mown and control plots only in the last year of the study, in 2011, p = 0.00726, 10 years after the start of the experiment (Figure 4).epigejos, "Number of species" = number of all vascular plant species.All coverage data are expressed as a percentage.For each variable, the same letter shows if the temporal difference between means is not statistically significant.In 2011, in the mown plots there was a considerable increase in the average cover of subordinate species (from the initial 33.16% to 104.41%).In the control plots, the average cover also increased from 38.71% to 54.85%.

Effects of Mowing on the Number of Species and Diversity
The total number of species increased during the study period.Comparing the data on the number of species on the study site, we found a significant difference between the mown and control plots only in the last year of the study, in 2011, p = 0.00726, 10 years after the start of the experiment (Figure 4).
A significant change in the control plots compared to the initial value in 2001 was first observed in 2009, p = 0.000328, and in 2011, p = 0.00592.
In the mown quadrats, the treatment in the first years did not bring significant changes in the number of species.However, a significant difference (p = 0.0396042) compared to the initial state was observed in 2006, and after that throughout the study (Figure 4).
The most important subordinate species characteristic of the study area is Bothriochloa ischaemum (Keng.), which became the fifth most dominant species with an average coverage of 7% in the mown plots after 10 years of the treatment.At the same time, it also reached the fourth rank in the control plots based on its average coverage, indicating a general drying and closing of the grassland in the entire study area.In the mown quadrats, the treatment in the first years did not bring significant changes in the number of species.However, a significant difference (p = 0.0396042) compared to the initial state was observed in 2006, and after that throughout the study (Figure 4).
The most important subordinate species characteristic of the study area is Bothriochloa ischaemum (Keng.), which became the fifth most dominant species with an average coverage of 7% in the mown plots after 10 years of the treatment.At the same time, it also reached the fourth rank in the control plots based on its average coverage, indicating a general drying and closing of the grassland in the entire study area.
At the same time, the Shannon diversity of mown plots increased only slightly.Comparing the control quadrats to 2001, we already noticed a significant difference in 2004, which, however, disappeared by the following year.It appeared again in 2007 and has been detectable since then.In the mown plots, the significant difference appeared as early as 2003, and remained throughout the entire period of the experiment (Figure 5).At the same time, the Shannon diversity of mown plots increased only slightly.Comparing the control quadrats to 2001, we already noticed a significant difference in 2004, which, however, disappeared by the following year.It appeared again in 2007 and has been detectable since then.In the mown plots, the significant difference appeared as early as 2003, and remained throughout the entire period of the experiment (Figure 5).

Effects of Mowing on Biomass Composition
Regarding the distribution of the important groups from the point of view of grassland management, it can be said that on Bükkös hill, a large amount of thorny, unpalatable species such as Eryngium campestre and Carlina vulgaris rarely occur.Otherwise, this component, which occurs mainly in grazed areas, is not typical.Mowing for 8 years (until 2009) had uniformly removed it from the area, in contrast to the grazing animals, which leave it and can, thus, promote its reproduction.Debris accumulation is significant in the case of plots 4, 5, and 6, which are located in the central region of the mountain.Dorycnium herbaceum, Astragalus glycyphyllos, and Securigeria varia species of the Fabaceae family account for a large proportion, while in more disturbed conditions, Vicia cracca is the most abundant (Figure 6).
Festuca rupicola and Brachypodium pinnatum are the most important grass species that are present in greater proportions as a result of mowing.Since their forage value is greater than that of C. epigejos, (Table S1), the total value of the grassland also increases.The cover of Kop also increases, but this does not contribute to the usability of the grassland, as its forage value is the same as that of C. epigejos.

Effects of Mowing on Biomass Composition
Regarding the distribution of the important groups from the point of view of grassland management, it can be said that on Bükkös hill, a large amount of thorny, unpalatable species such as Eryngium campestre and Carlina vulgaris rarely occur.Otherwise, this component, which occurs mainly in grazed areas, is not typical.Mowing for 8 years (until 2009) had uniformly removed it from the area, in contrast to the grazing animals, which leave it and can, thus, promote its reproduction.Debris accumulation is significant in the case of plots 4, 5, and 6, which are located in the central region of the mountain.Dorycnium herbaceum, Astragalus glycyphyllos, and Securigeria varia species of the Fabaceae family account for a large proportion, while in more disturbed conditions, Vicia cracca is the most abundant (Figure 6).In the control samples, the large amount of litter (dead biomass) is noticeable, with the average measured amount being 256 g/m 2 in the study site (Figure 7).Regarding the proportion of Dicotyledonous, the group of Fabaceae, and other Poaceae species, we measured a smaller biomass.In contrast, C. epigejos was present in large quantities in the control samples.Festuca rupicola and Brachypodium pinnatum are the most important grass species that are present in greater proportions as a result of mowing.Since their forage value is greater than that of C. epigejos, (Table S1), the total value of the grassland also increases.The cover of Kop also increases, but this does not contribute to the usability of the grassland, as its forage value is the same as that of C. epigejos.
In the control samples, the large amount of litter (dead biomass) is noticeable, with the average measured amount being 256 g/m 2 in the study site (Figure 7).Regarding the proportion of Dicotyledonous, the group of Fabaceae, and other Poaceae species, we measured a smaller biomass.In contrast, C. epigejos was present in large quantities in the control samples.In the control samples, the large amount of litter (dead biomass) is noticeable, with the average measured amount being 256 g/m 2 in the study site (Figure 7).Regarding the proportion of Dicotyledonous, the group of Fabaceae, and other Poaceae species, we measured a smaller biomass.In contrast, C. epigejos was present in large quantities in the control samples.

Discussion
In our mowing experiment, treatment twice a year effectively reduced the cover of C. epigejos.However, the significant decrease in the dominant species only occurred in the third year.A similar phenomenon was also reported from Germany, where Rebele and Lehmann [24] in their article wrote about a 2 year delay.The slow decline in C. epigejos coverage can be attributed to the nutrient reserves accumulated in the rhizomes [33,53,54].The dominant species lost a large amount of its biomass and, thus, its competitiveness as a result of the frequent cutting [53].Several studies have reported that C. epigejos becomes established in the early stages of succession [39,[55][56][57].These results suggest that regular mowing (twice a year) is likely to be a disturbance of sufficient magnitude to deplete storage organs and cause a state of nutrient deficiency.
In this study, we showed that mowing increased species number and Shannon diversity in regenerating, secondary grasslands.However, the response to the mowing

Discussion
In our mowing experiment, treatment twice a year effectively reduced the cover of C. epigejos.However, the significant decrease in the dominant species only occurred in the third year.A similar phenomenon was also reported from Germany, where Rebele and Lehmann [24] in their article wrote about a 2 year delay.The slow decline in C. epigejos coverage can be attributed to the nutrient reserves accumulated in the rhizomes [33,53,54].The dominant species lost a large amount of its biomass and, thus, its competitiveness as a result of the frequent cutting [53].Several studies have reported that C. epigejos becomes established in the early stages of succession [39,[55][56][57].These results suggest that regular mowing (twice a year) is likely to be a disturbance of sufficient magnitude to deplete storage organs and cause a state of nutrient deficiency.
In this study, we showed that mowing increased species number and Shannon diversity in regenerating, secondary grasslands.However, the response to the mowing treatment was slow: increases in species number occurred after several years and increases in Shannon diversity took even more years.We obtained similar results in another survey, where the exposition and other environmental conditions were different [47].Most studies have reported that mowing increases the species richness of abandoned grasslands [58][59][60][61].C. epigejos is a tall, clonally spreading grass with a large amount of standing dead matter.This dead biomass inhibits the germination and growth of other species.With the litter removed, subordinate species could grow.This is reflected in both the number of species and the values of diversity.Our results support the findings of previous similar studies and indicate that grasslands can adapt very successfully to different forms of human land use [57].
The functioning of the ecosystem is usually determined by the most important properties of the dominant species [58].Thus, moderate disturbance plays a key role in favoring subordinate species [59] and, thus, in maintaining plant diversity [60].
A rapid rearrangement of the dominance ranking of the species could be observed in the treated plots [61][62][63].The importance of long-term monitoring was highlighted also and has been previously reported [64,65].Parallel to the suppression of the dominant species that make up the matrix of the grassland, we observed the spread of subordinate grass and dicotyledonous species.It is an interesting phenomenon that the increase in the absolute coverage of the other (subordinate) species is also observable in the control plots.This was mainly due to the increase in the cover of shrubs.Analyzing the change in the proportion compared to the total cover (relative cover), we revealed that the subordinate species occurred in relatively greater numbers in the mown plots (Table 1).Due to mowing, the microclimate of the area also changes; it becomes drier.Because of this, the broad-leaved grass species lose their dominance, so the cover of narrow-leaved species, e.g., Festuca, increased.This observation is consistent with the results of other studies, e.g., [66,67].Due to regular mowing, the height of the grassland was reduced, and, therefore, more sunlight reached the soil surface, influencing drying [22,68].
Mowing twice a year significantly increased the number of plant species and, to a lesser extent, diversity.Several authors have reported similar results, e.g., Szépligeti et al. [69,70] and Piseddu et al. [71], suggesting that grassland vegetation has adapted to the traditional human land use.Other surveys show that moderate disturbance increases species richness [72].Increasing diversity, species richness, and grassland yield are important elements in the usability of grasslands, and maintaining a healthy structure plays an important role in reducing the risk of invasion.In addition to mowing, grazing is another important disturbing factor.Depending on the duration and timing of grazing, this treatment can strongly influence the competitiveness of plant species [73].
A well-chosen mowing method allows plants to complete their reproductive cycle and achieve maximum productivity [74].
Mowing is non-selective, while grazing belongs to the selective type of moderate disturbance.Individual animal species use the opportunities offered by the lawn according to their own needs.According to Catorci et al., both treatment methods increased species diversity in sub-Mediterranean grasslands, but mowing increased this to a greater extent [22].
At the same time, our present experiment could also be extended to investigate the additional effects of grazing since animal trampling, excrement, and the spread of seeds affect the species composition of the lawn differently.As the fodder value of the lawn increased due to regular mowing, species with higher yields became more prominent.Therefore, in the future, the area may be suitable primarily for grazing with sheep.

Conclusions
With this study, we have confirmed our observation that the retraction of C. epigejos occurs spontaneously in later stages of succession.Although C. epigejos began to decline spontaneously, its dominance remained for a longer period without treatment.
In our 10 year experiment, we found a significant impact of mowing on the abundance and dominance of C. epigejos, along with considerable changes in species richness and species composition.Species richness increased faster during succession when plots were mown.Nevertheless, the Shannon diversity of mown plots increased only slightly, and the difference was not constant over time.Regular mowing is recommended for preservation from the perspective of the richness and variety of grassland management functional groups and the functioning of the ecosystem in semi-arid regenerating grasslands.In the course of our work, we highlighted the importance of long-term studies, because the operating rules of complex ecological systems can only be revealed through regular and systematic studies.

Figure 1 .
Figure 1.The study area location map (a), blue color represents Hungary in Europe (b), blue dot represents the Cserhát Mountains in Hungary.

Figure 1 .
Figure 1.The study area location map (a), blue color represents Hungary in Europe (b), blue dot represents the Cserhát Mountains in Hungary.

Figure 2 .
Figure 2. Sampling method and different treatment.

Figure 2 .
Figure 2. Sampling method and different treatment.

Figure 3 .
Figure 3. Change of cover of C. epigejos in the control and mown plots on the west slopes of Bükk hill, during the 2001-2011 period, M = mown plot (grayish boxes), C = control plot (open boxe Significant differences between mown and control plots in the same year are marked by ** (p < 0.0 *** (p < 0.001).The circles represent outliers.

Figure 3 .
Figure 3. Change of cover of C. epigejos in the control and mown plots on the west slopes of Bükkös hill, during the 2001-2011 period, M = mown plot (grayish boxes), C = control plot (open boxes).Significant differences between mown and control plots in the same year are marked by ** (p < 0.01), *** (p < 0.001).The circles represent outliers.

Figure 4 .
Figure 4. Change in the number of species in the mown and control plots on the west slopes of Bükkös hill during the 2001-2011 period, M = mown plot (grayish boxes), C = control plot (open boxes).Significant differences between mown and control plots in the same year are marked by ** (p < 0.01).The circles represent outliers.A significant change in the control plots compared to the initial value in 2001 was first observed in 2009, p = 0.000328, and in 2011, p = 0.00592.In the mown quadrats, the treatment in the first years did not bring significant changes in the number of species.However, a significant difference (p = 0.0396042) compared to the initial state was observed in 2006, and after that throughout the study (Figure4).The most important subordinate species characteristic of the study area is Bothriochloa ischaemum (Keng.), which became the fifth most dominant species with an average coverage of 7% in the mown plots after 10 years of the treatment.At the same time, it also reached

Figure 4 .
Figure 4. Change in the number of species in the mown and control plots on the west slopes of Bükkös hill during the 2001-2011 period, M = mown plot (grayish boxes), C = control plot (open boxes).Significant differences between mown and control plots in the same year are marked by ** (p < 0.01).The circles represent outliers.

Figure 5 .
Figure 5. Change in Shannon diversity in the mown and control plots on the west slopes of Bükkös hill during the 2001-2011 period, M = mown plot (red boxes), C = control plot (open boxes).The circles represent outliers.

Figure 5 .
Figure 5. Change in Shannon diversity in the mown and control plots on the west slopes of Bükkös hill during the 2001-2011 period, M = mown plot (red boxes), C = control plot (open boxes).The circles represent outliers.

Figure 7 .
Figure 7. Average biomass composition of the mown and control plots (g/m 2 ).

Figure 7 .
Figure 7. Average biomass composition of the mown and control plots (g/m 2 ).