Ge Nature YS) Conservation Nature Conservation 55: 1-19 (2024) DOI: 10.3897/natureconservation.55.114385 Research Article Effects of management complexity on the composition, plant functional dominance relationships and physiognomy of high nature value grasslands Robert Kun'®, Daniel Babai?, Andras Istvan Csathd’, Arnold Erdélyi', Judit Hartdegen*, Attila Lengyel’, Nikoletta Kalman®, Andras Martonffy’, Alida Anna Habenczyus’, Zsofia Szegleti', Akos Vig’, Andras Maté’®, Akos Malatinszky'®, Timea Toth', Csaba Vadasz" 1 Hungarian University of Agriculture and Life Sciences, Institute for Wildlife Management and Nature Conservation, Department of Nature Conservation and Landscape Management, Pater Karoly u. 1., H-2100 Godollé, Hungary an oo FP WO LD Karoly u. 1., H-2100 Godollé, Hungary Research Centre for the Humanities, Hungarian Academy of Sciences, Toth Kalman u. 4., H-1097 Budapest, Hungary Kords-Maros National Park Directorate, Anna-liget 1., H-5540 Szarvas, Hungary OAKEYLIFE, Hungarian Ornithological and Nature Conservation Society, Kolt6 u. 21., H-1121 Budapest, Hungary Centre for Ecological Research, Institute of Ecology and Botany, H-2163 Vacratot, Hungary Hungarian University of Agriculture and Life Sciences, Institute for Wildlife Management and Nature Conservation, Department of Zoology and Ecology, Pater 7 E0otvos Lordnd University, Institute of Biology, Pazmany Péter sétany 1/C., H-1117 Budapest, Hungary 8 University of Szeged, Department of Ecology, Kozép fasor 52., H-6726 Szeged, Hungary 9 Tavasz u. 109., H-2890 Tata, Hungary 10 Dorcadion Kft., Harsfa u. 7., H-6000 Kecskemét, Hungary 11 Kiskunsdg National Park Directorate, Liszt Ferenc u. 1., H-6000 Kecskemét, Hungary Corresponding author: Robert Kun (rbert.kun@gmail.com) OPEN Qaceess Academic editor: Douglas Evans Received: 18 October 2023 Accepted: 1 December 2023 Published: 9 January 2024 ZooBank: https://zoobank. org/82EES50F6-2F55-4F3D-97C7- 413B3EB33BB3 Copyright: © Robert Kun et al. This is an open access article distributed under terms of the Creative Commons Attribution License (Attribution 4.0 International - CC BY 4.0). Abstract A significant proportion of Europe's species-rich grasslands are semi-natural habitats. They have a long history of traditional management. Several studies have been car- ried out to conserve them, resulting in the establishment of subsidised conservation management schemes. On the other hand, many of these conservation management schemes have failed to provide locally adaptive solutions to maintain the diversity and functional status of species-rich grasslands. In addition, few studies have compared the conservation effectiveness of different levels of management complexity. The levels of management complexity in our study are based on how different management types (e.g. grazing and mowing etc.) and how different herbage removal intensities (e.g. lower and higher grazing intensities) are combined within and between years. To investigate this, we compared the overall effects of management complexity, herbage removal in- tensity and management type on plant diversity, plant functional type dominance rela- tionships and plant physiognomy. Our field sampling was carried out in the sandy me- so-xeric grasslands of the Turjan Region of the Great Hungarian Plain (Central Hungary). We sampled nine 2 m x 2 m plots per grassland site (n = 12), recorded all the rooted plant species and estimated their percentage cover in each plot. High level of manage- ment complexity had significant positive effects on plant diversity, grazing had positive effects on plant diversity and phanerophyte density, while the studied levels of herbage removal intensity had no effect on diversity, plant functional types or plant physiogno- my. In parallel, mowing and/or low levels of management complexity had some negative effects on conservation value (e.g. lower Shannon and Simpson diversity). In this land- scape, the dominance of grazing and the more complex management is more optimal Robert Kun et al.: Effects of management complexity on HNV grasslands Citation: Kun R, Babai D, Csatho than relatively homogeneous mechanical mowing. The choice of management type and Al, Erdelyi A, Hartdégen J, intensity is an important tool in the conservation management system of this landscape, Lengyel A, Kalman N, Martontty but so too is its appropriate application in space and time. Through a detailed analysis A, Habenczyus AA, Szegleti Z, of the effects of management complexity levels compared to management types and vig A, Mate A, Malatinszky A, herbage removal intensity levels, we provide a new opportunity to make grassland man- Toth T, Vadasz C (2024) Effects ba ee ee agement practices more effective for conserving biodiversity in this region, but it would of management complexity on see Nat a frddiaaen ineguinositionyelanttuactional e important to investigate these in different landscapes and conditions. dominance relationships and physiognomy of high nature value Key words: Grassland conservation system, management effects, management efficiency grasslands. Nature Conservation 59: 1-19. https://doi.org/10.3897/ natureconservation.99.114385 Introduction A significant proportion of European landscapes are cultural landscapes that have been transformed and managed by humans (Plieninger et al. 2006; Dahl- strom et al. 2013). The majority of the grasslands in these landscapes are semi-natural, i.e. created and actively maintained by local communities (Maurer et al. 2006; Marini et al. 2007; Niedrist et al. 2009). Nevertheless, the number of endemic species in these grasslands is exceptionally high in Europe, underlining their high conservation value (Hobohm and Bruchmann 2009; Habel et al. 2013). Species-rich, semi-natural grasslands have been managed for centuries by small family farms to provide summer forage in pastures and to produce win- ter fodder in hay meadows for livestock (Babai and Molnar 2014). Economic, socio-cultural and political factors, such as intensification and abandonment, have led to the disappearance of extensive grassland management systems across Europe since the mid-twentieth century (Bakker and Berendse 1999; MacDonald et al. 2000; Ockinger et al. 2006). As a result, the habitat mosaic of the cultural landscapes has changed, with fragmentation and disappearance (scrub encroachment, afforestation) of species-rich semi-natural grasslands having occurred, while the biodiversity of these habitats has declined (Eriksson et al. 2002; Krauss et al. 2004; Ockinger and Smith 2006; Flynn et al. 2009; Lal- iberte et al. 2010). As semi-natural grasslands have been developed and maintained by human management, active and adaptive nature conservation management should be implemented to maintain the species composition and vegetation structure of these habitats. Through a long learning process, nature conservation aimed to mimic the patterns and disturbance regimes of former non-intensive, tradi- tional grassland management (Szépligeti et al. 2018; Kun et al. 2021). These principles recognise the importance of the co-formation of the vegetation and extensive management and the adapted nature of the species pool to the for- mer management practices (Poschlod et al. 1998). Conservation management of grasslands should also draw on the experience of local communities still practising traditional and adaptive grassland manage- ment (cf. Niedrist et al. (2009); von Glasenapp and Thornton (2011)), as in the case of experiments in conservation biology (e.g. Vadasz et al. (2016)). One of the most significant trends in conservation management research is the over- simplified ‘one-factor’ view, where ecologists study the effects of only one man- agement factor, mostly focusing on the general effects of management type (mostly grazing and/or mowing) and management intensity (Talle et al. 2016; Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 2 Robert Kun et al.: Effects of management complexity on HNV grasslands T6rdk et al. 2018; Kun et al. 2021). The explanatory power of these specialised and generic few-factor models often has major limitations in their applicability to specific and local grassland conservation practices (Vadasz et al. 2016; Kun et al. 2019). Therefore, more effective and practice-orientated nature conserva- tion also requires more detailed and comprehensive studies to fill the knowledge gaps on the complex, locally specific effects of different management factors (e.g. locally appropriate management types, regimes, spatial and temporal appli- cations etc.) on species-rich grasslands (Babai et al. 2015; Kun et al. 2021). One of the potentially important management factors for grassland conservation that should be investigated is how different management factors are applied spatial- ly and/or temporally on grasslands (Allan et al. 2014). Based on some previous studies (Vadasz et al. 2016; Kun et al. 2019), the spatial and temporal applica- tion of management types (e.g. grazing or mowing) or different herbage removal intensities (temporal speed of grazing or mowing, based on standard livestock unit and mowing frequency, see Table 1) can represent management complexi- ty. Levels of management complexity are based on how different management types (e.g. grazing or mowing) and different management intensities are varied within and between years on a given grassland (see Table 1 for details). With more knowledge about the appropriate application of levels of management complexity, we would be able to more effectively adapt our conservation objec- tives in different grassland conservation management cases (Kun et al. 2019). In this study, we aim to reveal the effects of management complexity, man- agement intensity levels and management types on plant diversity, plant func- tional type dominance relationships and plant physiognomy in species-rich meso-xeric, sandy grasslands of central Hungary. We hypothesise that high management complexity and low herbage removal intensity will positively af- fect plant diversity, plant functional state and physiognomy. We also hypothe- sise that grazing, in particular, has a positive effect on higher plant diversity and less graminoid (Poales) cover, more forbs and shrubs (Phanerophytes) cover. Our specific question is: How do low and high levels of management complexi- ty affect plant diversity, vegetation physiognomy and plant functional type cov- er in relation to management type and herbage removal intensity? Table 1. A list and an introduction to the management factors and their categories and sub-categories. Management factor categories Management factor subcategories Type of grassland management (T) | Mowing (M): Mechanical mowing at the end of June or the first half of July with 10-15 cm of Herbage removal intensity (/) Management complexity (C) stubble. See details of management complexity later in this Table. Grazing (G): Pastures are mainly grazed by cattle from the end of April to the beginning of October each year. Shepherds often work with them. Combined (C): Mowing and grazing are combined within the same year or between years. For more details, see management complexity later in this Table. Low: Grazing at < 0.5 Standard Livestock Units (SLU) per hectare or mown once a year. LUI value: 0.1 (Schneiders et al. 2011). High: Grazing at > 0.5 Standard Livestock Units (SLU) per hectare or mown once a year followed by grazing in the same year. LUI value: 0.2 (Schneiders et al. 2011). Low: Grazing with a standard sequence of two grazing units per year or one mowing with 10% uncut per year or one mowing per year combined with subsequent grazing. High: Mowing and grazing combined between years or grazing with different start times between years in a four-year rotation. Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 3 Robert Kun et al.: Effects of management complexity on HNV grasslands Methods Study area The study sites are located in the Turjan Region of the Great Hungarian Plain along the Danube in central Hungary, in the northern Kiskunsag area. The study sites are relatively close to each other, within a circle of about 10 km diameter around the neighbouring villages of Kunpeszér, TatarsszentgyOrgy and Kuna- dacs (Appendix 1). The climate is mainly continental with sub-Mediterranean influences. The average annual temperature is 10.5-11 °C, while the average annual precipitation is 500-550 mm (Kocsis 2018). The potential natural vege- tation is the Euro-Siberian steppic woods with Quercus spp. A significant part of the region consists of semi-natural Molinia meadows, which are mown or grazed by cattle and Pannonic sand steppes. These grasslands are mainly grazed by Hungarian Grey cattle and Charolais breeds and, to a lesser extent, by sheep. Most of the studied sites have been modified by local people in the past and present, through woodcutting and long-term grazing (Molnar et al. 2022). Some of the grasslands studied are old fields, abandoned several decades to a few cen- turies ago. These areas are fully regenerated and are well developed. Their spe- cies pool, species composition and physiognomy do not differ significantly from the other grasslands studied. Constant management is essential in these grass- lands to prevent reforestation and the spread of some native disturbance-toler- ant or invasive alien species (Erdélyi et al. 2023). Over the past century, a network of drainage canals has been constructed throughout the area, resulting in the drying out of wet grasslands and the creation of a significant amount of drier grassland (Télgyesi et al. 2022). All of the grasslands studied are meso-xeric habitats, representing the transitional zone between the Molinia meadows (Mo- linion caeruleae) and the dry Pannonic sand steppes (Festucion vaginatae), with a similar vegetation composition and state of development. This species-rich grassland covers a large area in the study area; but it is threatened in a regional and wider context. The meso-xeric grassland habitats are important for the whole Eurasian forest-steppe zone and can be considered as its species-rich grassland component (Mathar et al. 2016; Willner et al. 2019). The dominant and character- istic graminoid (Poales) species of the studied grasslands include Chrysopogon gryllus, Brachypodium pinnatum and Molinia caerulea and some forb species, such as Serratula tinctoria, Sanguisorba officinalis, Peucedanum cervaria, Beton- ica officinalis and Genista tinctoria, as well as some Hungarian protected forb species, such as Ophrys sphegodes, Iris spuria, Centaurea scabiosa subsp. sadle- riana etc. All the grasslands studied are part of the Kiskunsag National Park. As a result, these grasslands have been managed according to conservation princi- ples in the last decades which means a lower management intensity and a more complex management in space and time in general. Conservation is carried out throughout the study area by the Kiskunsag National Park Directorate. Data collection The surveys were conducted in June 2018 on 12 grassland sample sites, all of which were at least 5 ha and at most 10 ha in size (Appendix 1). Three of the sites were mown, six were pastures with varying levels of herbage remov- al intensity and complexity and, in three grasslands, these had combined use Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 4 Robert Kun et al.: Effects of management complexity on HNV grasslands (both mowing and grazing). For each study site, nine plots (2 m x 2 m each) were located in the inner zone of the grassland to exclude edge effects. In each grassland, a random starting point was chosen and the plots were sampled along two parallel line transects with a maximum length of 200 m and a mini- mum distance of 4 m between plots (Appendix 2). The coordinates of the plots were recorded by GPS. Data were collected from 108 plots in the 12 grassland sites mentioned above, nine plots per site (see Appendices 1 and 2). During sampling, we recorded each vascular plant spe- cies found in the sample plots and visually estimated its percentage cover. In addition, we visually estimated four vegetation physiognomic characteristics: 1) percent litter cover, 2) total plant cover, 3) the amount of bare soil surface and 4) average plant height. Average plant height was estimated using a tape measure and reported in centimetres. Due to overlapping layers of vegetation, total plant cover in plots could exceed 100%. We defined plant functional types (PFTs) as groups of species based on three growth forms: forbs including non-grassy herbs, graminoids (Poales) including grasses, sedges and rushes and phanerophytes including shrubs and small trees (Raunkiaer 1934; Box 1996; Kirdly 2009). We calculated the proportions of PFTs in each plot by summing the cover values of the species assigned to them. At each grassland site, we recorded three management factors at different levels, including intensity of herbage removal (I, with low and high levels), com- plexity of management (C, with low and high levels) and different types of man- agement (T, including grazing, mowing and combined types) (Table 1). Prior to our field sampling, we interviewed the conservation practitioners of the later sampled grasslands of the National Park Directorate and sampled grassland sites were selected, based on low and high levels of complexity and herbage re- moval intensity of management, as well as management types (grazing, mow- ing or combined). On each of the sampled grassland sites (n = 12), all three management factor categories (T, I, C) were applied, but only one subcategory of each management factor category was applied, for example, on a grazed site (one management type), only low or only high level of herbage removal in- tensity and only low or only high level of management complexity were applied (see Appendix 3). These management techniques on grasslands have been stable in the last decades and were only started by the Kiskunsag National Park Directorate in the Turjan Region (see Vadasz et al. (2016)). Data analysis We calculated diversity measures, namely species number, Shannon index and Simpson index, from the plant species and estimated percent cover data re- corded in each plot. The use of both diversity indices was important because the Shannon diversity index is more sensitive to the higher proportion of rare (often specialist) species, while the Simpson index is more sensitive to the bal- ance of more dominant species. We built linear and generalised linear mixed effects models (with ‘Imer’ and ‘glmer’ functions from the ‘Ime4’ package) to test the effect of management factors T, I and C as three fixed factors on plant diversity indices, on the abundance of PFTs and on vegetation physiognomy. Different families of distributions (Gaussian and Gamma) were used to treat each differently distributed dependent variable in the modelling (the ‘gamma_ Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 5 Robert Kun et al.: Effects of management complexity on HNV grasslands test’ function from the ‘goft’ package was used). In our analyses, site was a ran- dom factor in all models. To assess model fit, marginal R?,, was applied (Ives 2019) from the ‘MuMIn’ package in R 3.5.1. The beta R’,, statistic (Edwards et al. 2008) was applied using the ‘r.squaredLR’ function to assess the best-fitting model amongst those run with each factor (T, | and C) separately as a predic- tor. The levels of the fixed factors T, | and C were compared using the LMER Tukey post hoc test with the Bonferroni adjustment method (Hothorn et al. 2009) from the ‘multcomp’ package and with the ‘glht’ function. PERMANOVA analysis (with the ‘adonis’ function from the ‘vegan’ package) was used to in- vestigate general patterns in species composition via possible effects of man- agement factors. Principal component analysis (PCA) (using the ‘pca’ function from the ‘vegan’ package) was used to investigate the relationships between plant diversity, plant functional types and physiognomic factors in relation to different management factors. Our analyses were performed in the R 3.5.1 (R Core Team 2018) software environment (R Core Team 2018). Results Management type, levels of herbage removal intensity and management com- plexity had similarly strong effects on species number based on model fits (R? > 0.320, Table 2). There were no differences between low and high levels of herbage removal intensity for diversity, plant functional types and physiognomic factors. High levels of management complexity resulted in significantly higher Shannon and Simpson diversity (Fig. 1). In the case of T, grazing and combined management resulted in significantly higher Shannon and Simpson diversity than mowing and grazing had significantly higher phanerophyte cover than mow- ing, but no significant difference in species number was observed (Appendix 4). With PFT categories as dependent variables, management type showed a strong relationship with graminoid and forb cover (Table 3). Grazed sites had a significantly higher proportion of phanerophyte cover than mown sites and combined sites were between the two (Appendix 4). Herbage removal inten- sity showed a strong relationship with forb and graminoid cover, but a weaker a.) b.) c.) ——<— <<) #p ll aed | } E —-- — | 2 ee | eee T T ; Low High Low High Mown Grazed Combined Management complexity Management complexity Management types 25 a a b ab Shannon diversity Simpson diversity Phanerophytes cover (%) Figure 1. Significant differences in diversity and cover of phanerophytes in grasslands with low and high management complexity and different management types. Only mod- els with minimum R?’,, 2 0.100 fit (see Tables 2-4) and significant differences (Appen- dices 4-6) were selected for inclusion. Significance of differences between groups is based on the LMER Tukey post hoc tests. Significant differences (p < 0.05) between factor levels are indicated by letters (‘a’ and ‘b’) above the boxplots. Non-significant dif- ferences are indicated by the letters ‘ab’. Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 6 Robert Kun et al.: Effects of management complexity on HNV grasslands relationship with phanerophyte cover (Table 3). There were no significant dif- ferences between the levels of herbage removal intensity (Appendix 5). C had a stronger relationship with the forbs and graminoid groups, but a weaker rela- tionship with the phanerophyte group (Table 3). Apart from these relationships, no significant differences were found between C levels for PFTs (Appendix 6). T, | and C strongly influenced average plant height, litter cover and total plant cover, in general (Table 4). On the other hand, no significant differences in av- erage plant height, litter cover and total plant cover were observed between grasslands exposed to different levels of T, ] and C (Appendices 4-6). The two main components were presented in relation to forbs and graminoid (Poales) cover, based on principal component analysis. Higher graminoid cover was associated with mowing and higher forbs cover was mostly associated with grazing and combined management was intermediate between mowing and grazing (Fig. 2). High herbage-removal intensity was associated with higher graminoid cover and low herbage-removal intensity was associated with higher forbs cover (Fig. 3). A high level of management complexity was associated with higher forbs cover and a low level of management complexity was asso- Table 2. Effects of different management factors, namely T: management type; /: herb- age removal intensity of management; C: management complexity, on diversity mea- sures in terms of model fit. Goodness-of-fit is expressed as R’,, values. Species number Shannon diversity Simpson diversity Management factors R? R? R?2 ifs 0.324 0.096 0.057 ! 0.325 0.023 0.011 Cc 0.324 0.072 0.053 Table 3. Effects of different management factors, namely T: management type; I: herb- age removal intensity of management; C: management complexity in relation to forbs, graminoid and Phanerophyte cover. Goodness-of-fit is also presented in R’,, values. Forb species Graminoid species Phanerophyte species Management factors cover (%) cover (%) cover (%) R?2 R?2 R?2 ia 0.368 0.430 0.121 | 0.365 0.420 0.075 a 0.368 0.415 0.097 Table 4. Effects of different management factors, namely T: management type; I: herb- age removal intensity of management; C: management complexity in relation to physiog- nomic factors in relation to grasslands. Goodness-of-fit is also presented in R’, , values. : - Total plant Bare soil Average plant height Litter cover (%) ‘ “ ; Management factors cover (%) surface (%) in plots (cm) R? R? R? R? ip. 0.579 0.709 0.120 0.355 ! 0.572 0.705 0.090 0.298 c 0.559 0.703 0.115 0.318 Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 7 Robert Kun et al.: Effects of management complexity on HNV grasslands PC1 Figure 2. Principal Component Analysis of diversity indices, plant functional type cov- er and physiognomic factors across management types. The diversity indices exam- ined are species number (sp_num), Shannon (Sha) and Simpson (Sim) diversity. Plant functional type cover includes graminoids (Gram.), forbs and phanerophytes (Phane- ro.). Plant physiognomic factors are average plant height (height), total plant cover (full_cov), bare soil surface (bare_soil) and litter cover (litter_cov). Management types: mown, grazed and combined management. The direction, width and different colours of the ellipses in the figure show us the relationship between the samples of different management types. The length and direction of the arrows show the explanatory power and relationship of each variable studied with management types and other variables. -10 -5 0 3 PC1 Figure 3. Principal Component Analysis of diversity indices, plant functional type cover and physiognomic factors across herbage removal intensity levels. The meaning of the abbreviations used in this Figure is given in the legend to Fig. 2. ciated with higher graminoid cover (Fig. 4). Further details on the importance of the principal components, based on the proportion of variance explained by them, can be found in Appendix 8. Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 8 Robert Kun et al.: Effects of management complexity on HNV grasslands PC2 -10 5 0 5 PC1 Figure 4. Principal Component Analysis of diversity indices, plant functional type cover and physiognomic factors across management complexity levels. The meaning of the abbreviations used in this Figure is given in the legend to Fig. 2. Discussion Effects of different management, plant functional type cover and physiognomic factors on grassland diversity Different management types, mainly mowing and low and high levels of herb- age removal intensity and management complexity, significantly affected the species composition and dissimilarity ratios of the grasslands studied (Fig. 1, Appendix 7). In addition, we found a strong positive effect of high management complexity (C) on species number and, to a lesser extent, on Shannon and Simpson diversity and forbs and a negative effect on predominantly perennial and clonal graminoids (Figs 1, 4). The C increases when different management types (T) and herbage removal intensities (I) are varied in space and time (see Table 1; Vadasz et al. (2016)). Certain species or groups of species are likely to prefer certain combinations of T and J, while they may become locally extinct if other combinations are practised for a long time. When C is high, many combi- nations of T and / occur at least once within a time-frame of a few years, provid- ing opportunities for most species to experience a favourable year, preventing extinction (Catorci et al. 2014; Kun et al. 2021). For physiognomic factors (e.g. litter cover and average plant height), C levels did not play a significant role and these variables are better determined by the type of management. Although different T choices played a less important role in influencing com- positional diversity, the choice of the appropriate management type was also significant: grazing had a more positive effect on phanerophytes than mowing (Appendix 4). This difference can be explained by the most extensive, profes- sional cattle grazing on the studied grasslands and by the selective and struc- turing grazing behaviour of cattle (i.e. cattle avoid shrubs etc.) and/or other grazers (Dumont et al. 2012; Molnar et al. 2020). The presence of phanerophyte species and their adequate control by grazing can lead to greater structural or physiognomic heterogeneity of grasslands. The effect of grazing is in contrast Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 9 Robert Kun et al.: Effects of management complexity on HNV grasslands to that of mowing machines, which cut all plants uniformly in mid-summer with a low stubble height. As a result, mown sites could become more homogeneous in vegetation structure. By creating microhabitats and increasing structural variability by allowing a greater cover of phanerophyte species (mostly native shrubs, such as Crataegus monogyna, Prunus spinosa etc.), extensive grazing can contribute to the generative reproduction of herbaceous plants in grass- lands (Kelemen et al. 2017). Furthermore, the application of grazing, mowing or a combination of both also resulted in slight differences in Jaccard-based species composition (but not low or high I and C levels) (Appendix 7). We argue that these phenomena may positively influence species richness. The nurturing effect of shrub species may help the generative and vegetative reproduction of grassland species in the actively managed natural and semi-natural grass- land communities in the forest-steppe zone (Kelemen et al. 2017). On the oth- er hand, it is fundamental to keep the phanerophyte cover within an optimal range (~ 1-10%), which prevents reforestation. An extensive grazing regime can be an efficient way to optimally control the number of shrubs on grasslands. Like shrubs, many forbs can be considered important microhabitat and struc- ture-providing species, based on our field observations (e.g. Serratula tinctoria, Sanguisorba officinalis and Genista tinctoria). Due to several rare and special- ist members (e.g. /ris spuria, Centaurea scabiosa subsp. sadleriana and Ophrys spp., etc.), native, annual and characteristic forbs are also important conserva- tion targets. The occurrence and diversity of forbs in European steppe or for- est-steppe grasslands have a long evolutionary history (Brathen et al. 2020). The increase of clonal, often highly competitive graminoid species with high- er biomass production can reduce plant diversity (Deak et al. 2011; Hazi et al. 2011; Szentes et al. 2012) and suppress conservation target species in grass- lands (K6r6si et al. 2014; Szépligeti et al. 2018), for example, several native forb species and their proportions (Figs 2-4). Therefore, the optimal and continu- ous control of clonal, competitive graminoids and the maintenance of optimal proportions of native and often specialist forbs is important in conservation practices for high nature value grasslands (Kun et al. 2021). This is most likely to be facilitated by high levels of management complexity and low levels of herbage removal intensity grazing (Figs 2-4). On the other hand, there was no significant difference between low and high levels of spatio-temporal complex- ity, herbage removal intensity or management type on graminoid cover, based on linear mixed model post hoc tests and, therefore, further studies are needed to analyse these relationships. Importance and challenges of studying the management complexity and other management factors in grassland conservation locally and across regions Based on our results, special attention should be paid to the multiplicity of man- agement factors (e.g. different management types or herbage removal intensity levels), including their spatio-temporal variability (Kun et al. 2021). We argue that taking these aspects into account can provide practitioners and stakeholders with more straightforward guidelines for conserving and restoring grassland diversity in the Turjan Region. Local and regional scale case studies, as well as large-scale, comprehensive and comparative analyses of the effectiveness of different grass- Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 10 Robert Kun et al.: Effects of management complexity on HNV grasslands land conservation management techniques on different high nature value grass- land communities in different regions, should be carried out in the future to gain more detailed and broader knowledge (see, for example, Fischer and Wipf (2002); Socher et al. (2012); Vadasz et al. (2016); Kun et al. (2019); Rac et al. (2020)). This should provide a more complex view of the relationship between manage- ment practices and conservation objectives at the regional level, which could help to adapt grassland management to local conditions and challenges. Due to the often poor explanatory power of one-factor models, controversial management practices may arise in several cases (Babai et al. 2015; Kun et al. 2019), which may lead to locally ineffective conservation management (Vadasz et al. 2016). On the other hand, although we found that high management complexity is beneficial for grassland conservation, it may be difficult to apply such man- agement complexity and the same methods in practice in other regions, for example, for several individual farmers. Our conclusions are most relevant in terms of the exact management complexity which we have investigated in our study. Each region is different in terms of management possibilities and en- vironmental factors. It can be difficult to graze a site one year and mow it the next or to vary the intensity of management. It is also important to note that spatially and temporally complex management can be achieved in more ways than we have explored in our study. There are other and/or simpler ways, for ex- ample, mowing only every other year, mowing at the beginning of summer one year and at the end of summer the next. The use of different grazing animals and the leaving of uncut lines in different places on a grassland between years can also be effective tools for more complex management, depending on local conservation objectives and opportunities. However, there are often practical difficulties in applying multiple aspects of management to the modelling of community diversity. Including more explan- atory variables in a model requires larger sample sizes and a more balanced sample distribution (Harrison et al. 2018). Ideally, all possible factor combina- tions should be present in sufficient replicates without spatial autocorrelation across the study area. However, ongoing management plans are typically de- signed to meet different, often non-scientific, objectives and the actual man- agement design rarely satisfies statistical assumptions. One can sample what is out there and if certain combinations of factors simply do not exist in reality, they will not be present in the statistical model. This increases multicollinearity in the models and makes it more difficult to distinguish the effects of different management factors (Graham 2003). This is a likely explanation for why the explanatory power of management type, intensity and complexity was similar in our models. Balanced sampling designs are relatively easier to achieve in ex- periments where factor levels and spatial structure can be varied to meet sta- tistical requirements. On the other hand, more detailed assessments, based on multiple management factors in different parts and regions of Europe, would allow us to identify more comprehensively and accurately what should be in- cluded in conservation systems at larger scales, as well as in local practices. Implications Our aim was to collect, organise and compare the elements of the hard-to- compare, mosaic-like landscape of use according to various parameters, using Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 11 Robert Kun et al.: Effects of management complexity on HNV grasslands systematic sampling and to quantify and generalise the treatment results ob- tained mainly through experience. We must emphasise as an important mes- sage to legislators and developers of support schemes that because each site is different, generalisation is limited. High levels of management complexity and grazing as a management type are more positive and have a greater significance for grassland conservation (i.e. result in higher plant diversity, higher proportion of forbs etc.) than the intensity of herbage removal in our study area. At the same time, mowing and/or low levels of management complexity may have some negative effects on conservation val- ue. These analyses can be used to identify what are the strong or direct and less strong or indirect effects in the conservation of high nature value grasslands. Fur- ther research is needed to verify these relationships across a wider range of differ- ent study systems in order to provide generalisable guidelines for conservation. Acknowledgements We would like to thank the following colleagues for their help with the field- work: Judit Deme, Zsolt Molnar, Abolfazl Sharifiyan, Gantuya Batdelger, Nikolett Ponya, Gy6z6 Haszonits and David Schmidt. Additional information Conflict of interest The authors have declared that no competing interests exist. Ethical statement No ethical statement was reported. Funding During the study, Rébert Kun received the UNKP-19-3-I-SZIE-37 grant, Attila Lengyel was supported by the National Research, Development and Innovation Office of Hun- gary (PD-123997), Daniel Babai was supported by the MTA Premium Postdoctoral Re- search Fellowship Programme of the Hungarian Academy of Sciences [grant number: PPD008/2017], and was supported by the MTA-Lendiilet program (Lendulet_2020-56). Author contributions All authors have contributed equally. Author ORCIDs Robert Kun © https://orcid.org/0000-0002-9607-3110 Akos Malatinszky © https://orcid.org/0000-0001-6388-9191 Data availability All of the data that support the findings of this study are available in the main text. References Allan E, Bossdorf 0, Dormann CF, Prati D, Gossner MM, Tscharntke T, Bliithgen N, Bellach M, Birkhofer K, Boch S, Bohm S, Borschig C, Chatzinotas A, Christ S, Daniel R, Diekotter Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 12 Robert Kun et al.: Effects of management complexity on HNV grasslands T, Fischer C, Friedl T, Glaser K, Hallmann C, Hodac L, Hdlzel N, Jung K, Klein AM, Klaus VH, Kleinebecker T, Krauss J, Lange M, Morris EK, Miller J, Nacke H, Pasalié E, Rillig MC, Rothenwohrer C, Schall P Scherber C, Schulze W, Socher SA, Steckel J, Steffan-Dewenter |, Turke M, Weiner CN, Werner M, Westphal C, Wolters V, Wubet T, Gockel S, Gorke M, Hemp A, Renner SC, Schoning |, Pfeiffer S, Konig-Ries B, Buscot F, Linsenmair KE, Schulze E-D, Weisser WW, Fischer M (2014) Interannual variation in land-use intensity enhances grassland multidiversity. Proceedings of the National Academy of Sciences of the United States of America 111(1): 308-313. https://doi.org/10.1073/pnas.1312213111 Babai D, Molnar Z (2014) Small-scale traditional management of highly species-rich grasslands in the Carpathians. Agriculture, Ecosystems & Environment 182: 123- 130. https://doi.org/10.1016/j.agee.2013.08.018 Babai D, Toth A, Szentirmai |, Biro M, Mate A, Demeter L, Szépligeti M, Varga A, Molnar A, Kun R, Molnar Z (2015) Do conservation and agri-environmental regulations effec- tively support traditional small-scale farming in East-Central European cultural land- scapes? Biodiversity and Conservation 24(13): 3305-3327. https://doi.org/10.1007/ $10531-015-0971-z Bakker JP, Berendse F (1999) Constraints in the restoration of ecological diversity in grassland and heathland communities. Trends in Ecology & Evolution 14(2): 63-68. https://doi.org/10.1016/S0169-5347(98)01544-4 Box EO (1996) Plant functional types and climate at the global scale. Journal of Vegeta- tion Science 7(3): 309-320. https://doi.org/10.2307/3236274 Brathen K, Pugnaire Fl, Bardgett R (2020) The paradox of forbs in grasslands and their legacy of the Mammoth steppe. Frontiers in Ecology and the Environment 19(10): 584-592. https://doi.org/10.1002/fee.2405 Catorci A, Cesaretti S, Malatesta L, Tardella FM (2014) Effects of grazing vs mowing on the functional diversity of sub-Mediterranean productive grasslands. Applied Vegeta- tion Science 17(4): 658-669. https://doi.org/10.1111/avse.12103 Dahlstrom A, luga AM, Lennartsson T (2013) Managing biodiversity rich hay meadows in the EU: A comparison of Swedish and Romanian grasslands. Environmental Conser- vation 40(2): 194-205. https://doi.org/10.1017/S0376892912000458 Deak B, Valk6 O, Kelemen A, Tor6k P. Miglécz T, Olvedi T, Lengyel S, Tothmérész B (2011) Litter and graminoid biomass accumulation suppresses weedy forbs in grassland restoration. Plant Biosystems-An International Journal Dealing with all Aspects of Plant Biology 145(3): 730-737. Dumont B, Rossignol N, Loucougaray G, Carrere P Chadoeuf J, Fleurance G, Bonis A, Farruggia A, Gaucherand S, Ginane C, Louault F, Marion B, Mesléard F, Yavercovski N (2012) When does grazing generate stable vegetation patterns in temperate pas- tures? Agriculture, Ecosystems & Environment 153: 50-56. https://doi.org/10.1016/j. agee.2012.03.003 Edwards LJ, Muller KE, Wolfinger RD, Qagish BF, Schabenberger O (2008) An R2 statistic for fixed effects in the linear mixed model. Statistics in Medicine 27(29): 6137-6157. https://doi.org/10.1002/sim.3429 Erdélyi A, Hartdégen J, Malatinszky A, Vaddsz C (2023) Historical reconstruction of the invasions of four non-native tree species at local scale: A detective work on Ailanthus altissima, Celtis occidentalis, Prunus serotina and Acer negundo. One Ecosystem 8: e108683. https://doi.org/10.3897/oneeco.8.e108683 Eriksson O, Cousins SA, Bruun HH (2002) Land-use history and fragmentation of tradi- tionally managed grasslands in Scandinavia. Journal of Vegetation Science 13(5): 743-748. https://doi.org/10.1111/j.1654-1103.2002.tb02102.x Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 13 Robert Kun et al.: Effects of management complexity on HNV grasslands Fischer M, Wipf S (2002) Effect of low-intensity grazing on the species-rich vegetation of traditionally mown subalpine meadows. Biological Conservation 104(1): 1-11. https://doi.org/10.1016/S0006-3207(01)00149-5 Flynn DF, Gogol-Prokurat M, Nogeire T, Molinari N, Richers BT, Lin BB, Simpson N, May- field MM, DeClerck F (2009) Loss of functional diversity under land use intensification across multiple taxa. Ecology Letters 12(1): 22-33. https://doi.org/10.1111/j.1461- 0248.2008.01255.x Graham MH (2003) Confronting multicollinearity in ecological multiple regression. Ecol- ogy 84(11): 2809-2815. https://doi.org/10.1890/02-3114 Habel JC, Dengler J, JaniSova M, Torok P, Wellstein C, Wiezik M (2013) European grass- land ecosystems: Threatened hotspots of biodiversity. Biodiversity and Conservation 22(10): 2131-2138. https://doi.org/10.1007/s10531-013-0537-x Harrison XA, Donaldson L, Correa-Cano ME, Evans J, Fisher DN, Goodwin CE, Robin- son BS, Hodgson DJ, Inger R (2018) A brief introduction to mixed effects modelling and multi-model inference in ecology. PeerJ 23(6): e4794. https://doi.org/10.7717/ peerj.4794 Hazi J, Bartha S, Szentes S, Wichmann B, Penksza K (2011) Seminatural grassland man- agement by mowing of Calamagrostis epigejos in Hungary. Plant Biosystems-An In- ternational Journal Dealing with all Aspects of Plant Biology 145(3): 699-707. Hobohm C, Bruchmann | (2009) Endemische Gefaé&pflanzen und ihre Habitate in Euro- pa-Pladoyer fiir den Schutz der Grasland-Okosysteme. Berichte der Reinhold-Tix- en-Gesellschaft 21: 142-161. Hothorn T, Bretz F, Hothorn MT (2009) The Multcomp Package. The R Project for Statis- tical Computing. Ives AR (2019) R2s for Correlated Data: Phylogenetic Models, LMMs, and GLMMs. Sys- tematic Biology 68(2): 234-251. https://doi.org/10.1093/sysbio/syy060 Kelemen A, Télgyesi C, Kun R, Molnar Z, Vadasz C, Toth K (2017) Positive small-scale effects of shrubs on diversity and flowering in pastures. Tuexenia 37: 399-413. Kirély G [Ed.] (2009) Uj magyar fiivészkényv. Magyarorsz4g hajtasos ndévényei. Hatarozokulcsok. [New Hungarian Herbal. The vascular plants of Hungary. Identifica- tion key]. Aggteleki Nemzeti Park Igazgatosag, Josvaf6. Knipl |, Simegi P (2012) Life at the interface of two distinct landscapes-relationship of humans and environment in the periphery of the Danube-Tisza Interfluve between Ha- jOs and Csaszartoltés. Open Geosciences 4(3): 439-447. https://doi.org/10.2478/ $13533-011-0071-x Kocsis K [Ed.] (2018) National Atlas of Hungary — Natural environment. Budapest, MTA CSFK Geographical Institute, 183 pp. https://www.nemzetiatlasz.hu/MNA/2_en.html Kérési A, Szentirmai |, Batary P Kovér S, Orvéssy N, Peregovits L (2014) Effects of tim- ing and frequency of mowing on the threatened scarce large blue butterfly—A fine- scale experiment. Agriculture, Ecosystems & Environment 196: 24-33. https://doi. org/10.1016/j.agee.2014.06.019 Krauss J, Klein AM, Steffan-Dewenter |, Tscharntke T (2004) Effects of habitat area, iso- lation, and landscape diversity on plant species richness of calcareous grasslands. Biodiversity and Conservation 13(8): 1427-1439. https://doi.org/10.1023/B:BI- O0C.0000021323.18165.58 Kun R, Bartha S, Malatinszky A, Molnar Z, Lengyel A, Babai D (2019) “Everyone does it a bit differently!”: Evidence for a positive relationship between micro-scale land-use diversity and plant diversity in hay meadows. Agriculture, Ecosystems & Environment 283: €106556. https://doi.org/10.1016/j.agee.2019.05.015 Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 14 Robert Kun et al.: Effects of management complexity on HNV grasslands Kun R, Babai D, Csatho Al, Vadasz C, Kalman N, Maté A, Malatinszky A (2021) Simplici- ty or complexity? Important aspects of high nature value grassland management in nature conservation. Biodiversity and Conservation 30(12): 3563-3583. https://doi. org/10.1007/s10531-021-02262-z Laliberte E, Wells JA, DeClerck F, Metcalfe DJ, Catterall CP Queiroz C, Aubin |, Bonser SP, Ding Y, Fraterrigo JM, McNamara S, Morgan JW, Merlos DS, Vesk PA, Mayfield MM (2010) Land-use intensification reduces functional redundancy and response di- versity in plant communities. Ecology Letters 13(1): 76-86. https://doi.org/10.1111/ j.1461-0248.2009.01403.x MacDonald D, Crabtree JR, Wiesinger G, Dax T, Stamou N, Fleury P Lazpita JG, Gibon A (2000) Agricultural abandonment in mountain areas of Europe: Environmental conse- quences and policy response. Journal of Environmental Management 59(1): 47-69. https://doi.org/10.1006/jema.1999.0335 Marini L, Scotton M, Klimek S, Isselstein J, Pecile A (2007) Effects of local factors on plant species richness and composition of Alpine meadows. Agriculture, Ecosystems & Environment 119(3-4): 281-288. https://doi.org/10.1016/j.agee.2006.07.015 Mathar WP. Kampf I, Kleinebecker T, Kuzmin |, Tolstikov A, Tupitsin S, H6lzel N (2016) Floristic diversity of meadow steppes in the Western Siberian Plain: Effects of abiotic site conditions, management and landscape structure. Biodiversity and Conservation 25(12): 2361-2379. https://doi.org/10.1007/s10531-015-1023-4 Maurer K, Weyand A, Fischer M, Stocklin J (2006) Old cultural traditions, in addition to land use and topography, are shaping plant diversity of grasslands in the Alps. Biolog- ical Conservation 130(3): 438-446. https://doi.org/10.1016/j.biocon.2006.01.005 Molnar Z, Kelemen A, Kun R, Maté J, Safian L, Provenza F, Diaz S, Barani H, Biré M, Maté A, Vadasz C (2020) Knowledge co-production with traditional herders on cattle graz- ing behaviour for better management of species-rich grasslands. Journal of Applied Ecology 57(9): 1677-1687. https://doi.org/10.1111/1365-2664.13664 Molnar AP, Erdélyi A, Hartdégen J, Bird M, Panya |, Vaddsz C (2022) Historical analysis for nature conservation — the past three centuries of the Peszér forest. Journal of Landscape Ecology 20(1): 73-105. https://doi.org/10.56617/tI.3381 Niedrist G, Tasser E, Lith C, Dalla Via J, Tappeiner U (2009) Plant diversity declines with recent land use changes in European Alps. Plant Ecology 202(2): 195-210. https:// doi.org/10.1007/s11258-008-9487-x Ockinger E, Smith HG (2006) Landscape composition and habitat area affects butterfly species richness in semi-natural grasslands. Oecologia 149(3): 526-534. https://doi. org/10.1007/s00442-006-0464-6 Ockinger E, Eriksson AK, Smith HG (2006) Effects of grassland abandonment, resto- ration and management on butterflies and vascular plants. Biological Conservation 133(3): 291-300. https://doi.org/10.1016/j.biocon.2006.06.009 Plieninger T, Héchtl F, Spek T (2006) Traditional land-use and nature conservation in Eu- ropean rural landscapes. Environmental Science & Policy 9(4): 317-321. https://doi. org/10.1016/j.envsci.2006.03.001 Poschlod P Kiefer S, Trankle U, Fischer S, Bonn S (1998) Plant species richness in calcar- eous grasslands as affected by dispersability in space and time. Applied Vegetation Science 1(1): 75-91. https://doi.org/10.2307/1479087 R Core Team (2018) R: A Language and Environment for Statistical Computing. R Foun- dation for Statistical Computing, Vienna. https://www.R-project.org Rac |, Juvancié L, Erjavec E (2020) Stimulating collective action to preserve High Na- ture Value farming in post-transitional settings. A comparative analysis of three Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 15 Robert Kun et al.: Effects of management complexity on HNV grasslands Slovenian social-ecological systems. Nature Conservation 39: 87-111. https://doi. org/10.3897/natureconservation.39.51216 Raunkizer C (1934) The Life Forms of Plants and Statistical Plant Geography; Being the Collected Papers of C. Raunkizer. Schneiders A, Van Daele T, Van Landuyt W, Van Reeth W (2012) Biodiversity and ecosys- tem services: Complementary approaches for ecosystem management? Ecological Indicators 21: 123-133. https://doi.org/10.1016/j.ecolind.2011.06.021 Socher SA, Prati D, Boch S, Miller J, Klaus VH, H6lzel N, Fischer M (2012) Direct and pro- ductivity-mediated indirect effects of fertilization, mowing and grazing on grassland species richness. Journal of Ecology 100(6): 1391-1399. https://doi.org/10.1111/ j.1365-2745.2012.02020.x Szentes S, Sutyinszki Z, Szab6 G, Zimmermann Z, Hazi J, Wichmann B, Hufnagel L, Penksza K, Bartha S (2012) Grazed Pannonian grassland beta-diversity chang- es due to C 4 yellow bluestem. Open Life Sciences 7(6): 1055-1065. https://doi. org/10.2478/s11535-012-0101-9 Szépligeti M, Kérosi A, Szentirmai |, Hazi J, Bartha D, Bartha S (2018) Evaluating alter- native mowing regimes for conservation management of Central European mesic hay meadows: a field experiment. Plant Biosystems-An International Journal Dealing with all Aspects of Plant Biology 152(1): 90-97. https://doi.org/10.1080/11263504. 2016.1255268 Talle M, Deak B, Poschlod P, Valké O, Westerberg L, Milberg P (2016) Grazing vs. mowing: A meta-analysis of biodiversity benefits for grassland management. Ag- riculture, Ecosystems & Environment 222: 200-212. https://doi.org/10.1016/j. agee.2016.02.008 Todlgyesi C, Torma A, Batori Z, Seat J, Popovic M, Gallé R, Galle-Szpisjak N, Erdés L, Vinko T, Kelemen A, Torok P (2022) Turning old foes into new allies — harnessing drainage canals for biodiversity conservation in desiccated novel ecosystems. Journal of Ap- plied Ecology 59: 89-102. https://doi.org/10.1111/1365-2664.14030 Torok P Penksza K, Toth E, Kelemen A, Sonkoly J, Tothmérész B (2018) Vegetation type and grazing intensity jointly shape grazing effects on grassland biodiversity. Ecology and Evolution 8(20): 10326-10335. https://doi.org/10.1002/ece3.4508 Vadasz C, Maté A, Kun R, Vadasz-Besny6oi V (2016) Quantifying the diversifying poten- tial of conservation management systems: An evidence-based conceptual model for managing species-rich grasslands. Agriculture, Ecosystems & Environment 234: 134-141. https://doi.org/10.1016/j.agee.2016.03.044 von Glasenapp M, Thornton TF (2011) Traditional ecological knowledge of Swiss alpine farmers and their resilience to socioecological change. Human Ecology: an Interdis- ciplinary Journal 39(6): 769-781. https://doi.org/10.1007/s10745-01 1-9427-6 Willner W, Rolecek J, Korolyuk A, Dengler J, Chytry M, Janisova M, Lengyel A, Acic S, Becker T, Cuk M, Demina O, Jandt U, Kacki Z, Kuzemko A, Kropf M, Lebedeva M, Se- menishchenkov Y, Silc U, Stanéié Z, Staudinger M, Vassilev K, Yamalov S (2019) For- malized classification of semi-dry grasslands in central and eastern Europe. Preslia 91(1): 25-49. https://doi.org/10.23855/preslia.2019.025 Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 16 Robert Kun et al.: Effects of management complexity on HNV grasslands Appendix 1 oO Turjan region s | | Patches of Turjan region f . Ns + O Study area Hajés FF Csiszintihés ¥ SS Figure A1. Study area and location of the twelve sampled meso-xeric grassland sites (Knipl and Siimegi 2012). Appendix 2 Figure A2. Distances between an elevation of Molinia meadows and fens and a sandy steppe zone. For the field sampling, nine plots were established along two transects in each grassland site. The distance between two transects was at least 10 m and the distance between the quadrats was at least 4 m. Appendix 3 Table A1. Sampling design in the study area with factor combinations at each site and number of replicates. Number of sites | Management type Herbage removal Management Number of management Number of plots intensity complexity factor combinations per site 1 Mown Low Low 1 9 2 Mown Low Low 1 9 3 Mown Low Low 1 9 4 Grazed Low : High Zz 9 5 Grazed Low High 2 9 6 Grazed High Low 3 9 7, Grazed High Low 3 9 8 Grazed High | Low 3 9 9 Grazed High High 4 9 10 Combined Low Low 5 9 11 Combined Low High 6 9 12 Combined High High 7 9 Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 17 Robert Kun et al.: Effects of management complexity on HNV grasslands Appendix 4 Table A2. Differences in PFT cover and diversity indices between different management types (mowing: M, grazing: G and combined: C) of semi-natural grasslands. The Table shows means and standard deviations of PFT groups and diversity indices. Significant differences in LMER Tukey post hoc tests between different management types are in- dicated by the letters ‘a’, ‘b’ and ‘c’. MOWN GRAZED COMBINED Species number 34.143.2a 34.9+5.8a 35.443.5a Shannon diversity 1.640.4a 1.9+0.3b 1.8+0.3b Simpson diversity 0.6+0.2a 0.7+0.1b 0.7+40.1b Forbs cover (%) 19.0411.9a 24.4415.2a 31.1420.2a Graminoid cover (%) 82.2+11.3a 74.3418.2a 57.8+22.2a Phanerophytes cover (%) 2.24+1.7a 5.14+3.8b 4.0+4.0ab Mean plant height (cm) 31.4412.8a 26.7+8.4a 21.0+8.4a Total plant cover (%) 94.0+3.3a 95.6+3.1a 87.6+8.7a Bare soil surface (%) 0.6+0.4a 1.6+1.6a 1.841.9a Litter cover (%) 5.7+3.3a 3.8+2.7a 10.9+7.7a Appendix 5 Table A3. Effects of herbage removal intensity of management on plant diversity and cover of PFTs. Table shows means and standard deviations of PFT cover and diversity indices. Results are based on LMER Tukey post hoc tests. Significant differences be- tween different intensity levels are indicated by the letters ‘a’ and ‘b’. LOW HIGH Species number 34.344.7a 35.644.6a Shannon diversity 1.740.4a 1.840.4a Simpson diversity 0.740.1a 0.7+0.1a Forbs cover (%) 26.4419.1a 22.54+11.3a Graminoid cover (%) 66.9+21.3a 79.5+15.1a Phanerophytes cover (%) 4.44+3.9a 3.7+3.3a Mean plant height (cm) 25.94+10.8a 26.5+9.2a Total plant cover (%) 91.3+6.8a 96.043.3a Bare soil surface (%) 1.3+1.7a 1.6+1.6a Litter cover (%) 8.0+5.9a 3.2+2.4a Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 Robert Kun et al.: Effects of management complexity on HNV grasslands Appendix 6 Table A4. Differences between two levels of management complexity (low and high) on plant diversity and plant functional types. Table shows means and standard deviations of PFT cover and diversity indices. Results are based on LMER Tukey post hoc tests. Sig- nificant differences between different levels of management complexity are indicated by the letters ‘a’ and ‘b’. Species number Shannon diversity Simpson diversity Forbs cover (%) Graminoid cover (%) Phanerophytes cover (%) Mean plant height (cm) Total plant cover (%) Bare soil surface (%) Litter cover (%) Appendix 7 LOW 35.03.8a 1.740.4a 0.64+0.1a 24.24+17.2a 76.3#21.2a 3.442.9a 29.7410.2a 94.44.54 1.0+0.9a 5.243.9a HIGH 34.6+5.8a 1.940.4b 0.7+0.1b 25.6415.3a 66.3416.2a 9.2+4.3a 22.2+8.6a 91.547.5a 1.842.0a 7.346.8a Table A5. Differences in species composition dissimilarity between management types and levels of herbage removal intensity and management complexity, based on PERMANOVA analyses and Jaccard dissimilarity index. Df T 2 Residuals 105 Total 107 I 1 Residuals 106 Total 107 C 1 Residuals 106 Total 107 Appendix 8 33126 28.332 31.458 0.980 30.478 31.458 1.783 29.675 31.458 Sums of Sqs | Mean Sqs 1.563 0.270 0.980 0.288 1.783 0.280 F-test 5.792 3.409 6.368 R2 0.099 0.901 1.000 0.031 0.969 1.000 0.057 0.943 1.000 Pr(> F) 6.067 *** 0.001 *** COO st Table A6. Proportion of principal components expressed by eigenvalues, explained and cumulative proportions and their contribution to the variance. PC1 PC2 PC3 PC4 PC5 PC6 PC7 Eigenvalue 724.490 | 179.844 | 50.093 14.880 7.097 2.952 0.420 Explained share 0.739 0.184 0.051 0.015 0.007 0.003 0.000 Cumulative share 0.739 0.923 0.974 0.989 O97 0.100 1.000 Nature Conservation 55: 1-19 (2024), DOI: 10.3897/natureconservation.55.114385 19