Impact of land use changes on ecosystem services provision in grasslands: a case study from Slovakia

Abstract

Grasslands play a key role in biomass production and provide multiple ecosystem services. This study examines how land-use changes between 2001 and 2022 have affected the provision of ecosystem services in grasslands, reflecting two decades of agricultural transformation in the EU. The research focuses on the Veselianka River Basin in northern Slovakia, using the Common International Classification of Ecosystem Services (CICES) to select and assess relevant services. Land-use dynamics were analysed using GIS, while selected ecosystem services were evaluated using indicator-based approaches under the CICES. Grasslands covered 25.8% of the study area in 2001, declining slightly to 23.5% by 2022, primarily due to land abandonment, settlement expansion, and the development of recreational facilities. Transition analysis showed the largest increases in shrubland/small woodland and built-up areas, indicating progressive restructuring of the rural landscape. Ecosystem services assessment showed that extensively managed grasslands, particularly hay meadows, mesophilic pastures, and species-rich Nardus grasslands, had the highest multifunctional value. These habitats provided strong potential for biomass production, biodiversity conservation, pollination support, and cultural services linked to landscape aesthetics and recreation. In contrast, land abandonment and infrastructure expansion reduced habitat continuity and may weaken long-term ecosystem service capacity. The results highlight the importance of maintaining traditional extensive grassland management to preserve multifunctional rural landscapes. Integrating ecosystem service assessments into agri-environmental policy and spatial planning can support sustainable land management in mountain regions of Central Europe and comparable rural landscapes elsewhere.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-55552-3.

Introduction

Grasslands were often used for biomass production, providing multiple benefits and ecosystem services essential to both human society and the environment. Grasslands are globally threatened, either through intensification or, conversely, through agricultural abandonment^1–3. Most grassland habitats in Slovakia were transformed in the past to high-productive grasslands with low species richness to increase production⁴. However, after 1990, the collapse of socialism and the introduction of the Common Agricultural Policy (CAP) in 2004 changed grassland management, enabling the restoration of grassland biodiversity and an increase in plant diversity in some marginal regions of the Carpathian Mountains, through regular but less intensive management^5,6.

Ecosystem conditions affect the recovery and supply of ecosystem services⁷. The landscape with more natural and semi-natural habitats generally provides more diverse and higher-quality services than intensified systems^8–10. Grasslands deliver three categories of ecosystem services: provisioning, regulating, and cultural values, including forage and primary production; erosion and water regulation; climate regulation; biodiversity conservation; recreation; cultural heritage; and aesthetic values^11–14.

The general ecosystem benefits of grasslands are well understood; however, studies quantifying ecosystem services and benefits in greater detail are scarce¹⁵. Several indicators are used to assess ecosystem services. Valuation approaches include ecological methods that assess landscape characteristics, and economic methods that estimate values based on consumer preferences¹⁶. Recent studies have employed progressive methods using complex spatial units to define and map ecosystem services^17–19.

This study aims to assess changes in grassland land use between 2001 and 2022 and their effects on selected ecosystem services in the Veselianka River Basin (northern Slovakia). This period reflects two decades of socio-economic and policy-driven transformation in mountain agricultural landscapes, including developments following Slovakia’s EU accession and the implementation of related agricultural and environmental policies. The specific objectives are: (1) to quantify selected ecosystem services using indicator-based approaches and analyze spatial and temporal patterns of human activity affecting grasslands; and (2) to identify the main drivers and mechanisms underlying changes in ecosystem service provision.

Despite the growing body of research on ecosystem services, relatively limited attention has been given to long-term grassland dynamics and their ecosystem service implications in mountainous and sub-mountainous regions of Central Europe. By integrating land-use change analysis with an indicator-based assessment of ecosystem services at the regional scale of the river basin study area, this study provides new insights into how grassland management and landscape transformation influence ecosystem service provision over time. The results contribute to improving the regional understanding of ecosystem service dynamics and provide a methodological framework that can support landscape management, conservation planning, and sustainable agricultural policy in similar regions.

Study area

The research was conducted in the Veselianka River Basin, located in the Oravské Beskydy Mountains of the Western Carpathians in northern Slovakia. This region is characterized by a relatively high proportion of grassland and is undergoing land use changes. These dynamics, together with the availability of detailed long-term spatial and field data, make the area an ideal case study for analysing the relationships between land-use change and ecosystem service provision in Central European grassland landscapes. Except for the built-up areas of villages, most of the basin lies within the Horná Orava Protected Landscape Area. The study area covers 9,156.6 ha, of which approximately 40% is agricultural land dominated by grasslands, the primary focus of this study, while higher elevations are predominantly forested. The basin includes three municipalities: Oravská Jasenica, Oravské Veselé, and Mutné.

The region’s natural characteristics are shaped primarily by abiotic factors, including a dense hydrological network, rugged terrain, and variable slope gradients. Morphologically, it forms a rugged highland ranging from 630 to 1,557 m above sea level. Based on soil, climate, and topography, the area belongs to the mountain and sub-mountain region, a moderately cool, humid sub-region influenced by altitude, location, prevailing winds, and forest cover²⁰. Annual precipitation ranges from 800 to 1,200 mm. The dominant soil-forming substrates are flysch shale and sandstone, with acidic to slightly acidic soils, and some neutral areas with high carbonate content. Cambisols and their subtypes constitute 75.4% of the soils.

The area represents a typical sub-mountain agricultural region with a long-standing tradition dating back to the sixteenth century. Grasslands cover more than 20% of the area, and have historically been managed through grazing and mowing. Farming has been practiced and adapted to local climatic conditions. According to the Land Parcel Identification System²¹, land is cultivated by two large agricultural cooperatives, four smaller farms, and several private farmers (Oravská Jasenica – 9 farmers; Oravské Veselé – over 30; Mutné – 7). The dominant production system is livestock farming, with some crop cultivation, including cereals, silage maize, and perennial fodder crops (Fig. 1). Grasslands provide the main fodder base for cattle and horse breeding, while smaller farms focus on sheep and goat husbandry.

Fig. 1.

The study area of the Veselianka river basin and its localization within Slovakia (Source: The map was extracted in ArcMap 10.8.2 from the open-access WMTS layer of the ZBGIS of the Slovakia at the zoom level corresponding to the study area.²²).

Methods

This study integrates spatial analysis, field-based vegetation assessment and indicator-based evaluation of ecosystem services to examine the effects of land-use changes on grassland ecosystems (Fig. 2).

Fig. 2.

Methodological workflow of the study, illustrating the integration of spatial land-use analysis, field-based vegetation assessment, and indicator-based evaluation of ecosystem services.

Data collection

We assessed the impact of land-use changes between 2001 and 2022 on the provision of grassland ecosystem services in the Veselianka River Basin. Historical and current land-use maps were created in QGIS by integrating historical data (a land use map from 2001²³), manual polygon delineation from aerial photographs (2001, 2024), and field verification of land use categories. Land cover was classified into six main categories with subcategories as follows: 1. Forest; 2. Shrubland and small woodland (in an open agricultural landscape); 3. Floodplain woods; 4. Grassland (41. Grassland with trees, 42. Managed grassland); 5. Farmland (51. Narrow-strip mosaic of arable fields and grassland, 52. Farmland mosaic with woods, 53. Large-block arable fields); 6. Built-up area (61. Settlement, 62. Dispersed settlement; 63. Transport infrastructure).

Grassland habitats were investigated using our own phytocoenological records, following the Zurich-Montpellier methodology. Twenty-seven 5 × 5 m² plots were surveyed in 2001 and resampled in 2022 at the same coordinates. Plots were deliberately located to capture the full biodiversity of vascular plants, with only one record per homogeneous community (i.e., where roads or woody habitats divided similar communities). All taxa were consistently identified to species level, and plot data were stored in the TURBOVEG database²⁴ for further assessment. Habitats were classified according to the habitat catalog²⁵ and aligned with the EU Habitat Directive²⁶:

• Lk1 – Medium-altitude semi-natural hay meadows (Arrhenatherion elatioris; Habitat Directive (HD) – 6510; 3 records). Mesic, traditionally mown meadows on moderately nutrient-rich soils. Dominated by mesophyll-perennial grasses, with diverse forbs. Regular mowing maintains high species richness and prevents woody encroachment.

• Lk3 – Mesophilous pastures (Cynosurion cristati; 10 records): Structurally and compositionally distinct from hay meadows due to continuous grazing, though occurring under similar environmental conditions.

• Lk6 – Moist to wet mesotrophic–eutrophic hay meadows (Calthion palustris; 5 records): occupy periodically inundated, nutrient-rich floodplain soils. Herb layer dominated by tall, light-demanding species, with sedges and rushes (Carex, Juncus, Scirpus spp.), with minimal bryophyte cover.

• Tr8 – Species-rich Nardus grasslands on siliceous substrates (Nardo-Agrostidion; HD 6230; 8 records): Occur on acidic, nutrient-poor soils, forming low, dense swards dominated by Nardus stricta alongside montane to sub-montane grassland taxa. Extensive management promotes structural heterogeneity through scattered shrubs and trees (e.g. Picea abies, Juniperus communis).

• Kr1 – European dry heaths (Euphorbio cyparissiae–Callunion vulgaris; HD 4030; 1 record): Calluna vulgaris dominated heaths on acidic, oligotrophic soils, often with Vaccinium species, bryophytes, and lichens. Extremely rare in the study area, recorded only on steppe slopes in Oravská Jasenica. Due to a single record, it was excluded from quantitative analysis but included to highlight habitat rarity and vulnerability.

Data analysis

Land-use changes between 2001 and 2022 were analysed using geoprocessing tools in QGIS, specifically the Union tool, to compare land-use element composition, area, and land use element counts. Landscape diversity was quantified using edge density (ED) as a measure of spatial heterogeneity²⁷. Edge density is calculated as the ratio of total edge length (E, in meters) to total area (A, in hectares), ED = E/A. High edge density indicates numerous small or irregularly shaped patches, whereas low edge density reflects larger, more continuous patches. The statistical analysis of land use changes was performed in R, including a Paired t-test, percentage change per class, a one-way ANOVA test for change magnitude, and Tukey’s HSD.

Principal component analysis (PCA) performed by CANOCO program²⁸, was used to examine patterns of floristic similarity among sampling plots and to identify the main ecological gradients structuring grassland vegetation.

Ecosystem services were assessed using a biophysical approach that combined qualitative indicators and field data, following the CICES¹⁵. Ecosystem services were selected based on ecological relevance, their measurability with available field indicators, and their importance for local stakeholders. Priority was given to services directly linked to vegetation structure, biodiversity, and human use. Selected ecosystem services, along with their determinants and indicators, are summarized in Table 1.

Table 1.

Selected grassland ecosystem services in the Veselianka River Basin based on the CICES classification¹⁵.

	Division	Group	Class	Ecosystem services (Short descriptor)	Determinants/ Indicators
Provisioning (biotic/ biophysical)	Biomass	Cultivated terrestrial plants for nutrition, materials, or energy	P1. Plants (terrestrial) for nutrition, materials— fodder biomass, or medicinal use	The population of plants that can be used for breeding	The forage value of plants that increases yields Share of medicinal herbs
Regulation & Maintenance (Biotic/ Biophysical)	Regulation of physical, chemical, and biological conditions	Lifecycle maintenance, habitat, and gene pool protection	R1. Biodiversity promotion—maintaining nursery populations and habitats or breeding grounds (including gene pool protection)	Providing habitats for wild plants and animals that are threatened by intensive management	Habitat provision for wild plants and animals; presence of key species and ecological conditions
			R2. Pollination potential	Providing a habitat for native pollinators	The abundance of melliferous species, attractive to pollinators
Cultural (Biotic/ Biophysical)	Physical and experiential interactions with the natural environment	Direct, in-situ, and outdoor interactions with living systems that depend on presence in the environmental setting	R4. Recreation—elements of living systems that enable activities promoting health, recuperation, or enjoyment through recreation	Using the environment for sport and recreation	Attractiveness of the landscape for recreation (demand for recreation houses)

The species attributes^29,30 were applied to assess ecosystem services. Species cover was used to weight each determinant’s contribution per phytocoenological record, and mean values were calculated for each habitat type.

P1. Biomass – Wild plants (terrestrial) for nutrition and materials. Fodder biomass was quantified as the proportion of plants with different forage values, weighted by species cover. Values were categorized as positive (1–5), neutral (0), or negative ( − 1 to − 3). Medicinal value was assessed as the share of species used in pharmacy or common herbal medicine.

R1. Biodiversity promotion was evaluated based on four criteria²⁹:

• Species richness was expressed as the number of species as the simplest way of expressing species richness.

• Naturalness indicates the originality of the species composition or the degree of human intervention on habitats, expressed by the proportion of native versus non-native species, classified into five categories²⁹: A: native species with occurrence in natural habitats, B: native species with occurrence in natural or relatively far from natural habitats, C: archaeophytes, plants introduced in the area before 1500 A.D., often naturalized, and E: neophytes species (non-native with invasive status) according to Medvecka et al.³¹.

• Nature conservation significance indicates the proportion of protected, threatened, or rare plant species in individual habitats, determined by calculating a weighted conservation index E³², Red List status³³, the list of endemics³⁴, and the list of protected plants³⁵.

• Conservation status (expressed by conservation index) reflects Habitat quality assessed according to national regulations³⁵ and assesses their favourable conservation status: favourable conditions—maintained (A), good or recovered (B), unfavourable condition—partially destroyed (C) or destroyed (D).

The sum of these scores was used to classify biodiversity promotion into five categories, from very low to very high (Table 2). The thresholds for individual determinants were adopted according Halada²⁹ and the distribution of the underlying data.

Table 2.

Determinant values for biodiversity promotion.

Category	Species richness (Number of species)	Naturalness (proportion of naturalness categories)	Nature conservation significance (conservation index)	Conservation status (index Cs)	Overall biodiversity promotion (points)
Very low	up to 20	Anthropophytes more than 50%	0	0	5–8
Low	20–30	20–50% Anthropophytes	1–3	1–6	8.1–11
Medium	31–40	5–20% Anthropophytes	4–7	7–12	11.1–14
High	41–50	Up to 5% of Anthropophytes	8–15	13–20	14.1–17
Very high	More than 50	Only native species (A)	15 and more	20 and more	17.1–20

R2: Life cycle maintenance/pollination. Pollination potential was expressed as the percentage of melliferous species, weighted by nectar and pollen resources (scale 1–4), and did not account for spatial and temporal variability in flowering.

C1. Recreation & Tourism: Demand for recreation was indirectly reflected in accommodation demand, as evidenced by the number of cottages (weekend houses) built, often on former hayloft plots.

Results

Land use changes

Grasslands covered 25.8% of the study area in 2001. By 2022, the area of grassland without trees had decreased by more than 60 ha, primarily due to land abandonment, which resulted in the expansion of shrubland and small woodland in the open countryside, as well as an increase in built-up areas (Table 3, Fig. 3). Although the total forest area remained relatively stable, the number of forest patches increased, indicating a higher degree of forest fragmentation. Structural changes were also observed within forest habitats, where new forest stands developed through the natural succession of shrubland, small woodland, and formerly open grasslands that gradually became tree-dominated. Conversely, some forest areas were converted to shrubland following the windstorm in Oravská Jasenica and Pilsko Peak in 2004 and 2017, which caused significant disturbance to forest stands in the region. The land-use change matrix further indicates an overall decline in agricultural land, particularly arable land. In contrast, an increase was observed in built-up areas, both within existing settlements and in the surrounding rural landscape. This expansion is associated with the development of recreational infrastructure, such as huts and cottages, representing dispersed settlement patterns that have replaced former haylofts in grassland areas.

Table 3.

Changes in landscape structure between 2001 and 2022 (area in ha/% of the study area).

2001	2022
	Forest	Shrubland and small woodland	Floodplain woods	Grassland with trees	Grassland	Mosaic of arable fields and grassland	Farmland mosaic with woods	Large-block arable fields	Built-up area	Dispersed settlement	Area Total (2001) ha/ %	Count (2001)
Forest	4641.2	119.2	0	2.8	0	0	0	0	0	1.5	4765,4/ 52%	42
Shrubland and small woodland	37.1	149.2	0	0	0	8.6	0	0	0	1.2	196,1 2.1%	265
Floodplain woods	0	0	270.4	0	0	0	0	0	0	0	270,4 3%	32
Grassland with trees	71.1	86.8	0	421.5	0	2.4	0	0	2.6	6.5	590,9 6.5%	140
Grassland	1	9.1	0	35.2	1685.4	19.5	0	0.7	7	6.8	1764,7 19.3%	167
Mosaic of arable fields and grassland	0	5.1	0.3	0	4.8	906.5	0	0	70.8	0.6	988,0 10.8%	76
Farmland mosaic with woods	0	0	0	0	0	0	28.9	0	0.3	0	29,2 0.3%	7
Large-block arable fields	0	0	0	0	0	1.6	0	118.4	10	0	129,9 1.4%	8
Built-up area	0	0	0	0	0	0	0	0	407.7	0.8	408,5 4.5%	35
Dispersed settlement	0	0	0	0	0	0	0	0	0	13.7	13,7 0.1%	2
Area total (2022) ha/ %	4750.4 51.9%	369.4 4%	270.7 3%	459.5 5%	1690.3 18.5%	939.1 10.3%	28.9 0.3%	119.1 1.3%	498.4 5.4%	31.1 0.3	9156.6 100%	2775
Count (2022)	57	303	32	119	162	93	5	7	53	125	2978

Fig. 3.

Land cover of the Veselianka river basin in 2001 and 2022 (created by author³⁶).

Among the analysed land-cover categories, floodplain woods were the most stable landscape element. These habitats form narrow, linear stands of trees along streams and are typically surrounded by large blocks of grassland and arable land. Analysis of landscape structure using the edge density index²⁷ showed a slight decrease, from 144.435 m/ha in 2001 to 137.815 m/ha in 2022. This decline suggests a modest reduction in spatial heterogeneity within the landscape mosaic during the study period.

The results of a paired t-test indicated no statistically significant overall difference in land-use areas between 2001 and 2022 (t = − 0.0004, p = 0.9997). Similarly, one-way ANOVA did not reveal significant differences between the two years (F = 2.39 × 10⁻¹⁰, p = 0.99999), and Tukey’s HSD post-hoc test confirmed the absence of significant pairwise differences (p = 1.0). Despite the lack of statistical significance at the aggregated landscape level, the transition matrix analysis revealed considerable dynamics among individual land-use classes.

The largest increases were recorded for shrubland and small woodland (+ 173 ha) and built-up areas (+ 107 ha), while the most pronounced losses occurred in grassland with trees (− 205 ha). Persistence analysis showed that forest remained the most stable land-cover class, maintaining approximately 4641 ha throughout the study period. Overall, these trends indicate ongoing landscape restructuring driven by a combination of urban expansion and natural vegetation succession, reflecting both intensification of human land use and localized abandonment of traditional agricultural management.

Characteristics and changes in grassland habitats

Grassland vegetation in the study area consisted of several habitat types, including medium-altitude hay meadows (Lk1), mesophilic pastures (Lk3), moist or wet mesotrophic to eutrophic hay meadows (Lk6), species-rich Nardus grasslands on siliceous substrates in mountain areas (Tr8), and European dry heaths (Kr1) (table with the records in Online Appendix I, Supplementary Information). These communities reflect a range of ecological conditions and management regimes, including regularly mown meadows, grazed pastures, and semi-natural grasslands located near forest margins. Comparison of vegetation records between the two sampling periods (2001 and 2022) indicates noticeable changes in species composition and habitat structure in several sampling plots. The ordination diagram of PCA (Fig. 4) shows a clear separation of grassland habitats along these two axes, reflecting differences in vegetation structure, management intensity, and environmental conditions. The first PCA axis represents a gradient of management intensity and vegetation structure gradient. Plots located on one side of the axis are associated with regularly managed hay meadows, particularly Medium-altitude hay meadows (Lk1), which are typically dominated by competitive grasses such as Arrhenatherum elatius. In contrast, plots located on the opposite side of Axis 1 show increased representation of shrub and tree species, including Picea abies, Rosa canina, and Prunus spinosa. The presence of these species indicates reduced management intensity and the early stages of secondary succession, in which grassland habitats gradually transition toward shrubland or forest vegetation. The second PCA axis represents a floristic and ecological differentiation gradient among grassland habitat types, primarily associated with variation in environmental conditions, including soil moisture, nutrient availability, and substrate characteristics. This axis separates several habitat types (Lk3, Lk6, Tr8, Kr1), which differ substantially in species composition and ecological requirements.

Fig. 4.

Principal component analysis (PCA) of grassland vegetation recorded in two sampling periods (2001 – record number 101-127; and 2022 – record number 201-227). Each point represents a vegetation plot. Habitat types include Medium-altitude hay meadows (Lk1), Mesophilic pastures (Lk3), Moist or wet mesotrophic to eutrophic hay meadows (Lk6), Species-rich Nardus grasslands on siliceous substrates (Tr8), and European dry heaths (Kr1).

The PCA also reveals temporal shifts in vegetation composition between the two sampling periods. Several sampling plots recorded in 2001 shifted toward the ordination space occupied by regularly managed hay meadows in 2022, indicating increasing similarity to communities dominated by productive meadow grasses. This trend likely reflects changes in grassland management practices, particularly the influence of agri-environmental subsidy schemes promoting regular mowing or combined mowing and grazing. At the same time, some sampling plots located near forest margins moved toward the part of the ordination space characterized by shrub and tree species. This pattern suggests localized abandonment of traditional management practices, resulting in the gradual expansion of woody vegetation and the early stages of forest succession.

Assessment of provisioning ecosystem services

Provisioning (biotic) ecosystem services primarily involve terrestrial plants used for nutrition, specifically for grazing animals and mowing hay for winter use.

P1. Biomass: Plants (terrestrial) for fodder biomass

Fodder biomass for domestic or game animals was expressed by the forage values of plants composing different types of habitats (Fig. 5). The feeding value of above-ground plant parts is determined by water content, chemical composition, content of essential oils, toxic substances, vitamins, odour, morphological and anatomical texture (hairiness, spiciness, woodiness), and also by accessibility to animals, taste, resistance to fungal diseases, and other criteria³⁰.

Fig. 5.

Biodiversity promotion value of grassland habitats in Veselianka river basin. Lk1 Medium-altitude hay meadows; Lk3 Mesophilic pastures; Lk6 Moist or wet mesotrophic to eutrophic hay meadows; Tr8 Species-rich Nardus grasslands on siliceous substrates in mountain areas; Kr1 European dry heaths.

We observed the highest forage value in hay meadows and mesophilic pastures, due to the high proportion of species with high feeding value (80–86%) in both studied periods, with a higher overall value in 2022. The increase in positive feeding value for moist or wet mesotrophic to eutrophic hay meadows (Lk6) and species-rich Nardus grasslands on siliceous substrates in mountain areas (Tr8) reflects changes in grassland species. The decrease in forage value (from 81 to 73% of species with high feeding value) in heathland is due to an increase in the cover of dominant woody species (Calluna vulgaris, Vaccinium sp.)

The presence of medicinal herbs increased in hay meadows from 14% of species with medical value in 2001 to 19% in 2022. In other habitats, no significant changes in the proportion of medicinal herbs were observed.

Regulatory & maintenance ES – R1. Biodiversity promotion

The summary of the index of biodiversity promotion, evaluated by using four criteria: diversity, naturalness, vulnerability, and conservation status of habitats, is shown in Fig. 6. The overall biodiversity habitat value decreased in all studied habitats except for a slight increase in wet meadows (from 14 to 14.8) and medium-altitude hay meadows (from 13.7 to 14).

Fig. 6.

Changes in the proportion of species composition of grassland habitats with forage value and medicinal value recorded in 2001 and 2022. Lk1 Medium-altitude hay meadows; Lk3 Mesophilic pastures; Lk6 Moist or wet mesotrophic to eutrophic hay meadows; Tr8 Species-rich Nardus grasslands on siliceous substrates in mountain areas; Kr1 European dry heaths.

The overall value of biodiversity promotion was impacted by individual indicators. The highest species richness, expressed as the number of species, was recorded in species-rich Nardus grasslands and wet meadows.

For naturalness, almost all habitats were characterized by high degrees of autochthonous species, and we did not observe significant differences between communities. We distinguished habitats with natural species composition and habitats with almost or mostly natural species composition, in which some anthropophyte species occurred sporadically.

The vulnerability of occurrence of vulnerable and endangered species (a total of 13 species) significantly influenced the final biodiversity promotion value. The highest presence of vulnerable and rare species was observed in wet hay meadows and species-rich Nardus grasslands. The most significant is the occurrence of the Carpathian endemic Campanula serrata, which is also listed as a species of European importance²⁶. It was recorded in two records, and grows in the mountain pastures and grasslands between the villages of Oravské Veselé and Oravská Jasenica.

The overall conservation status of the studied habitats was favorable; most of the habitats are included among natural habitat types listed in Online Appendix I of Habitat Directive²⁶. Mesophilic pastures and wet meadows are significant only on a national level.

R2: Life cycle maintenance/pollination

Pollination potential was expressed based on the nectar and pollen stock profiles of melliferous plants within individual habitats. Species with very good pollen and nectar reserves include Trifolium repens, T. pratense, Taraxacum officinale, Calluna vulgaris, Petasites albus, and Melilotus albus. Species with a weak or medium pollen or nectar supply were the most frequent. Species with undetermined (not pollinated by insects) or absent value (for nectar, e.g., grasses) were also numerous.

The overall assessment of pollination potential showed that most habitats had a high abundance of species with nectar and pollen reserves (Fig. 7). However, its value varied over time, depending on the flowering period of each species. The highest honey-bearing potential, with very high values, was found in hay meadows, mesophilic pastures, and species-rich Nardus grasslands.

Fig. 7.

Pollination capacity expressed by the species composition and their value for honey nectar and pollen production. Lk1 Medium-altitude hay meadows; Lk3 Mesophilic pastures; Lk6 Moist or wet mesotrophic to eutrophic hay meadows; Tr8 Species-rich Nardus grasslands on siliceous substrates in mountain areas; Kr1 European dry heaths.

In contrast, the lowest potential (medium value) was found in wet meadows, where a greater number of species with missing pollen and nectar reserve values (species of the genus Carex, Juncus, Luzula, Dactylorhiza, and others) occurred.

Cultural ES

C1. Recreation & tourism

The study area consists of three compact rural settlements, surrounded by a mosaic of meadows, forests, and fields. The open agricultural landscape is dominated mainly by meadows, grassland with trees, and farmland mosaics with woods, creating a beautiful landscape scenery with high aesthetic value. Several hiking and cycling trails lead through the area. Pilsko Hill (1557 m above sea level), the second-highest peak of the Orava Beskydy Mountains, is the most popular hiking goal. The individual settlements are connected by cycle routes—the Beskydská magistrála, which runs through the basin, as well as side routes that follow the rugged terrain and the ridge of the study area. Forests and small woods are a popular destination for mushroom pickers and for collectors of other forest fruits.

We have recently observed an increase in demand for recreation and cultural ecosystem services, as indicated by proxies such as the number of huts in the open landscape. Over the past 20 years, more than 120 cottages have been built on the foundations of the former haylofts, surrounded by feature meadow pastures with scattered trees. Cottages are characterized by diverse architecture and sometimes disturb the landscape scene (Fig. 8).

Fig. 8.

Expansion of new huts impacts the landscape scenery (Photo by Jana Špulerová, 2022, Oravská Jasenica).

The construction of huts and the demand for temporary housing outside villages cause landscape fragmentation, as documented by the overall increase in land use elements and the decline in grazing areas (Table 3).

Discussion

Ecosystem services are influenced by landscape structure, spatial configuration, geographic context, and land use management intensity. Our results show that the diversity of different non-forest habitats, in combination with surrounding forests, supports high levels of plant and animal species richness and contributes to the maintenance of key ecological functions. The heterogeneous mosaic of grassland, small forests, and agricultural land creates a refuge of biodiversity in the agrarian landscape and provides important regulating and supporting ecosystem services.

The findings confirm broader trends observed in non-forest habitats across Central European landscapes, where grassland abandonment leads to natural succession and the gradual expansion of seminatural forests. These processes may increase ecosystem stability, but they are associated with a decline in species-rich semi-natural grasslands and associated ecosystem services, particularly those linked to biodiversity and pollination^37,38. Assessing land-use changes over the past 20 years was particularly important, as we focused on the most recent and rapid transformations in rural landscapes driven by socio-economic and policy shifts, and integration into broader policy frameworks such as the EU CAP. Evaluating changes over this period, therefore, provides critical insights into how contemporary drivers shape ecosystem service provision and biodiversity patterns, while also offering a relevant basis for informing current land management and policy decisions. Our findings supported the idea that extensively managed habitats, such as hay meadows, mesophilic pastures, and species-rich Nardus grasslands, had the highest potential for ecosystem service provision and were associated with agri-environmental management practices supported by the CAP. Traditional management practices are maintained in certain Central European regions^39–41, and extensive management intensity can enhance biodiversity, climate regulation, and water purification, even though it may trade off feed quality⁴². At the same time, increasing recreation development pressure, particularly through the expansion of recreational cottages and infrastructure, introduces additional drivers of landscape change and leads to fragmentation. These interacting processes highlight the importance of understanding the balance between traditional land management, grassland ecosystem service provision, and emerging socio-economic pressures for maintaining a healthy, sustainable rural environment and for sustainable agriculture and beekeeping.

Changes in land use are also reflected in shifting patterns of human interaction with the landscape. In the past, grasslands were primarily used for biomass production, and human presence was largely seasonal, linked to activities such as mowing or grazing. In contrast, the expansion of recreational housing has increased the time people spend in the countryside, particularly during weekends and holidays. While this increased presence may place additional pressure on natural resources, including soil and water, erosion processes it also highlights the growing importance of cultural ecosystem services associated with recreation, well-being, and human–nature interactions. Several studies have demonstrated that contact with natural environments provides significant benefits for human health, psychological well-being, and social relationships^43,44.

A key strength of this study lies in integrating long-term spatial analysis with field-based ecological data. The use of permanent plots allows for direct temporal comparison, reducing uncertainty associated with space-for-time substitutions commonly used in similar studies. By combining land-use mapping with repeated vegetation surveys, the study captures both landscape-scale transformations and fine-scale ecological responses. Balancing socioeconomic development with the maintenance of ecosystem service provision, therefore, represents a challenge for future regional development, ecological protection, and ecological restoration while eliminating negative environmental impacts. The regulation and management of ecosystem services are complex and require scientifically sound, widely understandable policies and governance models, based on detailed assessment methods¹⁸. The scenario-based studies of land use changes in the Hunshandake region demonstrated that prioritizing ecosystem services can provide valuable insights to inform land use decisions and facilitate sustainable development⁴⁵. Similarly, adaptive grassland management and restoration can help maintain or enhance the multifunctional benefits provided by grassland ecosystems^46,47.

Despite these strengths, several methodological limitations should be considered. The indicator-based approach used to assess ecosystem services relies primarily on vegetation characteristics as proxies, reflecting the potential of ecosystem service capacity rather than actual service delivery. These indicators are primarily derived from vegetation composition, habitat characteristics, and expert-based scoring, which introduces a degree of subjectivity and uncertainty into the assessment. Additionally, ecosystem services such as fodder production, pollination, or recreational value can vary significantly depending on seasonal conditions, management intensity, and interannual climatic variability. Consequently, the results should be interpreted as relative estimates of ecosystem service potential across habitat types rather than precise quantitative values. This highlights the need for more standardized and quantitative approaches, as suggested by recent research integrating field data with spatial modeling and remote sensing techniques^18,19.

Assessing cultural ecosystem services poses an additional challenge. In this study, recreation was approximated using proxy indicators such as the number of cottages, reflecting increasing demand for rural tourism. However, cultural services are affected by various environmental determinants, such as accessibility, landscape perception, and socio-economic factors, that extend beyond the indicator of demand for recreational houses. As a result, the applied approach captures only part of their variability and may underestimate their broader societal significance.

Overall, the findings contribute to a growing body of evidence demonstrating that extensively managed grasslands provide multiple ecosystem services, including biodiversity conservation, pollination support, and biomass production, while also supporting cultural values. Future research should integrate vegetation-based indicators with quantitative ecological measurements, spatial modelling, and long-term monitoring in order to better understand ecosystem service dynamics under changing land-use conditions. Such approaches may help support evidence-based landscape planning and contribute to the sustainable management of agrarian landscapes in Central Europe.

Conclusion

This study demonstrates that changes in land use within the Veselianka River basin between 2001 and 2022 have influenced both landscape structure and the capacity of ecosystems to provide multiple services. Although the overall statistical differences in land-use areas were not significant, the transition matrix analysis revealed important shifts in landscape composition, particularly the decline of grasslands and the expansion of shrubland, small woodland, and built-up areas. These findings highlight the critical role of land-use dynamics in shaping the capacity of grassland ecosystems to deliver multiple ecosystem services. The observed trends reflect broader processes occurring across European rural landscapes.

The assessment of ecosystem services based on vegetation indicators showed that extensively managed grasslands—such as hay meadows, mesophilic pastures, and species-rich Nardus grasslands—have the highest potential to supply ecosystem services. The results emphasize the ecological importance of extensively managed grasslands, which provide key ecosystem services including biodiversity conservation, pollination support, biomass production, and soil protection while also contributing to cultural ecosystem services associated with landscape aesthetics and recreation. In contrast, land-use changes associated with succession processes and infrastructure development may alter ecosystem service provision, particularly where habitat fragmentation and soil disturbance occur. At the same time, increasing demand for cultural ecosystem services associated with recreation and rural tourism is reshaping the relationship between people and the landscape. Maintaining the multifunctionality of rural ecosystems will therefore require balanced land management strategies that support traditional grassland management while regulating development pressures. Integrating ecosystem service assessments into landscape planning and agri-environmental policies may help guide sustainable land management and protect the multifunctional benefits provided by grassland ecosystems. These findings underscore the importance of targeted agri-environmental measures, landscape planning, and regulatory tools that balance production, conservation, and recreation. Supporting traditional low-intensity management practices, while guiding spatial development, will be essential to maintaining the multifunctionality of rural landscapes and ensuring the long-term resilience of ecosystem services.

Although this study focuses on a mountain landscape in Central Europe, the observed trends reflect broader processes affecting rural landscapes worldwide, including agricultural abandonment, land-use intensification, and increasing demand for recreational ecosystem services. Understanding these interactions is essential for developing management strategies that enhance ecosystem resilience and support sustainable rural development under changing environmental and socio-economic conditions. The study advances methodological approaches for linking land-use change with ecosystem service assessment and provides a foundation for future research aimed at improving the quantification and modeling of ecosystem service flows under changing environmental conditions.

Abbreviations

ANOVA

Analysis of variance

APA

Agricultural Paying Agency

CAP

Common Agricultural Policy

CICES

Common International Classification of Ecosystem Services

ED

Edge density

ES

Ecosystem services

EU

European Union

GIS

Geographic Information System

HD

Habitat Directive (Council Directive 92/43/EEC)

LPIS

Land Parcel Identification System

PCA

Principal component analysis

QGIS

Quantum Geographic Information System

R

R statistical software

TURBOVEG

Vegetation database software for phytosociological data

Author contributions

JS wrote the main manuscript, JM made analyses of land use changes and figures, SD did PCA analyses.

Funding

This work was supported by the Scientific Grant Agency of the Ministry of Education of the Slovak Republic [No. 2/0062/25, "Let’s bring flowers back to the meadows – changes and ecological restoration of the diversity of rare meadow habitats of European importance"].

Data availability

The datasets used and analysed during the current study available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.