Introduction
Forests, with their complex structures and dynamic interactions, provide a rich array of substrates and microhabitats for bryophytes (Müller et al. 2019). In forest ecosystems, bryophytes, along with other cryptogamic organisms, establish micro-communities on specific substrates such as exposed soil, the trunks of living trees, decaying wood, and rocky surfaces (Mölder et al. 2015; Stefańska-Krzaczek et al. 2022). Mosses and liverworts play an important role in forest ecosystems, with major contributions to the regulation of humidity and water levels (Zhang et al. 2023), as well as soil formation and protection (Ingerpuu et al. 2019; Rola et al. 2021; Gall et al. 2022). They form essential microhabitats for a diverse spectrum of organisms (Božanić et al. 2013; Peck 2006), and act as substrates for the seed bank of vascular plants and shelter for germinating plants (Rambo and Muir 1998; Glime 2024). In environments with minimal competition from phanerogamic plants, such as bare soil, rocky substrates, tree bark, and decaying wood, mosses and liverworts thrive well.
Forest bryophytes, due to their high sensitivity to various environmental conditions, could serve as valuable indicators. (Dierßen 2001; Oishi and Morimoto 2016; Czerepko et al. 2021; Diekmann et al. 2023; Van Zuijlen et al. 2023). Ongoing climate change is leading to prolonged periods of drought and reduced moisture levels in forest soils (Hänsel 2020; Nalevanková et al. 2020). Therefore, the diversity and distribution of bryophytes are often impacted (He et al. 2016; Táborská et al. 2015; Ingerpuu et al. 2019). Studies on species diversity and its changes over time can also indicate numerous anthropogenic influences, such as forest fragmentation (Pharo et al. 2004; Löbel et al. 2018), fertilization, wood extraction, or road construction (Friedel et al. 2006; Hofmeister et al. 2015b; Horvat et al. 2017; Bourgouin et al. 2022).
Advertisement
Large forest areas entirely missing traces of human activity are extremely rare in present-day Europe (Paciorek et al. 2016). The majority of woodlands exhibit features associated with managed and cultural forests, where multifunctional to intensive management is prevalent (Nabuurs et al. 2019). This leads to a decline in the species richness and diversity of forest bryophytes (Vellak and Ingerpuu 2005; Hofmeister et al. 2015b; Fojcik et al. 2019). However, the availability of various substrates in managed forests with diversified tree structures can result in bryophyte compositions similar to those in natural forests (Hofmeister et al. 2015b; Müller et al. 2019). Therefore, managed forests could be crucial habitats for a wide spectrum of species and provide important research sites for studying the impacts of human activities on ecosystems in the Central European landscape. Appropriate management can preserve a specific stage of forest development, while natural succession can result in forest homogenization (Santoro and Piras 2023).
The study of bryophyte distribution in various forest habitats helps to understand the dynamics and development of these ecosystems and provides valuable information for further research on their diversity and disturbances (Mölder et al. 2015; Stefańska-Krzaczek et al. 2022). In Slovakia, forests cover approximately 40% of the total territory, and the majority of them are managed (NLC 2024). The bryophytes of forest ecosystems in Slovakia have rarely been studied. Most studies focused on mountain regions (Petrášová et al. 2011; Schumacher 2000; Slezák et al. 2016; Širka et al. 2023), while data from lowland forests are lacking. Therefore, our attention was directed towards the significance of managed forests concerning bryophyte diversity at the regional level in a lowland environment. We aimed to provide answers to the following main questions: At the regional scale, does the type of forest stands determine the richness and composition of bryophyte species? Do the selected environmental variables significantly influence the species richness and composition? Do different forest types exhibit differences in bryophyte functional traits and substrate preferences?
Materials and methods
Study site
The present study was conducted in the southwestern region of Slovakia, in the cadastral territory of Stupava town (Fig. 1) within the Borská nížina Lowland geomorphological unit (midpoint 48°16’29.424"N, 16°59’6.175"E, elevation 150–165 m a.s.l.). Stupava is located approximately 12 km west of Bratislava (the capital city of Slovakia). The territory of the Borská nížina Lowland is mostly covered by Quaternary Aeolian sands. Regosols and podzols have developed in forests, umbrisols or organosols in moist depressions, and fluvisols in the alluvial plain of the Morava River. In terms of climate, Stupava has a warm lowland climate. During the period of 2013–2023, the annual average temperature was 11.1 °C (min. 10.2 °C, max. 11.9 °C), and the annual total precipitation was 668.6 mm (min. 527 mm, max. 792 mm) (Lapin et al. 2002; DATACube 2024). The region was categorised as a landscape with very low to moderate sensitivity to the impact of climate change (Pauditšová et al. 2021).
Fig. 1
Schematic map of the monitored area near the Stupava town in Borská nížina Lowland
×
![]()
Advertisement
The overall area of the observed forests covered 259.12 hectares. It consisted of several fragmented forest stands separated by arable land, meadows, abandoned orchards, and roads.
Figure 1 According to the geobotanical map of Slovakia, all monitored sites have the potential to be covered by a hardwood floodplain forest (Michalko et al. 1987). The present vegetation is primarily composed of managed deciduous and pine forests, with the dominant tree species Quercus robur, Q. petraea, Q. cerris, Q. rubra, Fraxinus excelsior, F. angustifolia, Robinia pseudoacacia, and Pinus sylvestris. Along with the prevailing trees, Acer spp., Alnus glutinosa, Betula pendula, Populus spp., Tilia cordata, T. platyphyllos, Ulmus laevis, and U. minor are primarily found in stands. The most abundant shrubs in the forests are Padus avium, Euonymus europaeus, and Sambucus nigra. The herbaceous understorey is usually continuous, with a high number of species that thrive in nutrient-rich environments, e.g., Allium ursinum, Bromus sterilis, Chelidonium majus, Corydalis cava, Galanthus nivalis, Galium aparine, Impatiens parviflora, Leucojum aestivum, Parietaria officinalis, Rubus caesius, and Urtica dioica. At present, we can classify these woodlands into three categories, namely, economic forests with wood production functions and forests focused on soil protection, and a limited percentage of the stands provide significant health, cultural, or recreational services (ISLHP 2024). Forest management practices include clear-cut logging, the removal of dead wood, and the planting of even-aged stands. The monitored forest area has been utilised for various purposes, including grazing, meadows, pheasantry, and timber logging, since at least the 18th century (StareMapy.sk 2024). Currently, an overabundance of deer in forests results in a negative effect on ground vegetation (Rooney and Waller 2003). Eutrophication and severe trampling have detrimental impacts on bryophytes, hindering the growth of all plant species along the paths frequently used by deer.
Selection of plots and sampling design
Field research was conducted for three years from 2020 to 2023. The sampling plots (SPs) were identified from the database “Forest Management Information System” (ISLHP 2024) following field visits. To eliminate the influences of environmental factors (i.e., slope, altitude, bedrock, and climatic factors) on bryophyte diversity, we focused on a relatively small area. This allowed us to evaluate, as the main variable, the tree composition in forest stands. In the Quercus, Fraxinus, Robinia, and Pinus stands, the dominant tree accounted for more than 60% of the total tree cover. If the proportion of individual tree species in the stand was less than 30%, we classified it as a mixed forest. In total, 37 SPs were selected in woods (Fig. 1), representing five forest types, namely, Quercus forests (SP Q1–Q8), Fraxinus forests (SP F1–F6), Robinia forests (SP R1–R7), Pinus forests (SP P1–P10) and mixed forests (SP X1–X6). In addition to the forest SP, five plots in anthropogenic habitats (SP A1–A5) and two in aquatic habitats (SP H1–H2) were included to evaluate the gamma diversity of the whole area. For each forest SP, we selected 1–4 sampling points. The selection was carried out using a non-random method, according to recommendations by Vondrák et al. (2018) and Malíček et al. (2019). As long as possible, we tried to select sampling points that contained around 3–6 trunks of dead wood that were thicker than 30 cm (Ódor et al. 2005). Log diameters on sampling plots R4, R5, R7, and X3 were smaller, ranging between 15 and 25 cm. Each of the sampling points comprised a 15 × 15 m area, and all bryophyte species occurring there were identified. Bryophytes were categorised into five frequency classes (Raunkiaer 1934) FC: V 81–100%, IV 61–80%, III 41–60%, II 21–40%, and I 1–20%. Species were collected from every substrate found, i.e., bark (1), rocks (2), dead wood (3), and soil (4). The stone substrates found in the surveyed area were only of anthropogenic origin, such as concrete and stone walls, chalets, abandoned military bunkers, drainage channels, and roads.
In addition to the type of forest stand, we selected the area size and age of the forest SPs, the anthropogenic impact, and the presence of water sources as environmental variables for analyses. The water sources in the area include the Stupava stream (“Stupavský potok”) and a small pond in the pine forests. The main characteristics of each SP, including area, age, number of SPs, and anthropogenic impact, are listed in detail in Supplementary Table 1.
Biological (life strategies/LS/, life forms/LF/) and ecological traits (i.e., Ellenberg’s indicator values/EIV/, hemeroby), which were used in the characteristics of the five forest types (Table 1), are based on Van Zuijlen et al. (2023), Hill et al. (2007), and Dierßen (2001). The EIV value N is given according to Simmel et al. (2021), who considered it an indicator of the general nutrient status of the sites in which the species was found. Categories of affinity to forest habitats follow Bernhardt-Römermann et al. (2018). The nomenclature of the bryophytes is based on the work of Hodgetts et al. (2020), and that of the vascular plants follows POWO (2024). The threat of species was evaluated according to Mišíková et al. (2020; 2021).
Table 1
Environmetal variables and functional traits of bryophytes used in analyses. SPs – sampling plots
Environmental variables | |
|---|---|
Habitat | Q – Quercus for., F – Fraxinus for., R – Robinia for., P – Pinus forests, X – mixed for., H – aquatic habitats, A – anthropogenic habitats |
Age of forest SPs (years) | |
Area size of forest SPs (ha) | |
Anthropogenic impact on forest SPs (AI) | 1 – weak (without any obvious human activities); 2 – moderate (a road, sidewalk, abandoned construction, and/or logging are on the edges of the SP); 3 – strong (a road, sidewalk, maintained construction, and/or logging are on the area of the SP) |
Number of water sources in forest SPs | |
Substrate | bark (1), rocks (2), dead wood (3), soil (4) |
Ecological traits | |
Ellenberg´s indicator values (EIV) | L – light, 1 (deep shade) to 9 (full light); T – temperature, 1 (cold indicator) to 9 (extreme warmth indicator); F – moisture, 1 (extreme dryness) to 9 (wet-site indicator); R – reaction, 1 (extreme acidity) to 9 (high pH); N – nutrients, 1 (nutrient very poor) to 9 (nutrient very rich); x (indifferent). (Dierßen 2001; Bernhardt-Römermann et al. 2018; Simmel et al. 2021) |
Affinity to forest habitats | M1.1 mainly in continuous forest; M1.2 primarily along edges and overlit parts; M2.1 in forest and open land; M2.2 primarily in open land. (Bernhardt-Römermann et al. 2018) |
Hemeroby | 1 (absent), 2 (absent to weak), 3 (weak), 4 (weak to moderate), 5 (moderate), 6 (moderate to strong), 7 (strong), 8 (strong to very strong), 9 (very strong). (Van Zuijlen et al. 2023) |
Biological traits | |
Life strategy (LS) | |
Life form (LF) | |
Data analysis
For all the statistical analyses and graph visualizations, the statistical software XLSTAT (Lumivero 2024) was used. Permutational multivariate analysis of variance (PERMANOVA) is a part of the R package Vegan (Anderson 2001, 2017) and is available via XLSTAT software. All p values ≤ 0.05 were considered as significant. Species richness was calculated as the sum of recorded species for each sampling plot (SP) and for each forest type. The matrix of “presence/absence species x forest SP” provided the basis for multivariate analyses.
The Shannon diversity index H´ (Shannon 1948; Shannon and Weaver 1949) and Margalef index R (Margalef 1958) were calculated for the five forest types (the alpha diversiy). The gamma diversity was assessed as the total number of species in all SPs. The distribution of the data was tested by the Kolmogorov‒Smirnov test with Lillefors correction.
The differences in bryophyte species richness, biological and ecological traits (Table 1), and substrate preference across the five observed forest types were tested using a non-parametric ANOVA (Kruskal‒Wallis test) and subsequent Dunn’s multiple comparisons. Principal component analysis (PCA) was used to visualise the functional traits with the most significant contributions to the variance across individual forest types.
The Spearman rank coefficient was determined to assess the correlation between species richness in forest SPs and environmental factors. In addition, the impact of environmental variables, including forest type, age and size of the study area, the presence of water sources, and human impacts on species richness, was examined using a generalised linear model (GLM) considering Akaike’s information criterion. Multiple comparisons were conducted using least squares means, and the forest types were compared pairwise with the Tukey and Bonferroni correction methods.
PERMANOVA was used to determine significant differences in species composition among the five forest types. We employed non-metric multidimensional scaling (NMDS) with the Bray‒Curtis dissimilarity coefficient to display the variations in species composition across the forest types. The stress level was 0.125, which falls below the criterion value (< 0.2), according to Clarke (1993). The extracted NMDS axis 1 and axis 2 scores of the SPs were compared among the five forest types with the Kruskal‒Wallis test with Bonferroni correction.
To investigate the effects of environmental variables on the bryophyte species composition in the forest SP, we performed a PERMANOVA with 999 permutations for each test. Canonical correspondence analysis (CCA) was used to summarise and visualise the relationships between species composition and environmental variables.
Results
Species diversity and functional traits
In total, 60 bryophyte species were identified (gamma diversity), of which five were liverworts and 55 were mosses (Table 2). The investigated area exhibited a low abundance of bryophytes, which can be attributed to the dense understorey of vascular plants. The number of individual SPs varied between 5 (R7) and 25 (F2) (Supplementary Table 1). The highest diversity indices were observed in mixed forests, whereas the lowest were in Robinia stands (Table 3). Non-parametric ANOVA (Kruskal‒Wallis test) showed that overall, there was no difference in species richness between the five forest types. On average, the number of species per SP was found in different forest types in the following order: mixed forests (18.3) > Fraxinus (16.7) > Quercus (12.4) > Pinus (11.8) > Robina (10.3) (Supplementary Table 1). The most frequent species were Amblystegium serpens (86%), Brachythecium rutabulum (97%), Hypnum cupressiforme (84%), Lewinskya affinis (84%), and Orthotrichum diaphanum (64%) (Table 2). In the surveyed area, no red-listed species were found.
Table 2
Bryophytes found in managed lowland forests near Stupava town, Borská nížina Lowland
Species | Frequency/FC | Habitat | Substrate | No. of plots |
|---|---|---|---|---|
Amblystegium serpens | 0.86/V | Q F R P X A | 1, 2, 3, 4 | 38 |
Anomodon viticulosus | 0.06/I | F X | 1 | 3 |
Atrichum undulatum | 0.27/II | Q P X | 4 | 12 |
Aulacomnium androgynum | 0.02/I | P | 3 | 1 |
Brachytheciastrum velutinum | 0.27/II | R P X | 1, 2, 3 | 12 |
Brachythecium rivulare | 0.02/I | H | 2 | 1 |
Brachythecium rutabulum | 0.97/V | Q F R P X A | 1, 2, 3, 4 | 43 |
Brachythecium salebrosum | 0.32/II | F R P X | 1, 3 | 14 |
Ceratodon purpureus | 0.39/III | R P X A | 2, 4 | 17 |
Ctenidium molluscum | 0.02/I | A | 2 | 1 |
Dicranum scoparium | 0.05/I | P | 4 | 2 |
Didymodon rigidulus | 0.02/I | A | 2 | 1 |
Drepanocladus aduncus | 0.02/I | H | 4 | 1 |
Eurhynchium angustirete | 0.02/I | P | 4 | 2 |
Eurhynchium striatum | 0.12/I | Q F R | 4 | 5 |
Fontinalis antipyretica | 0.02/I | H | 2 | 1 |
Frullania dilatata | 0.46/III | Q F X | 1 | 20 |
Funaria hygrometrica | 0.02/I | F P | 4 | 1 |
Homalia trichomanoides | 0.05/I | F | 1 | 2 |
Homalothecium lutescens | 0.02/I | A | 2 | 1 |
Hygroamblystegium varium | 0.02/I | H | 2 | 1 |
Hypnum cupressiforme | 0.84/V | Q F R P X A | 1, 2, 3, 4 | 37 |
Leptodictyum riparium | 0.02/I | H | 4 | 1 |
Leskea polycarpa | 0.50/III | Q F R X | 1 | 22 |
Leucodon sciuroides | 0.06/I | F | 1 | 3 |
Lewinskya affinis | 0.84/V | Q F R P X | 1 | 37 |
Lewinskya speciosa | 0.12/I | F X | 1 | 5 |
Lophocolea heterophylla | 0.10/I | P | 3 | 4 |
Metzgeria furcata | 0.02/I | Q | 1 | 1 |
Mnium hornum | 0.02/I | P | 4 | 1 |
Mnium stellare | 0.02/I | A | 2 | 1 |
Nyholmiella obtusifolia | 0.02/I | F | 1 | 1 |
Orthotrichum anomalum | 0.06/I | A | 2 | 3 |
Orthotrichum diaphanum | 0.64/IV | Q R P X | 1 | 28 |
Orthotrichum pumilum | 0.41/III | Q F R X | 1 | 18 |
Oxyrrhynchium hians | 0.27/II | F R X | 4 | 12 |
Physcomitrium pyriforme | 0.02/I | X | 1 | 1 |
Plagiomnium cuspidatum | 0.30/II | F P X A | 1, 3, 4 | 13 |
Plagiomnium rostratum | 0.46/III | Q F P X | 4 | 20 |
Plagiomnium undulatum | 0.14/I | Q F X | 4 | 6 |
Plagiothecium denticulatum | 0.02/I | P | 4 | 1 |
Plagiothecium nemorale | 0.02/I | P | 4 | 1 |
Platygyrium repens | 0.81/V | Q F R P X | 1 | 36 |
Pleurozium schreberi | 0.02/I | P | 4 | 1 |
Polytrichum formosum | 0.12/I | Q P X | 4 | 5 |
Porella platyphylla | 0.3/II | Q F R X | 1 | 13 |
Pseudanomodon attenuatus | 0.05/I | F X | 1 | 2 |
Pseudoscleropodium purum | 0.02/I | P | 4 | 1 |
Ptychostomum capillare | 0.43/III | Q F R P X | 3, 4 | 19 |
Ptychostomum imbricatulum | 0.09/I | P X | 4 | 4 |
Ptychostomum moravicum | 0.36/II | Q F R X | 1, 4 | 16 |
Pylaisia polyantha | 0.36/II | Q F X | 1 | 16 |
Radula complanata | 0.36/II | Q F X | 1 | 16 |
Rhynchostegium riparioides | 0.02/I | H | 2 | 1 |
Sciuro-hypnum populeum | 0.06/I | A | 2 | 3 |
Schistidium apocarpum | 0.09/I | A | 2 | 4 |
Syntrichia ruralis | 0.05/I | R A | 2, 3 | 2 |
Syntrichia virescens | 0.02/I | A | 2 | 1 |
Tortula muralis | 0.12/I | A | 2 | 5 |
Ulota crispa | 0.02/I | P | 1 | 1 |
Table 3
Alpha diversity indices for five forest types
Species richness | R | H´ | J | |
|---|---|---|---|---|
Q | 20 | 4.13 | 2 | 0.95 |
F | 26 | 5.44 | 2.01 | 0.96 |
R | 17 | 3.74 | 2.38 | 0.95 |
P | 26 | 5.23 | 2.61 | 0.90 |
X | 28 | 5.74 | 2.67 | 0.96 |
Significant differences were observed when assessing the biological (LF, LS) and ecological (EIV, hemeroby) traits of the bryophytes across the five forest stands (Figs. 2 and 3). However, when evaluating the preference for forest habitats, no significant differences were found. Species that occur in both forest and open land (M2.1, 70–80%) were dominant in all forest types. EIV-based indirect estimation of microclimatic conditions revealed a significant difference in substrate reaction (Fig. 3D) between Pinus stands and other forest types. Robinia stands showed differences in moisture compared to Fraxinus and mixed forests (Fig. 3B). Hemerophilous species (hemeroby degree 5–8) were found least in Fraxinus forests (31%) and most frequently in Robinia stands (60%) (Fig. 3F). The occurrence of hemerophobic species (hemeroby degree 2–3) was rather low, with the highest proportion observed in the Fraxinus (56%) and Quercus stands (45%). The percentage of hemerophobic bryophytes in the Robinia stands was 29%, which was the lowest among all forest types. Perennials (50–60%) and colonists (20–30%) were the most prevalent bryophyte life strategies across all forest types, while fugitives and annual species were found only occasionally (Supplementary Fig. 1A). Turfs dominated among the life forms, comprising 70–85% of the total (Supplementary Fig. 1B).
Fig. 2
Comparison of bryophyte biological traits across five forest types (Kruskal–Wallis test, p ≤ 0.05). Only significant values within life forms (LF) and life strategies (LS) are shown: A LF turf; B LS perennials; C LS colonists; Q – Quercus forests, F – Fraxinus forests, R – Robinia forests, P – Pinus forests, X – mixed forests. The box shows the interquartile range, the horizontal line in each box is the median value, the cross is the mean value, the whiskers show the range of the data, and “ω” are the outliers
×
![]()
Fig. 3
Comparison of bryophyte ecological traits across five forest types (Kruskal–Wallis test, p ≤ 0.05): A EIV for light; B EIV for moisture; C EIV for temperature; D EIV for reaction; E EIV for nutrients; EIV – Ellenberg´s indicator values; Q – Quercus forests, F – Fraxinus forests, R – Robinia forests, P – Pinus forests, X – mixed forests. The box shows the interquartile range, the horizontal line in each box is the median value, the cross is the mean value, the whiskers show the range of the data, and “♦” are the outliers
×
![]()
The functional traits that significantly contributed to the variance (Kruskal–Wallis test) across individual forest types were LS perennials (p = 0.02), colonists (p = 0.012), LF turf (p = 0.05), hemeroby (p < 0.001), and EIV for substrate reactions (p = 0.02) (Fig. 4A).
Fig. 4
Species composition across SPs in five forest types: A PCA diagram based on the functional traits of bryophytes, the length of vectors indicates their relative importance; B NMDS ordination diagram according to species composition; C relationship of forest SPs to the selected environmental variables displayed by CCA, Axes F1 and F2 explain 66.2% of the total variability. Hem – hemeroby, EIV – Ellenberg´s indicator values, L – EIV for light, F – EIV for moisture, T – EIV for temperature; R – EIV for reaction; N – EIV for nutrients, LS – life strategy, LS_c – colonists, LS_s – short lived shuttle, LS_l – long lived shuttle, LS_p – perennials, Life forms – turf, weft, cushion, mat, Q – Quercus forests, F – Fraxinus forests, R – Robinia forests, P – Pinus forests, X – mixed forests
×
![]()
Considering the impact of environmental variables on species richness in forest SPs, the Spearman coefficient revealed a significant correlation (rS = 0.486, p = 0.003) between species richness and area size. The age of the SP, presence of a water source, and anthropogenic impact had no significant impact on species richness. The impact of specific environmental variables on species richness was further examined using a generalised linear model (GLM) (Table 4). In addition to the area size, the number of species was also influenced by the forest type.
Table 4
Parameter estimates from GLM (generalised linear model) relating species richness in forest SPs to environmental variables
Explanatory variables | Estimate | SE | z value | Pr(>/z/) |
|---|---|---|---|---|
Area size of forest SPs | 0.406 | 0.089 | 4.540 | < 0.0001 |
Age of forest SPs | 0.043 | 0.195 | 0.221 | 0.825 |
Water source | -0.326 | 0.212 | -1.541 | 0.123 |
Pinus forest type | -0.905 | 0.222 | -4.084 | < 0.0001 |
Quercus forest type | -0.600 | 0.170 | -3.526 | 0.000 |
Robinia forest type | -0.724 | 0.301 | -2.405 | 0.016 |
Mixed forest type | -0.208 | 0.185 | -1.123 | 0.261 |
Species composition
PERMANOVA revealed significant differences in species composition among the five forest types (F.model = 8.261, R2 = 0.508, Pr/> F/= 0.001). The differences in species composition across forest types become evident when they are visualised using NMDS (Fig. 4B). Within individual forest stands, the Kruskal‒Wallis test showed significant differences in the NMDS axis 1 scores for Pinus stands compared to those for Quercus, Fraxinus, and mixed forests (Table 5). Additionally, Robinia stands exhibited dissimilarities compared to Quercus and Fraxinus forests. No differences were found in the scores of NMDS axis 2 across particular forest types.
Table 5
Differences in species composition across five forest types (Kruskal–Wallis test with Bonferroni correction)
Q | F | R | P | X | |
|---|---|---|---|---|---|
Q | No | No | p = 0.035 | p < 0.0001 | No |
F | No | No | p = 0.003 | p < 0.0001 | No |
R | p = 0.035 | p = 0.003 | No | No | No |
P | p < 0.0001 | p < 0.0001 | No | No | p = 0.003 |
X | No | No | No | p = 0.003 | No |
The species found only in Pinus forests were Aulacomnium androgynum, Dicranum scoparium, Lophocolea heterophylla, Pleurozium schreberi, and Pseudoscleropodium purum. However, epiphytic bryophytes, such as Frullania dilatata, Leskea polycarpa, Porella platyphylla, Ptychostomum moravicum, and Radula complanata, were not observed here. This forest type has the highest number of species represented in frequency class I, whereas species from FC III to FC V predominate in deciduous forests (Table 2).
To determine which particular environmental variables had a significant impact on species composition, PERMANOVA was performed. Table 6 shows that the area, age, and presence of water sources in the forest SPs had significant effects on the species composition. The anthropogenic influence and interaction effects had no impact.
Table 6
Impact of the environmental variables on species composition in forest SPs determined by PERMANOVA
Environmental variables | Df | SS | MS | F.Model | R2 |
|---|---|---|---|---|---|
Area | 1 | 0.389 | 0.389 | 6.293 | 0.118** |
Age | 1 | 0.408 | 0.408 | 6.612 | 0.124** |
Water source | 1 | 0.320 | 0.320 | 5.172 | 0.097** |
Anthropogenic impact | 2 | 0.187 | 0.094 | 1.516 | 0.057 |
The relationships between bryophyte species composition in forest SPs and environmental factors were determined by CCA (Fig. 4C) (Pseudo F = 2.32, p < 0.0001). The first two axes explained 66.24% of the total variability in the species composition of the forest SPs. Quercus stands primarily exhibited a positive correlation with stand age, whereas Robinia and Pinus forests showed a negative correlation (Fig. 4C). Although anthropogenic impact did not significantly affect species richness and composition, it was positively correlated with Pinus and Robinia stands and negatively correlated with Fraxinus and Quercus stands.
Substrate preference
Kruskal‒Wallis test showed significant differences in substrate preference across the five forest types (Fig. 5). Epigeic bryophytes were mostly found in coniferous forests, whereas epiphytes were the dominant group in broad-leaved and mixed forests (Supplementary Fig. 1C). Tree species such as Fraxinus excelsior, F. angustifolia, Quercus spp., Acer spp., Populus spp., and Betula pendula were the most favoured phorophytes. When compared to other substrates, the bryophyte diversity on dead wood was very low. Two species, Amblystegium serpens and Hypnum cupressiforme, were ubiquitous and grew on all substrates.
Fig. 5
Comparison of the species richness on various substrates across five forest types (Kruskal–Wallis test, p ≤ 0.05). The differences in relation to rock substrates were not statistically significant. Q – Quercus forests, F – Fraxinus forests, R – Robinia forests, P – Pinus forests, X – mixed forests. The box shows the interquartile range, the horizontal line in each box is the median value, the cross is the mean value, the whiskers show the range of the data, and “♦” are the outliers
×
![]()
The stone substrates found in the area were only of anthropogenic origin, specifically concrete structures, i.e., roads, ditches, bridges, and abandoned military bunkers. Hemerophilous epilithes (Orthotrichum anomalum, Schistidium apocarpum, Syntrichia ruralis, and Tortula muralis) represented the largest group, but several species of base-rich rock substrates also occurred there, e.g., Mnium stellare and Ctenidium molluscum.
Overall, aquatic bryophytes accounted for 10% of all the observed species. They grew in Stupava Creek (H1) and in a small pond (H2) in SP_P7. Due to the presence of a water source, this SP exhibited more species (23) than did the other Pinus plots (Supplementary Table 1). In addition to aquatic bryophytes, hygrophytic mosses such as Mnium hornum and Plagiothecium nemorale were also found here.
Discussion
The results of the present study are in agreement with observations of the species richness and composition in managed forests in Central Europe reported by Nowińska et al. (2009) and Brusa et al. (2021). Fojcik et al. (2019) identified 54 bryophyte species in these types of managed forests in Poland. Most of them were found on decayed wood, with the least number on tree bark. In contrast, we observed the highest species richness on bark and soil, while on dead wood, only a few species occurred. The most common mosses were the hemerophilous species Amblystegium serpens, Brachythecium rutabulum, Lewinskya affinis, and Orthotrichum diaphanum, which grow in nutrient-rich environments (Diekmann et al. 2023). Intense forest management, the construction of roads, concrete objects, and chalets, along with a dense population of deer, support the spread of synanthropic and nitrophilous bryophytes and the increase in hemerophilous species (Dittrich et al. 2016; Staniaszek-Kik et al. 2016). Nevertheless, artificial structures can enhance the diversity of microhabitats, and thereby creating suitable environments for even rare species (Paciorek et al. 2016; Staniaszek-Kik et al. 2016). Despite this fact, we were unable to verify the presence of rare or threatened bryophytes in the observed area. The region’s warm and dry climate, along with a lack of suitable substrates such as dead wood, has limited the presence of liverworts. The reduced availability of appropriate substrates, combined with competition from vascular plants, has contributed to the high representation of bryophytes with low frequency (FC I-II) (Fojcik et al. 2019; Müller et al. 2019; Stefańska-Krzaczek et al. 2022).
Bryophyte functional traits displayed differences across the five forest types in the investigated area. The variability was significantly impacted by EIV for reaction, hemeroby, LF turf, and LS. To some extent, the representation of LF and LS reflects habitat types and ecological conditions (Löbel et al. 2018; Spitale et al. 2020). LF turfs, which prevailed in all five forest types, frequently occur on sites with higher levels of sunlight and dryness, whereas wefts and mats are typical for more shaded and humid sites (Bates 1998). The high proportion of turfs may indicate intensive management in the forests, as well as the influence of the surrounding agricultural landscape. According to Spitale et al. (2020), turfs were more prevalent in broad-leaved and mixed forests in Northern Italy, with plagiotropic forms (mats and wefts) predominating in coniferous stands. Compared to other forest types, Robinia stands had the highest representation of LS colonists (48%). We assume that the shorter felling age of Robinia forests, and consequently their younger age, influenced the higher representation of LS colonists, which are characterized by a shorter lifespan, high sporophyte formation, and small spore size (Kürschner 2004).
We assumed that the anthropogenic impact, including the intensity of farming, man-made structures, but also the influence of forest animals, will have the greatest influence on the diversity of bryophytes among the selected environmental variables. Contrary to our assumptions, this variable proved to be the least significant, likely due to the similar intensity of forest management across the stands (Müller et al. 2019; Cacciatori et al. 2022). The type of forest SPs and their area size have a greater impact on bryophyte species richness and composition. SPs of smaller sizes, typically approximately 2 hectares, exhibited a lower number of species. In larger and more diverse areas, several additional environmental variables, including altitude, bedrock composition, rock cover, and forest management practices, might impact the variety of bryophytes. (Czerepko et al. 2021; Cacciatori et al. 2022; Kutnar et al. 2023; Širka et al. 2023). Species diversity is also affected by microclimate, i.e., light conditions (Diekmann et al. 2023) or nutrients (Dittrich et al. 2016). According to Ingerpuu et al. (2019), an increase in humidity is associated with greater species diversity within forest ecosystems. We observed this trend in SP_P7, in which the water source supported the presence of both hygrophytic and aquatic bryophytes.
One of the major factors influencing the species richness is the availability of suitable microhabitats and substrates (Wierzcholska et al. 2019; Glime 2024). Based on the results of the present study, epiphytic bryophytes are one of the most frequent groups, and they also exhibit the highest species diversity in Central European deciduous forests (Czerepko et al. 2021; Baumann 2023). Their survival depends on the presence of a wide variety of woody plant species and suitable light conditions (McGee and Kimmerer 2002; Mežaka et al. 2012; Király et al. 2013; Wierzcholska et al. 2024). The age of trees appears to be a major driver of epiphyte variety (Mežaka et al. 2012). However, we could not deal with this issue in our study, as SPs were even-aged with the absence of old trees. The majority of epiphytic bryophytes were observed in Fraxinus and mixed stands (Fig. 5), predominantly growing on F. angustifolia, F. excelsior, Betula pendula, Acer pseudoplatanus, and Quercus spp. Pinus sylvestris was the least frequent phorophyte. The acidic peeling bark of pines is unsuitable for the development of bryophyte assemblages. Furthermore, pine trunks retain only a small amount of precipitation (Király et al. 2013). According to Dittrich et al. (2016), Fraxinus excelsior and Acer pseudoplatanus are the two major phorophytes in Central European deciduous woods. Additionally, Ulmus glabra, Tilia cordata, and Populus tremula are additional suitable tree species for providing substrates for bryophytes (Gerra-Inohosa et al. 2023). Mixed forests can support greater bryophyte diversity than monoculture plantations because the specific properties of diverse tree species are well suited to various types of bryophytes (Mežaka et al. 2012; Rola et al. 2021). Among the rare epiphytes in the Borská nížina Lowland, only the species Ulota crispa (phorophyte Betula pendula) was identified. The presence of U. crispa indicates that sensitive species may recover in previously polluted areas (Baumann et al. 2022). Therefore, it can be assumed that Central European managed forests are currently in a state of equilibrium between the consequences of climate change and more sustainable forest management and pollutant reduction (Cacciatori et al. 2022).
For epixylic species, trunks with a minimum diameter of 30 cm were considered suitable substrates (Ódor et al. 2005; Kropik et al. 2021). Due to their removal during logging, their proportion was minimal in all forest types. In addition to substrates, moisture is an important factor for epixilic bryophytes. In Central Europe, an optimal level of annual precipitation is around 900 mm (Kropik et al. 2021). As we mentioned in methods section, the selected area encounters a notably reduced total precipitation, ranging from a low of 527 mm to a maximum of 792 mm (DATACube 2024).
Quercus and Fraxinus stands exhibited similar species composition, biological and ecological traits, and substrate preferences. Nevertheless, there was a substantial difference in the species composition across deciduous and coniferous forests. Due to the lower amount of fallen plant biomass in Pinus stands, soil bryophytes prevailed only in these forests. In addition, we observed here the highest abundance of epixylic bryophytes. This can be attributed to the unique characteristics of dead pine wood, including its high acidity, moisture retention, and low nutrient content (Kahl et al. 2017; Müller et al. 2019). However, epiphytic bryophytes, which grew on accompanying deciduous trees, were less prevalent. Stefańska-Krzaczek et al. (2022) listed the mosses Pleurozium schreberi and Dicranum scoparium among the species distinctive for coniferous forests, which we also observed only in Pinus stands. Contrary to our initial assumption, Pinus stands did not exhibit a substantial difference from deciduous stands in terms of moisture (EIV_F). However, in terms of temperature (EIV_T), pine stands showed a slightly greater relative abundance of cold-tolerant species. The current views on the application of EIV_T for plants in evaluating microclimatic differences within forest stands are not consistent. While Gril et al. (2024) do not assess EIV_T as a suitable indicator of microclimatic variations and changes in forests, Christiansen et al. (2022) suggest a correlation between the forest microclimate temperature and species composition. Zhang et al. (2023) evaluate temperature-related variables as the primary driver of bryophyte diversity and distribution.
The introduction of non-native trees, such as Robinia pseudoacacia or Quercus rubra, can have adverse effects on the diversity and composition of bryophytes (Woziwoda et al. 2017; Dyderski et al. 2020). In the present study, Robinia stands exhibited the lowest species diversity, a greater abundance of hemerophilous species, and significant differences in species composition across evaluated forest types. The microclimatic conditions were characterized by reduced moisture levels and an increased abundance of light-demanding and thermophilous species. According to Šibíková et al. (2019), non-native Robinia stands exhibited lower humidity and increased nutrient content in comparison to native forests. Furthermore, they found a loss of variability, indicating a greater degree of homogenization, an effect that we also observed for the moisture and substrate reaction.
Conclusions
Various macroclimatic, geological, geographical, and habitat factors influence the abundance and diversity of bryophytes in forests within large territories. In the present study, we tried to identify the drivers impacting bryophyte diversity in small-scale forest regions characterized by comparable environmental conditions. For the first time, we evaluated bryophytes in lowland managed forests on Borská nížina Lowland in Slovakia. The results highlight the importance of dominant tree species, the age and area size of forest stands, the presence of water sources, as well as the anthropogenic impact on species richness and composition. In addition to species richness and composition, functional traits and substrate preferences were analyzed. The forest type, area size, and age of the sampling plots were shown to be significant environmental variables. However, the anthropogenic impact turned out to be the least significant factor, probably due to its comparable effect on particular plots. The species richness in lowland managed forests was rather low, and no species of conservation interest were identified inside them. Nevertheless, these forest stands can serve as valuable refuges for specific hemerophobic and forest species.
To maintain the bryophyte diversity in intensively managed forests, we recommend, at least in part of the territory, appropriate methods of forest management, including planting broad-leaved stands with an appropriate area and an adequate combination of native trees, e.g., the genera Quercus, Fraxinus, Ulmus, Acer, and Tilia. Furthermore, we suggest protecting old specimens of trees as a source of microhabitats for various groups of organisms (Mežaka et al. 2012; Kozák et al. 2023) and maintaining forest retention patches for hemerophobic bryophytes. Recently, coniferous stands with Pinus sylvestris have been planted on sandy, nutrient-poor dry soil. However, to maintain and increase the species richness of epixylic and epigeic bryophytes in pine stands, we advise not removing a part of dead wood for epixylic species (Raabe et al. 2010; Müller et al. 2015) and preserving existing water resources.
Acknowledgements
We are grateful to both anonymous reviewers who helped to improve the quality of this paper.
Declarations
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
