FS Forest Systems FS 2171-9845 Instituto Nacional de Investigacion y Tecnologia Agraria y Alimentaria (INIA) 13488 10.5424/fs/2018273-13488 RESEARCH ARTICLE Effect of environmental factors on the composition of terrestrial bryophyte and lichen species in Scots pine forests on fxed sand dunes Tilk Mari Chair of Silviculture and Forest Ecology, Institute of Forestry and Rural Engineering, Estonian University of Life Sciences, Kreutzwaldi 5, Tartu, Estonia, 51014. Tallinn Botanic Garden, Kloostrimetsa Road 52, Tallinn, Estonia, 11913. Ots Katri Chair of Silviculture and Forest Ecology, Institute of Forestry and Rural Engineering, Estonian University of Life Sciences, Kreutzwaldi 5, Tartu, Estonia, 51014. Tullus Tea Chair of Silviculture and Forest Ecology, Institute of Forestry and Rural Engineering, Estonian University of Life Sciences, Kreutzwaldi 5, Tartu, Estonia, 51014. should be addressed to Mari.Tilk@botaanikaaed.ee 01 12 2018 2018 27 3 10.5424/fs/2018273-13488 21 05 2018 30 10 2018 © 2018 INIA 2018 This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 International (CC-by 4.0) License. Abstract

Aim of study : To investigate terrestrial bryophyte and lichen species richness and environmental factors affecting the composition of species. Area of study: : Four Boreal zone fxed dunes were selected in the coastal area of the Baltic Sea in southwest Estonia. Material and methods: Non-metric multidimensional scaling was performed to analyse distribution patterns and environmental factors like canopy cover, photosynthetically active radiation, soil organic horizon thickness and decomposition rates, soil volumetric water content, soil pH and electrical conductivity and soil nutrients correlated with bryophyte and lichen species composition. Main results : Thirty bryophytes and 22 lichens were found on 232 sample plots, the most frequent species were Pleurozium schreberi (Willd. ex Brid.) Mitt., Hylocomium splendens (Hedw.) Schimp., Dicranum polysetum Sw. ex anon., Cladonia arbuscula (Wallr.) Flot. and Cladonia furcata (Huds.) Schrad. The lichen species richness was highest on the slopes of the dunes and decreased towards the bottoms and tops; bryophyte species richness was higher on the bottoms and decreased towards the tops of the dunes. Research highlights : The composition of bryophytes and lichens is signifcantly influenced by the aspect and the location on the dune, light conditions, soil pH, soil salinity (measured as electrical conductivity) and volumetric water content, thickness of moderately decomposed organic horizon and vascular plant species cover.

Key words: Inland dunes; terrestrial bryophyte and lichen communities; environmental factors; topography.

Authors´ contributions: Original idea and study design: MT. Data collection: MT and KO. Data analysis: TT and MT. Manuscript preparation and revisions: MT, KO and TT.

Supplementary material: (Table S1) and (Figure S1) accompany the paper on FS's website.

Citation Tilk, M.; Ots, K.; Tullus, T. (2018). Effect of environmental factors on the composition of terrestrial bryophyte and lichen species in Scots pine forests on fxed sand dunes. Forest Systems, Volume 27, Issue 3, e015. https://doi.org/10.5424/fs/2018273-13488

This work was supported by institutional research funding IUT21-4 of the Estonian Ministry of Education and Research.

Competing interests: The authors have declared that no competing interests exist.

Introduction

During the last century in Europe, mainly due to urbanization and other human-related factors, a reduction of about 70% of the dune systems has been estimated ( Mc Lachlan & Brown, 2006 ). Vegetated dunes are complicated and fragile ecosystems which have a very high ecological and conservative value ( Lemauviel et al. , 2003 ; Van der Maarel, 2003 ). Lichens and bryophytes have principal functional importance on boreal zone inland dune forests ( Ahti & Oksanen, 1990 ); bryophytes dominate the forest floor vegetation and play an important role in the water and carbon budgets and improve microenvironment ( Bond-Lamberty & Gower, 2007 ; Márialigeti et al. , 2016 ) while lichens colonize open sand surfaces and create suitable conditions for vascular plants ( Ketner-Oostra & Sykora, 2000 ). Maintenance of biodiversity is an integral part of sustainable forest management which is a desirable goal for most forest-related initiatives and legislative bodies. Assessment of biological diversity is essential for understanding forest ecology and measuring sustainable forest management binds conservation and enhancement of biological diversity ( MCPFE Liaison Unit Vienna, 2002 ; Canullo et al. , 2013 ). According to Pharo et al. (1999) the bryophytes and lichens have different patterns of diversity compared to vascular plant species and therefore management practices and conservation actions should take into account the specialty.

Multiple studies have been conducted to clarify lichen and bryophyte distribution and richness and to give insight into the different factors that influence their growth in forests. These factors include substrate availability ( Pharo et al. , 1999 ; Ingerpuu, 2002 ), litter composition and quality ( Hill, 1979 ; Magnusson, 1982; ), soil nutrients (especially magnesium) and acidity (pH) ( Oechel & Van Cleve, 1986 ; Pausas, 1994 ; Jun & Rozé, 2005 ; Kösta & Tilk, 2008 ; Košuthovà et al. , 2015 ; Jüriado et al. , 2016 ), light conditions and soil moisture ( Ketner-Oostra & Sykora, 2000 ; Márialigeti et al. , 2016 ) and the cover of vascular plant species ( Löbel et al. , 2006 ). Also, studies have pointed out stress and disturbance as important factors affecting cryptogam communities ( Forey et al. , 2008 ; Cogoni et al. , 2011 ; Ciccarelli, 2015 ).

Topographical factors, especially the height and slope aspect, have been modestly discussed when bryophytes and lichens are concerned, whereas for the vascular plant species richness and composition the slope effect has been considered as an important factor ( Jenny, 1941 ; Sewerniak, 2016 ; Sewerniak et al. , 2017 ; Tilk et al. , 2017 ). In northern hemisphere north-facing slopes receive six times less solar radiation compared to south-facing slopes and therefore microclimatic conditions on the slopes vary greatly ( Auslander et al. , 2003 ; Mandre et al. , 2008 ; Sewerniak et al. , 2017 ; Sewerniak & Jankowski, 2017 ). Therefore, a respective variation in the species richness and composition of the understory vegetation can be expected as well ( Solon et al. , 2007 ; Sewerniak & Jankowski, 2017 ). Previous studies have provided data for the zonation of the vascular plant species composition along the dune profile ( Zoladeski, 1991 ; Sewerniak & Jankowski, 2017 ; Tilk et al. , 2017 ); however, precise information is insufficient for lichens and bryophytes. Besides, thorough knowledge of the abundance of lichens and bryophytes, as well as on topographical and environmental factors that affect their patterns in forests on fixed dunes, is still lacking.

This paper is continuation of earlier works of Tilk et al. (2011 ; 2017 ), where the vascular plant species richness and environmental factors were investigated. The main aim of this paper is to describe the variability of the bryophyte and lichen layer on Pinus sylvestris dominated fixed dunes and to give a new insight into specific environmental and topographical factors that influence bryophyte and lichen communities and species distribution. The specific objectives of the study were (1) to determine if bryophyte and lichen species richness varies along the topographical gradient on dunes; (2) to study whether zoning of the bryophyte and lichen species composition along the dune profile is possible; (3) to analyse the variability of bryophyte and lichen species in dune forests based on ecological indicator values ( Düll, 1991 ; Wirth, 2010 ); and (4) to analyse the influence of environmental and topographical factors on the composition of the bryophyte-lichen layer.

Material and methods Study area

The investigated dune system is situated in Southwest Estonia, in the coastal area of the Baltic Sea (Fig. 1). The dunes are located on the Uulu-Võiste landscape protection area and on the Luitemaa nature reserve where, the following priority habitats according to Council Directive 92/43/EEC (1992), are protected: wooded dunes of the Atlantic, Continental and Boreal region (type 2180), Western Taïga (type 9010), Fennoscandian deciduous swamp woods (type 9080), fixed coastal dunes with herbaceous vegetation (grey dunes) (type 2130) and humid dune slacks (type 2190).

Location of study area and study sites in Estonia.

For the investigation of bryophyte and lichen flora on wooded dunes of different height, four sampling sites were selected (Table 1). Sampling sites distance to Baltic Sea (Pärnu Gulf) is ranging between 2 and 3 km.

Study sites, number and direction of quadrats.

According to information from State Forest Management Centre database the dunes belong to the Cladonia and Vaccinium vitis-idaea forest site type, where the average stand age is 195 years; the average stand canopy cover is 55%; the average height of pines is 23 m; the site quality index is IV and the density of the understorey is low.

According to the closest weather station of the Estonian Weather Service in Pärnu, the main climate characteristics during the period of our investigations in 2010 were as follows: average annual temperature 5 °C with maximum average of 21.8 °C in July and minimum average of -12.3 °C in January; total average precipitation 908 mm with maximum of 122.8 mm in August and minimum of 27 mm in January.

Mean relative air humidity was 84% in the area. The length of the thermal growing period (mean temperature above 5 °C) was 203 days (from 25.04.2010 to 14.11.2010). During the winter of 2009/2010 there was permanent snow cover recorded of 112 days (11.12.2009 - 01.04.2010) and throughout winter there was altogether 143 days with snow.

 Methods

Field studies on four Pinus sylvestris dominated dunes were carried out in July 2010 for bryophyte and lichen species determination. On every dune, quadrats with the size of 1 m 2 formed a continuous transect over the dune, starting from the bottom in front of the dune and moving over the top to the bottom beyond with the distance of one meter between the quadrats ( n = 232). Additional quadrats ( n = 464) to determine species that did not occur on the basic quadrats of the transect were placed one meter to the left and one meter to the right of the basic quadrat. Data about species found on the additional quadrats were added into overall species list but were not included into the further statistical analysis.

On every quadrat, the total cover of the bryophyte and lichen layer, the total cover of vascular plants and the cover of each bryophyte and lichen species were estimated visually, using the scale 1-100%. Dominant species were determined based on Braun-Blanquet five-point cover-abundance scale: 5 - very abundant; 4 - abundant; 3 - plentiful; 2 - sparsely; 1 - single. The individuals that could not be identified in the field were collected and determined in the laboratory; species identification included using the Tallinn Botanical Garden bryophyte and lichen herbarium for comparative analysis. The nomenclature of lichens and bryophytes follows Randlane et al. (2016) and Ingerpuu et al. (1998) respectively.

On every quadrat, canopy cover was assessed visually by two evaluators as a measure of the percentage of forest floor covered by a vertical projection of the tree canopies expressed at scale 0 to 1 ( Masing, 1979 ; Pihelgas, 1983 ). Below-canopy photosyntheti­ca­lly active radiation (PAR, μmol m -2 s -1 ) was measured ( n = 10 per quadrat) with light interception device AccuPAR (Model PAR-80) in 23 of July 2008 at midday full sunshine. As light is highly variable, measurements were performed simultaneously during a very short time period (from 11:00 to 13:00) at all sites.

Absolute heights (meters above sea level, m.a.s.l) and aspects of the quadrats on the dune slopes (N; S; E; W) were determined using Garmin GPSMap 76CSx device and relative heights were calculated using the first quadrat on the bottom of the dune as a zero. An inclinometer was used to assess the incline of slopes and quadrats were classified into five groups according to angle value: 1 (1 - 10 degrees); 2 (11 - 20 degrees); 3 (21 - 30 degrees); 4 (31 - 40 degrees) and 5 (41 - 50 degrees).

Soil organic horizon (O-horizon) thickness and its decomposition rates: 1 - poorly decomposed (O i ) sub-horizon; 2 - medium decomposed (O e ) sub-horizon and 3 - well decomposed (O a ) sub-horizon, were assessed.

Volumetric water content (VWC, %) in the soil was determined in every quadrat with Field Scout TM TDR 300 at a depth of 20 cm ( n = 3). The data were collected in May, July and September 2008 and there was no rain recorded at least 3 days before measurements. Soil samples of every quadrat were collected in July 2010 from a depth of 20 cm from ground level (for measuring soil pH and electrical conductivity). Soil pH and electrical conductivity were measured from a soil-distilled water mixture (1:2.5 or for samples with high organic matter 1:5) with Eutech Instruments PC300 pH/conductivity meter.

Soil samples from mineral topsoil for analysing nitrogen (N), phosphorous (P), potassium (K), calcium (Ca) and magnesium (Mg) content were collected from a depth of 20 cm ( n = 15 per site) on the bottoms, slopes and tops of the dunes. The results are presented as N: P; N: K and Ca: Mg ratios and absolute values are presented in an article by Tilk et al. 2017 . The concentrations of N, P, K, Ca and Mg in the mineral topsoil samples ( n = 60) were determined in the Laboratory of Plant Biochemistry of the Estonian University of Life Sciences. The soil samples were analysed for their extractable concentration of P (ammonium lactate; by FiaStar5000 (Flow Injection Analyser)), K (ammonium lactate; Flame Photometric Method), Ca (ammonium acetate: Flame Photometric Method), Mg (ammonium acetate; by FiaStar5000 (Flow Injection Analyser)) and for total concentration of N (Copper Catalyst Kjeldahl Method).

Data analysis

Bryophytes and lichens were grouped based on the ecological indicator values of species according to Düll (1991) and Wirth (2010) .

For statistical analysis, quadrats were grouped according to location based on the relative height and the angle of the quadrat as follows: bottom ( n = 52 quadrats), slope ( n = 142 quadrats) and top ( n = 38 quadrats). Correlation analysis was performed with Microsoft Excel 2010 to determine relations between canopy cover and some topographical variables with the precondition of checked data normality. To clarify significant differences between groups one-way ANOVA of multiple groups followed by post-hoc Tukey HSD test using Tukey-Kramer formula for unequal observations was performed ( Vasavada, 2016 ).

A linear mixed model using free statistical software R Version 3.2.3 ( R Core Team, 2015 ) function “lmer” in package lme4 ( Bates et al. , 2015 ) (with a site as a random factor) was applied to clarify the effect of location on environmental factors and cryptogams characteristics. If a statistically significant effect of location was observed, a Tukey HSD test was applied to compare the group means. A level of significance of α = 0.05 was used to reject the null hypothesis after statistical tests.

The effect of grouping factors on bryophyte and lichen data was tested using Multiple Response Permutation Procedure (MRPP) ( Mielke et al. , 1976 ). To correct the p -values for multiple comparisons in MRPP, Bonferroni correction was applied. Indicator species analysis (ISA) was conducted to specify indicator species for different zones ( Dufrene & Legendre, 1997 ). The statistical significance of indicator values was proven using Monte Carlo simulation. MRPP and ISA were performed with PC-ORD Version 6 ( McCune & Mefford, 2011 ).

To analyse species distribution patterns and envi­ronmental variables correlated with bryophyte and lichen species composition, non-metric multidimensional scaling (NMDS) with free statistical software R Version 3.2.3 ( R Core Team, 2015 ) in the community ecology package Vegan ( Oksanen et al. , 2016 ) was performed. NMDS was run using the function “metaMDS” (default settings) and Bray-Curtis dissimilarities. For fitting environmental vectors and factors onto ordination, function “envfit” was used ( Oksanen, 2015 ).

The analyses of species composition (MRPP and NMDS) were performed twice: 1) based on the abundance data of bryophytes and lichens; 2) based on the abundance data of bryophytes. Separate analysis based on lichen data was not performed, as lichens were present on only 22% of the studied quadrats.

 Results Environmental factors

Soils are Haplic Podzols with A-horizon on the bottoms and Haplic Podzol on the slopes and on tops of the dune. Average values for light conditions and soil characteristics are presented in Table 2. Canopy cover showed a negative correlation ( r = -0.67) with the aspect of the quadrat, being higher on the northern and eastern quadrats and lower on southern and western quadrats. Canopy cover was significantly higher on the slopes (Table 2), while PAR showed no significant differences between locations on dunes. In addition, significant differences were found for soil volumetric water content (VWC), pH H2O and electrical conductivity between different locations (Table 2). Average soil VWC was highest on the bottoms of the dunes, being on average 48% higher compared to the top of the dunes. Average soil VWC on the north facing quadrats was significantly higher compared to the south, east and west facing quadrats ( p < 0.01), while differences between soil volumetric water contents on the southern, western and eastern quadrats were not significant ( p > 0.05).

Average values (±SE) for photosynthetically active radiation (PAR, μmol m<sup>-2</sup>s<sup>-1</sup>), canopy cover, soil volumetric water content (VWC, %), pH, electrical conductivity (EC, μS cm<sup>-1</sup>), average ratios of N:P, N:K and Ca:Mg and thicknesses of soil organic horizons with different decomposition rates (O<sub>i</sub> - poorly decomposed; O <sub>e</sub> - medium decomposed and O<sub>a</sub> - well decomposed, cm) from different locations on the dunes according to linear mixed model. Letters denote signifcant differences between locations according to Tukey HSD test. Average values of soil and light characteristics from different locations on the dunes for each study site are presented in Tilk <italic>et al</italic>. 2017: table 4.

Bottoms of the dunes also obtained higher electrical conductivity and lower soil pH values. The average values for N: P and N: K ratios were similar in all of the observed locations; only the Ca: Mg ratio showed statistically significant differences between bottoms and slopes (Table 2). Soil organic horizon averaged 9.1 cm on the bottom, 8 cm on the slope and 8.4 cm on the top of the dunes. The poorly decomposed organic layer made up 30.8% of the bottoms, 32.5% of the slopes and 32.1% of the tops of the dunes; the medium decomposed organic layer formed 58.2% of the bottoms, 56.3% of the slopes and 67.9% of the tops and the well decomposed organic horizon formed 11% of the bottoms, 11.3% of the slopes and was missing from the tops of the dunes.

 Species richness and composition changes along topographical and environmental gradients

Altogether, 52 species of bryophytes and lichens were distinguished on the quadrats (30 species of bryophytes and 22 species of lichens). Bryophytes were present on all 232 quadrats while lichens were recorded on 50 quadrats. Six bryophyte species were common to all four dunes while 17 species were recorded only on one dune. The most frequent bryophyte species were Pleurozium schreberi (present on 90% of the quadrats), Hylocomium splendens (64%) and Dicranum polysetum (51%). None of the lichen species was recorded on all dunes; the most frequent lichen species, Cladonia arbuscula (present on 6% of the quadrats), was found on three dunes. Fourteen lichen species were recorded on only one dune. There were four species found on the additional quadrats: lichen Cladonia cornuta (L.) Hoffm. and bryophytes Cephaloziella hampeana (Nees) Schiffn., Ptilidium pulcherrimum (Weber) Vain. and Scleropodium purum (Hedw.) Limpr. P. schreberi dominated on 128 quadrats, H. splendens on 76 and D. polysetum on 11 quadrats. Cladonia rangiferina (L.) F.H. Wigg. (on four quadrats) was the only dominated lichen species more than once.

Average lichen species richness was the highest on the slopes of the dunes and decreased towards the bottoms and tops (Table 3). Bryophyte species richness was significantly higher on the bottoms of the dunes, decreasing towards the tops of the dunes. The cover of the bryophyte and lichen layer was remarkably higher on the tops, contrary to the cover of vascular plants, which was the highest on the bottoms of the dunes (Table 3).

Average species richness for lichens (S<sub>Lichen</sub>), bryophytes (S<sub>Bryophyte</sub>), total cover of lichen and bryophyte layer (Cover<sub>LichBryo</sub>) and total cover of vascular plant species layer (Cover<sub>Vascular</sub>) (±SE) on different locations according to the linear mixed model. Letters denote signifcant differences between locations according to Tukey HSD test.

According to the ecological indicator values ( Düll, 1991 ), the highest number of bryophyte species were light-demanding species (with light value 8, 26% of all bryophyte species), while there were some (6% of bryophyte species) shade-tolerant species (Fig. S1 A [supplementary]). The moisture preference of bryophyte species was variable as the most frequent values of moisture index varied between 4 and 7, referring to the preference of moderately dry to moist habitats. Based on pH indices assigned to bryophyte species, the majority of species preferred acidic substrate; two species were with broad pH tolerance (Fig. S1 A [suppl.]).

Light index (according to Wirth, 2010 ) was not available for 36% of lichen species, however, according to available indices, lichens were mainly light-demanding species (the most frequent values of the light index were 7 and 8) (Fig. S1 B [suppl.]). Based on moisture and substrate pH indices, the highest number of lichen species was species with broad amplitude; however, it must be considered that moisture and substrate pH indices were unavailable for 8 lichen species in each case (Fig. S1 B [suppl.]).

According to the MRPP test based on zones on du­nes, bryophyte and lichen communities were significantly different on the bottoms versus tops ( p -value = 0.015) and slopes versus tops ( p -value = 0.012) while bottoms and slopes showed no statistically significant differences ( p -value = 0.535) (Table 4). When only bryophyte communities were concerned, slopes and tops were similar ( p -value = 0.263), while bottoms versus slopes ( p -value = 0.001) and bottoms versus tops ( p -value = 0.002) showed statistically significant differences. The distribution of bryophyte and lichen species in different locations on the dunes is shown in Table S1 ([suppl.]). Indicator species analysis based on bryophyte and lichen species data pointed out characteristic species for zones on dunes: Brachythecium erythrorrhizon, Brachythecium oedipodium and Hylocomium splen­dens for the bottoms, Ceratodon purpureus, Cladonia rangiferina and Cladonia stygia for the slopes of the dunes and Pleurozium schreberi for the tops.

The results of multiple response permutation procedure (MRPP) tests for the comparison of bryophyte and lichen species (<italic>p</italic>-value<sub>bryolich</sub>) and bryophyte species (<italic>p</italic>-value<sub>bryo</sub>) composition on different zones on dunes. Environmental factors affecting bryophyte and lichen species composition

NMDS analysis (Table 5) suggested that the most significant environmental factors (level of significance p ≤ 0.001) affecting epigeic bryophytes and lichens were site-specific factors such as aspect and height of the quadrat, but also vascular plant cover, PAR, soil VWC, pH, electrical conductivity and thickness of O e sub-horizon (Fig. 2). When only bryophyte communities were concerned, NMDS analysis pointed to the aspect, vascular plant cover, PAR, canopy cover, thickness of O e sub-horizon and soil VWC, pH, and electrical conductivity as the most significant factors (Table 5).

NMDS ordination (stress type 1), based on the abundance data of bryophytes and lichens. The arrows indicate environmental factors that were the most signifcantly (<italic>p</italic> ≤ 0.001) related to the ordination (see also Table 5). Environmental factors: Habs = absolute height (m.a.s.l); Asp = aspect of the quadrat (S; N; E; W); EC = average conductivity (μS cm<sup>-1</sup>); pH = average soil pH; VWC = average soil water content (%); PAR = photosynthetically active radiation (μmol m<sup>-2</sup>s<sup>-1</sup>); O<sub>e</sub> = average thickness of moderate decomposed organic horizon layer (cm); Vasccov = cover of vascular plants. List of lichen and bryophyte species: Aula. pal = <italic>Aulacomnium palustre</italic>; Aula. pim = <italic>Aulacomnium palustre var. imbricatum</italic>; Brac. alb = <italic>Brachythecium albicans</italic>; Brac. ery = <italic>Brachythecium erythrorrhizon</italic>; Brac. oed = <italic>Brachythecium oedipodium</italic>; Brac. sta = <italic>Brachythecium starkei</italic>; Camp. som = <italic>Campylium sommerfeltii</italic>; Camp. ste = <italic>Campylium stellatum</italic>; Ceph. bis = <italic>Cephalozia bicuspidata</italic>; Ceph. rub = <italic>Cephaloziella rubella</italic>; Cera. pur = <italic>Ceratodon purpureus</italic>; Cetr. isl = <italic>Cetraria islandica</italic>; Cirr. pil = <italic>Cirriphyllum piliferum</italic>; Clad. arb = <italic>Cladonia arbuscula</italic>; Clad. cil = <italic>Cladonia ciliata</italic>; Clad. mit = <italic>Cladonia mitis</italic>; Clad. por = <italic>Cladonia portentosa</italic>; Clad. ran = <italic>Cladonia rangiferina</italic>; Clad. ste = <italic>Cladonia stellaris</italic>; Clad. sty =<italic> Cladonia stygia</italic>; Clad. acu = <italic>Cladonia acuminata</italic>; Clad. car = <italic>Cladonia cariosa</italic>; Clad. chl = <italic>Cladonia chlorophaea;</italic> Clad. con = <italic>Cladonia coniocraea</italic>; Clad. cry = <italic>Cladonia cryptochlorophaea</italic>; Clad. fm = <italic>Cladonia fmbriata</italic>; Clad. fur = <italic>Cladonia furcata</italic>; Clad. och = <italic>Cladonia ochrochlora</italic>; Clad. pyx = <italic>Cladonia pyxidata</italic>; Clad. rnf = <italic>Cladonia rangiformis</italic>; Clad. rei = <italic>Cladonia rei</italic>; Clad. sca = <italic>Cladonia scabriuscula</italic>; Clad. squ = <italic>Cladonia squamosa</italic>; Clad. sub = <italic>Cladonia subulata</italic>; Dicr. bon = <italic>Dicranum bonjeanii</italic>; Dicr. fla = <italic>Dicranum flagellare</italic>; Dicr. maj = <italic>Dicranum majus</italic>; Dicr. pol = <italic>Dicranum polysetum</italic>; Dicr. sco = <italic>Dicranum scoparium</italic>; Drep. adu = <italic>Drepanocladus aduncus</italic>; Eurh. ang = <italic>Eurhynchium angustirete</italic>; Eurh. pra = <italic>Eurhynchium praelongum</italic>; Eurh. pul = <italic>Eurhynchium pulchellum</italic>; Hylo. spl = <italic>Hylocomium splendens</italic>; Loph. exc = <italic>Lophozia excisa</italic>; Plag. aff = <italic>Plagiomnium afne</italic>; Plag. med = <italic>Plagiomnium medium</italic>; Plag. lae = <italic>Plagiothecium laetum</italic>; Pleu. sch = <italic>Pleurozium schreberi</italic>; Pohl. nut = <italic>Pohlia nutans</italic>; Ptil. cil = <italic>Ptilidium ciliare</italic>; Ptil. crc = <italic>Ptilium crista-castrensis</italic>; Rhyt. tri = <italic>Rhytidiadelphus triquetrus</italic>. Relationships between species composition (NMDS ordination, <xref ref-type="fig" rid="F2">Fig. 2</xref>) and environmental factors on the dunes. H <sub>rel</sub> - relative dune height (m); H<sub>abs</sub> - absolute dune height (m.a.s.l); PAR - below-canopy photosynthetically active radiation (μmol m<sup>-2</sup>s<sup>-1</sup>); VWC<sub>aver</sub> - soil volumetric water content (%); AvpH - average pH<sub>H2O</sub>; EC - average electrical conductivity (μS cm<sup>-1</sup>); Cancov - canopy cover; O<sub>i</sub> - poorly decomposed soil organic horizon thickness (cm); O<sub>e</sub> - medium decomposed soil organic horizon thickness (cm); O<sub>a</sub> - well decomposed soil organic horizon thickness (cm); A - humic horizon thickness (cm); N - nitrogen content (%); P - phosphorus content (mg kg<sup>-1</sup>); K - potassium content (mg kg<sup>-1</sup>); Ca - calcium content (mg kg<sup>-1</sup>); Vasccov - vascular plant species coverage (%); Loc - location on dune; Asp - aspect of the quadrat (N; S; E; W). sc signifcance codes: '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 'n.s.' not signifcant. Discussion

Coastal sand dunes are one of the most threatened ecosystems in the world because of human activities and plant invasion ( Cogoni et al. , 2011 ; Vaz et al. , 2014 ). Changes in traditional land use and invasions of non-native species affect the composition and structure of native vegetation and may lead to loss of biodiversity ( Latorre et al. , 2013 ; Rivis et al. , 2016 ). Most of the Estonian coastal vegetation consists of semi-natural plant communities with moderate human management ( Rivis et al. , 2016 ). Lichens and bryophytes are considered to be pioneer species on open dune areas; they form habitats for other plants, collect humus and wither the substrate. They also play an important role in protecting the surface of the dunes from water and wind erosion ( Cogoni et al. , 2011 ) as well as provide stability, nutrients and moisture to the soil (Fernández & Barradas, 1997). Therefore, the significance of lichens and bryophytes on dunes should not be underestimated. In northern and boreal temperate zones, lichen-rich assemblages are restricted to edaphically specific stressful environments with nutrient insufficiency ( Košuthovà et al. , 2015 ).

In Estonia 594 species of bryophytes (according to Vellak et al. , 2015 ) and 1176 species of lichens (accor­ding to Randlane et al. , 2016 ) have been identified. In this study, a total of 30 species of epigeic bryophytes and 22 species of epigeic lichens were distinguished on the quadrats. In forests of the northern hemisphere, epigeic bryophyte communities are primarily dominated by a few common species with very broad ranges of tolerance such as P. schreberi ( Frego, 2007 ). As regards this point, the dune forest in our study area is not an exception and P. schreberi was observed in 90% and H. splendens in 64% of the quadrats respectively. The species composition on Rannametsa dunes seems to have been quite stable over three decades as the most common species in our study were also mentioned by örd (1972) , who pointed out different Cladonia species and Cetraria islandica (L.) Ach. as the most common species in the same dune area. Two lichen species are according to the IUCN Red List near threatened (NT) and vulnerable (VU) in Estonia: Cladonia portentosa (Dufour) Coem. (NT) and Cladonia scabriuscula (Delise) Nyl. (VU).

Soil and vegetation units of the coastal landscape in Estonia strongly depend on the topography, geological structure and water regime ( Rivis et al. , 2016 ). Haplic Podzols, characteristic of Estonian coastal areas, are nutrient-poor soils. The most important plant growth-limiting elements in the terrestrial ecosystems are nitrogen and phosphorus ( Güsewell, 2004 ). The high value of the N: P ratio indicates P limitation in the observed dune sands. Our results showed that the N: P and N: K ratios did not differ between the different locations. Only the Ca: Mg ratio was significantly different between locations, varying from 3.9 to 6.0. Dune soils on the bottoms of the dunes showed lower pH values which can be explained by accumulation of soil organic material (debris of coniferous trees, shrubs, accumulation of acid humus) ( Sewerniak & Jankowski, 2017 ; Sewerniak et al. , 2017 ). Nutrient limitation and nutrient availability affect the competition between different species and the species composition of plant communities ( Chapin et al. , 1986 ; Koerselman & Meuleman, 1996 ).

According to ordination analysis, bryophyte and lichen species composition on fixed P. sylvestris dominated dunes was most significantly determined by aspect, vascular plant species cover, amount of photosynthetically active radiation, thickness of the O e sub-horizon and soil volumetric water content, acidity and electrical conductivity. As observed by Sewerniak et al. (2017) , on dune areas topography plays an important role in soil development, thereby affecting the variability and distribution of vegetation. Our results confirmed this statement as far as epigeic bryophytes and lichens were concerned. As to an earlier finding by Sewerniak (2016) , the height and slope aspect play an important role in determining the light and moisture regime on dune forests, our results confirmed the importance of the slope aspect in modifying the soil moisture regime on sites with different cardinal directions.

Zonal variation of ground vegetation is considered typical for dunes, although it has been described mostly for vascular plant species ( Lane et al. , 2008 ; Tilk et al. , 2017 ). The study on Rannametsa dunes showed that the species richness of bryophytes and lichens varied along the dune profile. According to our results, bryophytes preferred lower and moister bottoms of the dunes where their species richness was the highest. In addition, the species composition of bryophytes on the bottoms differed significantly from the species composition of the slopes and tops, according to the results of MRPP tests. This can be explained by more humid soils on the bottoms of the dunes. Higher moisture availability on dunes raises plant-specific diversity and the number of plant groups increases ( Latorre et al. , 2013 ). Soil moisture is important to moss species composition ( Proctor, 2008 ), and fine-scale differences in soil moisture probably have a major effect on the bryophyte community ( Grytnes et al. , 2006 ). Analysis of indicator species also pointed out Brachythecium erythrorrhizon and B. oedipodium (species preferring moist habitats according to Düll (1991) ) as indicator species for dune bottoms.

At the same time, the lichen species richness was the highest on the slopes of the dunes where patches without vascular plants occurred and bryophyte species were not dominated. In stable and stress-free areas, slow-growing lichens are usually crowded out by vascular plants and by the more vigorous bryophytes because of lichens inability to cope with shading by shrubs or tree canopies, accumulation of humus and leaf litter or substrate instability ( Hale, 1974 ). According to our results, the composition of bryophyte and lichen species was affected by the cover of vascular plant species; similar results were registered by Pharo et al. (1999) , who found that the cover of vascular plants explains significant variations in fern, bryophyte and lichen species richness.

The light radiation are another key factor determining the diversity and composition of terrestrial lichens ( Palmquist, 2000 ; Márialigeti et al. , 2016 ) because lichens require sufficient light for photosynthesis and growth, while mosses are able to efficiently take advantage of the low irradiances in the shade of the tree canopies ( Kolari et al. , 2006 ). Terrestrial lichens are probably limited by light availability rather than soil moisture ( Palmquist, 2000 ). Based on our results, certain lichens (species of Cladonia ) seem to prefer open dune areas, where PAR is high and soil moisture is low. The proportion of light-demanding species was high among lichen species (the most frequent values of light indices 6-8). Bryophyte species showed more variable light preferences (the values of light indices varied between 3 and 9) and both shade-demanding and full-light requiring bryophyte species were represented on dunes.

Acidity preference of bryophyte species was also variable, ranging on a scale from acid to weakly acid/weakly neutral. Although bryophytes receive nutrients mainly from precipitation (including leachates from tree canopies and plant leaves) ( Oechel & Van Cleve, 1986 ), the nutrient properties and pH of the substrate can also be important and affect bryophyte and lichen species composition ( Pausas, 1994 ). Ketner-Oostra & Sykora (2000) found that when carbon content and acidity increase, the vegetation will change from lichen-poor to moss-rich communities. The number of moss species is more limited by soil moisture than by soil nutrients and a positive relationship for the pH was found by Pausas (1994) . Based on NMDS ordination it can be concluded that lichen and bryophyte species composition was dependent on the soil pH, although a previous work compiled in the dune area in south-west Estonia ( Kösta & Tilk, 2008 ) found no influence of the soil pH on the distribution of bryophyte and lichen species.

In addition, a previous study ( Sun et al. , 2013 ) indicates that bryophyte distribution depends on the depth of litter. Our results agree with this statement, as the bryophyte and lichen species composition was affected by the thickness of the moderately decomposed organic layer. Magnusson (1982) found that certain lichens (species of Cladonia ) colonise coastal dune areas in southern Sweden where litter has accumulated. Considering future research, we suggest that lichens and bryophytes should be separately analysed because of their different responses to environmental factors.

Conclusions

Altogether 30 species of bryophytes and 22 species of lichens were found on the study area. The average lichen species richness was the highest on the slopes of the dunes, while the highest average bryophyte species richness occurred on the bottoms of the dunes. The highest number of lichen species was identified as light-demanding while the light preference of bryophyte species was more variable. The distribution of bryophyte and lichen species between zones showed significant differences, and therefore it can be concluded that zonation of bryophyte and lichen species can be found on dunes just as it has been described for vascular plant species. In conclusion, it can be said that factors affecting the bryophyte and lichen species composition are related to the dune slope effect, which causes differentiation of soils and therefore a complex of different microhabitats populated by different species. According to our results, some of the most important factors are the aspect of the dune, vascular plant species cover, light conditions, a thickness of the medium decomposed soil organic horizon and soil pH, electrical conductivity and volumetric water content.

 Acknowledgments

The authors would like to thank Leiti Kannukene for her assistance in the determination of bryophytes. Thanks to Ede Oja and Ave Suija for their assistance in the determination of some unusual lichen species.

 References Ahti T Oksanen J 1990 Epigeic lichen communities of Taiga and Tundra regions. Vegetatio 86 39 70 https://doi.org/10.1007/BF00045134 Auslander M Nevo E Inbar M 2003 The effects of slope orientation on plant growth, developmental instability and susceptibility to herbivores. J Arid Environ 55 405 416 https://doi.org/10.1016/S0140-1963(02)00281-1 Bates D Maechler M Bolker B Walker S 2015 Fitting Linear Mixed-Effects Models Using lme4. J Stat Softw 67 1 48 https://doi.org/10.18637/jss.v067.i01 Bond-Lamberty B Gower ST 2007 Estimation of stand-level leaf area for boreal bryophytes. Oecologia 151 584 592 https://doi.org/10.1007/s00442-006-0619-5 Canullo R Starlinger F Giordani P 2013 Chapter 13 - Diversity and composition of plant and lichen species. Dev Environ Sci 12 237 250 Chapin F Vitousek P van Cleve K 1986 The nature of nutrient limitation in plant communities. Am Nat 127 48 58 https://doi.org/10.1086/284466 Ciccarelli D 2015 Mediterranean coastal dune vegetation: Are disturbance and stress the key selective forces that drive the psammophilous succession? Estuar Coast Shelf Sci 165 247 253 https://doi.org/10.1016/j.ecss.2015.05.023 Cogoni A Brundu G Zedda L 2011 Diversity and ecology of terricolous bryophyte and lichen communities in coastal areas of Sardinia (Italy). Nova Hedwigia 92 159 175 https://doi.org/10.1127/0029-5035/2011/0092-0159 Dufrene M Legendre P 1997 Species assemblages and indicator species: the need for a flexible asymmetical approach. Ecol Monogr 67 345 366 https://doi.org/10.2307/2963459 Düll R 1991 Zeigerwerte von Laub-und Lebermoose Scripta Geobotanica 18 175 214 In: Zeigerwerte von Pflanzen in Mitteleuropa; Ellenberg H, Weber H, Düll R, Wirth V, Werner W Paulissen D (eds). Forey E Chapelet B Vitasse Y Tilquin M Touzard B Michalet R 2008 The relative importance of disturbance and environmental stress at local and regional scales in French coastal sand dunes. J Veg Sci 19 493 502 https://doi.org/10.3170/2008-8-18392 Frego KA 2007 Bryophytes as potential indicators of forest integrity. Forest Ecol Manag 274 65 75 https://doi.org/10.1016/j.foreco.2007.01.030 Grytnes JA Heegaard E Ihlen PG 2006 Species richness of vascular plants, bryophytes, and lichens along an altitudinal gradient in western Norway. Acta Oecol 29 241 246 https://doi.org/10.1016/j.actao.2005.10.007 Güsewell S 2004 N:P ratios in terrestrial plants: Variation and functional significance. New Phytol 164 243 266 https://doi.org/10.1111/j.1469-8137.2004.01192.x Hale ME 1974 The Biology of Lichens. William Clowes & Sons Limited, London. 192 Hill MO 1979 The development of a flora in even-aged plantations. Institute of Terrestrial Ecology, Cambridge. 175 192 In: The Ecology of Even-aged Forest Plantations; Ford ED, Malcolm DC, Atterson J (eds). Ingerpuu N 2002 Bryophyte diversity and vascular plants. Dissertationes Biologicae Universitatis Tartuensis, Tartu. Ingerpuu N Kalda A Kannukene L Krall H Leis M Vellak K 1998 Key-book of Estonian bryophytes. EPMü Zooloogia ja Botaanika Instituut, Tartu. Jenny H 1941 Factors of Soil Formation. McGraw Hill, New York Jun R Rozé F 2005 Monitoring bryophytes and lichens dynamics in sand dunes: example on the French Atlantic coast. VLIZ Special Publication, Koksijde. In: Dunes and Estuaries 2005; Herrier J-L, Mees J, Salman A, Seys J, Van Nieuwenhuyse H, Dobbelaere I (eds). Jüriado I Kämärä M-L Oja E 2016 Environmental factors and ground disturbance affecting the composition of species and functional traits of ground layer lichens on grey dunes and dune heaths of Estonia. Nord J Bot 34 244 255 https://doi.org/10.1111/njb.00936 Ketner-Oostra R Sykora KV 2000 Vegetation succession and lichen diversity on dry coastal calcium-poor dunes and the impact of management experiments. J Coast Conserv 6 191 206 https://doi.org/10.1007/BF02913815 Koerselman W Meuleman AF 1996 The vegetation N:P ratio: A new tool to detect the nature of nutrient limitation. J Appl Ecol 33 1441 1450 https://doi.org/10.2307/2404783 Kolari P Pumpanen J Kulmala L Ilvesniemi H Nikinmaa E Grönholm T Hari P 2006 Forest floor vegetation plays an important role in photosynthetic production of boreal forests. Forest Ecol Manag 221 241 248 https://doi.org/10.1016/j.foreco.2005.10.021 Kösta H Tilk M 2008 Variability of bryophytes and lichens on a forested coastal dune Tõotusemägi in Southwestern Estonia. Forestry Studies 49 71 80 Košuthovà A Svitkovà I Pišùt I Senko D Valachovic M Zaniewski PT Hàjek M 2015 Climatic gradients within temperate Europe and small-scale species composition of lichen-rich dry acidophilous Scots pine forests. Fungal Ecol 14 8 23 https://doi.org/10.1016/j.funeco.2014.10.005 Lane C Wright SJ Roncal J Maschinski J 2008 Characterizing environmental gradients and their influence on vegetation zonation in a subtropical coastal sand dune system. J Coastal Res 24 213 224 https://doi.org/10.2112/07-0853.1 Latorre EC Fagúndez C Dacosta E Canavero A 2013 Composition and vegetation structure in a system of coastal dunes of the "de la Plata" river, Uruguay: a comparison with Legrands's descriptions (1959). Braz J Bot 36 9 23 https://doi.org/10.1007/s40415-013-0009-2 Lemauviel S Gallet S Rozé F 2003 Sustainable management of fixed dunes: Example of a pilot site in Brittany (France). C.R. Biol 326 183 191 https://doi.org/10.1016/S1631-0691(03)00056-8 Löbel S Dengler J Hobohm C 2006 Species richness of vascular plants, bryophytes and lichens in dry grasslands: The effects of environment, landscape structure and competition. Folia Geobot 41 377 393 https://doi.org/10.1007/BF02806555 Magnusson M 1982 Composition and succession of lichen communities in an inner coastal area in southern Sweden The Lichenologist 14 153 163 https://doi.org/10.1017/S0024282982000292 Mandre M Tilk M Kõresaar P 2008 Chemical characteristics of soils in Scots pine forests of Cladina and Vaccinium vitis-idaea site types on coastal dunes of Baltic Sea. Forestry Studies 49 5 12 Masing V 1979 Botaanika III. Valgus, Tallinn, 414 Márialigeti S Tinya F Bildó A Ódor P 2016 Environmental drivers of the composition and diversity of the herb layer in mixed temperate forests in Hungary. Plant Ecol 217 549 563 https://doi.org/10.1007/s11258-016-0599-4 McCune B Mefford MJ 2011 PC-ORD. Multivariate Analysis of Ecological Data. Version 6 MjM Software, Oregon MCPFE Liaison Unit Vienna 2002 Improved Pan-European indicators for sustainable forest management. Liaison Unit Vienna, Vienna, 6 McLachlan A Brown AC 2006 The Ecology of Sandy Shores Academic Press, Burlington. Mielke PW Berry KJ Johnson ES 1976 Multi-response permutation procedures for a priori classifications. Commun Stat- Theor M 5 1409 1424 Oechel WC Van Cleve K 1986 The role of bryophytes in nutrient cycling in the Taiga. Springer, Berlin 121 137 In: Forest ecosystems in the Alaskan taiga; Van Cleve K, Chapin IF, Flanagan PW, Viereck LA, Dyrness CT (eds). https://doi.org/10.1007/978-1-4612-4902-3_9 Oksanen J 2015 Multivariate Analysis of Ecological Communities in R: vegan tutorial http://cc.oulu.fi/~jarioksa/opetus/metodi/vegantutor.pdf Oksanen J Blanchet FG Friendly M Kindt R Legendre P McGlinn D Minchin PR O'Hara RB Simpson GL Solymos 2016 Vegan: Community Ecology Package. R package version 2.4-1 https://cran.r-project.org/web/packages/vegan/index.html örd A 1972 On soils of dune forests in South-western Estonia Forestry Studies 9 207 221 [In Estonian]. Palmquist K 2000 Carbon economy in lichens New Phytol 148 11 36 https://doi.org/10.1046/j.1469-8137.2000.00732.x Pausas JG 1994 Species richness patterns in the understorey of Pyrenean Pinus sylvestris forest J Veg Sci 5 517 524 https://doi.org/10.2307/3235978 Pharo EJ Beattie AJ Binns D 1999 Vascular plant diversity as a surrogate for bryophyte and lichen diversity. Conserv Biol 13 282 292 https://doi.org/10.1046/j.1523-1739.1999.013002282.x Pihelgas E 1983 Metsabioloogia. Valgus, Tallinn 223 Proctor MC 2008 Physiological ecology. Cambridge University Press, Cambridge. 237 268 In: Bryophyte Biology 2; Goffinet B, Shaw AJ, (eds). https://doi.org/10.1017/CBO9780511754807.007 R Core Team 2015 R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna https://www.r-project.org/ Randlane T Saag A Suija A 2016 Lichenized, lichenicolous and allied fungi of Estonia Ver. December 31, 2016 http://esamba.bo.bg.ut.ee/checklist/est/home.php Rivis R Kont A Ratas U Palginõmm V Antso K Tõnisson H 2016 Trends in the development of Estonian coastal land cover and landscapes caused by natural changes and human impact J Coast Conserv 20 199 209 https://doi.org/10.1007/s11852-016-0430-3 Sewerniak P 2016 Differences in early dynamics and effects of slope aspect between naturally regenerated and planted Pinus sylvestris woodland on inland dunes in Poland. iForest 9 875 882 Sewerniak P Jankowski M 2017 Topographically-controlled site conditions drive vegetation pattern on inland dunes in Poland. Acta Oecol 82 52 60 https://doi.org/10.1016/j.actao.2017.06.003 Sewerniak P Jankowski M Dabrowski M 2017 Effect of topography and deforestation on regular variation of soils on inland dunes in the Torun Basin (N Poland). Catena 149 318 330 https://doi.org/10.1016/j.catena.2016.10.008 Solon J Degòrski M Roo-Zielinska E 2007 Vegetation response to a topographical-soil gradient. Catena 71 309 320 https://doi.org/10.1016/j.catena.2007.01.006 Sun S-Q Wu Y-H Wang G-X Zhou J Yu D Bing H-J Luo J 2013 Bryophyte species richness and composition along an altitudinal gradient in Gongga Mountain, China. PLoS ONE 8 https://doi.org/10.1371/journal.pone.0058131