Origin of a boreal birch bog woodland and landscape development on a warm low mountain summit at the Carpathian

Abstract

Hilly regions along the Western Carpathian–Pannonian border are phytogeographically important, but their vegetation history remains largely unknown. We analysed two peat cores of Late Glacial origin from a bog woodland in the Malé Karpaty Mts (SW Slovakia) using plant macrofossil, pollen, peat chemistry and charcoal analyses to trace local successional patterns, regional vegetation development and occurrence of rare species. The small distance between the two profiles situated within homogeneous vegetation enabled us to explore small-scale differences in local vegetation history. The sediment started to accumulate at the end of the Allerød (ca. 12950 cal. yr BP), when a shallow oligotrophic/mesotrophic lake with macrophytes developed. Open pine-birch forests dominated in the landscape. During the early Holocene, the lake was infilled, mire vegetation appeared and broad-leaved forests spread in the surroundings. Two fire events indicated by increases in number of macroscopic charcoal particles were recorded. The first one, which occurred at the end of the Late Glacial, was found only in one of the profiles, while the second one affected entire mire and probably caused a hiatus spanning the middle and late Holocene. Fagus started to spread no later than 5800 cal. yr BP. Open mire vegetation reappeared after the fire (ca. 400 cal. yr BP). During the 19th century, the mire was overgrown by a birch bog woodland. The two profiles showed basically the same successional patterns, but some local events and occurrences of rare species (Potamogeton alpinus, Potamogeton praelongus, Scorpidium scorpioides and Pleurospermum austriacum) were traced only in one of them.

Keywords

charcoal lake sediment macrofossil peat chemistry phytogeography pollen

Introduction

The Western Carpathian–Pannonian border is an important contact area between contrasting phytogeographical zones. To understand the vegetation development in the area, it is necessary to cover the entire Holocene. The vegetation history of this area is therefore insufficiently known. In the wider region, palaeoecological research has concentrated either on mountainous regions (e.g. Hájková et al., 2015a; Jankovská, 1988; Rybníčková and Rybníček, 2006) or on the Pannonian lowland (e.g. Jamrichová et al., 2014; Magyari et al., 2010; Pišút et al., 2010) and tended to avoid the contact zone between these two regions. The few studies that are available lack reliable absolute chronologies (e.g. Krippel, 1965; Rybníčková and Rybníček, 1972). Some radiocarbon-dated profiles from this contact zone do exist from the Czech Republic (Rybníček and Rybníčková, 2008), Slovakia (Hájková et al., 2013; Petr et al., 2013) and Hungary (Magyari et al., 2001; Willis et al., 1998), but most of them do not cover the entire Holocene, and the studies carried out on them are focused on a single proxy. No study has dealt with summit parts of low mountains rising from the warm Pannonian lowland. One of the main reasons why so little is known about their Holocene history is the scarcity of old sediments in these mountains. The climate and geology do not support the formation of ombrotrophic bogs, unlike the more mountainous regions of northern Slovakia, where they are an important source of palaeoecological data (Rybníček and Rybníčková, 2002). Moreover, minerotrophic mires are generally subjected to higher rates of mineralization, and their sediments are not sufficiently preserved in warmer regions. In addition, they have recently become scarce in the study region because of drainage and agricultural or other land uses. Lakes ceased to exist in the study area as well (Petr et al., 2013; Rybníčková and Rybníček, 1972; Svobodová, 1997). Thus, sediments suitable for palaeoecological studies are often buried, overlain by sediments of anthropogenic origin and missing a record of younger Holocene periods (Hájková et al., 2015b; Jamrichová et al., 2014).

The Malé Karpaty Mts are a low mountain range sharply rising from the Pannonian plain. To the west, they are surrounded by the Borská lowland, a marginal part of the Pannonian area. Northwards, only small hills connect the Malé Karpaty Mts to the Carpathian flysch zone (the White Carpathians Mts). The Malé Karpaty Mts are therefore a crucial region for understanding the vegetation development in the Pannonian-Carpathian contact zone. Our study site (Nad Šenkárkou Nature Reserve) is a rare boreal-oceanic birch bog woodland habitat (Eriophoro vaginati–Betuletum pubescentis; Valachovič and Hegedüšová, 2013). The isolated presence of a boreal-like mire habitat might suggest an old age of the sediment and thus provide a rare and excellent opportunity to explore the entire Holocene history of the region.

Birch bog woodlands in Central Europe are considered to be successionally young forests, often occurring at mire margins, sometimes in a mosaic with Pinus bog woodlands (Chytrý et al., 2013). However, not much is known about their Holocene history and dynamics, and palaeoecological studies are rare (e.g. Karpińska-Kołaczek et al., 2014). Some studies from Central and Eastern Europe show phases rich in Betula already in the early Holocene or the Late Glacial (e.g. Bešta et al., 2009 in Czech Republic; Karpińska-Kołaczek et al., 2013 and Kołaczek et al., 2015 in Poland), but there is no evidence about whether contemporary birch mires could have a relic origin. On the other hand, Karpińska-Kołaczek et al. (2014) show the possibility of their long-term persistence (continuity of 2500 years), so it cannot be excluded. Therefore, concerning the local development, we postulate the hypothesis that the boreal-like birch mire woodland on its southern boundary is a cold-stage relict rather than only temporary successional stage.

In minerotrophic mires and fens, studies based on two or more profiles per site are rare (e.g. Hájková et al., 2012). Studies from lakes (Bos et al., 2006; Muller et al., 2003) and ombrotrophic bogs (Barber et al., 1998; Chiverrell, 2001; Streel et al., 2014) have shown that analysing more than one core per site is useful and it gives a better picture about the past development and spatial structure of the study site. Moreover, in bog woodlands and carrs, stratigraphy may be complicated because of sediment degradation by the activity of tree roots, which can cause hiatuses like in alder carrs (Pokorný et al., 2000). Even though some papers indicate that expansion of birch did not influence pollen preservation and peat accumulation (e.g. at Lake Czarne in NE Poland; Karpińska-Kołaczek et al., 2014), we decided to compare two profiles in order to compensate for this risk and to obtain better insight into site history.

We took a multi-proxy approach comprising pollen, macrofossil and charcoal analysis and geochemical analyses including loss on ignition (LOI). While the pollen analysis provided information about vegetation composition on a broader regional scale, other proxies helped to trace the history of the habitat (Birks and Birks, 2000). We aimed to reconstruct the Late Glacial and Holocene history of the site and to trace the past occurrence of some rare and relic plant species.

Materials and methods

Site description

The study site ‘Nad Šenkárkou’ is located about 5 km from the village of Limbach (Bratislava district, western Slovakia; Figure 1). It belongs to the Inner Western Carpathians, being situated in the southern part of the Malé Karpaty Mts, in the saddle between the hills Tri kamenné kopce (584 m a.s.l.) and Hrubý vrch (572 m a.s.l.). The mire lies at an elevation of 558–569 m a.s.l. and has been a Nature Reserve (10.92 ha) since 1984. The geological bedrock is formed by granitoids of a crystalline core. The climate of the region is slightly warm with mean annual temperatures of 6–7°C (mean in January −4 to −5°C, mean in July 14 to 16°C), and mean annual precipitation of 700–800 mm (http://geo.enviroportal.sk/atlassr). The mire developed in a terrain depression, where rainwater and surface water from adjacent slopes accumulated. The central part of the mire is vegetated by a birch carr (Eriophoro vaginati–Betuletum pubescentis Hueck 1931) with hollows dominated by Sphagnum fallax and hummocks dominated by Eriophorum vaginatum and Polytrichum commune. Its marginal parts are overgrown by an alder carr (Carici elongatae-Alnetum Schwickerath 1933).

Figure 1.

Position of the study site Nad Šenkárkou and position of other peat profiles in the region discussed in the text.

Sediment sampling and dating

To explore the heterogeneity of past development, we studied two different peat profiles, both sampled in the central part of the mire where the sediment is deepest and supposedly oldest. The idea to explore the site arose independently in two research groups (one of them led by the third author and the other composed of the rest), who later decided to combine their research efforts. This is the reason why the coring equipment and chemical treatment of pollen samples differ between the two cores.

One of the cores was longer (109 cm), while the other was slightly shorter (100 cm). Profile 1 was taken from the deepest point (hereafter referred to as S1; 48°18′48.5″N, 17°10′43.6″E). The section was obtained using a single gouge auger (6 cm diameter, 100 cm length). Profile 2 (hereafter referred to as S2; 48°18′48.2″N, 17°10′44.5″) was situated about 20 m from Profile 1. This profile was obtained using a chamber corer (5 cm diameter, 50 cm length). Selected samples of macrofossils of terrestrial plants (seeds of taxa specified in Table 1, spindles of Eriophorum, Betula bark) and ephippia of Cladocera were sent for AMS dating to the Centre for Applied Isotope Studies, University of Georgia, Athens, USA.

Table 1.

Results of ¹⁴C dating (AMS method) from the peat profiles studied. The calibrated ages are median values and intervals of the calibrated 2σ range BP.

Samples	Depth (cm)	Dating method	Age in uncal. BP	Cal. yr BP (interval)	Cal. yr BP (median)	Material
Šenkárka 1
UG-13295	22–25	AMS	390 ± 20	505–331	454	Bark
UG-12422	37–40	AMS	5130 ± 50	5991–5746	5858	Seeds (Betula, Ranunculus)
UG-12423	52–55	AMS	9060 ± 30	10,249–10,194	10,222	Spindles (Eriophorum)
UG-12424	70–73	AMS	9320 ± 30	10,650–10,422	10,529	Seeds (Ranunculus)
UG-12425	88–91	AMS	10,240 ± 30	12,090–11,826	11,976	Seeds (Menyanthes)
UG-11632	106–109	AMS	10,300 ± 30	12,376–11973	12,093	Seeds (Menyanthes, Carex)
Šenkárka 2
UG-18787	20–25	AMS	1510 ± 20	1516–1342	1388	Spindles (Eriophorum)
UG-15701	30–35	AMS	320 ± 20	458–307	383	Bark
UG-18788	40–45	AMS	720 ± 25	695–571	672	Ephippia (Cladocera), Seeds (Typha)
UG-18789	50–55	AMS	10,040 ± 30	11,749–11,360	11,535	Seeds (Menyanthes, Ranunculus, Eleocharis)
UG-15700	80–85	AMS	10,510 ± 30	12,562–12,406	12,484	Seeds (Menyanthes, Carex, Sparganium)

An age–depth model with 1-cm resolution was constructed for the S1 profile based on a P_Sequence function with the k parameter equal to 0.5 cm⁻¹ and log10(k/k0) equal to 0.3 using OxCal 4.2.4. (Bronk Ramsey, 2009). To incorporate into the model the horizons of potential changes in sedimentation rate, such as the contact of different types of deposits, the command Boundary was applied. The boundaries were placed at 80 cm (lake-mire transition) and at 10 cm (transition from decomposed to undecomposed peat). In the text below, we use mean values of modelled data in the range of 95.4% rounded to the nearest 50. In the S2 profile, we used linear interpolation between middle values of calibrated dates.

Palaeoecological analyses

Samples for pollen analysis (1 cm³) were treated by acetolysis (profile S1; Faegri and Iversen, 1989) or 10% KOH and sodium hexametaphosphate (profile S2; Pacltová, 1963). A minimum of 500 terrestrial pollen grains were counted and determined using pollen keys (Beug, 2004; Punt and Clarke, 1984; Reille, 1992) and the reference collection of the Institute of Botany of the Czech Academy of Sciences. Non-pollen palynomorphs were identified according to Van Geel et al. (1980). The nomenclature of all determined pollen types follows Beug (2004) except for the Daucaceae (Punt and Clarke, 1984). Percentage pollen diagrams were constructed using the total sum (TS) comprising arboreal and non-arboreal pollen. Aquatic and clearly local plants, algae, fungi and other non-pollen palynomorphs were excluded from the TS. The microscopic charcoal particles (fraction 0.01–0.1 mm) were expressed as absolute value counts (numbers of particles).

Material for plant macrofossil analysis was sub-sampled at 3 cm (in profile S1) and 5 cm (in profile S2 and upper 20 cm of profile S1) intervals along the profiles. Because the analysed volume of sediment differed among the samples and profiles depending on the amount of available material, final macrofossil counts were standardized to 100 mL. The samples were wet-sieved using a set of sieves with mesh diameters of 1, 0.5 and 0.2 mm. Seeds and plant remains were identified under a stereoscopic microscope according to Cappers et al. (2006), Velichkevich and Zastawniak (2006, 2008) and other available identification literature. They were also checked against the reference seed collection of the Institute of Botany of the Czech Academy of Sciences. Wood fragments and bryophytes were determined using an optical microscope and suitable identification literature (Schweingruber, 1978; Smith, 1996; and others). Identification of macroscopic charcoal particles (pieces larger than 0.5 mm) was carried out under an episcopic microscope with the help of an interactive identification key (Heiss, 2000) in addition to the standard literature (Schweingruber, 1978). Values of macrofossils are presented as absolute numbers per sample for countable fossils (seeds, needle tips, Cladocera, etc.) or as volume percentages for bryophytes and wood fragments. The nomenclature follows Marhold and Hindák (1998) for vascular plants and Kučera et al. (2012) for bryophytes.

Macrofossil and macro-element results are presented using C2 software (Juggins, 2003). Pollen diagrams were created in Tilia v. 1.7.16 (Grimm, 2011). Pollen data from the two profiles were compared using detrended correspondence analyses (DCA) as implemented in Canoco 5 (Šmilauer and Lepš, 2014). Percentage data were arcsine-transformed; algae were passive in the analysis.

Chemical analyses

The weight percentage of organic matter was determined by means of LOI according to Heiri et al. (2001) and Holliday (2004) in each sample. The samples were dried at 105°C for 24 h, and the combustion at 550°C took 3 h. The chemical analyses mainly covered ICP analyses of the main macro-elements. The acidic leachable fraction of macro-elements was measured in each sample. Each precisely weighed sample (1 g) was stepwise combined with 20 mL of 20% aqueous hydrochloric acid (ultrapure, Merck) in a plastic test tube (Apollo type, 50 mL volume) covered by a watch glass. After the reaction ceased, the content was left to stand overnight at room temperature. The remaining solid material was filtered through a sheet of filter paper (blue strip) and extensively washed with water. Finally, the volume was adjusted to 100 mL in a volumetric flask. The chemical analyses were performed using the ICP EOS technique on an Intrepid DUO spectrometer (ThermoFisher) at the Institute of Geology ASCR in Prague. The amounts of macro-elements (Ca, Fe, Al, K, Mg, Mn, Na, P, S and Si) were measured using the standard experimental conditions recommended by the manufacturer. The instrument was set to dual plasma view mode and 2.5 mL min⁻¹ sample intake rate. For calibration, mixed standard solutions of all analysed elements with nominal concentrations of 0, 2.5, 5, 10, 20 ppm were prepared by combining commercial 1000 ppm solutions of single elements (produced by Analytika s.r.o., Prague) in a 2% nitric acid solution. The following wavelengths were used to quantify the elements: 396.1 nm for Al, 317.9 nm for Ca, 259.9 nm for Fe, 766.4 nm for K, 285.2 nm for Mg, 259.3 nm for Mn, 589.5 nm for Na, 189.5 nm for P, 182.0 nm for S and 212.4 nm for Si. A quality control sample was inserted into each analytical run.

Results and interpretations

Chronology

The model reached the agreement value of 98%. A reliable chronology was available in the bottom part of each profile (100–55 cm), holding a continuous record of sediment accumulation in a shallow lake at the end of the Late Glacial and the beginning of the early Holocene. Also, the error values of modelled dates in S1 are rather low in this part of the profile (ca. 30–80 years), and the accumulation rate is relatively high (ca. 0.04–0.13 cm yr⁻¹) compared to other parts of the profile (with the exception of the depth of 71.5–80 cm with errors of 200–300 years and a lower accumulation rate of about 0.007 cm yr⁻¹). Even though there are no reversal dates in the S1 profile (see Table 1), which allowed us to create an age–depth model (Figure 2), we suspect a discontinuity in the peat and detected a strongly decomposed layer rich in macroscopic charcoal particles in the middle/late Holocene section. A probable hiatus is indicated by a ca. 4350-year difference between the layers at 52–55 cm, dated to ca. 10,250 cal. yr BP, and at 37–40 cm, dated to ca. 5850 cal. yr BP. As the sediment contains a high amount of macroscopic charcoal particles (mostly Pinus with an admixture of Betula) at the depth between 35 and 70 cm, peaking at 50 cm, it is highly probable that this hiatus was at least partially caused by a local fire (see also Van Geel et al., 2014). The next zone, characterized by a dominance of Fagus pollen (S1-p3), spans ca. 5400 years over a 15 cm thick layer. For this section, we suppose hiatus(es) caused by high decomposition rather than fire because of the absence of charcoal and the presence of strongly decomposed peat without determinable macrofossils. The high values of errors (250–900 years) in this part of the profile (ca. 24–53 cm) indicate rather weak reliability of the modelled dates. Despite all mentioned problems with hiatuses, the age–depth model is reliable as a whole, reaching an agreement value of 98%, but it is important to be careful when interpreting the data from sections with high errors of modelled dates. Parts of the profile with the supposed hiatus are also characterized by very low accumulation rate of 0.003–0.007 cm yr⁻¹ compared to undisturbed parts (0.04–0.13 cm yr⁻¹).

Figure 2.

Age–depth model for the S1 profile.

In the case of the S2 profile, we do not have an absolute chronology for the peat section at the depth of 0–50 cm, where there are three reversal dates. The date of 720 ± 25 BP for the depth of 40–45 cm (see Table 1) is not reliable because it does not correspond with the pollen record. The higher representation of mesophilous trees suggests early Holocene age (ca. 10,500–10,000 cal. BP; compare S1 profile), which is also supported by DCA ordination (Figure 3). As this zone (S2-p2; 35–50 cm) is characterized by a large amount of charcoal, we suppose that it represents the rest of the sediment destroyed by a local fire. Fire is probably also one of the reasons for the presence of a large probable hiatus at the depth of around 35 cm, before a steep onset of Fagus. Other explanations of the big age difference between nearby peat layers are decomposition and/or slow sediment accumulation. As concerns the other two reversal dates, we cannot unequivocally decide which of them is more reliable, but the date of 320 ± 20 BP (depth 30–35 cm) seems to suit better than 1510 ± 20 BP (depth 20–25 cm) when compared with the S1 profile, which is also supported by the onset of anthropogenic indicators. Thus, we used this date for linear interpolation, the results of which we used in the diagrams and in the text.

Figure 3.

DCA ordination diagram of pollen samples from both profiles. We used arcsine-transformed percentage data. First axis: Eigenvalue 0.203, explained variability 26.3%. Second axis: Eigenvalue 0.066, explained variability 8.5%. Full circles are samples from S2 profile; empty circles are samples from S1.

Ordination analysis

Indirect ordination analyses (DCA) of all pollen samples from the S1 and S2 profiles show how the samples from the two profiles mutually correspond (Figure 3). Late Glacial and early Holocene samples from both profiles are located in one separate group, but the samples from both profiles are not shuffled. This could be caused by different degrees of preservation of particular herb taxa because the fossil record was rich in herb taxa in this stage. The second group of samples from burned layers confirmed that two samples from the S2 profile (at 38 and 43 cm) correspond well with the samples of early Holocene age from the S1 profile (dated to ca. 10,000 cal. yr BP). Thus, the results of ordination analyses have helped us reject the radiocarbon date of 720 ± 25 BP (UG-18788, Table 1) for this layer. The Fagus-stage group of samples represents the rest of the middle and late Holocene sediment, and the last group of upper samples (15 cm in the S1 profile or 23 cm in the S2 profile) represents the last ca. 200–250 years.

Regional vegetation development

As indicated by the pollen results (zones S1-p1, ca. 12,050–10,800 cal. yr BP and zone S2-p1, ca. 12,650–11,550 cal. yr BP), the landscape was dominated by pine-birch forests in a mosaic with open steppes (pollen of Artemisia, Chenopodiaceae, Poaceae, Thalictrum) at the end of the Late Glacial and at the beginning of the early Holocene. Some scattered temperate broad-leaved trees (especially Ulmus and Quercus; both S1 and S2) and shrubs (Cornus sanguinea t.; S1) could also occur in microclimatically and/or edaphically suitable habitats. Pleurospermum austriacum probably occurred in the herb layer of open pine-birch forests or in meadow steppes, like recently Pleurospermum uralense, a closely related species, in the southern Siberia (pers. obs.). The recent distribution of Pleurospermum austriacum is split into island-like parts, situated mostly in Eastern and Central Europe (Meusel et al., 1978), but its pollen grains (for photo, see Appendix S1, available online) are regularly found in Glacial (e.g. Jankovská et al., 2002; Kołaczek, 2007) and early Holocene sediments even outside the present-day area, which suggests its relic character (Hadač et al., 1967). The early Holocene sequence was dominated by pollen of Betula and Corylus, and a continual presence of pollen of other mesophilous trees (Fraxinus excelsior t., Quercus, Tilia cordata t. and Ulmus). Values of Pinus pollen strongly decreased, and Fagus pollen occurred in very low abundance. This developmental stage is better preserved in the S1 profile. The very steep increase of Fagus pollen values in the next stage (zones S1-p3, ca. 5850 cal. yr BP and S2-p3, ca. 390 cal. yr BP) is probably connected with a hiatus in the fossil record, which spanned an obviously longer period in S2 (almost the whole middle Holocene). Alongside the onset of Fagus, pollen of Carpinus also appeared. This stage is further characterized by a decrease of pollen of other mesophilous trees, an increase of the Poaceae pollen and at the end also by the appearance of pollen of cereals (Secale cereale, Hordeum t.) and other human indicators such as Plantago lanceolata, Rumex acetosa t., Urtica and so on (Gaillard, 2013). The last stage (zones S1-p4, ca. 190 cal. yr BP and S2-p4, ca. 120 cal. yr BP) represents a birch bog woodland indicated mainly by a steep increase in Betula pollen.

Local vegetation development

The bottom layer of Late Glacial age (zones S1-m1, S2-m1; ca. 12,100–11,900 and ca. 12,950–11,900 cal. yr BP; S1-p1, S2-p1, ca. 12,100–11,900 and ca. 12,650–11,900 cal. yr BP, respectively; Figures 4 –7) is characterised by large amounts of macrofossils and pollen of aquatic higher plants (Batrachium cf. trichophyllum, Myriophyllum cf. spicatum, Potamogeton pusillus, Potamogeton alpinus, Sparganium natans) and algae (Botryococcus, Nitella sp., Pediastrum boryanum, Pediastrum duplex, Pediastrum integrum). This species composition, together with peat chemistry, indicates the presence of a small shallow lake with rather oligotrophic and clean water (Freedman et al., 1989; Komárek and Jankovská, 2001; Søndergaard et al., 2010). Macrofossils of littoral species indicate that the littoral zone was occupied mostly by Carex rostrata, Menyanthes trifoliata and bryophytes, namely, Sphagnum palustre, Sphagnum sect. Subsecunda, Warnstorfia exannulata and, in very low quantities, also Calliergon cf. richardsonii and Scorpidium revolvens agg. (in S1 only). At the end of the zone, seeds of other species of muddy shores Eleocharis palustris agg., Ranunculus flammula and Ranunculus sceleratus appeared, indicating advancing terrestrialization and more frequent drying. The presence of trees in the close vicinity of the shallow lake is indicated by Betula, Picea and Pinus charcoals as well as seeds and needles, but in low abundances. Later, during the Late Glacial/early Holocene transition and shortly after it (zones S1-m2, S2-m2, ca. 11,900–11,250 and ca. 11,950–11,600 cal. yr BP respectively; upper part of S1-p1, S2-p1, ca. 11,900–10,800 and ca. 11,900–9000 cal. yr BP, respectively), the terrestrialization had continued, and seeds of littoral species started to dominate (Eleocharis palustris agg., Hippuris vulgaris, Ranunculus species). Macrophytes strongly declined and algae disappeared. Local fires in the vicinity of lake are documented by macroscopic charcoal particles (cf. Whitlock and Larsen, 2001), which were recorded especially in the S2 profile. Early and middle Holocene sequences (S1-m3, S2-m3, ca. 11,250–4600 and ca. 11,600–3200 cal. yr BP, respectively; S1-p2, S2-p2, ca. 10,800–4600 and ca. 9000–6300 cal. yr BP, respectively) are very poorly preserved, and their fossil record is fragmentary. The zone is characterised by a large amount of macroscopic charcoal particles (Betula, Pinus, Picea/Larix, dwarf shrubs of Vacciniaceae), indicating one or more local fires during this period. In herb layer, Eriophorum vaginatum (larger amount of spindles in S1), Juncus effusus type (seeds in S2), Molinia caerulea (seeds in S2) and Typha sp. (seeds in S2 and pollen in S1) occurred. A larger amount of ephippia of Daphnia was further recorded in S2. The sedimentation started to accelerate again after the youngest detected fire ca. 400 cal. yr BP (the end of the zone S1-m4, zone S2-m4). The open acidic mire is indicated by macrofossils of Carex rostrata (S2), Juncus effusus (S2), Eriophorum vaginatum (S1 and S2), Molinia caerulea (S1 and S2) and bryophytes Sphagnum cf. fallax and Polytrichum sp. (S1, cf. Hájek et al., 2002). The last developmental stage represents a transition to a closed-canopy birch woodland mire (S1, S2-m5, S1, S2-p5) that took place approximately during the 19th century, which is indicated by an increase in Betula macrofossils (seeds, scales and wood) and pollen. In the herb and moss layers, Molinia caerulea (seeds in S1 and S2), Carex canescens (seeds in S1) and Sphagnum sp. (leaves in S1 and S2, in S1 determined as S. fallax) dominated. Eriophorum vaginatum still occurred, even if only a small number of spindles were found.

Figure 4.

Pollen diagram of the S1 profile. Pollen, spores and algae not shown in the diagram: Acer (1/49), Apium (2/97), Boraginaceae (1/40, 1/82), Botrychium (1/73), Carduus (1/15, 1/106), Fabaceae (1/10, 1/15, 1/103), Juniperus (1/76, 1/79, 1/100), Lythrum (1/31, 1/70), Myosotis arvensis (1/61, 1/64), Pediastrum duplex var. gracilinum (1/85), Plantago major (1/15), Potamogeton (1/91), Rumex t. (1/25), Rumex obtusifolius t. (1/97), Sambucus (1/15), Sanguisorba minor t. (1/103), Scenedesmus (3/76), Scrophulariaceae (1/43), Trifolium t. (1/82), Vaccinium (1/67), Valeriana montana t. (1/31, 1/43, 1/82). The first value is abundance, and the second one is depth (in cm). Identified by L. Petr.

Figure 5.

Macrofossil and charcoal diagram of the S1 profile. When not specified otherwise (%), macrofossil counts are absolute numbers. Not shown in diagram: Aulacomnium palustre (1 st + lvs/28–31 cm), Calliergon cf. richardsonii (4 lvs/103–106 cm), Carex nigra (1 s/79–82 cm), Cicuta virosa (2 s/97–100 cm), Juncus effusus (1 s/34–37 cm), Picea abies (1 scale/97–100 cm), Plagiothecium sp. (1 st + lvs/31–34 cm), Scorpidium revolvens agg. (1 lf/106–109 cm), Sphagnum sp. (1 opp./70–73 cm, 2 opp./100–103 cm). Abbreviations: s (seed), st + lvs (stem and leaves), opp. (opperculum). The first value is abundance, and the second one is depth (in cm). Macrofossils identified by P. Hájková. Charcoals identified by J. Novák.

Figure 6.

Pollen diagram of S2 profile. Pollen, spores and fungi not shown in diagram: Abies (1/28, 2/33), Anthemis t. (3/78), Brassicaceae (1/33), Centaurea jacea t. (1/23), Cornus mas (2/13), Ephedra fragilis t. (3/68), Fabaceae (2/63), Juniperus (2/48, 2/78), Lathyrus vernus (2/48), monolete spores (4/3), Rhamnus (2/13, 1/18, 2/63), Sporormiella (1/33, 1/78), Vaccinium (1/28, 1/38, 2/48). The first value is abundance and the second one is depth (in cm). Identified by M Čierniková.

Figure 7.

Macrofossil and charcoal diagram of S2 profile. When not specified otherwise (%), macrofossil counts are absolute numbers. Not shown in diagram: Alnus glutinosa (2s/5–10 cm), Cirsium sp. (1s/15–20 cm), Rubus sp. (2s/35–40 cm). The first value is abundance, and the second one is depth (in cm). Macrofossils identified by A. Gálová. Charcoals identified by J. Novák.

Peat chemistry

The basal sediment samples from both profiles contain a higher admixture of sand (Table 2), which is reflected in low LOI values (see Figure 8). They also contain higher concentrations of certain elements (iron, potassium, magnesium and manganese; 106–97 cm in S1 and 95–90 cm in S2). The very low calcium concentrations of up to 20 mg kg⁻¹ in the lake sediments and up to 50 mg kg⁻¹ in the terrestrial sediments reflect the siliceous calcium-poor bedrock and correspond to soil calcium concentrations in Carpathian poor fens (Hájek et al., 2002). This result fits well with the continual presence of Sphagnum spores in the pollen record because Sphagna avoid the sites with high calcium concentrations from physiological reasons (Vicherová et al., 2015). Lake sediments in the bottom part of each profile (100–80 cm in S1 and 95–55 cm in S1) are characterised by higher amounts of silica, probably of biotic origin (diatoms), reflecting the organic productivity of the lake. The small amount of indicators of clay minerals transported in suspension (potassium, sodium) and low aluminium concentration in the silicate fraction indicate an absence of erosion and absence of transport of allogenic material from the lake’s vicinity. LOI increased simultaneously with decreasing silica, which indicates accumulation of organic matter and the lake/mire transition. Slightly increased calcium concentrations, released from burned organic material, characterize zones with large amounts of charcoal (70–50 cm in S1 and 50–35 cm in S2). Neither of the sediments in the youngest layers (30–0 cm) contains any indicator of erosion (aluminium and silica). The upper layers representing the last centuries (ca. 25–0 cm, recent to 400 cal. yr BP) are characterised especially by increased values of phosphorus.

Table 2.

Lithological zones of studied profiles.

Profiles	Description	Colour
Šenkárka 1
0–10 cm	Undecomposed remains of plants and bryophytes, root zone	–
10–80 cm	Strongly decomposed organic sediment with wood remains and charcoal particles, root zone; diffuse transition to next zone	Dark brown
80–106 cm	Moderately decomposed organic sediment with sand admixture	Brown
106–109 cm	Clayish-organic sediment, 30–60% of coarse sand	Grey brown
109–116 cm	Sand	Yellow
Šenkárka 2
0–10 cm	Undecomposed remains of plants and bryophytes, root zone	–
10–35 cm	Slightly decomposed organic sediment with remains of plants and bryophytes	Dark brown
35–70 cm	Strongly decomposed organic sediment with charcoals	Dark brown
70–100 cm	Strongly decomposed organic sediment with sand admixture	Grey brown

Figure 8.

Peat chemistry and LOI of profiles S1 and S2.

Discussion

Regional vegetation history

In the Late Glacial and early Holocene, Pinus and Betula dominated in the arboreal and Artemisia and Poaceae in the non-arboreal pollen spectra. These taxa were frequent dominants of Late Glacial and early Holocene landscapes across Central Europe (Feurdean, 2005; Hájková et al., 2013; Magyari et al., 2010, 2014; Margielewski, 2006; Theuerkauf and Joosten, 2009). Open Pinus and Betula forests and Artemisia steppes were probably the dominant habitats, forming the expositional forest-steppe biome that we know from analogous landscapes in southern Siberia (Chytrý et al., 2008). Contrary to pollen spectra from some mountainous regions of the Carpathians, such as the northern Carpathian basins (Hájková et al., 2015a; Jankovská, 1988) or the Southern Carpathians (Magyari et al., 2012), pollen of Larix and Pinus cembra is missing from the Late Glacial (Younger Dryas) and early Holocene pollen record at the study site. This is in concordance with the results of other studies carried out in southwestern Slovakia (Hájková et al., 2013, 2015b; Jamrichová et al., 2014). These tree species were present in northern Pannonia, including southern Moravia (Magyari et al., 1999; Rybníčková and Rybníček, 2014; Sümegi et al., 2013; Willis and Van Andel, 2004) in the Full Glacial and older part of the Late Glacial (at least up to ca. 16,000 cal. yr BP), but later retreated to higher altitudes. Recently, Pinus cembra occurs only in the Tatra Mts within the whole Western Carpathians (Futák, 1966).

In the whole of northern Pannonia (Jamrichová et al., 2014; Kuneš et al., 2015; Magyari et al., 2010; Petr et al., 2013; Willis et al., 1995; for a detailed comparison, see Table 3), pollen of temperate broad-leaved trees occurred already at the end of the Late Glacial (from ca. 12,000 cal. yr BP at the study site) or at the very beginning of the Holocene (11,500–11,200 cal. yr BP). They started to be more abundant between 10,500 and 10,000 cal. yr BP with exception of localities situated on the thick aeolic sands, which supports Pinus and where this process started much later 7500–4500 cal. yr BP (Table 3). At the study site, around 10,500 cal. yr BP, distinct changes in tree pollen composition took place. Betula increased together with an increase or appearance of Corylus, Quercus, Ulmus, Tilia, Fraxinus and Alnus. This spectrum represents the vegetation of mixed deciduous forests, hazel scrub and alder carrs. The expansion of these species in the period between 11,000 and 10,000 cal. BP occurred not only in the northern Pannonia (cf. Table 3, Magyari et al., 2010) but also at some other sites along the margins of the Carpathians (NW Romania: Feurdean and Bennike, 2004; Feurdean, 2005; S Romania – Retezat: Magyari et al., 2012; E Carpathians – Vihorlat Mts: Petr, unpublished data) being related to the increasing Northern Hemisphere temperature (e.g. Blockley et al., 2012; Tóth et al., 2015). In the Eastern Carpathians, the vicinity of Full Glacial refugia, verified recently in the Hargita Mts (Magyari et al., 2014), was probably crucial for the early expansion of mesophilous broad-leaved trees. In the case of the Western Carpathians, local glacial refugia still have not been unambiguously proven, and the nearest known glacial refugia are in SE Hungary (Sümegi et al., 2013). Nevertheless, pollen of some temperate trees, especially Tilia, Corylus and Ulmus, was found in Late Glacial sequences even in mountain regions of the northwestern Carpathians (Břízová et al., 2002; Pánek et al., 2010). They might have been re-deposited, but may also suggest refugia directly in the Western Carpathians.

Table 3.

Comparison of the onset and expansion of pollen of selected tree species at the study site and other sites in the region. Temperate broad-leaved trees include Ulmus, Tilia and Quercus species. The term ‘onset’ refers to the beginning of a continual presence. The dates are given in cal. yr BP.

	Nad Šenkárkou	Šúr	Parížske močiare	Malé Bielice (Partizánske)	Vracov	Dúbrava
Reference	This paper	Petr et al. (2013)	Jamrichová et al. (2014)	Hájková et al. (2013)	Kuneš et al. (2015)	Jamrichová et al. (unpublished data)
Geomorphological unit	Malé Karpaty Mts	Danubian lowland (western margin)	Danubian lowland	Nitra Basin	Lower Morava valley	Lower Morava valley
Prevailing bedrock in landscape	Granitoids	Fluvial sands and gravels	Fluvial sands and gravels	Fluvial sands	Aeolic sands	Aeolic sands
Altitude (m a.s.l.)	560	130	123	185	190	190
Settlement since	Half of 18th century	Neolithic	Neolithic	Neolithic	Neolithic	Neolithic
Onset of temperate trees	ca. 12,000	11,500	11,200	11,500	11,500	11,300
Increase in temperate trees	10,500	9500	ca. 10,300	10,000	4500	7600
Onset of Fagus	10,500	9700	9600	9000	4500	7600
Fagus expansion	>5800	5000	no	5500	3500	5500
Distinct decrease of Pinus	10,500	10,000	no	9000–10,000	5500	5600

Fagus pollen was continually (but in low abundances) present from ca. 10,500 cal. yr BP and started to expand rapidly no later than in 5800 cal. yr BP according to the S1 profile. However, it is impossible to accurately date the onset of this species expansion because of a highly probable hiatus in the fossil record, which is also the reason of the later increase of Fagus pollen in the S2 profile. Simultaneously with the Fagus expansion, pollen of Carpinus had appeared. The fossil pollen record from the lake Šúr, situated ca. 9 km SSE from the summits of the Malé Karpaty Mts (for position, see Figure 1), captured a continual presence of Fagus even from ca. 9700 cal. yr BP and its expansion around 5000 cal. yr BP (Petr et al., 2013; see also Table 3), which largely fits our results. Moreover, the values of Fagus pollen accumulation rates were very similar between the study site and profiles from the Šúr lake (data not shown). The Fagus expansion at other sites along the interface between the Carpathians and northern Pannonia started at a similar time or slightly later between 5500 and 4500 cal. yr BP (Bodnariuc et al., 2002; Feurdean, 2005; Magyari et al., 2010). During the last two centuries, Fagus pollen significantly decreased while Betula pollen rapidly increased. Because the high pollen percentage of Betula reflects recent local vegetation of a mire birch woodland, a steep decrease in the amount of Fagus pollen is merely an artefact of local change and does not mirror changes in the matrix of upland vegetation. Indeed, the whole mountain range is currently overgrown by extensive beech forests. When the mire got overgrown by birch trees, Betula pollen suppressed the accumulation of pollen of taxa surrounding the mire.

In the past, the steep slopes of the Malé Karpaty Mts were unsuitable for human settlement, which was concentrated in the Pannonian lowland (Farkaš et al., 2008; Vavák, 2010). The absence of any settlement in the close vicinity of the study site is evidenced by a scarcity of ruderal, weed and crop pollen in both profiles. The Wallachian colonization wave (AD 14th–17th century), which strongly influenced other areas of the Western Carpathians, did not affect the study area (Cojocaru, 2014). Even today, the nearest meadows are located at the foothills of mountains by the villages of Limbach and Borinka. From the half of the 18th century onwards, the forests of the Malé Karpaty Mts were scarcely settled by German settlers (called ‘huncokári’), who worked with wood and had permission to graze sheep in local forests (Horváthová, 2002). However, their impact on the landscape was probably very restricted.

Local vegetation history: Late Glacial lake phase

The sedimentation started in Younger Dryas (ca. 12,950 cal. yr BP) in a shallow and oligo/mesotrophic lake with macrophyte vegetation and algae. Similar communities were recorded in many other Late Glacial deposits in the Pannonian lowland (e.g. Břízová, 2007; Petr et al., 2013; Potůčková, 2014; Rybníček, 1983) and beyond (e.g. Bos et al., 2006; Gałka, 2014; Gałka and Sznel, 2013), although some differences exist. As the water in the lake was poor in minerals (especially calcium) and nutrients, some species like Najas marina, Nymphaea alba or Nuphar lutea were missing. In the plant macrofossil spectra, we detected several species which are now rare in Central Europe: macrophytes Sparganium natans, Potamogeton alpinus and P. praelongus; littoral species Hippuris vulgaris; and boreal bryophytes Scorpidium scorpioides and Calliergon cf. richardsonii. Sparganium natans, a species of shallow peaty waters, used to be recorded in the Late Glacial sediments (e.g. Hanšpile site, Hájková et al., 2015b), but also in younger sediments before the intensification of human impact (e.g. in 1000–1300 cal. yr BP; Pokorný et al., 2000). Both Potamogeton praelongus and Potamogeton alpinus no longer occur in the study area because they require clear transparent water with oligotrophic to mesotrophic, rarely eutrophic conditions (Søndergaard et al., 2010; Šumberová, 2011). They are often found in Late Glacial sediments (e.g. Bos et al., 2006; Gałka and Sznel, 2013; Potůčková, 2014), which confirms the relic status of these species in Central Europe. We found only one seed of Potamogeton praelongus in the oldest layer of the S1 profile, which suggests that it probably disappeared early from our study site. Boreal bryophytes Scorpidium scorpioides and Calliergon richardsonii, which probably occupied peaty lake margins, are traditionally considered to be relicts from the cold phases (Rybníček, 1966) and are currently known only from a single locality in Slovakia (Scorpidium scorpioides; Dítě and Šoltés, 2010) or do not occur in Central Europe at all (Calliergon richardsonii). Both species are more common in Northern Europe (Hedenäs, 2003). Our finding suggests that the distribution of the species reached more southerly areas in the Late Glacial and early Holocene. These species have been found in Late Glacial sediments, for example, in Hungary (Scorpidium scorpioides, Sümegi et al., 2008, Calliergon richardsonii, Magyari et al., 1999) and SW Slovakia (S. scorpioides, Hájková et al., 2015b). The end of the Late Glacial and the beginning of the early Holocene did not see a change in nutrient levels, but the succession gradually continued because of terrestrialization, as indicated by the increasing number of macrofossils of littoral-zone species. Occasional seasonal drying of the lake (or its margins) is indicated by abundant seeds of Ranunculus sceleratus. Around 11,500 cal. yr BP, most of the lake was probably almost dry because the amount of aquatic macrophytes strongly decreased in both profiles. Terrestrialization of Late Glacial shallow lakes at nearby lowland sites was considerably slower. At the former lake Šúr (Figure 1), open water persisted until the Sub-Boreal period (3870 cal. yr BP; Potůčková, 2014); the former lake Vracov in southeastern Moravia also had open water throughout the middle Holocene (Rybníček, 1983). The main reason behind early terrestrialization is probably the small size and depth of the studied lake.

Fire activity, hiatus(es) and spatial differences

Fire dynamics probably played an important role in the subsequent development of the site. Despite the small distance between the two profiles, the recorded macroscopic and microscopic charcoal peaks differed in some points. The S2 profile shows that a minor local fire in the lake’s vicinity probably took place at the very end of the Late Glacial, which is reflected by an additional peak in both macroscopic and microscopic charcoal particles. The subsequent zone characterized by a great abundance of macroscopic charcoal particles in both profiles (70–35 cm in S1 and 50–35 cm in S2, cf. Figures 5 and 7) suggests that larger fire event(s) probably took place later. However, the section with the most important charcoal peaks in both profiles contains a probable hiatus, so a precise chronology of the fire event(s) is impossible and the situation might have been more complex than just one single large fire. Previous studies have found important small-scale differences between nearby charcoal samples dated to the same time period, both in palaeoecological profiles (Edwards and Whittington, 2000; Innes and Simmons, 2000) and in surface samples (Blackford, 2000). The size and intensity of individual fires can be hard to deduce, even on a very fine scale, because of the complexity of charcoal taphonomy (Innes and Simmons, 2000).

The macroscopic charcoal particles belonged especially to Pinus and Betula, a possible sign that the mire was already encroached by trees by the time of the fire event, but still with a bog understorey (Eriophorum vaginatum, Vacciniaceae) and some remaining pools (Daphnia). However, small pools at least temporarily filled by water could appear subsequently after fire in the burned layer. The reason why trees encroached into the site and why at least one fire subsequently broke out, might reside in the drying up of the mire supplied exclusively by rain and surface water, not by groundwater. When the mountain range got overgrown by a dense deciduous forest, the increased transpiration in the catchment might be related to lower water supply to the mire. Some mountain mires started to accumulate more water after the deforestation of the surrounding slopes (Speranza et al., 2000). Increasing evaporation towards the middle Holocene because of rising temperatures (Jamrichová et al., 2014) might also have had an effect on the lowering of the water table. Maybe the geographic position connected with a drier climate in the Carpathian–Pannonian borderland might have played a certain role by supporting fires. A higher concentration of ephippia of Daphnia, also observed in this zone, could indicate a higher rate of stress for these organisms (e.g. Nevalainen et al., 2011; Sarmaja-Korjonen, 2003, 2004), which could also be connected with the lowered water level in pools (probably of windthrow origin) in the S2 profile. Under the conditions of drying and tree succession, the mire was probably not only prone to fires, but a considerable degree of mineralization and slow accumulation of new sediments can also be expected. It is likely that the hiatus is a result of several factors acting synergistically. Above the hiatus in the late Holocene (ca. 400 cal. yr BP), the succession proceeded through a Juncus effusus marsh to an acidic mire with Eriophorum vaginatum and Carex rostrata. The species composition suggests the presence of open or semi-open vegetation (cf. Hájek et al., 2002) before the expansion of birch in the youngest phase, but unfortunately, we do not know how long this open mire phase was because of the complicated chronology caused by the hiatus.

Concerning the spatial structure of the site, the S1 profile showed a phase with Eriophorum vaginatum dominance in the early Holocene, which could not be observed in the S2 profile, suggesting differences in the progress of the infilling process. On the other hand, the S2 profile showed a phase with dominant Typha sp. and Daphnia sp. (charred zone S2-m3), which were completely absent in the S1 profile. Seeds of Typha sp. are small, numerous and dispersed by wind, so it is very likely they would appear at both sites, at least in small counts. This indicates that different parts of the sediment were stochastically preserved in particular profiles. Gałka (2014) has found similar differences in the timing of the succession and distribution of individual species among the three macrofossil profiles, but the distances between them were larger than in our study (about 200–300 m). Hájková et al. (2012) found important differences between the two profiles situated only 43 m apart; however, the sites were chosen because their recent vegetation was different, unlike the site ‘Nad Šenkárkou’. If the multiple profiles are deliberately chosen to reflect certain topographical features of the site, the results of such analysis help reconstruct its past zonation (e.g. Bos et al., 2006; Muller et al., 2003). However, the zonation of former lakes may not always be simple and straightforward; the infilling process can progress in a mosaic pattern rather than in a simple manner from the edges to the centre, as shown by Muller et al. (2003) and Potůčková (2014). One of the reasons for such a mosaic pattern could be the existence of patches where species with higher biomass production grow, accelerating the infilling process, as was observed at the nearby lowland site Šúr (Potůčková, 2014). Decomposition patterns may also diversify infilled lakes (Ammann et al., 2013). Our results suggest that a small-scale mosaic pattern of succession from a lake to a mire probably existed also at the study site.

Birch mire woodland – Relict vegetation or a succession stage?

An abrupt increase in Betula pollen and macrofossils in the most recent developmental phase (ca. the last 100–200 years; the exact dating is beyond the resolution of our analyses) suggests that the mire birch woodland in its present state has a relatively short history at the study site. Valček (2011) reported that the site was first mapped as a non-forested patch (second military mapping in the first half of the 19th century), then as a patch with shrubby vegetation (third military mapping, second half of the 19th century) and finally as forested (military map from 1930). This corresponds well with our results about succession from (semi-)open vegetation to a birch woodland. Scheffer (1933) described the existence of an open patch with Eriophorum vaginatum in the central part of the mire with shrub and tree birches in the surroundings. The onset of birch expansion roughly falls at the end of the ‘Little Ice Age’ (ca. AD 1850, Brázdil, 1994), a possible natural cause of the lowered water table because of decreasing precipitation (Büntgen et al., 2011). Direct anthropogenic impact on the mire is another possible factor, as Valček (2011) described the existence of a drainage ditch of unknown age across the mire.

However, our results have shown that phases of drought and tree dominance (Pinus and Betula) occurred also in the past even though we are unable to determine either their exact age and causes or their frequency because of the destruction of peat. Deforestation because of fire might have played an important role in the dynamics of bog woodlands, including the alternation of open and forest stages like in zonal boreal forests (Goldammer and Furyaev, 1996). Macrofossils indicate a continual presence of Betula pubescens/pendula since the Late Glacial (especially in the S2 profile, but with a hiatus in the fossil record), even though its quantities never reached the levels of the most recent zone. The high abundance of Betula pollen (20–40%) is also evidence of local birch occurrence since the early Holocene, with exception of the zone right after the fire (S1-p3, S2-p3, from ca. 400 cal. yr BP). The lower abundance of Betula pollen together with the presence of Carex rostrata seeds suggests wetter conditions and at least a semi-open character of the mire. Karpińska-Kołaczek et al. (2014) found a continuity of 2500 years of birch bog woodland under low human impact in NE Poland, which means this type of habitat can show a long-time stability. Betula stages are sometimes found already in the early Holocene or even the Late Glacial in palaeoecological studies in Central and Eastern Europe (e.g. Czech Republic: Bešta et al., 2009; Poland: Fajer et al., 2012; Karpińska-Kołaczek et al., 2013, 2014), even though it can be hard to distinguish whether birch was locally present or more distant from the study site because winged Betula seeds can be easily transported by wind (Greatrex, 1983). On the other hand, birch often grows on mire margins, whereas most palaeoecological cores are taken from the central parts of the mires. In this context, our findings of Betula charcoal particles and wood fragments are remarkable and suggest the possibility of existence of a mire woodland at the study site already in Late Glacial and early Holocene, even if their continuity may have been interrupted by several large disturbances (especially fires, climate induced or anthropogenic water level changes).

Furthermore, the even age of trees and lack of a young generation of birch in recent vegetation (Hrbatý, 2000) may propose a hypothesis about the existence of cyclic succession similar to that described for alder carrs (Barthelmes et al., 2010; Pokorný et al., 2000), where phases of Alnus glutinosa dominance alternate with phases dominated by the Cyperaceae. We detected a zone similar to the wet and open phase of a cyclic succession (low Betula, high Sphagnum, presence of C. rostrata) as well as one analogous to its dry and closed phase (high Betula, low Sphagnum, presence of Molinia), but unfortunately, the older fossil record is erased, so we cannot make any conclusion about whether these phases repeated in some type of a cycle or rather progressed in a linear manner. Moreover, the existence of cyclic succession is still a subject of scientific debate, and there is some evidence that the alternation of Alnus and Cyperaceae vegetation might coincide with changes in the intensity of human impact in the surroundings (Barthelmes et al., 2010).

Conclusion

In this study, we outline the Holocene history of a birch bog woodland at the phytogeographical border of Western Carpathians and Pannonia. This boreal-like bog woodland established after terrestrialization of the Late Glacial lake, after the semi-aquatic fen phase. From the early Holocene until recent times, birch bog woodlands occurred locally, so we verified the relic character of the birch bog woodland, although its history was rather complex. More open and more closed vegetation probably alternated at the site, but repeating fires disturbed continuity of fossil record, and we cannot describe fire dynamics in more detail. A rather dense bog woodland developed during the 19th century after the water table dropped. The species-poor nature of the study bog woodland may coincide with middle Holocene hiatuses: drought, fires and peat mineralization might have caused the extinctions of some mire specialists.

We detected fire dynamics connected with hiatuses in the fossil record. We further documented past occurrences of several rare species, especially in Late Glacial and early Holocene peat sequences. Our results based on two peat profiles confirm that small-scale variability of the palaeoecological record is important for understanding the patterns of past vegetation development, for example, in the process of terrestrialization and in the local fire activity, even when today’s vegetation seems to be homogeneous. If only one peat profile from a site is studied, it can be hard to distinguish local from whole-site and broader scale events.

Footnotes

Acknowledgements

We are thankful to Milan Valachovič for his assistance with profile sampling, Ondřej Hájek for preparing the map and Zuzana Formánková for laboratory work.

Funding

Our research was supported by project P504/11/0429 of the Czech Science Foundation, institutional support and project MUNI/A/1048/2015 of Masaryk University and long-term research development project RVO 67985939. M Čierniková was supported by project APVV-0866-12 of the Slovak Research and Development Agency.

References

Ammann

Wright

Stefanova

. (2013) The role of peat decomposition in patterned mires: A case study from the central Swiss Alps. Preslia 85(3): 317–332.

Barber

Dumayne-Peaty

Hughes

. (1998) Replicability and variability of the recent macrofossil and proxy-climate record from raised bogs: Field stratigraphy and macrofossil data from Bolton Fell Moss and Walton Moss, Cumbria, England. Journal of Quaternary Science 13(6): 515–528.

Barthelmes

Gerloff

de Klerk

. (2010) Short-term vegetation dynamics of Alnus dominated peatlands: A high resolution palaeoecological case study from Western Pomerania (NE Germany). Folia Geobotanica 45(3): 279–302.

Bešta

Šafránková

Pouzar

. (2009) Late Pleistocene–early Holocene transition recorded in the sediments of a former shallow lake in the Czech Republic. Hydrobiologia 631(1): 107–120.

Beug

(2004) Leitfaden der Pollenbestimmung für Mitteleuropa und angrenzende Gebiete. München: Verlag Dr. Friedrich Pfeil.

Birks

HJB

(2000) Future uses of pollen analysis must include plant macrofossils. Journal of Biogeography 27(1): 1–35.

Blackford

(2000) Charcoal fragments in surface samples following a fire and the implications for interpretation of subfossil charcoal data. Palaeogeography, Palaeoclimatology, Palaeoecology 164: 33–42.

Blockley

SPE

Lane

Hardiman

.; INTIMATE members (2012) Synchronisation of palaeoenvironmental records over the last 60,000 years, and an extended INTIMATE event stratigraphy to 48,000 b2k. Quaternary Science Reviews 36: 2–10.

Bodnariuc

Bouchette

Dedoubat

. (2002) Holocene vegetational history of the Apuseni mountains, central Romania. Quaternary Science Reviews 21: 1465–1488.

10.

Bos

JAA

Bohncke

SJP

Janssen

(2006) Lake-level fluctuations and small-scale vegetation patterns during the late glacial in the Netherlands. Journal of Paleolimnology 35(2): 211–238.

11.

Brázdil

(1994) Climatic fluctuation in the Czech Lands during the last millennium. GeoJournal 32(3): 199–205.

12.

Břízová

(2007) Vodní a mokřadní ekosystémy v průběhu pozdního glaciálu a holocénu na základě pylových analýz [Aquatic and wetland ecosystems during the late glacial and Holocene based on pollen analyses]. Zprávy České botanické společnosti, Materiály 22: 85–97.

13.

Břízová

Hradecký

Pánek

(2002) Application of pollen analysis in the problematic of landslide chronology in the Silesian Beskydy Mts. Zprávy o geologických výzkumech v roce 2002: 65–68 (in Czech).

14.

Bronk Ramsey

(2009) Bayesian analysis of radiocarbon dates. Radiocarbon 51(1): 337–360.

15.

Büntgen

Tegel

Nicolussi

. (2011) 2500 years of European climate variability and human susceptibility. Science 331(6017): 578–582.

16.

Cappers

RTJ

Bekker

Jans

JEA

(2006) Digitale Zadenatlas van Nederland [Digital Seed Atlas of the Netherlands]. Groningen: Barkhuis Publishing.

17.

Chiverrell

(2001) A proxy record of late Holocene climate change from May Moss, northeast England. Journal of Quaternary Science 16(1): 9–29.

18.

Chytrý

Danihelka

Kubešová

. (2008) Diversity of forest vegetation across a strong gradient of climatic continentality: Western Sayan Mountains, southern Siberia. Plant Ecology 196: 61–83.

19.

Chytrý

Zelený

Navrátilová

. (2013) Boreokontinentální jehličnaté lesy [Boreo-continental coniferous forests]. In: Chytrý

(ed.) Vegetace České republiky 4. Lesní a křovinná vegetace [Vegetation of the Czech Republic 4. Forest and Scrub Vegetation]. Praha: Academia, pp. 380–432.

20.

Cojocaru

(2014) Valasi na Slovensku a v Poľsku. Predbežné úvahy. In: Marza

Syrný

(eds) Valašská kolonizácia na Slovensku a slovenská kolonizácia v Rumunsku. Zborník príspevkov z 10. Zasadnutia Komisie historikov Slovenska a Rumunska. Banská Bystrica: Múzeum Slovenského národného povstania, pp. 62–71.

21.

Dítě

Šoltés

(2010) Rozšírenie druhu Scorpidium scorpioides na Slovensku: minulosť a súčasnosť. Bryonora 46: 66–69.

22.

Edwards

Whittington

(2000) Multiple charcoal profiles in a Scottish lake: Taphonomy, fire ecology, human impact and inference. Palaeogeography, Palaeoclimatology, Palaeoecology 164: 67–86.

23.

Faegri

Iversen

(1989) Textbook of Pollen Analysis. Chichester: John Wiley & Sons.

24.

Fajer

Waga

Rzetala

. (2012) The Late Vistulian and Holocene evolution of Jezioro Lake: A record of environmental change in southern Poland found in deposits and landforms. Journal of Paleolimnology 48: 651–667.

25.

Farkaš

Vavák

Wittgrúber

(2008) Dejiny Pezinka od praveku po formovanie sa Uhorského štátu. In: Pospechová

(ed.) Bozen – Pezinok 1208–2008. Pezinok: Mesto Pezinok, pp. 1–15.

26.

Feurdean

(2005) Holocene forest dynamics in northwestern Romania. The Holocene 13(3): 435–446.

27.

Feurdean

Bennike

(2004) Late Quaternary palaeoecological and palaeoclimatological reconstruction in the Gutaiului Mountains, northwest Romania. Journal of Quaternary Science 19(8): 809–827.

28.

Freedman

Maessen

Bauchamp

. (1989) Trace elements in water, sediment, and biota of three oligotrophic lakes in Queens and Annapolis counties, Nova Scotia. Proceedings of the Nova Scotian Institute of Science 39(3): 123–132.

29.

Futák

(ed.) (1966) Flóra Slovenska II. Bratislava: Slovenská akadémia vied.

30.

Gaillard

M-J

(2013) Pollen methods and studies – Archaeological applications. In: Elias

Mock

(eds) Encyclopedia of Quaternary Science. 2nd Edition. Amsterdam: Elsevier, pp. 880–904.

31.

Gałka

(2014) Pattern of plant succession from eutrophic lake to ombrotrophic bog in NE Poland over the last 9400 years based on high-resolution macrofossil analysis. Annales Botanici Fennici 51: 1–25.

32.

Gałka

Sznel

(2013) Late Glacial and Early Holocene development of lakes in northeastern Poland in view of plant macrofossil analyses. Quaternary International 292: 124–135.

33.

Goldammer

Furyaev

(1996) Fire in Ecosystems of Boreal Eurasia. Dordrecht: Kluwer Academic Publishers.

34.

Greatrex

(1983) Interpretation of macrofossil assemblages from surface sampling of macroscopic plant remains in mire communities. Journal of Ecology 71: 773–791.

35.

Grimm

(2011) Tilia Software v.1.7.16. Springfield, IL: Illinois State Museum.

36.

Hadač

Slavík

Richterová

(1967) The distribution of Pleurospermum austriacum (L.) Hoffm. in Czechoslovakia. Preslia 89: 375–391.

37.

Hájek

Hekera

Hájková

(2002) Spring fen vegetation and water chemistry in the Western Carpathian flysch zone. Folia Geobotanica 37: 205–224.

38.

Hájková

Grootjans

Lamentowicz

. (2012) How a Sphagnum fuscum-dominated bog changed into a calcareous fen: The unique Holocene history of a Slovak spring-fed mire. Journal of Quaternary Science 27: 233–243.

39.

Hájková

Horsák

Hájek

. (2015a) Using multi-proxy palaeoecology to test a relict status of refugial populations of calcareous-fen species in the Western Carpathians. The Holocene 25(4): 702–715.

40.

Hájková

Jamrichová

Horsák

. (2013) Holocene history of a Cladium mariscus-dominated calcareous fen in Slovakia: Vegetation stability and landscape development. Preslia 85: 289–315.

41.

Hájková

Petr

Horsák

. (2015b) Interstadial inland dune slacks in south-west Slovakia: A multi-proxy vegetation and landscape reconstruction. Quaternary International 357: 314–328.

42.

Hedenäs

(2003) The European species of the Calliergon-Scorpidium-Drepanocladus complex, including some related or similar species. Meylania 28: 1–117.

43.

Heiri

Lotter

Lemecke

(2001) Loss on ignition as a method for estimating organic and carbonate contents in sediments: Reproducibility and comparability of results. Journal of Paleolimnology 25: 101–110.

44.

Heiss

(2000) Anatomy of European and North American Woods – An Interactive Identification Key (version 2009-05-14). Available at: http://www.holzanatomie.at/.

45.

Holliday

(2004) Soils in Archaeological Research. Oxford: Oxford University Press.

46.

Horváthová

(2002) Nemci na Slovensku. Etnokultúrne tradície z aspektu osídelnia, remesiel a odievania. Komárno; Dunajská Streda: Fórum inštitút pre výskum menšín, Lilium Aurum.

47.

Hrbatý

(2000) Významné rašeliniská Chránenej krajinnej oblasti Malé Karpaty [Important peatlands of the protected landscape area Malé Karpaty]. In: Stanová

(ed.) Rašeliniská Slovenska. Bratislava: DAPHNE, Inštitút aplikovanej ekológie, pp. 143–148.

48.

Innes

Simmons

(2000) Mid-Holocene charcoal stratigraphy, fire history and palaeoecology at North Gill, North York Moors, UK. Palaeogeography, Palaeoclimatology, Palaeoecology 164: 151–165.

49.

Jamrichová

Potůčková

Horsák

. (2014) Early occurrence of temperate oak-dominated forest in the northern part of the Little Hungarian Plain, SW Slovakia. The Holocene 24(12): 1810–1824.

50.

Jankovská

(1988) A reconstruction of the Late Glacial and Early Holocene evolution of forest vegetation in the Poprad Basin. Folia Geobotanica et Phytotaxonomica 23: 303–319.

51.

Jankovská

Chromý

Nižnianská

(2002) Šafárka – First palaeobotanical data of the character of Last Glacial vegetation and landscape in the West Carpathians (Slovakia). Acta Palaeobotanica 42(1): 39–50.

52.

Juggins

(2003) C2 User Guide, Version 1.5: Software for Ecological and Paleoecological Data Analysis and Visualization. Newcastle upon Tyne: University of Newcastle.

53.

Karpińska-Kołaczek

Kołaczek

Stachowicz-Rybka

(2013) Palaeobotanical studies on Late Glacial and Holocene vegetation development and transformations of the ‘Wielkie Błoto’ mire near Gołdap (north-eastern Poland). Acta Palaeobotanica 53(1): 53–67.

54.

Karpińska-Kołaczek

Kołaczek

Stachowicz-Rybka

(2014) Pathways of woodland succession under low human impact during the last 13,000 years in northeastern Poland. Quaternary International 328–329(1): 196–212.

55.

Kołaczek

(2007) Late Glacial and Holocene vegetation changes in the western part of Rzeszów foothills (Sandomierz basin) based on the pollen diagram from Krasne near Rzeszów. Acta Palaeobotanica 47(2): 455–467.

56.

Kołaczek

Gałka

Karpińska-Kołaczek

(2015) Succession of arboreal taxa during the Late Glacial in south-eastern Poland: Climatic implications. Palaeogeography, Palaeoclimatology, Palaeoecology 421: 1–14.

57.

Komárek

Jankovská

(2001) Review of the green algal genus Pediastrum: Implication for pollen-analytical research. Bibliotheca Phycologica 108: 1–127.

58.

Krippel

(1965) Postglaciálny vývoj lesov Záhorskej nížiny. Bratislava: Vydavateľstvo Slovenskej akadémie vied.

59.

Kučera

Váňa

Hradílek

(2012) Bryophyte flora of the Czech Republic: Updated checklist and Red List and a brief analysis. Preslia 84: 813–850.

60.

Kuneš

Svobodová-Svitavská

Kolář

. (2015) The origin of grasslands in the temperate forest zone of east-central Europe: Long-term legacy of climate and human impact. Quaternary Science Reviews 116: 15–27.

61.

Magyari

Chapman

Passmore

. (2010) Holocene persistence of wooded steppe in the Great Hungarian Plain. Journal of Biogeography 37(5): 915–935.

62.

Magyari

Jakab

Bálint

. (2012) Rapid vegetation response to Lateglacial and early Holocene climatic fluctuation in the South Carpathian Mountains (Romania). Quaternary Science Reviews 35: 116–130.

63.

Magyari

Jakab

Rudner

. (1999) Palynological and plant macrofossil data on Late Pleistocene short-term climatic oscillations in north-eastern Hungary. Acta Palaeobotanica Suppl. 2: 491–502, Proceedings 5th EPPC.

64.

Magyari

Sümegi

Braun

. (2001) Retarded wetland succession: Anthropogenic and climatic signals in a Holocene peat bog profile from north-east Hungary. Journal of Ecology 89: 1019–1032.

65.

Magyari

Veres

Wennrich

. (2014) Vegetation and environmental responses to climate forcing during the Last Glacial Maximum and deglaciation in the East Carpathians: Attenuated response to maximum cooling and increased biomass burning. Quaternary Science Reviews 106: 278–298.

66.

Margielewski

(2006) Records of the Late Glacial-Holocene palaeoenvironmental changes in landslide forms and deposits of Beskid Makowski and Beskid Wyspowy Mts area (Polish Outer Carpathians). Folia Quaternaria 76: 6–137.

67.

Marhold

Hindák

(1998) Zoznam nižších a vyšších rastlín Slovenska [Checklist of Non-Vascular and Vascular Plants of Slovakia]. Bratislava: Veda.

68.

Meusel

Jäger

Rauschert

. (1978) Verrgleichende Chorologie der Zentraleuropäischen Flora. Karten, Band II. Jena: VEB Gustav Fischer Verlag.

69.

Muller

Richard

PJH

Larouchel

(2003) Holocene development of a peatland (southern Québec): A spatio-temporal reconstruction based on pachymetry, sedimentology, microfossils and macrofossils. The Holocene 13(5): 649–664.

70.

Nevalainen

Luoto

Kultti

. (2011) Do subfossil Cladocera and chydorid ephippia disentangle Holocene climate trends? The Holocene 22(3): 291–299.

71.

Pacltová

(1963) Metody paleobotanického výzkumu. Praha: Státní pedagogické nakladatelství.

72.

Pánek

Hradecký

Smolková

. (2010) The largest prehistoric landslide in northwestern Slovakia: Chronological constraints of the Kykula long-runout landslide and related dammed lakes. Geomorphology 120(3): 233–247.

73.

Petr

Žáčková

Grygar

. (2013) Šúr, a former late-glacial and Holocene lake at the westernmost margin of the Carpathians. Preslia 85: 239–263.

74.

Pišút

Břízová

Čejka

. (2010) Paleofloristic and paleofaunistic analysis of Dudváh River oxbow and implication for Late Holocene paleoenvironmental development of the Žitný ostrov Island (SW Slovakia). Geologica Carpathica 61(6): 513–533.

75.

Pokorný

Klimešová

Klimeš

(2000) Late Holocene history and vegetation dynamics of a floodplain alder carr: A case study from eastern Bohemia, Czech Republic. Folia Geobotanica 35: 43–58.

76.

Potůčková

(2014) Rekonstrukce paleo-environmentálních poměrů zaniklého jezera Šúr v pozdním glaciálu a holocénu pomocí analýzy makrozbytků [Reconstruction of palaeo-environmental conditions of a former lake Šúr during the late-glacial and Holocene using macrofossil analysis]. Diploma Thesis, Charles University in Prague.

77.

Punt

Clarke

GCS

(eds) (1984) The Northwest European Pollen Flora, IV. Amsterdam: Elsevier.

78.

Reille

(1992) Pollen et spores d’Europe et d’Afrique du nord. Marseilles: Laboratoire de botanique historique et palynologie.

79.

Rybníček

(1966) Glacial relics in the bryoflora of the highlands Českomoravská vrchovina (Bohemian-Moravian Highlands); their habitat and cenotaxonomic value. Folia Geobotanica et Phytotaxonomica 1: 101–119.

80.

Rybníček

(1983) The environmental evolution and infilling process of a former lake near Vracov (Czechoslovakia). Hydrobiologia 108: 247–250.

81.

Rybníček

Rybníčková

(2002) Vegetation of the Upper Orava region (NW Slovakia) in the last 11000 years. Acta Palaeobotanica 42: 153–170.

82.

Rybníček

Rybníčková

(2008) Upper Holocene dry land vegetation in the Moravian-Slovakian borderland (Czech and Slovak Republics). Vegetation History and Archaeobotany 17: 701–717.

83.

Rybníčková

Rybníček

(1972) Erste Ergebnisse palaogeobotanischer Untersuchungen des Moores bei Vracov, Südmähren. Folia Geobotanica et Phytotaxonomica 7: 285–308.

84.

Rybníčková

Rybníček

(2006) Pollen and macroscopic analyses of sediments from two lakes in the High Tatra mountains, Slovakia. Vegetation History and Archaeobotany 15: 345–356.

85.

Rybníčková

Rybníček

(2014) Palaeovegetation in the Pavlovské vrchy hills region (South Moravia, Czech Republic) around 25,000 BP: The Bulhary core. Vegetation History and Archaeobotany 23: 719–728.

86.

Sarmaja-Korjonen

(2003) Chydorid ephippia as indicators of environmental change – Biostratigraphical evidence from two lakes in southern Finland. The Holocene 13(5): 691–700.

87.

Sarmaja-Korjonen

(2004) Chydorid ephippia as indicators of past environmental changes – A new method. Hydrobiologia 526: 129–136.

88.

Scheffer

(1933) Horné rašeliny v Malých Karpatoch. Věda Přírodní 14: 301–303.

89.

Schweingruber

(1978) Mikroskopische Holzanatomie. Birmensdorf: Swiss Federal Institute of Forestry Research.

90.

Šmilauer

Lepš

(2014) Multivariate Analysis of Ecological Data Using CANOCO 5. 2nd Edition. Cambridge: Cambridge University Press.

91.

Smith

AJE

(1996) The Moss Flora of Britain and Ireland. Cambridge: Cambridge University Press.

92.

Søndergaard

Johansson

Lauridsen

. (2010) Submerged macrophytes as indicators of the ecological quality of lakes. Freshwater Biology 55(4): 893–908.

93.

Speranza

Hanke

van Geel

(2000) Late-Holocene human impact and peat development in the Černá Hora bog, Krkonoše Mountains, Czech Republic. The Holocene 10(5): 575–585.

94.

Streel

Beghin

Gerrienne

. (2014) Late Subatlantic history of the ombrotrophic Misten Bog (Eastern Belgium) based on high resolution pollen, testate amoebae and macrofossil analysis. Geologica Belgica 17(2): 148–160.

95.

Šumberová

(2011) Vegetation of aquatic plants rooted in the bottom. In: Chytrý

(ed.) Vegetation of the Czech Republic, Vol. 3: Aquatic and Wetland Vegetation. Praha: Academia, pp. 100–247.

96.

Sümegi

Gulyás

Jakab

(2008) Holocene paleoclimatic and paleohydrological changes in Lake Balaton as inferred from a complex quantitative environmental historical study of a lacustrine sequence of the Szigliget embayment. Documenta Praehistorica 35: 33–43.

97.

Sümegi

Magyari

Dániel

. (2013) Responses of terrestrial ecosystems to Dansgaard-Oeshger and Heinrich-events: A 28,000-year record of environmental changes from SE Hungary. Quaternary International 293: 34–50.

98.

Svobodová

(1997) Die Entwicklung der Vegetation im Südmähren (Tchechien) während des Spätglazials und Holozäns. (Eine Palynologische Studie). Verhandlungen der Zoologisch-Botanischen Gesellschaft in Österreich 134: 317–356.

99.

Theuerkauf

Joosten

(2009) Substrate dependency of Lateglacial forests in north-east Germany: Untangling vegetation patterns, ecological amplitudes and pollen dispersal in the past by downscaling regional pollen. Journal of Biogeography 36: 942–953.

100.

Tóth

Magyari

Buczkó

. (2015) Chironomid inferred Holocene temperature changes in the South Carpathians (Romania). The Holocene 25(4): 569–582.

101.

Valachovič

Hegedüšová

(2013) Lesné spoločenstvá Jurského jazera (Malé Karpaty) [Forest communities of the Jurske jazero (Malé Karpaty Mts)]. Naturae Tutela 17(1): 15–19.

102.

Valček

(2011) Pôdno-vegetačné pomery prírodnej rezervácie Nad Šenkárkou (Malé Karpaty). Diploma Thesis, Comenius University in Bratislava.

103.

Van Geel

Bohncke

Dee

(1980) A palaeoecological study of an upper Late Glacial and Holocene sequence from ‘de Borchert’, The Netherlands. Review of Palaeobotany and Palynology 31: 367–448.

104.

Van Geel

Heijnis

Charman

. (2014) Bog burst in the eastern Netherlands triggered by the 2.8 kyr BP climate event. The Holocene 24(11): 1465–1477.

105.

Vavák

(2010) Nové poznatky k najstarším dejinám Svätého Jura. In: Vavák

(ed.) Zo starších dejín Svätého Jura. Svätý Jur: Mesto Svätý Jur, pp. 7–30.

106.

Velichkevich

Zastawniak

(2006) Atlas of the Pleistocene Vascular Plant Macrofossils of Central and Eastern Europe, Part 1: Pteridophytes and Monocotyledons. Kraków: W. Szafer Institute of Botany, Polish Academy of Sciences.

107.

Velichkevich

Zastawniak

(2008) Atlas of the Pleistocene Vascular Plant Macrofossils of Central and Eastern Europe, Part 2: Herbaceous Dicotyledons. Kraków: W. Szafer Institute of Botany, Polish Academy of Sciences.

108.

Vicherová

Hájek

(2015) Calcium intolerance of fen mosses: Physiological evidence, effects of nutrient availability and successional drivers. Perspectives in Plant Ecology, Evolution and Systematics. Epub ahead of print 28 June. DOI: 10.1016/j.ppees.2015.06.005.

109.

Whitlock

Larsen

(2001) Charcoal as a fire proxy. In: Smol

Birks

HJB

Last

(eds) Tracking Environmental Change Using Lake Sediments. Vol. 3: Terrestrial, Algal, and Siliceous Indicators. Dordrecht: Kluwer Academic Publishers, pp. 75–98.

110.

Willis

Van Andel

(2004) Trees or no trees? The environments of central and eastern Europe during the Last Glaciation. Quaternary Science Reviews 23: 2369–2387.

111.

Willis

Sümegi

Braun

. (1995) The Late Quaternary environmental history of Bátorligert, N.E. Hungary. Palaeogeography, Palaeoclimatology, Palaeoecology 118: 25–47.

112.

Willis

Sümegi

Braun

. (1998) Prehistoric land degradation in Hungary: Who, how and why? Antiquity 72: 101–113.

Origin of a boreal birch bog woodland and landscape development on a warm low mountain summit at the Carpathian–Pannonian interface

Abstract

Keywords

Introduction

Materials and methods

Site description

Sediment sampling and dating

Palaeoecological analyses

Chemical analyses

Results and interpretations

Chronology

Ordination analysis

Regional vegetation development

Local vegetation development

Peat chemistry

Discussion

Regional vegetation history

Local vegetation history: Late Glacial lake phase

Fire activity, hiatus(es) and spatial differences

Birch mire woodland – Relict vegetation or a succession stage?

Conclusion

Footnotes

Acknowledgements

Funding

References