Multi-proxy evidence of Holocene climate variability in Volhynia Upland (SE Poland) recorded in spring-fed fen deposits from the Komarów site

Abstract

Radiocarbon-dated spring-fed fen deposits from the Komarów site (Volhynia Upland, SE Poland) with its multi-proxy data (macrofossils, molluscs, geochemistry, pollen, stable isotopes of oxygen and carbon) enable us (1) to distinguish four main stages of fen evolution, which reflected a distinct variability of water supply conditions and (2) to reconstruct the Holocene humidity–temperature changes. The beginning of peat–tufa deposition took place in a Boreal phase, after a significant cool fluctuation of climate occurring ca. 9.4 ka cal. BP. We suggest that climate was the most important factor conditioning the development of the spring-fed fen. Permafrost degradation, and then wet periods, intensified the activity of ascending springs. The impact of humans was possible since the Neolithic period and increased during the Middle Ages: therefore, the anthropogenic influence could have partially overlapped with the regional tendencies of climate changes. Autogenic development of deposit succession in the studied fen was definitely conditioned by hydrological changes induced by climate. Based on the multi-proxy data, 12 cold events of different ranks were identified. They are also recorded in other Polish and European sites. A record of distinct variability of depositional conditions at ca. 9.4, 8.2, 5.9, 4.6, 2.8, 1.4 and 0.55 ka cal. BP corresponds to quasi-periodical global climate changes in the Holocene named the Bond events. The majority of the cold events recorded in δ¹³C and δ¹⁸O of carbonates can be correlated to the Greenland oxygen isotope curve.

Keywords

calcareous tufa Holocene climate changes multi-proxy south-eastern Poland spring-fed fen Volhynia Upland

Introduction

Knowledge of the Holocene climate fluctuations has been considerably expanded in the last decades with the increasing interest in global climate changes. It has been possible owing to many multiparameter palaeoclimate proxy records gathered from many sites on both hemispheres. Such studies have been undertaken because an understanding of modern, dynamic changes of temperature and humidity requires better knowledge of climate in the recent past. Multi-proxy data from limnic or/and telmatic sediments investigated with high-resolution enable a detailed description of the dynamics of past climate changes at global, continental and regional scales (De Vleeschouwer et al., 2009; Jonsson et al., 2010; Magny, 2004; Magny et al., 2007). Spring mires are a special group of palaeoenvironmental data archives, though they have been rarely used for such purposes. They belong to the rare group of soligenous mires supplied with ascending groundwater outflow. Due to their origin, they cover small areas and form cupolas, which occur mostly within larger mire complexes with different natures of supply. Such sites are known from many areas in the northern hemisphere, well recognized particularly in Europe, which differ in environmental conditions (Boyer and Wheeler, 1989; Kovanda, 1971; Succow, 1988; Succow and Jeschke, 1986). They were found in mountains (e.g. Grootjans et al., 2006; Hájek et al., 2002; Hájková et al., 2012), carbonate uplands (i.a. Alexandrowicz et al., 1994; Dobrowolski, 1994; Dobrowolski et al., 2002, 2005), young glacial moraine plateaux (i.a. Dobrowolski et al., 2010; Osadowski et al., 2009), and more rarely in old glacial landscapes (Żurek, 1990). The development of spring mires is usually related to the activation of ascending groundwater supplies resulting from extraglacial permafrost degradation (Dobrowolski et al., 1996). As permafrost thawing occurred in Europe at different times, depending on the distance from retreating ice sheets and changing climatic conditions, the beginning of such fens’ functioning was not simultaneous (Dobrowolski et al., 2005; Grootjans et al., 2006; Hájek et al., 2002).

Besides the specific vegetation composition and morphology, spring-fed fens are distinguished by a special lithology of deposits, that is, the alternating occurrence of peat and calcareous tufa layers (Dobrowolski, 2011; Kovanda, 1971; Succow, 1988), stratigraphically representing often the whole Holocene (Alexandrowicz et al., 1994; Dobrowolski et al., 2005, 2012). Due to the continuous record of biogenic-carbonate deposition such fens are highly suitable for detailed palaeoenvironmental studies, including palaeoclimatic reconstructions (Dobrowolski et al., 2002; Mazurek et al., 2014; Pazdur et al., 2002). As the course of carbonate deposition is closely connected with the surrounding environment, the tufa layers can be an important indicator of humidity–temperature changes in the Holocene (Dobrowolski et al., 2002). On the other hand, spring fens are dynamic ecological and hydrological systems, and climatic signals can be modified by feedbacks inherent in peat and/or tufa formation and hydrology (Grootjans et al., 2012). One cannot exclude a priori human role in the development of supply and drainage conditions in catchment and post-depositional chemical changes occurring, among other things, due to leaching of the near-surface part of peat deposits by rainwater, decomposition of organic matter (OM) or bioaccumulation of some elements. However, even if the geochemical features are modified in peat–tufa deposits, it may be possible to take advantage of the effect of diagenetic changes on these records and use this information, in connection with other proxies, to infer changes in peatland hydrology as it relates to anthropogenic disturbances or/and climate changes (Bindler, 2006). That is why multi-proxy studies are essential for detailed palaeoclimatic reconstructions in order to separate global or regional signals from local ones (Birks and Birks, 2006; De Vleeschouwer et al., 2009; Kulesza et al., 2012).

The Komarów spring-fed fen in the Volhynia Upland is the first peatland of this type in the eastern part of Europe for which multi-proxy data have been gathered to reconstruct past ecological changes. We focus on an attempt to reconstruct the Holocene climatic changes on a regional scale in relation to autogenic vegetation and human impacts. The aim of our study is to describe the main phases of fen evolution and discuss whether a climate signal can be recorded in the deposits of peat–tufa rhythmite.

Study site

The study site is located in the western, marginal part of the Volhynia Upland, within the Hrubieszów Basin sub-regional unit (Figures 1 and 2). Its location is unique in many respects: geographical (the borderland between two European mega-regional units – East-European Lowlands and Non-Alpine Central Europe), tectonic (the borderland between the East-European Craton and Teisseyre–Tornquist Zone), geomorphologic, hydrological (second-order watershed zone between the Wieprz and Bug rivers) and climatic (transitional zone between oceanic and continental climates). The area is characterized by the shallow occurrence of Upper Cretaceous, carbonate, strongly fissured rocks. Opokas and marly opokas occur in higher hypsometric positions forming the sub-parallel structural ridges – Grabowiec Height and Horodło Ridge (210–230 m a.s.l.) and Sokal Ridge (240–260 m a.s.l.), which surround the Hrubieszów Basin from the north and the south (Figure 2). They are covered by thick (up to 40 m) loess deposits representing several cycles of the Pleistocene periglacial deposition (Dolecki, 2002). Marls and chalk are found in the valley and basin bottoms (180–200 m a.s.l.).

Figure 1.

Location of the study area in relation to the main units of Central Europe: (a) tectonic; (b) physico-geographical.

Figure 2.

Numerical model of the land surface of the Hrubieszów Basin with (a) the location of the Komarów site (in rectangle), and with (b) the distribution of archaeological sites and with (c) the distribution of boreholes.

The Komarów spring-fed fen (50°38′30″N; 23°30′02″E; 213.8 m a.s.l.) is located in the vast mire complex (ca. 1300 ha) filling the basin of the Sieniocha River heads (third-order river in the Horton classification). The spring fen forms a distinct peat–tufa cupola, which is about 1 ha in area (Figure 2), and rises 1.5 m above the peat plain (210–212 m a.s.l.).

The modern vegetation in the site is mostly composed of meadow communities of the Molinion alliance. Small patches of low-sedge rush (Caricion davallianae alliance) occur locally. Fen phytocoenoses are characterized by a high proportion of species typical of the Caricion davallianae, such as Carex davalliana, Carex lepidocarpa, Carex flava, Epipactis palustris and Swertia perennis. Single patches of Cladium mariscus are also found. Within the meadow phytocoenoses the density of the moss layer does not exceed 30%, while low-sedge rushes, in places classified to the Carex davalliana association, are characterized by a considerably higher density of moss layers (up to 90%). Phytocoenoses belonging to the Calthion association occur at the foot of the cupola.

Near the study site, traces were found of settlements from the Neolithic to early Middle Ages (Sadowski, 2008). The oldest traces of human activity in the study area are the remains of Neolithic and early Bronze Age settlements, the highest concentration of which has been found about 1–2.5 km to the south of the site (Figure 2). On the other hand, a small number of archaeological sites with artefacts representing the Roman period indicates that the scale of Roman settlement in the study area was small. The intensity of settlement increased in the early Middle Ages, when the margins of the mire complex were settled in the northern and western parts of the study area (Figure 2), and in modern times.

Material and methods

Field methods

Sampling and sedimentological analysis

Geological investigations of peat deposits (determination of thickness and biogenic deposit succession) were carried out every 10–50 m along orthogonally oriented sections (NW-SE) over the study site. Additionally, in the immediate vicinity of the spring cupola, core drillings were made using a Russian peat corer (length: 50 cm; diameter: 10.0 cm). In total, 14 drillings were made (Figure 2). Each core was sedimentologically analysed – the Troels-Smith method (Troels-Smith, 1955), modified by Dobrowolski (2011), was used for the description of peat–tufa sequences. A core with a total thickness of 8 m, denoted as KOM-9, was taken from the central part of the cupola (γ50°38′30″N; λ23°30′01″E) for detailed laboratory analyses.

Laboratory methods

Palaeobotanical analysis

Macrofossils analysis

Samples (about 30 cm³) for macrofossils analysis were taken every 5 cm. In total, 160 samples were prepared using the method described by Tobolski (2000), and analysed using the accessible botanical keys and atlases (Grosse-Brauckmann, 1972, 1974; Grosse-Brauckmann and Streitz, 1992). Names of vascular plants were given according to Mirek et al. (2002), and of Bryophyta according to Ochyra et al. (2003). As far as possible, the macrofossils were identified to species level and their percentages in samples were calculated.

Pollen analysis

Peat layers of the peat–tufa rhythmite were sampled for pollen analysis. A total of 10 samples (2 cm³) were prepared by the standard Erdman’s acetolysis method (Berglund and Ralska-Jasiewiczowa, 1986) after being treated with HCl to remove carbonates and with 38% HF to remove silica. Pollen, due to its low frequency, was counted on the whole slide of 400 mm² at a magnification of 400×. Pollen frequency varied from several dozen to 500 grains.

Malacological analysis

A total of 159 samples, each 5 cm thick, were taken for malacological analysis. As the number of samples was very high, some of them were combined by taking into account the lithology of distinguished layers. Ultimately, the standard malacological analysis (Alexandrowicz, 1987; Alexandrowicz and Alexandrowicz, 2011; Ložek, 1964) was made in 51 bulk samples. Besides the completely preserved specimens, the fragments and juvenile forms were identified to species level if possible. The number of shell fragments was then converted into the number of complete specimens according to the formula developed by SW Alexandrowicz (1987). Molluscs were identified using keys and comparative collections. Individual species were classified into ecological groups, and their ecological requirements were determined based on literature (e.g. Alexandrowicz, 1987, 2004; Ložek, 1964; Piechocki, 1979; Piechocki and Dyduch-Falniowska, 1993).

Geochemical analysis

Chemical composition was determined in 79 samples taken every 10 cm. After drying at 105°C and homogenization in agate mortar, the samples were calcined at 550°C until a constant weight in order to calculate the contents of mineral and OM. The remains after calcination were treated with aqua regia at room temperature for 16 h and then boiled in a water bath for 2 h. In the obtained solution, the following micro- and macroelements were determined using an atomic absorption spectrometer SpectraAA 20 Plus; Varian: Ca, Mg, Na, K, Fe, Mn, Cu, Zn, Pb and Ni. The results, expressed in mg or g of an element per 1 kg of dry mass, as well as several so-called geochemical environmental indicators, were presented in diagrams prepared using the CONISS software. The content of calcium carbonate was determined with the volumetric Scheibler method (Bednarek et al., 2004).

Identification of sources from which the components were supplied to the spring-fed fen deposits was based on principal component analysis. Standardized values of the contents of OM, CaCO₃, and 10 macro- and microelements were used as input variables.

Radiocarbon dating

Radiocarbon dating was performed for 11 samples of peat layers from the KOM-9 core. Bulk sediments were dated, each from a 2-cm-thick layer. Although it is commonly accepted that short-lived plant remains would yield reliable dating results, the bulk sediment can also lead to acceptable age estimates in the case of spring mire deposits, as was proven by Dobrowolski et al. (2012) and also for other types of non-ombrotrophic peat deposits, for example, organic-minerogenic deposits of fen-type peat mire (Margielewski et al., 2011), poor fen (Fiałkiewicz-Kozieł et al., 2014a, 2014b) or basal peats (Holmquist et al., 2016). Moreover, the location and lithology of the core indicates that there is little likelihood that the OM in the sediment was subjected to redeposition. In the case of spring-fed deposits, the most problematic contamination would be carbonates, which may have much older ages. The sample size was not a limiting factor in the present study, and the application of more economical conventional techniques rather than AMS allowed for dating of more samples from the investigated core, improving the chronological frames for all results. The samples after visual inspection and removal of roots were treated with 2% HCl to remove carbonates, and converted to benzene for Liquid Scintillation Counting with the use of a Quantulus 1220 spectrometer (GdS samples, see Table 1; Pazdur et al., 2003) or ICELS (GdC samples; Tudyka and Pazdur, 2012).

Table 1.

Description of the deposits from the KOM-9 core according to Troels-Smith (1955) and Dobrowolski (2011).

Depth (cm)	Lithology	Units	T-S formula
0–45	Sedge peat, strongly decomposed, dark brown to black, in places with amorphous calcium carbonate precipitations, abundant malacofauna	4	Th⁴4(Car.), Cp(min.)++, sicc.2, nig.4, str.0, test. (moll.)
45–50	Sedge peat, strongly decomposed, light brown to light grey, abundant malacofauna		Th⁴2(Car.), Cp(min.)2, sicc.2, nig.2-3, str.1, test. (moll.), lim.1
50–65	Sedge–moss peat, slightly decomposed, light brown, impregnated with amorphous calcium carbonate, with malacofauna		Th¹2(Car.), Tb¹2,Cp(min.)++, sicc.2, nig.2-3, str.0, test. (moll.), lim.1
65–90	Sedge–moss peat, slightly decomposed, light brown, strongly impregnated with amorphous calcium carbonate, with malacofauna		Th¹2(Car.), Tb¹1,Cp(min.)++, sicc.2, nig.2, str.0, test. (moll.), lim.0
90–100	Sedge peat, slightly decomposed, with inserts of slightly decomposed wood peat, impregnated with amorphous calcium carbonate, with malacofauna		Th¹3(Car.), Tl³1, Cp(min.)++, sicc.2, nig.3, str.0, sicc.2, test. (moll.), lim.0
100–125	Sedge–moss peat, with silty tufa incrustations, with malacofauna		Th²2, Tb²1, Cp(min.)1, sicc.2, nig.2, str.1, test. (moll.), lim.0
125–150	Sedge–moss peat, medium decomposed, brown, impregnated with amorphous calcium carbonate, with malacofauna		Th²3(Car.), Tb³1, Cp(min.)+, sicc.2, nig.3, str.0, test. (moll.), lim.1
150–235	Fine-grained calcareous tufa, in places silty, with inserts of sedge–reed peat, light grey to white	3	Cp(min.)2, Cp(maj.)1, Th¹1, sicc.2, nig.0-1, str.2, lim.1
235–240	Sedge–moss peat, slightly decomposed, light brown, strongly impregnated with amorphous calcium carbonate, abundant malacofauna in the bottom		Th¹2, Tb¹1, Cp(min.)1, sicc.2, nig.2, str.0, test. (moll.), lim.0
240–250	Sedge–moss peat, medium decomposed, brown, slightly impregnated with calcium carbonate		Th²3,Tb²1,Cp(min.)+, sicc.2, nig.3, str.0, lim.0
250–284	Silty calcareous tufa, in places fine-grained, forming incrustation in sedge–reed peat, medium decomposed, olive green–grey		Cp(min.)3, Th²1, sicc.2, nig.2-3, str.1, part test. (moll.), lim.1
284–313	Calcareous tufa, fine-grained, light grey to white, with lamination, abundant malacofauna		Cm(min.)2, Cp(maj.)2, Th³++, sicc.2, nig.0-1, str.2, lim.1
313–337	Sedge peat, medium decomposed, brown, with amorphous calcium carbonate, scarce malacofauna		Th²4(Car.), Cp(min.)+, sicc.2, nig.3-4, str.0, part test. (moll.), lim.1
337–413	Silty calcareous tufa, fine-grained in the bottom, light grey to white, with lamination, abundant malacofauna and inserts of sedge peat in the top		Cm(min.)2, Cp(maj.)2, Th²(Car.)++, sicc.2, nig.0-1, str.2, test. (moll.), lim.0
413–507	Sedge peat, strongly decomposed, in places medium decomposed, dark brown, scarce malacofauna	2	Th^3-44, sicc.2, nig.4, str.0, part test. (moll.), lim.1
507–520	Sedge peat incrusted with calcareous tufa		Th³2, Cm(min.)2, sicc.2, nig.3, str.1, part test moll., lim.0
520–545	Sedge peat, strongly decomposed, in places medium decomposed, dark brown		Th²4, Cp(min.)+, sicc.2, nig.4, str.0, lim.1
545–585	Sedge peat, strongly decomposed, strongly impregnated with amorphous calcium carbonate, abundant malacofauna		Th²3, Cp(min.)1, sicc.2, nig.3, str.0, test. (moll.), lim.1
585–650	Calcareous tufa, fine-grained, with inserts of medium decomposed sedge–reed peat	1	Cp(maj.)3, Th²1, Dh.++, sicc.2, nig.1-2, str.1, lim.1
650–675	Sedge–reed peat, strongly decomposed, impregnated with amorphous calcium carbonate, with malacofauna		Th²4, Cp(min.)+, sicc.2, nig.3-4, str.2, test. (moll.), lim.0
675–700	Silty calcareous and fine-grained tufa, with sedge–reed detritus		Cp(maj.)2, Cp(min.)1, Dh1, Sh++, sicc.2, nig.2, str.0, test. (moll.), lim.1
700–722	Sedge peat, strongly decomposed, slightly silty in the bottom, dark brown to black, scarce malacofauna		Th⁴4, Cs⁴+, sicc.2, nig.4, str.0, lim.1
722–758	Calcareous tufa, fine-grained, grey, with inserts of strongly decomposed sedge peat		Cp(maj.)4, Th⁴+, sicc.2, nig.2, str.0, part test. (moll.), lim.1
758–775	Sedge peat, strongly decomposed, silty in the bottom, with single grains of sand		Th⁴3, As1, Ag+, sicc.2, str.0, nig.4, lim.1
775–800	Fine-grained sand with massive structure, light grey, with organic matter in the top	Bedrock	Gmin4, Sh++, Ag+, sicc.2, str.0, nig.0-1, lim.1

Stable isotopes

An especially important source of palaeoclimate information is the stable isotope composition of the allogenic calcite from peat–tufa deposits. Stable carbon and oxygen isotopes are sensitive indicators of past changes in temperature and precipitation (Jonsson et al., 2010; Leng and Marshall, 2004; Różański et al., 2010). Isotopic analysis was carried out in the Mass Spectrometry Laboratory of Maria Curie-Skłodowska University in Lublin using a dual inlet and triple collector mass spectrometer (modified and modernized MI1305 model). Carbonate samples were analysed on CO₂ produced by reaction with 100% H₃PO₄ in a glass vacuum line connected to the inlet system of the mass spectrometer. The reaction proceeded at an electronically controlled temperature of 25 ± 0.2°C to achieve δ¹⁸O in the PDB scale. For normalization of both δ¹³C and δ¹⁸O values, the international standard NBS-19 was analysed in each series of samples. The analytical uncertainty of both delta values in terms of standard deviation was better than 0.06‰.

Results

Sedimentological analysis

The peat–tufa deposits are 775 cm thick. They consist of four lithostratigraphic units: (unit 1) peat–tufa rhythmite (775–600 cm), (unit 2) sedge peat, medium and strongly decomposed, in places with carbonate incrustations (600–413 cm), (unit 3) tufa–peat rhythmite (413–100 cm), and (unit 4) sedge and sedge–moss peat, in places with carbonate incrustations (100–0 cm). Bedrock is composed of the Vistulian, fluvial–periglacial, fine sands. A detailed description of deposits from the KOM-9 core is presented in Table 1.

Palaeobotanical analysis

Macrofossils analysis

Based on the macrofossils analysis of 160 samples taken from a depth of 5–800 cm, we distinguished six distinct zones (Figure 3) with different species composition (coded Mac 1–6):

Zone Mac 1 (800–705 cm). The occurrence of Carex flava (single nuts) and Carex elata, accompanied by Bryales, is found in the bottom layer. In this zone, the proportion of Phragmites australis is low (up to 5%) but constant, while the percentages of vascular plants (such as Equisetum fluviatile and Menyanthes trifoliate) are very low. Macrofossils of Typha, Schoenoplectus lacustris, Eleocharis acicularis, Chara sp. and Sphagnum sp. occur sporadically in some layers.

Zone Mac 2 (705–605 cm). This zone is dominated by reed macrofossils, accompanied by sedge (up to 20%) and Equisetum fluviatile (up to 1%). The proportion of Bryales does not exceed 1%. Equisetum fluviatile and the oospores of Chara sp. occur in the whole zone. Sphagnum cuspidatum is found in two layers.

Zone Mac 3 (605–470 cm). Macrofossils of sedge vegetative organs predominate. A few fruits of Carex elata are also found. Fruits of Cladium mariscus occur in most of the analysed layers. The following species are identified among the moss macrofossils: Campylium stellatum, Liprichtia cossoni, Calliergonella cuspidata, Scorpidium scorpioides, Warnstorfia revolvens and Drepanocladus sendtnerii. The oospores of Chara sp. and macrofossils of Equisetum fluviatile are also found.

Zone Mac 4 (470–415 cm). In this zone, the macrofossils of Carex sp. predominate. They are accompanied by Bryales with Campylium stellatum as well as Equisetum fluviatile while other plant macrofossils are very scarce.

Zone Mac 5 (415–110 cm). Sedge macrofossils predominate. Among them the most frequent are fruits of Carex elata. Nuts of other sedge species, that is, Carex flava and Carex lasiocarpa, occur sporadically. The nuts of Cladium mariscus are found in many layers. The continuous occurrence of Equisetum fluviatile is found in the bottom part. The macrofossils of Thelypteris palustris occur in several layers. Among Bryophyta, three species dominate, especially in the bottom layers (Campylium stellatum, Liprichtia cossoni, Calliergonella cuspidata), while Hamatocaulis vernicosus, Scorpidium scorpioides and Warnstorfia exannulata appear periodically or sporadically. The frequencies of Chara sp. are periodically high.

Zone Mac 6 (110–5 cm). The predominant Carex elata is accompanied by periodically abundant Cladium mariscus. The proportion of Bryales increases in particular layers to 75%. The continuous occurrence of Equisetum fluviatile is found. Single oospores of Chara sp. are found in several layers.

Figure 3.

Macrofossil and pollen diagram for the KOM-9 core. The bars show the following: for macrofossils – number of macrofossils per sample; for pollen – percentage number of pollen grains per sample (black bars – %; empty bars – ‰). Zonation code – explanation in text.

Pollen analysis

The results of pollen analysis of the object under study are presented in Figure 3 as the percentages of pollen. The pollen spectra are characterized by a low content of sporomorphs and their small taxonomic diversity. This fact confirms that peat–tufa deposits are not favourable to preservation of pollen (Alexandrowicz et al., 1994; Pidek et al., 2012). Degradation of sporomorphs was favoured by the oxidizing depositional environment connected with the supply of bicarbonate-calcium water to the spring-fed fen cupola (Alexandrowicz et al., 1994). Slightly better conditions of sporomorphs deposition are recorded only in the spectra representing sedge–reed peat impregnated with amorphous calcium carbonate (with Phragmites australis) and sedge–moss peat with Camphyllum stellatum and Calliergonella cuspidata. The accumulation layer was less saturated with oxygen in more wet environments and under moss carpets. The obtained results are very poor material to reconstruct the changes of vegetation cover; however, the varying frequency of sporomorphs may indicate palaeohydrological changes. According to the results of radiocarbon dating, the first pollen grains of thermophilous trees (Quercus, Alnus and Corylus) appeared only in the Atlantic period. They do not give a representative picture of vegetation cover in the area under study (Bałaga, 1998) as pollen of Ulmus, Tilia and Fraxinus (species that had their development optima in this period) is absent (compare Bałaga, 1998). Pollen of Ulmus and Tilia is found only in the samples representing the Subboreal period. The deposit layers with Carpinus, Fagus and Abies pollen were undoubtedly deposited in the younger Holocene. Due to the relatively low frequency of sporomorphs, in the detailed palaeoecological interpretation we used also the published pollen diagrams obtained for mires and lakes occurring in the immediate vicinity of the study site (Bałaga, 1998).

Malacological analysis

Malacofauna found in the Komarów site is rich and differentiated with respect to species composition and assemblage structure (Figure 4). The species identified in the analysed material belong to five ecological groups of molluscs: shade-loving snails (F), open-country (meadow) snails (O), mesophilous snails (M), hygrophilous snails (H) and water molluscs (W). The last group is divided into two sub-groups: species typical of intermittent water bodies (T) and species of perennial water bodies (P). The whole analysed material comprises 34 species – 18 taxa of terrestrial snails, 8 taxa of water snails and 6 species of bivalves. The number of taxa in individual samples ranges from 6 to 20 and the number of specimens from 21 to 709. Precise analysis of environmental changes is difficult because of the relatively low number of shells (below 100) in some samples; however, the main trends in depositional conditions are reconstructed based on the analysis of the whole sequence and succession of malacocoenoses in longer sections of the profile. The presence of some samples with a lower frequency of fauna does not significantly affect the interpretation of the entire profile. The following malacological zones were distinguished in the profile under study (Figure 4):

Zone Mal 1 (800–580 cm). The bottom part of this zone is dominated by terrestrial taxa, mostly meadow and mesophilous ones. The shade-loving species are also present in this part. Species typical of intermittent water bodies, mostly Galba truncatula (Müll.), are more numerous in the upper part of the zone.

Zone Mal 2 (580–470 cm). This zone is characterized by a considerable proportion of species of intermittent water bodies and hygrophilous taxa. The frequency of water species strongly fluctuates. The composition and structure of fauna indicate that wet and periodically flooded habitats prevailed.

Zone Mal 3 (470–410 cm). This malacocoenosis is characterized by the disappearance of water species and the predominance of meadow and hygrophilous taxa. It is evidence of distinct drying of habitats.

Zone Mal 4 (410–100 cm). This zone is characterized by the predominance of water species and the limited occurrence of terrestrial snails. The predominant taxa, typical of intermittent water bodies (Galba truncatula (Müll.)), are accompanied by taxa preferring shallow, strongly overgrown but perennial water bodies: Anisus contortus (L.), Pisidium obtusale (Lam.) and Bithynia tentaculata (L.) with the predominance of shells over opercula. The ratio of Bithynia tentaculata (L.) shells to opercula (so-called ‘Bithynia-index’) is commonly used to characterize the features of water bodies (Alexandrowicz, 1999a, 1999b, 2007, 2013).

Zone Mal 5 (100–0 cm). The top zone of the profile is mostly represented by the species of meadow biotopes (Vallonia pulchella (Müll.), Vertigi pygmaea (Drap.)) and mesophilous taxa such as Succinella oblonga (Drap.).

Figure 4.

Malacofauna in the KOM-9 core: D-1, D-2 – two-component diagrams showing the ratios of selected molluscan groups, which indicate the changes in environmental conditions of deposit accumulation: D-1 – ratio of terrestrial species (L) to water species (W), D-2 – ratio of species of intermittent water bodies (T) to those of perennial water bodies (P); F-1 – changes in habitat characteristics: (L) mainly terrestrial habitats, (W) mainly water habitats; F-2 – changes in water habitats’ characteristics: (T) mainly habitats of intermittent water bodies, (P) mainly habitats of perennial water bodies; N – number of taxa (N_T) and specimens (N_S); Mf – malacofauna, Sm – malacospectrum of species; F – shade-loving species, O – open-country species, M – mesophilous species, H – hygrophilous species, W – water species. Zonation code – explanation in text.

Geochemical analysis

Based on the varying contents of two main components of deposits in the analysed core – OM and calcium carbonate – five main geochemical zones are distinguished, which also differ in the macro- and microelements contents as well as the geochemical indices values (Figure 5):

Zone Chem 1 (800–775 cm). This zone represents mineral substratum composed of fine-grained sands with massive structure. It is characterized by a low content of OM (up to about 6%), calcium carbonate (up to 4.5%), and also of most elements. Only potassium and iron reach higher contents in this zone. This fact results in very low values of the Na/K ratio and in high values of Fe/Mn and Fe/Ca ratios. The content of OM increases in the transition zone between the mineral substratum and the overlying sedge peat layer.

Zone Chem 2 (775–585 cm). The highest content of CaCO₃ (up to 93%) is found in the bottom part of the zone. The OM content gradually increases upward in the profile reaching the maximum values in peat layers. The deposits of this zone are characterized by high contents of Fe, Mn and Ni, which also reach their maximum concentrations in the whole profile: 9.26 g kg⁻¹ DM, 578 mg kg⁻¹ DM and 34.9 mg kg⁻¹ DM, respectively.

Zone Chem 3 (585–413 cm). This zone is distinguished by the highest content of OM and a low content of calcium carbonate (below 32%). The concentrations of Na, K, Fe, Mn and Cu are also low, while the contents of other elements fluctuate around their average values calculated for the whole core.

Zone Chem 4 (413–150 cm). This zone is composed of different peat types with very varying content of OM (25–85%), which reaches three maxima exceeding 70%. The course of the CaCO₃ content curve is reversed. The interrelation between these two components is strong, with a correlation coefficient r = −0.97 (p < 0.001). The concentrations of lithophilic elements (Ca, Mg and Na) increase in this zone. The copper content reaches high values, while the concentrations of potassium, iron and manganese are low. The Fe/Mn index reaches the lowest value (below 2) in the whole core.

Zone Chem 5 (150–0 cm). The calcium carbonate content reaches the lowest values in the bottom part of this zone and increases to about 30% in its top part. The content of OM is high. The concentrations of calcium, iron, manganese and nickel decrease, and those of copper and potassium are average. The deposits of this zone are characterized by high contents of Zn (11.3 mg kg⁻¹ DM) and Pb (7.1 mg kg⁻¹ DM) with the maximum values in the subsurface layer.

Figure 5.

Geochemistry of deposits in the KOM-9 core. Geochemical zones are based on stratigraphically constrained cluster analysis CONISS. Data values were standardized before cluster analysis. Zonation code – explanation in text.

Based on the contents of 12 geochemical components of deposits, three complementary eigenvectors are distinguished that together explain 65.4% of the geochemical variability of the deposits (Figure 6). The first eigenvector (PC1), which explains 32.51% of the total variation, is inversely proportional to the content of OM, and directly proportional to the contents of carbonates, calcium and manganese. The second eigenvector (PC2), which explains 18.99% of the total variation, is directly proportional to the content of magnesium, and is inversely proportional to the content of potassium. The third eigenvector (PC3) is less important (it explains 13.9%). It mostly reflects the variability of micro-components, such as copper and zinc.

Figure 6.

Principal component analysis based on geochemical classification of sediment samples by the first and second components.

The first eigenvector (PC1) is partially connected with the autochthonous part of deposits, that is, OM. Its presence affects redox conditions and pH of depositional environment, which in turn influence geochemical processes and changes in the concentrations of some elements in deposits (e.g. Fe). The occurrence of K (second eigenvector) and Zn (third eigenvector) in organic deposits may also be related to autochthonous bioaccumulation of these elements.

Positive values of the first and second eigenvectors reflect the occurrence of allogenic (in terms of the alimentation area of the spring-fed fen) products of weathering of carbonate, strongly fissured rocks (opokas and marly opokas, marls and chalk) and loess (Ca, Mg, Na), which are supplied to the mire by groundwater (Janiec, 1997; Michalczyk, 2001). Fe and Mn occurring in the spring-fed fen could also have been leached by groundwater from carbonate deposits. The possibility of their migration in water varies depending on redox conditions. In the studied case, the proximity of the first two eigenvectors in the diagram (Figure 6) indicates the accumulation under oxidizing conditions and high pH (for Ni it is similar). High mean content of Cu in the studied deposits, and also the location of the first two eigenvectors (Figure 6) indicate that this element has also been supplied by chemical denudation occurring in the mire catchment.

The second eigenvector (PC2) is also identified with potassium, the content of which is often interpreted as an indicator of mechanical denudation (Wojciechowski, 2000) and supply of clastic material from the associated catchment. A part of Zn has also been supplied in this way (third eigenvector). The increased concentration of heavy metals, especially Zn and Pb, in the top part of the profile may indicate allogenic atmospheric supply and influence of anthropogenic activity. However, Shotyk (1996) notes that in the case of a mire supplied by both groundwater and precipitation it is impossible to distinguish between atmospheric and hydrospheric metal inputs.

Radiocarbon dating and age–depth model

The measurement results are expressed as radiocarbon ages BP with associated uncertainty in Table 2. The radiocarbon dating results were subjected to calibration with the use of the IntCal09 calibration curve (Reimer et al., 2009) and OxCal v.4 software (Bronk Ramsey, 2009). An age–depth model has been constructed using a P_Sequence function (Bronk Ramsey, 2008), with the parameter k = 0.5. The boundaries of the most pronounced changes of sediment characteristics have been introduced at depths of 150, 415 and 585 cm. Between those depths, the sections are assumed to have a constant deposition rate. Moreover, the top of the core (depth 0 cm) was assigned the collection year AD 2009 and the model was extrapolated to a depth of 800 cm.

Table 2.

Results of radiocarbon dating and calibration for the KOM-9 core (bulk organic sediment). Dates considered as outliers are marked by asterisks.

No.	Sample name	Lab. No.	T (¹⁴C BP)	Calibrated age (yr cal. BP) (range 68%)	Calibrated age (yr cal. BP) (range 95%)
1	KOM-9/45 (43–45 cm)	GdS-980	445 ± 30	525 (68.2%) 490	535 (95.4%) 460
2	KOM-9/100 (98–100 cm)	GdC-267	1120 ± 55	1080 (68.2%) 955	1175 (95.4%) 930
3	KOM-9/127 (125–127 cm)	GdC-279	1740 ± 35	1705 (68.2%) 1605	1730 (95.4%) 1545
4	KOM-9/150 (148–150 cm)	GdS-983	2090 ± 35	2120 (51.8%) 2035	2155 (93.4%) 1985
				2030 (16.4%) 2000	1980 (0.9%) 1970
					1965 (1.1%) 1951
5	KOM-9/242 (240–242 cm)	GdC-269	2075 ± 65^*	2130 (64.7%) 1970	2305 (6.4%) 2240
5	KOM-9/242 (240–242 cm)			1965 (3.5%) 1951	2180 (89.0%) 1885
6	KOM-9/315 (313–315 cm)	GdS-972	3110 ± 60	3395 (68.2%) 3255	3460 (93.0%) 3200
6	KOM-9/315 (313–315 cm)				3195 (2.4%) 3160
7	KOM-9/337 (335–337 cm)	GdC-272	3245 ± 60	3560 (15.8%) 3520	3620 (95.4%) 3360
				3510 (5.1%) 3495
				3490 (47.3%) 3395
8	KOM-9/415 (413–415 cm)	GdS-971	4290 ± 80	5035 (2.4%) 5015	5265 (2.4%) 5185
				4980 (55.6%) 4810	5060 (92.9%) 4575
				4760 (9.6%) 4705
				4670 (0.5%) 4660
9	KOM9/532 (530–532 cm)	GdS-1114	5140 ± 110	6005 (68.2%) 5730	6185 (95.4%) 5650
10	KOM-9/675 (673–675 cm)	GdS-1049	7780 ± 120^*	8720 (68.2%) 8415	8985 (95.4%) 8385
11	KOM-9/722 (720–722 cm)	GdS-1024	7720 ± 150	8720 (68.2%) 8365	8995 (93.3%) 8300
11	KOM-9/722 (720–722 cm)				8260 (2.1%) 8210

During the analysis, two dates were identified as outlying (GdC-269 and GdS-1049, see Table 2). Although they do not clearly demonstrate the age inversion, including these two dates in the model would lead to unreasonably high sedimentation rates for the intervals 150–240 and 675–722 cm and a very poor agreement index of the whole model. The final age–depth model presented in Figure 7 was obtained with an agreement index of 57%, and indicates that the whole record reaches ca. 9740 yr cal. BP.

Figure 7.

The age–depth model for the KOM-9 core obtained with the use of the P_Sequence function in the OxCal programme. Grey-shaded rectangles represent calibrated radiocarbon ages; 68.4% probability is marked with the darker areas, 95.4% probability with the lighter colours.

The uncertainty of the age–depth model generally increases with time, and is characterized by the average value of 144 years for the whole core. However, for the depths between 0 and 450 cm (up to 5020 cal. BP), the average is only 72 years, while for the deeper levels it increases at an almost constant rate from 115 years at 450 cm up to 330 years for the lowermost, extrapolated part of the age–depth model.

The sediment accumulation rate is almost constant, and the average value corresponds to the resolution of 12.2 ± 3.0 yr cm⁻¹. Taking into account sample thickness, the samples for ¹⁴C dating covered ca. 25 years each, which is less than model uncertainty. Furthermore, the 5-cm-thick sample sizes for macrofossil and malacological analysis correspond to ca. 60-year resolution.

Stable isotopes

The stable isotope records (δ¹³C and δ¹⁸O) in the studied profile are highly variable with a general trend towards higher values with development of the alkaline spring, excluding unit 3 (tufa–peat rhythmite) when this trend was opposite. It should be noted that δ¹³C is much more variable than δ¹⁸O, being directly related to δ¹⁸O of water and its temperature (Craig, 1963; Demény et al., 2010; O’Neil et al., 1969) in which calcite was precipitated as calcareous tufa. Minor variations of δ¹⁸O in the precipitated calcite reflect a compensation effect: during climate warming, δ¹⁸O in precipitation increases, whereas the magnitude of isotope fractionation diminishes. According to the global correlation line reported by Dansgaard (1964), δ¹⁸O in precipitation increases by 1‰ when mean annual air temperature increases by 1.5°C. Hence, in our case, the temperature increase was at least 3°C, considering 2‰ increase of δ¹⁸O during the period of travertine formation.

On the other hand, calcite δ¹³C in the profile varies significantly from −9‰ to −12‰. For such large variations, the influence of plant growth may be responsible. Plants favour the consumption of isotopically light carbon isotope in the assimilation process; hence, with an enhanced development of plant cover around the spring (warmer climate) more ¹²CO₂ is extracted from the solution and the remaining bicarbonate is enriched in heavy isotopes and δ¹³C of the precipitated calcite will be less negative. In such periods, more intensive calcite precipitation is expected, which is the case. Other factors like solar radiation and atmospheric humidity also influence δ¹³C, but to a lesser extent.

Discussion

Fen development

The results of the multi-proxy studies enable us to reconstruct the evolution of the spring-fed fen ecosystem. We distinguish four main evolution phases (terminology after Dobrowolski, 2011), which show a distinct variability of the water supply and ecological conditions.

Phase I – Crenous – 9740–6440 cal. BP (lithological units: 1 and bottom part of 2; Mac: 1, 2 and bottom part of 3; Mal: 1; Chem: 1 and 2)

The initial phase of the spring mire development occurred at the beginning of the Boreal chronozone and was related to the activation of artesian spring water. This fact suggests that the activation of ascending groundwater circulation, resulting from the final degradation of permafrost, occurred relatively late, ca. 1000 years later than in the northern Lublin Upland foreland (Dobrowolski et al., 1996, 2015). Therefore, we may suppose that in the areas characterized by specific conditions – geographical/climatic (= eastern part of Central Europe with more continental climate), geologic (= elevated blocks of the Upper Cretaceous massif), and morphologic (= upper part of the catchment, near watershed) – permafrost could have still been preserved during the whole Preboreal chronozone up until a considerable warming at the beginning of the Boreal chronozone. This is indirectly confirmed by the sedimentary succession and pollen spectrum from the adjacent site on the Volhynia Upland (Bałaga, 1998). This episode in the KOM-9 core is recorded in the macrofossils composition by a considerable predominance of sedges, indicating that rushes with Carex flava disappeared at the time. These communities were developing on alkaline basement, strongly supplied with bicarbonate-calcium water (Zarzycki et al., 2002). With the decrease of the ascending groundwater supply, the sedge communities passed into the high-sedge rushes with Carex elata. Terrestrial snails (meadow, mesophilous and shade-loving species) were predominant in this community. The occurrence of the latter indicates that the developing fen was surrounded by shrubs and varied deciduous forests with Quercus, Fraxinus, Alnus, Tilia and pine forests (Bałaga, 1998). Vertigo geyeri, an indicator species of a cool climate, appeared at this time. From ca. 8500 cal. BP the malacological succession was characterized by the increased proportion of snails preferring dry open habitats.

About 8100 cal. BP, the vegetation succession changed considerably. Sedges and Bryales were still present but reeds became predominant. The predominance of reeds lasted about 1500 years. Strongly wet habitat in this phase of fen development is also evidenced by the occurrence of Equisetum fluviatile, Sphagnum cuspidatum and Chara sp.

Reeds had been disappearing since ca. 6800 cal. BP. With the change of water conditions into a drier environment the sedge communities, with a low proportion of Bryales, were developing again. Similar trends in fen evolution at that time are also recorded in other neighbouring sites, including Tarnawatka and Krasnobród, approximately 20 km away from the study site (see Figure 1). Shade-loving snails disappeared, probably because the open surface of the fen increased considerably, though it was surrounded by climax forest communities with elm, oak, ash, lime and alder of Atlantic chronozone (Bałaga, 1998).

Phase II – Helocrenous – 6440–4570 cal. BP (lithological unit: upper part of 2; Mac: upper part of 3 and 4; Mal: 2 and bottom part of 3; Chem: 3)

In this period, the high-sedge rushes were still developing with Carex elata. This species predominated between 6200 and 5900 cal. BP. Equisetum fluviatile, Phragmites australis and Thelypteris palustris were the most abundant in comparison with the other phases. Bryales were still present as well. Cladium mariscus appeared for the first time. The Chara developed with a different intensity during the whole phase indicating changes in habitat moisture. Malacofauna of wet and periodically flooded habitats existed under these conditions with varying moisture (Figures 4 and 8).

Figure 8.

Comparison of multi-proxy records from the Komarów spring-fed fen (KOM-9); blue-shaded areas mark the ‘cold periods’ (= Bond events) during the Holocene and orange-shaded areas mark the ‘warm periods’ over the same time period.

A considerable drying of the habitat started ca. 4800 cal. BP, and is recorded by a gradual decrease of the Bryales proportion in the vegetation succession. The contribution of boggy species, such as Equisetum fluviatile, also decreased, while Chara sp. and Cladium mariscus completely disappeared. Water snails also disappeared, while meadow and hygrophilous species became predominant (Figures 4 and 8).

Phase III – Limnocrenous – 4570–2000 cal. BP (unit: 3 and bottom part of 4; Mac: 5 and upper part of 6; Mal: 4 and bottom part of 5; Chem: 4)

Generally, three periods of intensive groundwater supply to the fen are recorded in the deposits of this phase (= 3 maxima of CaCO₃ concentration); however, they did not considerably influence the succession in vegetation communities and molluscan assemblages. The fen was still covered by sedges with Carex elata, and periodically with a considerable proportion of Cladium mariscus. The latter species overgrew the spring cupola mostly between 2500 and 1360 cal. BP. The community had a rich moss layer composed of the following species: Campylium stellatum, Limprichtia cossoni, Calliergonella cuspidata, Scorpidium scorpioides, Warnstorfia revolvens and Drepanocladus sendtnerii. The occurrence of Chara sp. and Equisetum fluviatile indicates that the habitat was wet in this phase. Water species were dominant among snails living in these communities. The taxa typical of intermittent water bodies, for example, Galba truncatula (Müll.), predominated. Taxa preferring shallow, strongly overgrown but more perennial water bodies also appeared. The existence of a paralimnic small pool is evidenced by the occurrence of Anisus contortus (L.) and Pisidium obtusale (Lam.) (Figures 4 and 8). The species composition of molluscan assemblages indicates that this period was the most humid in the whole history of the spring-fed fen development.

Phase IV – Helocrenous – 2000 cal. BP – The present (lithological unit: upper part of 4; Mac: upper part of 6; Mal: upper part of 5; Chem: 5)

In this phase, the pool was overgrown, and then wet boggy habitats were replaced by drier meadow biotopes (Figures 4 and 8). The beginning of the phase was characterized by the development of sedge–moss communities. Since ca. 1150 cal. BP, during the next 400 years, Equisetum fluviatile had become predominant. Then sedge rushes developed, at first with the Cladium mariscus and abundant Bryales, and since ca. 560 cal. BP with Carex elata. Bryales gradually disappeared. The occurrence of Chara sp. indicates periodical water logging of the habitat. The species composition of snails also considerably changed during this phase. The predominant water species were gradually replaced by terrestrial species. The only exception was a very wet, short episode ca. 900 cal. BP, which was characterized by the growing importance of water species.

Climate changes, human impact or autogenic development

Regional climatic conditions, including long-term climatic regimes and short-term weather events, are considered to be the most responsible for peatland development (e.g. Charman, 2002; Lamentowicz et al., 2008), although it should be stressed that other allogenic disturbances (human impact) and autogenic factors – self-regulation mechanisms (peat growth, vegetation succession) – can also modify water supply conditions (e.g. Swindles et al., 2013). In this discussion, special attention should be focused on spring-fed fens because peat–tufa series accumulate in more complex depositional environments than deposits of bogs or/and kettle bogs. Therefore, the key issue for palaeoenvironmental reconstructions, based on the peat–tufa rhythmite of the spring-fed fen in the Komarów site, is an attempt to distinguish local signals, including human impact, from regional and over-regional climate signals, examined for the main stages of mire development.

The small area and size of the cupola of the spring-fed fen at Komarów, which is embedded in a vast peat plain, its location in the upper part of the catchment (near the watershed) and outside the direct influence of anthropogenic activity (also in the past), as well as its deposit sequence typical of artesian supply, indicate a greater role of regional factors in the mire development rather than local changes in supply. The Komarów mire is supplied from the aquifer with a fissure-layer-type of water circulation, which developed in Cretaceous rocks occurring near the ground surface. Over 50 years of observation of the spring regime in this part of the Volhynia Upland indicates high stability of spring discharge and quick response to changes in supply (Michalczyk, 2001); therefore, we can suppose that the changes in water supply to the mire substantially reflect regional changes in precipitation regime. Moreover, the location of the Komarów spring-fed fen in an area with continental climate influences indicates that in this case, as suggested by Schoning et al. (2005), temperature might be the most important parameter governing the peatland hydrology, and changes in temperature control oscillation of the groundwater table.

Early Holocene (before 8.3 ka cal. BP)

The early Holocene was a very dynamic period in wetland development (Roberts, 1998). Climatic changes resulted in a substantial restructuring of supply and drainage conditions, and radical morphogenetic changes, which occurred after complete degradation of permafrost. In the Volhynia Upland, these phenomena took place distinctly later than in most areas of Western and Central Europe (Dobrowolski et al., 2005; Grootjans et al., 2006; Kokfelt et al., 2010; Starkel et al., 2013), most probably because of more severe, continental climate conditions in the study area that resulted in permafrost preservation until the end of the Boreal period.

The oldest cold and dry Holocene climatic episode, preceding the development of the spring-fed fen at Komarów, is recorded in the bottom part of the KOM-9 core as the relatively low values of oxygen and carbon stable isotopes. This phase, lasting till ca. 9.0 cal. BP, probably corresponded to the cool Erdalen event (= Bond event 6) closing the Preboreal chronozone (Bond et al., 1997; Dahl et al., 2002; Nesje and Dahl, 1991; Figures 8 and 9).

Figure 9.

The results of stable isotope measurements for carbonates from the KOM-9 core compared with the δ¹⁸O curve for NGRIP ice core (North Greenland Ice Core Project Members, 2004) placed on the GICC05 time scale (Vinther et al., 2006), Lake Gościąż, Central Poland (Ralska-Jasiewiczowa et al., 1998) and Lake Purwin, NE Poland (Gałka and Apolinarska, 2014). Main events discussed in text are marked by a dotted line and description. The probability density function for calcareous tufa (CPDF) and a summary of temperature changes from Starkel et al. (2013) are presented (with permission).

The activation of ascending groundwater supply at the beginning of the Boreal chronozone resulted from an abrupt increase in temperature and humidity on an over-regional scale (recorded in the KOM-9 core as a considerable increase in the δ¹³C and δ¹⁸O values – Figures 8 and 9) after a distinctly dry and cool phase. These changes caused an intensive deposition of calcareous tufa (with high concentrations of Ca, Fe, Mn and Ni). A similar tendency of intensified leaching of carbonates in catchments, and their precipitation in seepage spring areas is recorded in the whole belt of Polish carbonate uplands (Dobrowolski et al., 1996, 2002, 2005; Pazdur et al., 1988; Starkel et al., 2013).

‘8.2 ka event’ (8.3–8.1 ka cal. BP)

The initial Atlantic chronozone, recorded in the deposits of the KOM-9 profile, was warm and humid and lasted ca. 350–380 years. It ended with a gradual deterioration of temperature–humidity conditions since ca. 8400 cal. BP with the maximum of distinct cooling recorded in the δ¹⁸O curve ca. 8230 cal. BP. The deposits accumulated at that time in the Komarów site are characterized by a considerably higher content of autochthonous OM, lower content of carbonates, as well as by the minimum of the δ¹⁸O curve (dependent on mean annual temperature and humidity) and of δ¹³C curve (connected with water temperature and the escape rate of CO₂ released by vegetation to the atmosphere). Both δ¹³C and δ¹⁸O values changed simultaneously in this time range (Figures 8 and 9). A decreasing trend in humidity in the same period, expressed as a rapid lithofacial change of biogenic deposition, was also recorded in other peat profiles of the Volhynia Upland, among others in the Krasnobród and Tarnawatka sites (Bałaga, 1998), as well as in lacustrine profiles of the Volhynian Polesie, in the Czerepacha, Okunin (Dobrowolski et al., 2001), Perespilno (Bałaga, 2004) and Bezedna (Dobrowolski et al., 2015; see Figure 1) sites. This phase can be correlated with the cold event occurring 8.2 ka BP (Bond event 5), which is related with the final melting of the Laurentian ice-sheet and with outflows of freshwater into the northern part of the Atlantic Ocean (e.g. Barber et al., 1999; Bond et al., 1997; Muscheler et al., 2004; Renssen et al., 2007; Wanner et al., 2011). The cool event is also distinctly recorded in the oxygen isotope curve obtained for ice cores (North Greenland Ice Core Project Members, 2004), which is presented in Figure 9. An episode of cooling, a decrease in humidity and the consequent lowering of the water table around 8.2 ka BP were recorded in many profiles of lacustrine deposits in Central and Northern Europe (Alexandrowicz, 1999b, 2007, 2013; Digerfeldt, 1988; Sarmaja-Korjonen and Seppä, 2007; Seppä et al., 2005; Starkel, 1999; Szeroczyńska and Zawisza, 2011; Wojciechowski, 2000). In Western Europe, this period is described as cool but rather humid, with increasing water tables in lakes (Magny et al., 2007; Wenninger et al., 2006). The palaeoclimatic data obtained for the Central European lakes indicate that at that time the mean temperature of the warmest month was approximately 2°C lower than current values (Magny et al., 2007; Szeroczyńska and Zawisza, 2011; Wenninger et al., 2006).

Mid Holocene (8.1–2.0 ka cal. BP)

In the KOM-9 core, after the cooling about 8.2 ka BP, a warm and humid phase is recorded that lasted ca. 1500 years. That period was characterized by the next intensification of tufa deposition (higher values of the CaCO₃/OM index) resulting from a more intensive supply of groundwater to the spring-fed fen and a change of water chemical composition reflecting the intensity of chemical denudation processes in the catchment. Moreover, small fluctuations of δ¹⁸O and a gradual increase in δ¹³C are recorded (Figures 8 and 9). A similar trend of climate warming in Poland, lasting from 8100 to 7300 cal. BP and evidenced by speleothems’ growth and calcareous tufa deposition, was found by Starkel et al. (2013).

In the analysed profile, a slight, short-lived deterioration of temperature–humidity conditions was also recorded in biotic proxies and stable isotopes composition, ca. 7700 and 7250 cal. BP. These climate changes were also reflected by an increase of water level in the Volhynian Polesie lakes – Lake Słone (Kulesza et al., 2012) and Lake Bezedna (Dobrowolski et al., 2015), similarly as in Lake Linówek and Lake Purwin in NE Poland (Gałka and Apolinarska, 2014).

The next cooling period, 7250–6200 cal. BP, with very distinct drying about 6750 cal. BP, is recorded as an abrupt decrease in the δ¹³C and δ¹⁸O values. A gradual decrease in humidity is evidenced by an increase of OM content (= sedentation of sedge peat) and simultaneous decrease of CaCO₃ content. This period is characterized by a minimum in the calcareous tufa deposition record for Poland (see Figure 9). It can be correlated with a change of biogenic accumulation in the Tarnawatka site ca. 6200 cal. BP (Bałaga, 1998). A decrease in humidity contributed to the disappearance of water bodies and the development of a mire in this site. According to Żurek et al. (2002), the period ca. 6700–6000 cal. BP (5100–5800 BP) is widely regarded as the driest on mires of Eastern Poland.

The successive improvement of temperature–humidity conditions is recorded in all proxies (KOM-9) and dated between 6200 and 6000 cal. BP (Figure 8). A similar trend of climate changes in Polish territory, recorded in different proxies but within a slightly longer time-frame (from 6400 to 5600 cal. BP), is reported by Starkel et al. (2013). For Central and North-Eastern Europe, this period is named the Holocene Thermal Maximum (HTM) (Heikkilä and Seppä, 2003; Heikkilä et al., 2010; Starkel et al., 2013).

After a warm and relatively humid phase, there was a distinct cooling ca. 5900 cal. BP with an abrupt decrease in humidity. In the KOM-9 core, it is recorded as medium and strongly decomposed sedge peat with low concentrations of Na, Fe, Mn and Cu. A low concentration of Ca, connected with a low content of calcium carbonate in the deposits, indicates that the supply of chemical denudation products to the spring-fed fen decreased. It may be related to a decrease in the groundwater supply and lower intensity of leaching processes in the underground drainage basin, suggesting that geochemistry is largely controlled by allogenic processes (Lamentowicz et al., 2009). OM was gradually deposited in the environment with limited access to oxygen, under Mn-reducing conditions, which could have resulted in its lower concentration in deposits. A similar situation was found in the Okunin site in the Volhynian Polesie, where this period was characterized by the minimum rate of biogenic accumulation (Dobrowolski et al., 2001). In the Tarnawatka site, this dry period is recorded as a change in peat succession – moss is replaced by sedge (Bałaga, 1998). Similar, very dry climatic conditions during this period are described for the whole northern hemisphere based on different proxies (Cremaschi, 1998; Parker et al., 2006; Wanner et al., 2011). They were also recorded, based on the isotope records, in lacustrine cores from the territory of NE Poland (Gałka and Apolinarska, 2014); however, the termination of that period was not synchronous in all described lakes. Generally, the cold and dry episode ca. 5900 cal. BP from the KOM-9 core can be correlated with Bond event 4, related to a minimum in solar activity (Bond et al., 1997, 2001).

Subboreal chronozone, recorded in the KOM-9, started with a distinct increase in the δ¹⁸O values corresponding with the maximum of δ¹³C, which indicate a considerable though short climate warming (Figures 8 and 9). However, the first 1500 years of this period were generally regarded as cool, with only several episodes characterized by relatively higher humidity, which were separated by distinctly drier phases. These oscillations are clearly recorded in the deposit sequence, malacofauna and macrofossils composition, and fluctuations of δ¹³C and δ¹⁸O. A considerable cooling, expressed as a distinct decrease in δ¹³C and δ¹⁸O, occurred ca. 5200–5400 cal. BP. The widespread, strong climate cooling and moistening ca. 5600–5000 cal. BP is well recorded in all of Europe (Magny et al., 2009). At that time there was also recorded an intensification of catastrophic events in the Polish Carpathians, such as landslides and debris flows, as well as the appearance of minerogenic horizons in mires (Kotarba, 2006; Margielewski, 2006; Starkel et al., 2013). The earliest traces of Neolithic settlements, which were found near the studied site, date from this period (Figure 2). Activity of the Neolithic groups in the study area can also be indirectly inferred from the pollen diagrams from other peat sites in the region (Bałaga, 1998). In these diagrams are recorded the maxima of Pteridium aquilinum spore, the occurrence of meadow and nitrophilous herbs, together with a decline of Pinus and Ulmus curves, as well as small fluctuations of deciduous trees at the end of the Atlantic period. These changes could have been connected with the settlement or stay of Neolithic tribes in this region (Bałaga, 1998). Although the traces of settlement in the study area were found 1–2.5 km outside the alimentation zone of the spring-fed fen (Figure 2), we should consider the possible impact of anthropogenic deforestation and progressive development of agriculture. These potential changes in land use should have led to an increase of the water table and consequently to a change of trophic states and vegetation. However, it is not confirmed by the results of geochemical analysis in the KOM-9 core. For the time ca. 5900 cal. BP, they indicate that the supply of chemical denudation products to the spring-fed fen decreased because of a lower intensity of leaching processes in the underground drainage.

A dry phase, the most distinct in the whole analysed profile, was dated at 4800–4400 cal. BP. It is recorded as a series of massive peat, which are characterized by a high amount of OM and low content of CaCO₃ and an increase in the concentrations of lithophilous elements: Mg, Na and K. A simultaneous decrease in both δ¹⁸O and δ¹³C values, indicating climate cooling, can be observed at ca. 4.8 and 4.6 ka BP. A distinct decrease in water levels was detected at that time in the lakes of the Jura mountains, the northern French Pre-Alps and the Swiss Plateau (Magny, 2004), north-eastern Poland (Gałka and Apolinarska, 2014), northern Estonia (Punning et al., 2003), southern Sweden (Digerfeldt, 1998) and southern Finland (Väliranta et al., 2007). This period also follows the temperature decrease and a period of high frequency of fluvial extremes reported for Polish territory by Starkel et al. (2013; see Figure 9).

As is reported by Booth et al. (2005), several authors relate this phase to strong dryness in North America, which is postulated to have been the result of reduction of solar activity defined as the 4.2 event or/and Bond event 3 (Bond et al., 2001). In Northern and Eastern Europe, where continental climates prevailed, cooling and drying in the middle of the 4th millennium was considerably stronger, and occurred a little later than in Western Europe, where it is rather weakly recorded and its palaeoclimatic interpretation is often dubious (Hughes et al., 2000; Swindles et al., 2013). The cold event recorded in the KOM-9 core can probably be related to the distinct long-term cooling dated at 4.6–4.8 cal. BP.

A warm and humid phase, dated at 4400–2000 cal. BP, is evidenced in the KOM-9 core by peat–tufa rhythmite with the predominance of calcareous layers. Three visible maxima of calcareous tufa deposition are correlated with the maximum concentrations of heavy carbon isotope in CaCO₃. The occurrence of carbonates also indicates oxidizing depositional conditions and alkaline water in shallow paralimnic small pools favouring the accumulation of chemical denudation products and resulting in high concentrations of Ca, Na, Cu and Mn in deposits (Borówka and Tomkowiak, 2010; Davison, 1993; Wojciechowski, 2000). These conditions changed at least three times as indicated by an increased accumulation of OM and higher values of the Fe/Mn index evidencing a less oxidizing environment. The oxygen isotope curve fluctuates between −8.4‰ and −4.7‰, with lower δ¹⁸O values probably indicating cooler periods at ca. 3740, ca. 3240 and 2910–2750 cal. BP. Deterioration of climate conditions in a similar period was recorded on a larger regional scale. The low water level in lakes of north-eastern Poland is recorded at 3800–3500 cal. BP (Gałka and Apolinarska, 2014) and in lakes of mid-Europe at 3150–2750 cal. BP (Magny, 2004). The dry episode, dated at ca. 3150–2900 cal. BP, was also recorded in peatbogs of southern Finland (Väliranta et al., 2007). Slightly later (2900–2500 cal. BP) cold and dry phases were reported by Żurek et al. (2002) from mires in eastern Poland. It corresponds to the last of the mentioned cold/dry phases recorded in the KOM-9 core.

More visible traces of deforestation are shown in all published pollen diagrams from the studied region (Bałaga, 1998) by a decline of the Pinus, Ulmus, Quercus and Fraxinus curves and to a considerable extent also Carpinus, and by greater abundance of anthropogenic indicators coincided with the cultures of the Bronze Age. However, a small increase of the total content of K, Na and Mg (i.e. lithophile elements treated as geochemical indicators of erosion) in relation to the content of Ca in the KOM-9 core may suggest that human economic activity contributed only marginally to the changes of the groundwater table, which occurred in this period.

Late Holocene (2.0 ka cal. BP–present)

The last two millennia of the Subatlantic chronozone in the KOM-9 profile were characterized by relatively stable temperature conditions and relatively lower humidity. Calcium carbonate content in the spring-fed fen deposits was considerably lower because of a smaller supply of chemical denudation products. All proxies indicate gradually more frequent dry conditions. Distinctly cooler and dry climate oscillations are recorded between 1400 and 900 cal. BP (AD 500–1050) and between 450 and 200 cal. BP (AD 1500–1750). Similar changes are indicated by the ¹⁸O record in Lake Purwin (NE Poland). Significant increases in the δ¹⁸O and δ¹³C values occurred between ca. 400 and 200 cal. BP (Gałka and Apolinarska, 2014).

These two episodes, which are distinguished in the KOM-9 core, can be related to Bond event 1 (1.4 ka cal. BP) and Bond event 0 (LIA2), respectively. The latter one is the coldest and driest phase of the ‘Little Ice Age’ (LIA) (Bradley and Jones, 1993; De Vleeschouwer et al., 2009; Gałka and Apolinarska, 2014; Pauling et al., 2005; Starkel et al., 2013). The LIA is a consequence of changes in solar activity (Bond et al., 2001) and/or the transformation of the thermohaline circulation in the Atlantic Ocean (Broecker, 2000). It is also documented by the δ¹⁸O curve for NGRIP ice core (North Greenland Ice Core Project Members, 2004), as shown in Figure 9. In the KOM-9 core, these cool episodes were separated by a warmer phase (higher values of δ¹³C and δ¹⁸O), which corresponded to the ‘Medieval Warm Period’. At first – 675–900 cal. BP (AD 1040–1275) – this phase was slightly more humid (higher proportion of water snails, higher content of CaCO₃), and then – 560–675 cal. BP (AD 1275–1390) – it was considerably drier. This climatic deterioration might have been reinforced by human impacts and forest opening. The strongest increase in human impacts on the region was noted with the beginning of the early Middle Ages (Figure 2) and afterwards to contemporary times.

The successive intensification of the economic activities of man in the whole region resulted in high frequencies of NAP, anthropogenic indicators and pine, and in a considerably lower proportion of deciduous trees including elm, oak, ash and hornbeam. In the KOM-9 profile, this phase is distinctly shown by increased values of iron, potassium and zinc, as well as a decrease of OM, which confirm our assumptions about more intensive soil erosion.

Conclusion

Based on the multi-proxy data from the spring-fed fen in the Komarów site, we distinguished four main stages of fen evolution that reflected a distinct variability of water supply conditions and were connected with regional changes of humidity–temperature conditions in the Holocene: I – crenous – 9740–6440 cal. BP; II – helocrenous – 6440–4570 cal. BP; III – limnocrenous – 4570–2000 cal. BP; and IV – helocrenous – 2000 cal. BP–the present. We suggest that climate was the most important determinant of development of the fen in almost all stages. The influence of local conditions, including human impacts, which led to changes in the supply dynamics of the mire, should be considered less important, only modifying the results of regional trends in climate changes. Only in the last century could the anthropogenic factors have been more important in creating supply conditions of the mire. In general, we may assume that climate-induced hydrological changes determined the autogenic development of the succession in the studied spring-fed fen during the Holocene. The cycle of changes of humidity–temperature conditions is recorded in (a) peat–tufa depositional sequence, (b) distribution of δ¹³C and δ¹⁸O values, (c) contents of OM and CaCO₃, (d) succession of molluscan assemblages and (e) indicator macrofossils. Based on the record of multi-proxy data, 12 Holocene cold events are distinguished, which correspond well to the results obtained for other sites in Poland and Europe. Some distinct changes of individual proxies, dated at ca. 9.4, 8.2, 5.9, 4.6, 2.8, 1.4 and 0.55 ka cal. BP, clearly correspond to quasi-periodical global climate changes in the Holocene with a rhythm of ca. 1470 ± 500 years (Bond et al., 1997, 2001; Wanner et al., 2011). Other changes that are less clear (ca. 6.7, 5.2, 3.7, 3.4, 3.0 ka cal. BP) could be considered as events of local or at most regional importance, resulting from the restructuring of water supply conditions in the area with special geological structure. The majority of the cold events recorded in δ¹³C and δ¹⁸O of carbonates can be correlated to the Greenland oxygen isotope curve. Nevertheless, the uncertainties in the chronology of the investigated core together with decadal resolution of samples have some potential consequences for the climatic interpretation. In particular, for the correlation of peaks in isotope curves from Komarów site the NGRIP 80-year averaged dataset was chosen (Figure 9) to compare records of similar resolution in time. Although there is a possibility of so-called ‘suck-in and smear effects’ (sensu Baillie, 1991) in particular for older part, the correlated events are distinct excursions of isotope values in Komarów core, happening at the same time, within age model error band, as in NGRIP curve.

The chronology for Komarów site was not tuned in any way to other sites, therefore it is fully independent and any drawbacks of tuning can be excluded, as recommended by Blaauw (2012). This implies that some conclusions about leads and lags may be cautiously made. The time differences between relevant minima of isotope records from Komarów and NGRIP (Figure 9) are systematic, at least for events older than 3.0 ka cal. BP, and suggest a delay in continental response to global changes. Although it cannot be proved statistically because of low resolution and chronological issues, the delay can be estimated to last for tens to hundred years.

Footnotes

Acknowledgements

We thank the editor and anonymous reviewers for their constructive comments, which helped us to improve the manuscript. We are also grateful to Jacek Łojek and Michał Dobrowolski for their help in field works.

Funding

This work was carried out as a part of the project No N N 306 279 035 financed by the Polish Ministry of Science and Higher Education.

References

Alexandrowicz

(1987) Analiza malakologiczna w badaniach osadów czwartorzędowych. Geologia, Kwartalnik AGH 12(1–2): 3–240.

Alexandrowicz

(1999a) Bithynia tentaculata (Linnaeus, 1758) as an indicator of age and deposition environment of Quaternary deposits. Folia Malacologica 7(2): 79–88.

Alexandrowicz

(2011) Analiza malakologiczna. Metody badań i interpretacji (Rozprawy Wydziału Przyrodniczego, vol. 3). Krakow: Polska Akademia Umiejętności (PAU), 302 pp.

Alexandrowicz

Dobrowolski

Bałaga

(1994) Etapy rozwoju torfowiska kopułowego Krzywice w okolicach Chełma Lubelskiego [Stages of development of cupola mire in Krzywice near Chełm Lubelski]. Geologia, Kwartalnik AGH 20(3): 259–273.

Alexandrowicz

(1999b) Evolution of the malacological assemblages in North Poland during the Late Glacial and Early Holocene. Folia Quaternaria 70: 39–69.

Alexandrowicz

(2004) Molluscan assemblages of Late Glacial and Holocene calcareous tufa in Southern Poland. Folia Quaternaria 75: 3–309.

Alexandrowicz

(2007) Malakofauna późnoglacjalnych i holoceńskich węglanowych osadów jeziornych Północnej Polski. Geologia, Kwartalnik AGH 33: 395–420.

Alexandrowicz

(2013) Late Glacial and Holocene molluscan assemblages in deposits filling palaeolakes in Northern Poland. Studia Quaternaria 30(1): 5–17.

Baillie

MGL

(1991) Suck-in and smear: Two related chronological problems for the 90s. Journal of Theoretical Archaeology 2: 12–16.

10.

Bałaga

(1998) Post-glacial vegetational changes in the Middle Roztocze (E Poland). Acta Palaeobotanica 38(1): 175–192.

11.

Bałaga

(2004) Changes of vegetation in Lake Perespilno environs (Lublin Polesie) in the Late Glacial and Holocene. Acta Palaeobotanica 44(2): 147–166.

12.

Barber

Dyke

Hillaire-Marcel

. (1999) Forcing of the cold event of 8,200 years ago by catastrophic drainage of Laurentide lakes. Nature 400: 344–348.

13.

Bednarek

Dziadowiec

Pokojska

. (2004) Badania ekologiczno-gleboznawcze. Warszawa: Państwowe Wydawnictwo Naukowe (PWN), 344 pp.

14.

Berglund

Ralska-Jasiewiczowa

(1986) Pollen analysis and pollen diagrams. In: Berglund

(ed.) Handbook of Holocene Palaeoecology and Palaeohydrology. Chichester: John Wiley & Sons, pp. 455–484.

15.

Bindler

(2006) Mired in the past – Looking to the future: Geochemistry of peat and the analysis of past environmental changes. Global and Planetary Change 53: 209–221.

16.

Birks

HJB

(2006) Multi-proxy studies in palaeolimnology. Vegetation History and Archaeobotany 15: 235–251.

17.

Blaauw

(2012) Out of tune: The dangers of aligning proxy archives. Quaternary Science Reviews 36: 38–49.

18.

Bond

Kromer

Beer

. (2001) Persistent solar influence on North Atlantic climate during the Holocene. Science 294: 2130–2136.

19.

Bond

Showers

Cheseby

. (1997) A pervasive millennial-scale cycle in North Atlantic. Science 278: 1257–1266.

20.

Booth

Jackson

Forman

. (2005) A severe centennial-scale drought in midcontinental North America 4200 years ago and apparent global linkages. The Holocene 15(3): 321–328.

21.

Borówka

Tomkowiak

(2010) Skład chemiczny osadów z profilu torfowiska Żabieniec [Chemical composition of sediments profile from the Żabieniec peat bog]. In: Twardy

Żurek

Forysiak

(eds) Torfowisko Żabieniec: warunki naturalne, rozwój i zapis zmian paleoekologicznych w jego osadach. Poznań: Bogucki Scientific Publishers, pp. 163–172.

22.

Boyer

MLH

Wheeler

(1989) Vegetation patterns in spring-fed calcareous fens: Calcite precipitation and constraints on fertility. Journal of Ecology 77: 597–609.

23.

Bradley

Jones

(1993) ‘Little Ice Age’ summer temperature variations: Their nature and relevance to recent global warming trends. The Holocene 3: 367–376.

24.

Broecker

(2000) Was a change in thermohaline circulation responsible for the Little Ice Age? Proceedings of the National Academy of Sciences of the United States of America 97: 1339–1342.

25.

Bronk Ramsey

(2008) Deposition models for chronological records. Quaternary Science Reviews 27(1–2): 42–60.

26.

Bronk Ramsey

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

27.

Charman

(2002) Peatlands and Environmental Change. Chichester: John Wiley & Sons, 312 pp.

28.

Craig

(1963) The isotopic geochemistry of water and carbon in geothermal areas. In: Tongiorgi

(ed.) Nuclear Geology in Geothermal Areas, Spoleto. Pisa: Laboratorio di Geologia Nucleare, pp. 17–53.

29.

Cremaschi

(1998) Late Quaternary geological evidence for environmental changes in south-western Fezzan (Libyan Sahara). In: Cremaschi

Di Lernia

(eds) Wadi Teshuinat: Palaeoenvironment and Prehistory in South-Western Fezzan (Libyan Sahara). Firenze: Ed. All’Insegna del Giglio, pp. 13–47.

30.

Dahl

Nesje

Lie

. (2002) Timing, equilibrium-line altitudes and climatic implications of two early-Holocene glacier readvances during the Erdalen Event at Jostedalsbreen, western Norway. The Holocene 12: 17–25.

31.

Dansgaard

(1964) Stable isotopes in precipitation. Tellus 16: 436–468.

32.

Davison

(1993) Iron and manganese in lakes. Earth-Science Reviews 34: 119–163.

33.

De Vleeschouwer

Piotrowska

Sikorski

. (2009) Multiproxy evidence of ‘Little Ice Age’ palaeoenvironmental changes in a peat bog from northern Poland. The Holocene 19(4): 625–637.

34.

Demény

Kele

Siklòsy

(2010) Empirical equations for the temperature dependence of calcite-water oxygen isotope fractionation from 10 to 70°C. Rapid Communications in Mass Spectrometry 24(24): 3521–3526.

35.

Digerfeldt

(1988) Reconstruction and regional correlation of Holocene lake-level fluctuations in Lake Bysjön, South Sweden. Boreas 17: 165–182.

36.

Digerfeldt

(1998) Reconstruction of Holocene lake-level changes in southern Sweden: Technique and results. Palaöklimaforschung 25: 87–98.

37.

Dobrowolski

(1994) Tektoniczne uwarunkowania rozwoju źródliskowego torfowiska kopułowego Krzywice koło Chełma [Tectonic conditions of Krzywice cupola spring mire near Chełm]. Przegląd Geologiczny 7: 532–535.

38.

Dobrowolski

(2011) Problemy klasyfikacyjne osadów torfowisk źródliskowych [Problems with classification of deposits of spring-fed fens]. Studia Limnologica et Telmatologica 5(1): 3–12.

39.

Dobrowolski

Durakiewicz

Pazdur

(2002) Calcareous tufas in the soligenous mires of eastern Poland as an indicator of the Holocene climatic changes. Acta Geologica Polonica 52(1): 63–73.

40.

Dobrowolski

Mazurek

Osadowski

(2010) Geological, hydrological and phytosociological conditions of spring mires development in the Parsęta River catchment (Western Pomerania, Poland). Geologija 52(1–2): 37–44.

41.

Dobrowolski

Bałaga

Bogucki

. (2001) Chronostratigraphy of the Okunin and Czerepacha lake-mire geosystems (Volhynia Polesiye, NW Ukraine) during the Late Glacial and Holocene. Geochronometria 20: 107–115.

42.

Dobrowolski

Durakiewicz

Pazdur

. (1996) Chronostratygrafia i fazy rozwoju źródliskowego torfowiska kopułowego Krzywice koło Chełma, Wyżyna Lubelska [Chronostratigraphy and development stages of the spring peat – Knoll at Krzywice near Chełm, Lublin Upland, SE Poland]. Geochronometria 14: 19–29.

43.

Dobrowolski

Hajdas

Melke

. (2005) Chronostratigraphy of calcareous mire sediments at Zawadówka (eastern Poland) and their use in palaeogeographical reconstruction. Geochronometria 24: 69–79.

44.

Dobrowolski

Kulesza

Łojek

. (2015) Origin and evolution of the Bezedna lake-mire complex in the Lublin area (East Poland): A case study for permafrost lakes in karstic regions. Journal of Paleolimnology 53(2): 191–213.

45.

Dobrowolski

Pidek

Alexandrowicz

. (2012) Interdisciplinary studies of spring mire deposits from Radzików (South Podlasie Lowland, East Poland) and their significance for palaeoenvironmental reconstructions. Geochronometria 39(1): 10–29.

46.

Dolecki

(2002) Main Profiles of the Neopleistocene Loesses on the Horodło Plateau-Ridge and Their Lithological-Stratigraphical Interpretation. Lublin: Maria Curie-Skłodowska Press, 263 pp.

47.

Fiałkiewicz-Kozieł

Kołaczek

Piotrowska

. (2014a) High resolution age-depth model of a peat bog in Poland as an important basis for paleoenvironmental studies. Radiocarbon 56(1): 109–125.

48.

Fiałkiewicz-Kozieł

Śmieja-Król

Piotrowska

. (2014b) Carbon accumulation rates in two poor fens with different water regimes: Influence of anthropogenic impact and environmental change. The Holocene 24(11): 1539–1549.

49.

Gałka

Apolinarska

(2014) Climate change, vegetation development, and lake level fluctuations in Lake Purwin (NE Poland) during the last 8600 cal. yr BP based on a high-resolution plant macrofossil record and stable isotope data (δ¹³C and δ¹⁸O). Quaternary International 328–329: 213–225.

50.

Grootjans

Šefferová Stanová

Jansen

(2012) Calcareous Mires of Slovakia – Landscape Setting, Management and Restoration Prospects. Zeist: KNNV Publishing, 110 pp.

51.

Grootjans

Adema

Bleuten

. (2006) Hydrological landscape settings of base-rich fen mires and fen meadows: An overview. Applied Vegetation Science 9: 175–184.

52.

Grosse-Brauckmann

(1972) Über pflanzliche Macrofossilien mitteleuropäischer Torfe I. Gewebereste krautiger Pflanzen und ihre Merkmale. Telma 2: 19–55.

53.

Grosse-Brauckmann

(1974) Über pflanzliche Macrofossilien mitteleuropäischer Torfe II. Weitere Reste (Früchte und Samen, Moose u.a.) und ihre Bestimmungsmöglichkeiten. Telma 4: 51–117.

54.

Grosse-Brauckmann

Streitz

(1992) Pflanzliche Macrofossilien mitteleuropäischer Torfe III. Fruchte, Samen und einige Gewebe (Fotos von fossilen Pflanzenesten). Telma 22: 53–102.

55.

Hájek

Hekera

Hajkova

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

56.

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(3): 233–243.

57.

Heikkilä

Seppä

(2003) A 11 000 yr palaeotemperature reconstruction from the southern boreal zone in Finland. Quaternary Science Reviews 22: 541–554.

58.

Heikkilä

Edwards

TWD

Seppä

. (2010) Sediment isotope tracers from Lake Saarikko, Finland, and implications for Holocene hydroclimatology. Quaternary Science Reviews 29: 2146–2160.

59.

Holmquist

Finkelstein

Garneau

. (2016) A comparison of radiocarbon ages derived from bulk peat and selected plant macrofossils in basal peat cores from circum-arctic peatlands. Quaternary Geochronology 31: 53–61.

60.

Hughes

PDM

Mauquoy

Barber

. (2000) Mire-development pathways and palaeoclimatic records from a full Holocene peat archive at Walton Moss, Cumbria, England. The Holocene 10(4): 465–479.

61.

Janiec

(1997) Transformacje i translokacje jonowe w wodach Roztocza Zachodniego [Ionic Transformations and Translocations in Natural Waters of Western Roztocze]. Lublin: Maria Curie-Skłodowska Press, 214 pp.

62.

Jonsson

Andersson

Rosqvist

. (2010) Reconstructing past atmospheric circulation changes using oxygen isotopes in lake sediments from Sweden. Climate of the past 6: 49–62.

63.

Kokfelt

Reiss

Struyf

. (2010) Wetland development, permafrost history and nutrient cycling inferred from late Holocene peat and lake sediment records in subarctic Sweden. Journal of Paleolimnology 44: 327–342.

64.

Kotarba

(2006) The Little Ice Age in the High Tatra mountains. Studia Quaternaria 23: 47–53.

65.

Kovanda

(1971) Quaternary limestones of Czechoslovakia. Antropozoikum A7: 7–236.

66.

Kulesza

Suchora

Pidek

. (2012) The Holocene palaeoenvironmental changes reflected in the multi-proxy studies of Lake Słone sediments (SE Poland). Palaeogeography, Palaeoclimatology, Palaeoecology 363–364: 79–98.

67.

Lamentowicz

Obremska

Mitchell

EAD

(2008) Autogenic succession, land-use change, and climatic influences on the Holocene development of a kettle hole mire in Northern Poland. Review of Palaeobotany and Palynology 151: 21–40.

68.

Lamentowicz

Balwierz

Forysiak

. (2009) Multiproxy study of anthropogenic and climatic changes in the last two millennia from a small mire in central Poland. Hydrobiologia 631: 213–230.

69.

Leng

Marshall

(2004) Paleoclimate interpretation of stable isotope data from lake sediment archives. Quaternary Science Reviews 23: 811–831.

70.

Ložek

(1964) Quartärmollusken der Tschechoslowakei. Rozpravy Ustředniho Ustavu Geologického 31: 3–374.

71.

Magny

(2004) Holocene climate variability as reflected by mid-European lake-level fluctuations, and its probable impact on prehistoric human settlements. Quaternary International 113: 65–80.

72.

Magny

de Beaulieu

Drescher-Schneider

. (2007) Holocene climate changes in the central Mediterranean as recorded by lake-level fluctuations at Lake Accesa (Tuscany, Italy). Quaternary Science Reviews 26: 1736–1758.

73.

Magny

Galop

Bellintani

. (2009) Late-Holocene climatic variability south of the Alps as recorded by lake-level fluctuations at Lake Ledro, Trentino, Italy. The Holocene 19: 575–589.

74.

Margielewski

(2006) Records of the Late Glacial-Holocene palaeo-environmental changes in landslide forms and deposits of the Beskid Makowski and Beskid Wyspowy Mts. Area (Polish Outer Carpathians). Folia Quaternaria 76: 1–149.

75.

Margielewski

Kołaczek

Michczyński

. (2011) Record of the meso- and neoholocene palaeoenvironmental changes in the Jesionowa landslide peat bog (Beskid Sądecki MTS. Polish Outer Carpathians). Geochronometria 38(2): 138–154.

76.

Mazurek

Dobrowolski

Osadowski

(2014) Geochemistry of deposits from spring-fed fens in West Pomerania (Poland) and its significance for palaeoenvironmental reconstruction. Geomorphologie: Relief, Processus, Environnement 4: 323–342.

77.

Michalczyk

(2001) Źródła Wyżyny Lubelskiej i Roztocza [Spring of the Lublin Upland an the Roztocze Region]. Lublin: Maria Curie-Skłodowska Press, 298 pp.

78.

Mirek

Piękoś-Mirek

Zając

. (2002) Vascular Plants of Poland: A Checklist (Polish botanical studies: Guidebook series, 15). Kraków: W. Szafer Institute of Botany, Polish Academy of Sciences, 303 pp.

79.

Muscheler

Beer

Vonmoos

(2004) Causes and timing of the 8200 yr BP event inferred from the comparison of the GRIP 10Be and the tree ring Δ¹⁴C record. Quaternary Science Reviews 23: 2101–2111.

80.

Nesje

Dahl

(1991) Late-Holocene glacier fluctuations Bevringsdalen, Jostedalsbreen region, western Norway (c. 3200–1400 BP). The Holocene 1: 1–7.

81.

North Greenland Ice Core Project Members (2004) High-resolution record of Northern Hemisphere climate extending into the last interglacial period. Nature 431(7005): 147–151.

82.

Ochyra

Żarnowiec

Bednarek-Ochyra

(2003) Census Catalogue of Polish Mosses. Kraków: W. Szafer Institute of Botany, Polish Academy of Sciences.

83.

O’Neil

Clayton

Mayeda

(1969) Oxygen isotope fractionation in divalent metal carbonates. Journal of Chemical Physics 51: 5547–5558.

84.

Osadowski

Mazurek

Dobrowolski

(2009) Structure and development conditions of spring mires in the Parsęta basin (Western Pomerania). In: Łachacz

(ed.) Wetlands – Their Functions and Protection. Olsztyn: Department of Land Reclamation and Environmental Management, University of Warmia and Mazury in Olsztyn, pp. 107–124.

85.

Parker

Goudie

Stokes

. (2006) A record of Holocene climate change from lake geochemical analyses in southeastern Arabia. Quaternary Research 66(3): 465–476.

86.

Pauling

Luterbacher

Casty

. (2005) Five hundred years of gridded high-resolution precipitation reconstructions over Europe and the connection to large-scale circulation. Climate Dynamics 26(4): 387–405.

87.

Pazdur

Szulc

(1988) Radiocarbon dating of Holocene calcareous tufa from south Poland. Radiocarbon 30(2): 133–146.

88.

Pazdur

Dobrowolski

Durakiewicz

. (2002) δ¹³C and δ¹⁸O time record and palaeoclimatic implications of the Holocene calcareous tufa from South-Eastern Poland and Eastern India (Orissa). Geochronometria 21: 97–108.

89.

Pazdur

Fogtman

Michczynski

. (2003) Precision of ¹⁴C dating in Gliwice Radiocarbon Laboratory: FIRI programme. Geochronometria 22: 27–40.

90.

Pidek

Noryśkiewicz

Dobrowolski

. (2012) Indicative value of pollen analysis for deposits of spring-fed fens. Ekológia 31(4): 430–458.

91.

Piechocki

(1979) Mięczaki (Mollusca). Ślimaki (Gastropoda) (Fauna słodkowodna Polski, 7). Warszawa: Państwowe Wydawnictwo Naukowe (PWN), pp. 7–187.

92.

Piechocki

Dyduch-Falniowska

(1993) Mięczaki (Mollusca). Małże (Bivalvia) (Fauna słodkowodna Polski, 7a). Warszawa: Państwowe Wydawnictwo Naukowe (PWN), pp. 7–204.

93.

Punning

J-M

Kangur

Koff

. (2003) Holocene lake-level changes and their reflection in the paleolimnological records of two lakes in northern Estonia. Journal of Paleolimnology 29: 167–178.

94.

Ralska-Jasiewiczowa

Goslar

Madeyska

. (eds) (1998) Lake Gościąż, Central Poland: A Monographic Study. Kraków: W. Szafer Institute of Botany, Polish Academy of Sciences, 340 pp.

95.

Reimer

Baillie

MGL

Bard

. (2009) IntCal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon 51: 1111–1150.

96.

Renssen

Goose

Fichefet

(2007) Simulation of Holocene cooling events in a coupled climate model. Quaternary Science Reviews 26: 2019–2029.

97.

Roberts

(1998) The Holocene: An Environmental History. 2nd Edition. London: Blackwell Publishers Ltd, 316 pp.

98.

Różański

Klisch

Wachniew

. (2010) Oxygen-isotope geothermometers in lacustrine sediments: New insights through combined δ¹⁸O analyses of aquatic cellulose, authigenic calcite and biogenic silica. Geochimica et Cosmochimica Acta 74: 2957–2969.

99.

Sadowski

(2008) Sprawozdanie z badań powierzchniowych na obszarze AZP 90-90 [The report on the surface area study (Archaeological Surface Surveys 90-90)]. Manuscript, The Archives of the Provincial Heritage Monuments Protection Office in Lublin, Zamość Regional Office, Zamość.

100.

Sarmaja-Korjonen

Seppä

(2007) Abrupt and consistent responses of aquatic and terrestrial ecosystems to the 8200 cal. yr cold event: A lacustrine record from Lake Arapisto, Finland. The Holocene 17: 457–467.

101.

Schoning

Charman

Wastegoard

(2005) Reconstructed water tables from two ombrotrophic mires in eastern central Sweden compared with instrumental meteorological data. The Holocene 15: 111–118.

102.

Seppä

Hammarlund

Antonsson

(2005) Low-frequency and high-frequency changes in temperature and effective humidity during the Holocene in south-central Sweden: Implications for atmospheric and oceanic forcings of climate. Climate Dynamics 25: 285–297.

103.

Shotyk

(1996) Natural and anthropogenic enrichments of As, Cu, Pb, Sb, and Zn in ombrotrophic versus minerotrophic peat bog profiles, Jura Mountains, Switzerland. Water, Air, and Soil Pollution 90: 375–405.

104.

Starkel

(1999) The role of the Holocene in the environment’s evolution and its stratigraphy. In: Starkel

(ed.) Geography of Poland: Natural Environment. Warsaw: Panstwowe Wydawnictwo Naukowe (PWN), pp. 103–105.

105.

Starkel

Michczyńska

Krąpiec

. (2013) Progress in Holocene chrono-climatostratigraphy of Polish territory. Geochronometria 40(1): 1–21.

106.

Succow

(1988) Landschaftsökologische Moorkunde. Berlin; Stuttgart: Gebrüder Borntraeger, 340 pp.

107.

Succow

Jeschke

(1986) Moore in der Landschaft. Leipzig; Jena; Berlin: Urania Verl, 268 pp.

108.

Swindles

Lawson

Mattews

. (2013) Centennial-scale climate change in Ireland during the Holocene. Earth-Science Reviews 126: 300–320.

109.

Szeroczyńska

Zawisza

(2011) Records of the 8200 cal BP cold event reflected in the composition of subfossil Cladocera in the sediments of three lakes in Poland. Quaternary International 233: 185–193.

110.

Tobolski

(2000) Przewodnik do oznaczania torfów i osadów jeziornych [Handbook for the Determination of Peat and Lake Deposits] (Vademecum Geobotanicum). Warszawa: Państwowe Wydawnictwo Naukowe (PWN), 508 pp.

111.

Troels-Smith

(1955) Characterization of unconsolidated sediments. Danmarks Geologiske Undersøgelse 4(3–10): 1–73.

112.

Tudyka

Pazdur

(2012) ¹⁴C dating with the ICELS liquid scintillation counting system using fixed-energy balance counting window method. Radiocarbon 54(2): 267–273.

113.

Väliranta

Korhola

Seppä

. (2007) High-resolution reconstruction of wetness dynamics in a southern boreal raised bog, Finland, during the late Holocene: A quantitative approach. The Holocene 17: 1093–1107.

114.

Vinther

Clausen

Johnsen

. (2006) A synchronized dating of three Greenland ice cores throughout the Holocene. Journal of Geophysical Research 111: D13102.

115.

Wanner

Solomina

Grosjean

. (2011) Structure and origin of Holocene cold events. Quaternary Science Reviews 30: 3109–3123.

116.

Wenninger

Alram-Stern

Bauer

. (2006) Climate forcing due to the 8200 cal yr BP event observed at Early Neolithic sites in the eastern Mediterranean. Quaternary Research 66: 401–420.

117.

Wojciechowski

(2000) Zmiany paleohydrologiczne w środkowej Wielkopolsce w ciągu ostatnich 12000 lat w świetle badań osadów jeziornych rynny kórnicko-zaniemyskiej [Palaeohydrological Changes in the Central Wielkopolska Lowland during the Last 12 000 Years on the Basis of Deposits of the Kórnik-Zaniemyśl Lakes]. Poznań: Adam Mickiewicz University Press, 236 pp.

118.

Zarzycki

Trzcińska-Tacik

Różański

. (2002) Ecological Indicator Values of Vascular Plants of Poland. Kraków: W. Szafer Institute of Botany, Polish Academy of Sciences, 183 pp.

119.

Żurek

(1990) Związek procesu zatorfienia z elementami środowiska przyrodniczego wschodniej Polski [Relationship between peat formation process and natural environment elements in East Poland]. Roczniki Nauk Rolniczych, Seria D: Monografie 220: 1–174.

120.

Żurek

Michczyńska

Pazdur

(2002) Time record of palaeohydrologic changes in the development of mires during the Late Glacial and Holocene, North Podlasie Lowland and Holy Cross Mts. Geochronometria 21: 109–118.