Late Glacial and early Holocene development of an oxbow lake in Central Europe (Poland) based on plant macrofossil and geochemical data

Abstract

Sediments from an oxbow lake located in the Prosna River valley (Poland) were analysed to investigate the developmental history of the wetland ecosystem and any response to abrupt climatic changes. High resolution plant macrofossil analysis and radiocarbon dating were undertaken on two cores, with lower resolution geochemical analysis conducted on one of these cores. We provide evidence of a palaeolake with a late Glacial origin (older than 12,500 years). Abundant fossil presence of macrophytes (e.g. multiple Potamogeton species) in the studied palaeomeander may indicate that the north–south orientation of the Prosna valley made it an important route for the spreading of aquatic plants during the late Glacial. Chara sp., Batrachium sp. and Potamogeton spp. were the pioneer plants that colonised cold water with a high Ca²⁺ content. Early Holocene warming trigged a decrease in water level at the oxbow lake and facilitated the expansion of thermophilous water plants, for example, Ceratophyllum demersum, Typha sp. and Lemna trisulca, which usually occur in shallow water. A decreasing water level resulted in the gradual isolation of the study site from the influence of groundwater, leading to acidification of the habitat and the development of a Sphagnum population, with S. contortum and S. teres as dominant species. The presence of S. contortum (the oldest occurrence in the European lowlands) and S. teres during the early Holocene may indicate that river valleys and the peatlands that developed in this region acted as an important habitats (and possibly refugia) for some minerotrophic Sphagnum species.

Keywords

climate change peatland plant macrofossils plant succession Prosna River refugia Sphagnum succession

Introduction

Rapid climate warming during the transition from the last Glacial period to the Holocene resulted in a shift in biogeographic zones and the creation of meridional migration gateways for plants. River valleys are important habitats for many wetland plant species and provide suitable pathways for their expansion (cf. Naiman and Décamps, 1997), particularly since the deglaciation of northern Europe (Kołaczek et al., 2018; Kolstrup, 2007; Szafer, 1946). The late Glacial and early Holocene (ca. 15,000–8500 years ago) were characterised by major climate fluctuations and a transition from a cold to warm climate (Rasmussen et al., 2014). This climatic warming influenced the fluvial processes occurring in European lowland river valleys (Starkel et al., 2006). Climate fluctuations in the late Glacial have been directly linked to changes in river bed morphology in Polish lowlands and changes in flora associated with the formation of meanders, oxbow lakes and peatlands (Pawłowski et al., 2016a, 2016b; Słowik et al., 2019; Starkel et al., 2006). Sediments deposited in river valleys produced a stratigraphic record and are surrounded by former oxbow lakes and peatlands. Detailed palaeoecological studies of the deposits accumulated in the river valleys provide useful information about past environmental changes (Kołaczek et al., 2018; Magyari et al., 2008; Pawłowski et al., 2015a, 2016a, 2016b). Moreover, reconstructing the response of riparian and wetland biota to previous rapid climate changes (e.g. late Glacial and early Holocene transition) can inform the prediction of future vegetation shifts in response to changing temperature, precipitation and flood frequency (Dwire et al., 2018; Garssen et al., 2014; Thodsen et al., 2016).

Previous studies of river valleys in the central part of the Polish lowlands have focused mainly on their evolution, palaeohydrology and the geochemical composition of organic deposits (Borówka et al., 2015; Kittel et al., 2016; Michno, 2004, 2005; Słowik, 2011, 2013; Starkel et al., 2006). Recent palaeoecological studies of late Glacial and Holocene sediments using fossil biotic data to investigate the influence of climate change (e.g. temperature) have been carried out in the Ner river (Kittel et al., 2016; Płóciennik et al., 2011), Grabia river (Pawłowski et al., 2015b, 2016a, 2016b) and Bug river (Kołaczek et al., 2018). However, detailed knowledge regarding local plant succession, especially macrophytes, and ecosystem adaptation to abrupt climate changes in river ecosystems is lacking.

The Prosna valley has primarily been studied in terms of its geology and geomorphology (Dąbrowski, 1991; Dyjor, 1985; Młynarczyk and Rotnicki, 1989), alongside archaeological research aimed at understanding the relationship between settlement distribution and landform development (Stupnicka et al., 2006).

Here, we present the first detailed plant macrofossil and geochemical records dated by radiocarbon analysis in the Prosna River valley. The Prosna valley is ideal for palaeoecological reconstructions because of unique and well-preserved palaeomeanders filled with organic sediment. The extent and scale of the resulting landforms appear to be linked with the establishment of fortified settlements, for example, near the village of Grodzisko (Kaczmarek and Szczurek, 2015).

Focussing on the long-term dynamics of an oxbow ecosystem, we aim to (a) date the beginning of organic sediment deposition, (b) reconstruct the initial stage of development of local vegetation in the oxbow lake, (c) determine the impact of climatic changes on local plant succession, (d) detect the presence of possible refugia for some minerotrophic Sphagnum species and (e) assess the variability of physicochemical parameters in the oxbow lake and investigate any potential relationship with plant succession.

Study site

The study site is located in Central Poland, near the village of Grodzisko, ~750 m southwest of the current Prosna River (Figure 1). In the Prosna River valley on the outskirts of the village, there are remains of a settlement called ‘Szwedzkie szańce’. The fortified settlement is located in a marshy area of a former river valley, which enhances its natural defensive position (Śmigielski and Szczurek, 2013). Establishment of the settlement is associated with the defensive construction movement at the beginnings of the Iron Age in Wielkopolska (Kaczmarek and Szczurek, 2015).

Figure 1.

Study site: (a) regional setting with view of the glacial extent over much of Europe during the last ice age (source https://commons.wikimedia.org/wiki/File: Weichsel-W%C3%BCrm-Glaciation.png); (b) LIDAR digital elevation model showing the sampling site, with the core locations marked by grey stars in the vicinity of the Grodzisko stronghold.

The first phase of development for the Prosna fossil valley took place towards the end of the Pliocene and has evolved during subsequent glacial and interglacial cycles through to the Pleistocene and Holocene, creating the present day sediment record (Piszczygłowa, 2014). The youngest sediments filling the valley are comprised of glacial clay, the sand-gravel series and organic deposits including gyttja and peat (Młynarczyk and Rotnicki, 1989; Piszczygłowa, 2014). The climate is temperate and affected by both maritime and continental influences. The average annual temperature is around 8.5°C, with a small number of frost days across the year. The winter period is the most variable in terms of temperature, ranging between –8.8°C and 2.2°C. The growing season lasts 210 days. The Prosna valley, especially in the southern and eastern parts, has one of the lowest annual rainfalls in Poland at 500–550 mm (Lorenc, 2005; Woś, 1999).

The natural vegetation of the Prosna valley has been strongly modified by humans, with the area of palaeomeander converted to an agricultural meadow. The main tree species in the valley are Alnus glutinosa and Salix spp., while in the drier places, Pinus sylvestris is the dominant species. Among vascular plants, Carex spp., Juncus spp. and Ranunculus spp. are the dominant species.

Materials and methods

Cores were taken using a Russian peat corer, 5 cm in diameter and 50 cm in length. The cores GRI (186 cm long) and GRII (240 cm) were placed in PVC tubes following extraction, individually wrapped and stored at 4°C until subsampling. In the laboratory, the sediment was unpacked, cleaned and sliced into 1-cm slices using a surgical scalpel.

To determine the start of organic sediment accumulation in the study area, macrofossils of terrestrial plants from four samples were selected for AMS radiocarbon dating. Radiocarbon dating was carried out in the Poznan Radiocarbon Laboratory. The resulting conventional radiocarbon dates were calibrated using OxCal 4.1 software (Bronk-Ramsey, 2009).

Plant macrofossils were analysed at 1-cm intervals contiguously in the two cores for the time period of interest (late Glacial/early Holocene). In the GRI Core, 100 samples were analysed for plant macrofossil remains from 85 to 186 cm, and in the GRII, 105 samples were analysed from interval 135 to 240 cm. The total volume of material analysed for plant macrofossils per sample was ca. 8 cm³. The samples were rinsed under a warm-water spray over 0.20 mesh screens. The vascular plant composition was determined on the basis of carpological remains and vegetative fragments (leaves, rootlets, epidermis) using the available identification keys (Tobolski, 2000; Velichkevich and Zastawniak, 2006, 2008). Mosses were identified using the keys prepared by Hölzer (2010). The reference collection of plant macrofossils and recent plant material gathered by Mariusz Gałka was also used. The macrofossils were identified using a Nikon SMA 800 stereoscopic microscope under 10–200 A and an Olympus CX 41 biological light microscope. Fossil fruits and seeds are expressed in absolute numbers, and the contribution of mosses (e.g. Sphagnum contortum) and vascular plant rootlets are expressed in percentage values of the total volume of a sieved sediment sample. A total of 205 samples were analysed. The names of the plant species were adopted following Mirek et al. (2002).

In order to determine the concentration of 59 chemical elements, a 5110 ICP-OES (Agilent, USA) inductively coupled plasma with optical emission spectrometry was used. For multi-elemental determination, the common conditions were as follows: radio frequency (RF) power = 1.2 kW, nebulizer gas flow = 0.7 L min⁻¹, auxiliary gas flow = 1.0 L min⁻¹, plasma gas flow = 12.0 L min⁻¹, viewing height for radial plasma observation = 8 mm, detector CCD (charge coupled device) temperature = −40°C and signal acquisition time = 5 s for three replicates. The detection limits were 0.01 mg kg⁻¹ dry weight for all elements (at three-sigma criteria). The uncertainty for the total analytical procedure (including sample preparation) was 20%. The traceability was checked using reference materials: CRM S-1 – loess soil; CRM NCSDC (73349) – bush branches and leaves; CRM 2709 – soil; CRM 405 – estuarine sediments and CRM 667 – estuarine sediments. The recovery (80–120%) was acceptable for most elements. For uncertified elements, recovery with the standard addition method was defined.

Plant macrofossils and geochemical data are presented in diagrams and were plotted using the C2 software (Juggins, 2007). Stratigraphic clustering and ordination analysis was conducted in R version 3.4.3 (R Core Team, 2018). Stratigraphic zones in the plant macrofossil and geochemical records were defined using constrained incremental sum of squares cluster analysis (CONISS; Grimm, 1987) with the Gower dissimilarity index to allow for zero values in the packages vegan (Oksanen et al., 2019) and rioja (Juggins, 2018). We conducted ordination analysis to explore the relationship between corresponding geochemical and plant macrofossil data in the GRI record. Plant macrofossil species with <5 occurrences in the record were removed from ordination analysis to reduce clustering. As a result of the mixture of plant macrofossil relative abundance and count data, we used the ‘capscale’ function in vegan (Oksanen et al., 2019), a variant of constrained analysis of principle coordinates (CAP; Anderson and Willis, 2003), with Gower (1971) dissimilarity (see Birks, 2014). Correlation analysis (Spearman’s Rank) was conducted on geochemical variables to identify redundant variables; these were then removed from the CAP analysis to improve clarity of the bi-plot.

Because of a hiatus spanning the middle Holocene at the study site, we present only the lower part of the collected profiles. The lower profiles span the Younger Dryas and the early Holocene, while the upper profile representing the late Holocene has a broader archaeological context and will be presented in another paper.

Results

Lithostratigraphy and chronology

The details of the sediment lithostratigraphy are presented in Table 1. At both sampling sites, detritus-calcareous gyttja accumulated on top of a silt and sand layer. Terrestrial plant macrofossil material for radiocarbon dating was selected from 1-cm core slices. Four samples were taken in total, with two from each core, respectively (Table 2). Two of the samples (GRI, 145.5 cm; GRII, 226.5 cm) were selected in order to date the beginning of organic sediments (gyttja) accumulation. However, because of the lack of reliable plant macrofossils, it was impossible to date the bottommost part of the sediments in both cores. Nevertheless, our interpretation is that organic sediments were deposited in the final stage of Younger Dryas period and after several hundreds of years into the early Holocene peat started forming (GRI 144 cm; GRII 172 cm) on top of the detritus gyttja layer.

Table 1.

Lithostratigraphic description of the sediment sequence.

Sampling site/core	Description of sediments
GRI	186–178 cm silt with sand 178–149 cm detritus-calcareous gyttja 149–144 cm coarse detritus gyttja 144–125 cm herbaceous peat 125–97 cm moss–herbaceous peat 97–85 cm strongly decomposed herbaceous peat
GRII	240–235 cm silt with sand 235–172 cm detritus-calcareous gyttja 172–150 cm brown moss–herbaceous peat 150–132 cm strongly decomposed herbaceous peat

Table 2.

Radiocarbon dates from Grodzisko, core GRI and GRII.

Depth (cm)	Material	Nr. Lab.	AMS date	Age (cal. yr BP)
GRI 92.5	Menyanthes trifoliata seeds, Pinus sylvestris seed	Poz-94372	9250 ± 50 BP	10,560–10,269
GRI 145.5	Pinus sylvestris needles and periderm	Poz-94373	10,000 ± 60 BP	11,750–11,263
GRII 154.5	Menyanthes trifoliata seeds	Poz-94853	9620 ± 50 BP	11,173–10,773
GRII 226.5	Charred wood	Poz-94854	10,260 ± 50 BP	12,374–11,767

Plant macrofossils

Core GRI

We identified five zones in the local vegetation development for plant macrofossils in GRI, confirmed by CONISS (Gower; Figure 2). The zone GRI-pm-1 (186–165 cm) is characterised by a dominance of Chara sp. alongside Potamogeton perfoliatus, P. fresii, P. natans and Myriophyllum sp. GRI-pm-2 (165–147 cm) sees a decline in Chara sp. and is the last zone where Potamogeton spp. is present, while the macrophytes Sparganium minimum and Typha sp. appear. P. sylvestris macrofossils were also observed for the first time in GRI-pm-2. GRI-pm-3 (147–120 cm) is characterised by Cyperaceae, herbs, Carex spp., Menyanthes trifoliata and tree macrofossils. In GRI-pm-4 (120–97 cm), S. contortum dominates alongside periods of Meesia triquetra and S. teres. In the zone GRI-pm-5 (97–85 cm), mosses disappear, while numerous M. trifoliata seeds were present and Alisma plantago-aquatica appeared for the first time.

Figure 2.

Plant macrofossil diagram: core GRI, Grodzisko. Taxa with (%) are given in estimated volume percentages, and others are given in counted numbers (with X-axis scale labels; note scale differences).

Core GRII

Six zones in the local vegetation development, confirmed by CONISS (Gower), were determined for the core GRII (Figure 3). GRII-pm-1 (240–217 cm) is composed of Chara sp., Myriophyllum sp., P. natans and P. fresii. In GRII-pm-2 (217–202 cm), Chara sp. are still present, alongside Carex sp. and P. sylvestris. During GRII-pm-3 (202–187 cm), Ceratophyllum demersum, Nuphar sp. and Typha sp. appear with Carex sp. and tree macrofossils (P. sylvestris and Betula sp.) are present throughout. In zone GRII-pm-4 (187–171 cm), the presence of tree macrofossils continues, while Batrachium sp., S. minimum, Lycopus europaeus, Ranunculus sceleratus and charcoal pieces are recorded for the first time. GRII-pm-5 (171–149 cm) is characterised by numerous M. trifoliata seeds, Calliergon cordifolium/giganteum leaves, herbs and the presence of M. triquetra. In GRII-pm-6 (149–135 cm), Cyperaceae rootlets dominate the record, with limited P. sylvestris and Betula sp. remains.

Figure 3.

Plant macrofossil diagram: GRII, Grodzisko. Taxa with (%) are given in estimated volume percentages, and others are given in counted numbers (with X-axis scale labels; note scale differences).

Geochemical analysis

Geochemical analysis was performed on the GRI core. Four stratigraphic zones were identified using CONISS (Gower; Figure 4). A substantial concentration of Ca²⁺ characterises zone GRI-ge-1 (186–174 cm). Zone GRI-ge-2 (174–144 cm) is characterised by a decrease in Ca and increase in the concentration of Al and correlating variables. In the zone GRI-ge-3 (144–100 cm), there is a large decrease in Al, Zn, K, Mg and Cr with a parallel increase in Si and correlating variables. During zone GRI-ge-4 (100–85 cm), an increase in Al, Zn, Cr, Fe, Cu and Pb took place.

Figure 4.

Results of the geochemical analysis of core GRI (mg/kg⁻¹).

Constrained analysis of principle coordinates (CAP)

Ordination analysis (CAP) of plant macrofossil and depth data explains a degree of variation on both axis 1 (CAP1; 10.7%) and axis 2 (CAP2; 4%) (Figure 5). Correlation analysis (Spearman’s Rank) found that Al significantly correlated with Cr (p < 0.001), Cu (p < 0.001), Fe (p < 0.05), K (p < 0.001), Mg (p < 0.001), Mn (p < 0.001), Nd (p < 0.001), P (p < 0.01), Pb (p < 0.001), Ti (p < 0.01) and Zn (p < 0.001), while Si correlated with As (p < 0.01) and Na (p < 0.01). Ca correlated with Mg (p < 0.01), but to a lesser extent than Al and Mg (p < 0.001). Therefore, redundant correlating variables were removed and only Ca, Al and Si were retained as environmental variables in the CAP analysis. Higher Si concentrations show a clear association with Cyperaceae and a slight association with brown mosses and herbs and some sites in GRI-pm-3. Chara sp. and GRI-pm-1 are clearly associated with increased Ca concentrations. Al shows a very slight association with S. teres and GRI-pm-2.

Figure 5.

Constrained analysis of principle coordinates (CAP) species-depth bi-plot of plant macrofossils, depth and select geochemical variables. Stratigraphic zones are grouped for CONISS analysis of plant macrofossils in GRI.

Discussion

The first stage of oxbow lake development during cold climate conditions

The oxbow lake at our study site is a palaeomeander of the Prosna River and formed during the late Glacial period. AMS dating indicates that the accumulation of the organic-rich deposits began sometime before 12,374–11,767 cal. yr BP, most likely during the Younger Dryas, but potentially earlier. Many palaeomeanders in this part of the Europe have a late Glacial origin, associated with the transition of river valleys from braided channels to a main meandering channel (Duda and Borówka, 2007; Forysiak, 2012; Forysiak et al., 2010; Gonera and Kozarski, 1987; Kozarski and Rotnicki, 1977; Młynarczyk and Rotnicki, 1989; Pawłowski et al., 2015b, 2016a; Płóciennik et al., 2011; Słowik et al., 2019; Starkel et al., 2015).

Pawłowski et al. (2016a) suggest the summer temperature during Younger Dryas in central Poland oscillated ca. 14°C, which would likely have facilitated the growth of some macrophyte species in the palaeolake. Chara sp., Batrachium sp. and Potamogeton spp. were the pioneer plants colonising the cold alkaline water, rich in Ca and Mg (Figures 5 and 6), with the relationship between Chara sp. and increased Ca concentrations further supported by ordination analysis (Figure 5).

Figure 6.

Comparison of selected plant taxa and geochemical results, core GRI. Geochemical results are presented in mg kg^-1.

The presence of these submerged plants during the late Glacial in the first stage of macrophyte succession is in agreement with records from many other European lowland lakes (Amon et al., 2010; Fajer et al., 2012; Gałka et al., 2014; Gałka and Sznel, 2013; Kołaczek et al., 2015; Mortensen et al., 1997). In particular, stoneworts oospores (e.g. Chara sp.) have been shown to be common in the first stage of macrophyte succession, emphasising their pioneering role in colonising new water basins, such as palaeolakes formed in river valleys (Kołaczek et al., 2018; Pawłowski et al., 2016a; Płóciennik et al., 2011). During the period, when oxbow lake existed at the sampling site, the Potamogeton species of P. fresii, P. praelongus and P. alpinus appeared (GRI-pm-2; GRII-pm-2), suggesting quite shallow and eutrophic water conditions (Zalewska-Gałosz, 2008). Only a single endocarp of P. praelongus was identified in the GRI core, supporting previous observations from other lakes located in European lowland and mountain sites that this species does not have competitive abilities to disperse compared with other Potamogeton species (Gałka et al., 2017a, and references quoted there). The meridional orientation of the valley and relatively rich fossil presence of macrophytes in the palaeo-record may indicate that the Prosna River was an important pathway for the spreading of aquatic plants during the late Glacial.

Early Holocene plant succession during warm climate conditions

Early Holocene (ca. 11,300 cal. yr BP, Kołaczek et al., 2018) mean June temperatures ca. 18°C likely led to the expansion of thermophilous macrophytes such as C. demersum and Typha sp. at both sampling sites (Figure 2; GRI-pm-2 and 3; GRI-pm-3). These plants typically grow in shallow water (Podbielkowski and Tomaszewicz, 1996) and are often recorded in the transitional zone between lakes and peatlands during the early Holocene (Gałka and Sznel, 2013; Hannon and Gaillard, 1997; Kołaczek et al., 2015, 2018).

The warmer temperature appears to have had a positive effect on forest development and its density in areas surrounding river valleys. Both P. sylvestris and Betula pubescens macrofossils were observed in early Holocene sediments from both GRI and GRII. Those trees likely occupy mineral soils in the river valley and along river banks. However, typical riparian trees such Salix and Populus were not detected in the studied cores. Nevertheless, river valleys were important routes for deciduous trees spreading in the European lowlands (Kołaczek et al., 2018; Kolstrup, 2007) and vegetation cover was an important factor in controlling river system evolution (Turner et al., 2013). Higher temperatures and potentially lower precipitation contributed to increased fire events during the early Holocene in river valleys and their surrounding areas. This is evidenced by the presence of early Holocene charcoal pieces in both our cores (Figures 2 and 3) as well as in other sites located in river valleys across the Polish lowlands, for example, Grabia (Pawłowski et al., 2016a, 2016b), Ner (Kittel et al., 2016) and San (Kołaczek et al., 2018).

At ca. 11,750–11,260 cal. yr BP in the GRI core (Figure 4, GRI-ge-3), there is an abrupt decrease in concentrations of Al, K, Mg, Ti, Fe and Cr and an increase in Si concentrations. This geochemical shift is associated with a disappearance of aquatic species and development of typical peatland plants population with the appearance of Carex rostrata, C. lasiocarpa, M. trifoliata, brown mosses, Cyperaceae and herbs (Figure 2, GRI-pm-3). Ordination analysis further supports the relationship between increased Si concentrations and brown mosses, Cyperaceae and herbs (Figures 5 and 6). Such a visible change of geochemical composition might indicate isolation of the study site from the influence of the Prosna River, as has been recorded in other early Holocene palaeomeanders located in central Poland (Pawłowski et al., 2016a, 2016b; Płóciennik et al., 2011).

A decrease in water level during the early Holocene has been documented in several central European lowland sites (cf. Gałka et al., 2015b, and references quoted there). A decreased water table may explain the development of a Sphagnum population dominated by S. contortum in zone GRI-pm-4, preceded by a short phase of M. triquetra (cf. Gałka et al., 2015a, 2017a, 2017b) and followed by a short stage of S. teres (Figures 3 and 6). The presence of S. contorum in our early Holocene dated deposits from the Prosna River valley is the oldest such example in the European lowlands, with S. contortum being among the rarest of Sphagnum mosses encountered in the fossil state. To date, its fossil presence has only been confirmed at a few sites in the northern hemisphere (Gałka and Lamentowicz, 2014). S. contortum macrofossils were recorded, for example, in late Glacial deposits at Scragh Bog, Central Ireland (O’Connell, 1980). In addition, the early Holocene presence of S. teres in GRI is important from a biogeographic perspective as one of the earliest fossil examples in Central Europe. Previously, S. teres had only been recorded in late Glacial peat layers from SE Germany (Hölzer and Hölzer, 1994), NE Poland (Gałka and Sznel, 2013) and in early Holocene (ca. 10,000 years ago) lake and peat deposits from the Eastern Carpathians, N Romania (Gałka et al., 2017a). S. contortum and S. teres are minerotrophic species that often grow together and can be found in rich fen habitats with a pH ca. 6 (Hájková and Hájek, 2004; Hölzer, 2010; Tahvanainen, 2013). Their macrofossils are usually found in the peat layer preceding the development of poor fen conditions (Gałka et al., 2018; Gałka and Lamentowicz, 2014). The presence of early Holocene Sphagnum in this river valley peatland suggests that river ecosystems have played an important role for the survival and spreading of some minerotrophic Sphagnum species, especially in areas with limited topographic depressions for peatlands to develop. The disappearance of mosses in the upper part of both peat profiles (Figures 2 and 3, GRI-pm-5; GRII-pm-6) may be linked to a changing interaction with the Prosna River. Increased heavy metal concentrations in sediments, especially Cu, Cr and Mn (Figure 6), might indicate an increased influence of the Prosna River water and gradual flooding of this peatland. River waters are usually rich in Mn²⁺ ions or they transport this chemical element in a form of suspension colloid (Kabata-Pendias and Pendias, 1979). However, the increase in Mn could also have been caused by chemical weathering, as a result of preferential leaching of Mn from the soil under slightly acidic conditions. An increase of the water level at sampling sites is also supported by the low value of the Fe/Mn ratio (cf. Boyle, 2001; Pawłowski et al., 2015a). A large increase in water level would have been required to cause the decline in minetrophic mosses such as M. triquetra and Sphagnum species that are tolerant to wet and periodically submerged conditions (Hölzer, 2010; Montagnes, 1990). However, this decline in mosses could also have been caused by erosion during flood events, which is quite a common phenomenon in river valleys (Starkel, 2002).

Summary

We present detailed plant macrofossil data from two sediment sequences from a palaeolake in the Prosna River valley with supporting geochemical analysis and radiocarbon dating. We draw the following conclusions from our analysis:

The studied palaeomeander was formed in the late Glacial (most likely in Younger Dryas). This is in agreement with records from other Polish lowland river valleys that document the development of oxbow lakes during this period.

Rich fossil presence of macrophytes in the studied palaeomeander, along with the meridional orientation of the valley may indicate that the Prosna valley was an important route for the spreading of aquatic plants in the late Glacial and an important habitat for rare mosses such as M. triquetra.

Early Holocene warming likely led to a decrease in water level, triggering the terrestrialization process of the oxbow lake – as has been observed across many Polish lowland sites.

Organic sediments deposited in river valleys are a valuable source of biogeographic knowledge regarding the past distribution of mosses. S. contortum and S. teres macrofossils were found in the early Holocene deposits of the Prosna River valley. These early Holocene Sphagnum findings suggest that river valleys – and the peatlands that developed there – were important habitats for the survival and spread of some minerotrophic Sphagnum species, potentially even functioning as refugia.

Footnotes

Acknowledgements

We would like to thank Jakub Sypniewski, Karol Jakubowski and Jędrzej Proch for their help during laboratory work.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Radiocarbon dating was supported by Fundacja Ochrony Zabytków.

ORCID iD

Mariusz Gałka

References

Amon

Heinsalu

Veski

(2010) Late glacial multiproxy evidence of vegetation development and environmental change at Solova, southeastern Estonia. Estonian Journal of Earth Sciences 59: 151–163.

Anderson

Willis

(2003) Canonical analysis of principal coordinates: A useful method of constrained ordination for ecology. Ecology 84: 511–525.

Birks

HJB

(2014) Challenges in the presentation and analysis of plant-macrofossil stratigraphical data. Vegetation History and Archaeobotany 23: 309–330.

Borówka

Tomkowiak

Okupny

et al. (2015) Skład chemiczny osadów bagiennych z Doliny Luciąży (Torfowisko Bęczkowice na równinie piotrkowskiej). Folia Quaternaria 83: 5–23.

Boyle

(2001) Inorganic geochemical methods in palaeolimnology. In: Last

Smol

(eds) Tracking Environmental Change Using Lake Sediments, Physical and Geochemical Methods, vol.2. Dordrecht: Kluwer Academic Publishers, pp. 83–141.

Bronk-Ramsey

(2009) Dealing with outliers and offsets in radiocarbon dating. Radiocarbon 51: 1023–1045.

Dąbrowski

(1991) Rozwój doliny Prosny w plejstocenie. Geneza, litologia i stratygrafia utworów czwartorzędowych, Wydawnictwo Naukowe UAM, Poznań. Geografia 50: 35–46.

Duda

Borówka

(2007) Zmiany w krajobrazie doliny dolnej Odry na te rozwoju paleogeograficznego regionu. Prace Komisji Krajobrazu Kulturowego 7: 210–218.

Dwire

Mellmann-Brown

Gurrieri

(2018) Potential effects of climate change on riparian areas, wetlands, and groundwater-dependent ecosystems in the Blue Mountains, Oregon, USA. Climate Services 10: 44–52.

10.

Dyjor

(1985) Systemy Kopalnych dolin Polski Zachodniej, Problemy młodszego neogenu i eoplejstocenu w Polsce. Warszawa: Wydawnictwo PAN.

11.

Fajer

Waga

Rzetala

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

12.

Forysiak

(2012) Record of changes in the natural environment of the Late Weichselian and Holocene preserved in the sediments of peatlands of the Łódź Region. Acta Geographica Lodziensia 90: 1–116.

13.

Forysiak

Obremska

Pawłowski

et al. (2010) Late Vistulian and Holocene changes in the Ner river valley in light of geological and palaeoecological data from the Ner-Zawada peatland. Geologija 52: 25–33.

14.

Gałka

Lamentowicz

(2014) Sphagnum succession in a Baltic bog in Central-Eastern Europe over the last 6200 years and paleoecology of Sphagnum contortum. The Bryologist 117: 22–36.

15.

Gałka

Sznel

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

16.

Gałka

Miotk-Szpiganowicz

Marczewska

et al. (2015a) Palaeoenvironmental changes in Central Europe (NE Poland) during the last 6200 years reconstructed from a high resolution multi-proxy peat archive. The Holocene 25: 421–434.

17.

Gałka

Swindles

Szal

et al. (2018) Response of plant communities to climate change during the Late-Holocene: Palaeoecological insights from peatlands in the Alaskan Arctic. Ecological Indicators 85: 525–536.

18.

Gałka

Tanţău

Feurdean

(2017a) Plant succession in a peatland in the Eastern Carpathian Mts. (CE Europe) during the last 10,200 years: Implications for peatland development and palaeoclimatic research. Review Palaeobotany and Palynology 244: 203–216.

19.

Gałka

Tobolski

Bubak

(2015b) Late Glacial and Early Holocene lake level fluctuations in NE Poland tracked by macro-fossil, pollen and diatom records. Quaternary International 388: 23–38.

20.

Gałka

Tobolski

Lamentowicz

et al (2017b) Unveiling exceptional Baltic bog ecohydrology, autogenic succession and climate change during the last 2000 years in CE Europe using replicate cores, multi-proxy data and functional traits of testate amoebae. Quaternary Science Reviews 156: 90–106.

21.

Gałka

Tobolski

Zawisza

et al. (2014) Postglacial history of vegetation, human activity and lake-level changes at Jezioro Linówek in northeast Poland, based on multi-proxy data. Vegetation History and Archaeobotany 23: 123–152.

22.

Garssen

Verhoeven

JTA

Soons

(2014) Effects of climate-induced increases in summer drought on riparian plant species: A meta-analysis. Freshwater Biology 59: 1052–1063.

23.

Gonera

Kozarski

(1987) River channel changes and rough paleodischarge estimates for the Warta river, west-central Poland. Geografiska Annaler Series A Physical Geography 69: 163–171.

24.

Gower

(1971) A general coefficient of similarity and some of its properties. Biometrics 28: 857–871.

25.

Grimm

(1987) CONISS: A FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares.Computers & Geosciences 13: 13–35.

26.

Hájková

Hájek

(2004) Bryophyte and vascular plant responses to base-richness and water level gradients in western Carpathian Sphagnum-rich mires. Folia Geobotanica 39: 335–351.

27.

Hannon

Gaillard

(1997) The plant-macrofossil record of past lake-level changes. Journal of Paleolimnology 18: 15–28.

28.

Hölzer

(2010) Die Torfmoose Südwestdeutschlands und der Nachbargebiete. Jena: Weissdorn.

29.

Hölzer

(1994) Studies on the vegetation history of the Lautermoor in the Upper Rhine valley (SW Germany) by means of pollen, macrofossils, and geochemistry. Dissertationes Botanicae 234: 309–336.

30.

Juggins

(2007) C2 User Guide: Software for Ecological and Paleoecological Data Analysis and Visualization. Newcastle upon Tyne: University of Newcastle.

31.

Juggins

(2018) Rioja: Analysis of Quaternary Science Data. R Package version 0.9-15.1. Available at: https://cran.r-project.org/package=rioja.

32.

Kabata-Pendias

Pendias

(1979) Pierwiastki śladowe w środowisku biologicznym. Warszawa: Wydawnictwo Geologiczne.

33.

Kaczmarek

Szczurek

(2015) The early iron age fortified settlements in Wielkopolska (western Poland) – The past and present perspective of archaeological research. Praehistorische Zeitschrift 90: 245–270.

34.

Kittel

Płóciennik

Borówka

et al. (2016) Early Holocene hydrology and environments of the Ner River (Poland). Quaternary Research 85: 187–203.

35.

Kołaczek

Gałka

Karpińska-Kołaczek

et al. (2015) Late Pleniglacial and Late Glacial lake-mire transformations in south-eastern Poland reflected in aquatic and wetland vegetation changes. Quaternary International 388: 39–50.

36.

Kołaczek

Płóciennik

Gałka

et al. (2018) Persist or take advantage of global warming: A development of Early Holocene riparian forest and oxbow lake ecosystems in Central Europe. Quaternary Science Review 200: 191–211.

37.

Kolstrup

(2007) Water-borne macroscopic plant particle transport through central and northern Europe during warming phases: A hypothetical spreading mechanism for climatic pioneers. GFF 129: 307–313.

38.

Kozarski

Rotnicki

(1977) Valley floors and changes of river channel patterns in the North Polish Plain during the Late-Würm and Holocene. Quaestiones Geographicae 4: 51–93.

39.

Lorenc

(red.) (2005) Atlas klimatu Polski. Warszawa: Instytut Meteorologii i Gospodarki Wodnej.

40.

Młynarczyk

Rotnicki

(1989) Późnovistuliańskie i holoceńskie formy i osady środkowej Prosny i ich paleohydrograficzna interpretacja. Poznań: Wydawnictwo Naukowe UAM.

41.

Magyari

Jakab

Sümegi

et al. (2008) Holocene vegetation dynamics in the Bereg Plain, NE Hungary – The Báb-tava pollen and plant macrofossil record. Acta GGM Debrecina Geology, Geomorphology, Physical Geography 3: 33–50.

42.

Mirek

Piękoś-Mirkowa

Zając

et al. (2002) Fowering Plants and Pteridophytes of Poland: A Checklist. Kraków: W. Szafer Institute of Botany, Polish Academy of Sciences.

43.

Montagnes

RJS

(1990) The habitat and distribution of Meesia triquetra in North America and Greenland. The Bryologist 93: 349–352.

44.

Mortensen

Birks

Christensen

et al. (1997) The ecology of interfaces: Riparian zones. Annual Review of Ecology and Systematics 28: 621–658.

45.

Naiman

Decamps

(1997) The ecology of interfaces: Riparian zones. Annual Review of Ecology and Systematics 28: 621–658.

46.

O’Connell

(1980) The developmental history of Scragh Bog, Co. Westmeath, and the vegetation of its hinterland. New Phytologist 83: 301–309.

47.

Oksanen

Blanchet

Friendly

et al. (2019) vegan: Community ecology package. Available at: https://cran.r-project.org/package=vegan.

48.

Płóciennik

Self

Birks

HJB

et al. (2011) Chironomidae (Insecta: Diptera) succession in Żabieniec bog and its palaeo-lake (central Poland) through the Late Weichselian and Holocene. Palaeogeography, Palaeoclimatology, Palaeoecology 307: 150–167.

49.

Pawłowski

Borówka

Kowalewski

et al. (2016a) Late Weichselian and Holocene record of the paleoenvironmental changes in a small river valley in Central Poland. Quaternary Science Reviews 135: 24–40.

50.

Pawłowski

Borówka

Kowalewski

et al. (2016b) The response of flood-plain ecosystems to the Late Glacial and Early Holocene hydrological changes: A case study from a small Central European river Valley. Catena 147: 411–428.

51.

Pawłowski

Milecka

Kittel

et al. (2015a) Palaeoecological record of natural changes and human impact in a small river valley in Central Poland. Quaternary International 370: 12–28.

52.

Pawłowski

Płóciennik

Brooks

et al. (2015b) A multiproxy study of Younger Dryas and Early Holocene climatic conditions from the Grabia River paleo-oxbow lake (central Poland). Palaeogeography, Palaeoclimatology, Palaeoecology 438: 34–50.

53.

Piszczygłowa

(2014) Geologia i ewolucja Doliny Prosny w rejonie Kalisza w świetle analizy litologiczno-strukturalnej osadów. Archives of Institute of Civil Engineering 17: 51–60.

54.

Podbielkowski

Tomaszewicz

(1996) Zarys Hydrobotaniki. Warszawa: Wydawnictwo Naukowe PWN.

55.

R Core Team (2018) R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing.

56.

Rasmussen

Bigler

Blockley

SPE

et al. (2014) A framework for robust naming and correlation of past abrupt climatic changes during the recent glacial period based on three synchronised Greenland ice-cores. Quaternary Science Review 106: 14–28.

57.

Słowik

(2011) Reconstructing migration phases of meandering channels by means of ground-penetrating radar (GPR): The case of the Obra River, Poland. Journal of Soils and Sediments 11: 1262.

58.

Słowik

(2013) Transformation of a lowland river from a meandering and multi-channel pattern into an artificial canal: Retracing a path of river channel changes (the Middle Obra River, W Poland). Regional Environmental Change 13: 1287–1299.

59.

Słowik

Gałka

Marciniak

(2019) The evolution and disappearance of “false delta” multi-channel systems in postglacial areas (central Europe). Global and Planetary Change.

60.

Śmigielski

Szczurek

(2013) Wyniki badań wykopaliskowych przeprowadzonych w 1973 roku na grodzisku ludności kultury łużyckiej w Grodzisku. In: Szczurek

Różański

(eds) Grodzisko z wczesnej epoki żelaza i wczesnego średniowiecza w Grodzisku, gm. Pleszew, woj. Wielkopolskie. Poznań: Fundacja Ochrony zabytków w poznaniu, pp. 11–47.

61.

Starkel

(2002) Change in the frequency of extreme events as the indicator of climatic change in the Holocene (in fluvial systems). Quaternary International 91: 25–32.

62.

Starkel

Michczyńska

Gębica

et al. (2015) Climatic fluctuations reflected in the evolution of fluvial systems in Central-Eastern Europe (60-8 ka cal. BP). Quaternary International 388: 97–118.

63.

Starkel

Soja

Michczyńska

(2006) Past hydrological events reflected in Holocene history of Polish rivers. Catena 66: 24–33.

64.

Stupnicka

Baranowski

Bendet

(2006) Wpływ czynników klimatycznych na procesy osadnicze w dolinach rzek środkowej Polski w okresie rzymskim i we wczesnym średniowieczu. Archeologia Polski 1–2: 93–120.

65.

Szafer

(1946) Zarys historii rozwoju flory Holarktydy. Annales Societatis Geologorum Poloniae 16: 179–241.

66.

Tahvanainen

(2013) Eräiden lehtisammallajien suhde suoveden pHarvoon. In: Aapala

Similä Penttinen

(eds) Ojitettujen soiden Ennallistamisopas. Vantaa; Helsinki: Metsähallitus, pp. 54–55.

67.

Thodsen

Baattrup-Pedersen

Andersen

et al. (2016) Climate change effects on lowland stream flood regimes and riparian rich fen vegetation communities in Denmark. Hydrological Sciences Journal 61: 344–358.

68.

Tobolski

(2000) Przewodnik do oznaczania torfów i osadów jeziornych. Warszawa: Państwowe Wydawnictwo Naukowe.

69.

Turner

Tolksdorf

Viehberg

et al. (2013) Late glacial/early Holocene fluvial reactions of the Jeetzel river (Elbe valley, northern Germany) to abrupt climatic and environmental changes. Quaternary Science Reviews 60: 91–109.

70.

Velichkevich

Zastawniak

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

71.

Velichkevich

Zastawniak

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

72.

Woś

(1999) Klimat Polski. Warsaw: Wydawnictwo Naukowe PWN.

73.

Zalewska-Gałosz

(2008) Rodzaj Potamogeton L. w Polsce taksonomia i rozmieszczenie. Kraków: Instytut Botaniki Uniwersytetu Jagiellońskiego.