Landscape development at Lina myr fen,Eastern Gotland,9000−2500 cal. yr BP

Abstract

Using diatoms, pollen, and geochemistry, we explore human habitation around Lina myr, Gotland, in relation to shore displacement. Archeological evidence has shown that Lina myr was an important area for its prehistoric human inhabitants. We investigate if and when Lina myr was connected to the sea and could therefore have been part of an inland water system useful for transport. A chronology was based on ¹⁴C AMS dating of terrestrial macrofossils and bulk sediments with dates ranging between 9100 and 2360 cal. yr BP. The initiation of the Littorina transgression was dated to 8500 cal. yr BP. A twofold pattern for the maximum sub-phase of the Littorina Sea is suggested from 8100 to 7500 cal. yr BP and from 6500 to 6000 cal. yr BP. The onset of cultivation and grazing was indicated by the presence of Hordeum and Plantago lanceolata in the pollen record during the Late Neolithic, at about 4580 cal. yr BP. During this time sea level was relatively higher than today and the Lina myr basin was connected with the Littorina Sea, which it continued to be until isostatic uplift caused it to become isolated at about 3820 cal. yr BP. After about 3000 cal. yr BP, human-made landscape changes intensified, grasslands increased, and shrublands decreased.

Keywords

archeology Baltic Sea diatoms Gotland human impacts Littorina transgression pollen shore displacement

Introduction

Lina myr, a drained and cultivated area of fenland in central-eastern Gotland, is an area of archeological significance with a complex history of shoreline displacement. Archeological excavations have indicated that people have been active in the local area since the Mesolithic period (Lund, 1996; Österholm, 1989; Seving, 1986; Wallin, 2010; Welinder, 1975) and during the Neolithic period (Hallström, 1971; Sundberg, 2008). Nowadays these archeological sites are located inland but were perhaps once situated next to the Littorina Sea shorelines. Archeological excavations at Gothemshammar, 7 km northeast of Lina myr (Figure 1), unearthed the foundations of a rampart, consisting of a 500-m-long cavity wall. Radiocarbon dating of the domestic materials found intermixed with the wall foundations indicate that the rampart was built 2850–2540 cal. yr BP (Wallin, 2010; Wallin and Martinsson-Wallin, 2018). A large Bronze Age stone cairn and stone ship setting (a type of burial custom common on Gotland) 5 km northeast of Lina myr (Martinsson-Wallin and Wehlin, 2017) also confirms the historical importance of the area for Bronze Age inhabitants of Gotland. Around 400 stone ship settings are found on Gotland and are usually located along the Ancylus Lake and Littorina Sea shorelines of Gotland (Hansson, 1927). According to Wehlin (2010), around 15% of the stone ships on Gotland can be found around Lina myr. Martinsson-Wallin and Wehlin (2017) stated that several of these stone ships are located along the northern side of the Gothem River and Lina myr and are positioned in an eastwest direction. This differs from the majority of stone ships settings on Gotland, which are found in coastal locations around the island placed with the bow toward the south. According to Ohlsson (1984) at least up to medieval times, a historical waterway, comprising the Gothem River and some areas of open water, could have been used for transport through Lina myr, into central Gotland. This important waterway could explain the difference in the orientation in the ship settings. The outlet of this water system is located at Vitviken bay, northeast of Lina myr. This sea connection would have made Lina myr an important center for transport, trade, and settlement.

Figure 1.

(a) The location of Gotland, (b) Lina myr, and (c) surface geology (SGU, 2017).

Previous paleoecological studies have used fossil pollen to research the vegetation history of Gotland (Eriksson, 1992; Munthe et al., 1927; Österholm, 1989; Påhlsson, 1977; Pettersson, 1958; Sernander, 1894). A few studies have focused on the paleoecology of Lina myr. Thomasson (1927) described the diatom stratigraphy of Lina myr; Lundqvist (1928) described diatoms and mollusks in a layer of gyttja, which he attributed to Littorina Sea deposits. Svensson (1989) investigated the shore displacement of Gotland using pollen stratigraphy as a method of correlating sites all over the island. However, there have been no studies directly aimed at investigating the prehistoric human impacts on vegetation at Lina myr.

In order to understand when Lina myr may have been connected to the sea, and therefore possibly used as an inland waterway for transport, it is important to explore the history of shore displacement in this area. However, the history of sea-level change and isostatic uplift around the Baltic basin is complex. Discrepancies between shore displacement curves from different parts of the Baltic basin have previously been discussed by Björck and Svensson (1992) and Uścinowicz (2003). There are also apparent inconsistencies around the timing of the Littorina transgression and number of sub-phases.

Based on studies in Blekinge, southeastern Sweden, Berglund (1971), Berglund and Sandgren (1996), Berglund et al. (2005), and Risberg et al. (2005) suggested that the Littorina transgression onset occurred at about 8500 cal. yr BP and distinguished five separate phases. Several transgressive sub-phases have also been seen in Denmark (Christensen and Nielsen, 2008; Wohlfarth et al., 2008). However, in Finland only one uniform transgression has been identified (Eronen, 1974; Miettinen et al., 2007). Similarly, only one main transgression has been detected in Estonia (Raukas et al., 1995). At Vääna lagoon, Estonia, the Littorina transgression began about 8300 cal. yr BP and continued until about 7000 cal. yr BP. A similar timing was observed on Saaremaa Island, Estonia, where the transgression began about 8300–8200 cal. yr BP and lasted until 7300 cal. yr BP (Saarse et al., 2009a, 2009b). A twofold transgression started at 8400 cal. yr BP on the Karelian Isthmus is suggested by Miettinen et al. (2007). Rosentau et al. (2013) proposed that the Littorina transgression occurred as a single event and started 8300 cal. yr BP based on coastal sediments in the eastern Gulf of Finland.

The detailed shoreline displacement and vegetation history in relation to archeological sites, climate, and uplift is poorly understood on Gotland. Our overall aim was to investigate the interplay between sea-level change, the development of waterways, human settlement, and vegetation changes on Gotland. Sea-level fluctuations were determined using diatom and geochemical proxies from new sediment cores from Lina myr. The timing and impact on vegetation by humans was investigated using fossil pollen.

Materials and analysis

Study site

Lina myr (57° 34′ 31.764″ N, 18° 37′ 59.124″ E), situated 6.3 km southwest of Åminne and 21 km east of Visby, is a 9-km² area of cultivated, drained fenland. The area stretches for 6 km from southwest to northeast and 2 km in an east to west direction. Lina myr drains in a northeast direction toward the east coast of Gotland via the Gothem River. The Gothem River was straightened during the draining of the fen in the 1940s. Today, the elevation of the Lina myr fen is about 9 m a.s.l. The original basin threshold is estimated to be 9.5–9.9 m a.s.l., based on a cross-section established from modern digital evaluation models (Barliaev, 2017). To the west of the fen the surface geology consists of marl and further to the west of glacial till (Figure 1c). To the east, limestone bedrock is exposed. The island of Gotland consists of Silurian limestone (Hede, 1925). The bedrock of Lina myr is classified as the Halla Formation (Laufeld, 1974).

Coring, sampling, and stratigraphy

Fieldwork was conducted in August 2016 and coring was carried out along three transects (Figure 2) to investigate the stratigraphy of the fen. A representative master core was collected at site 7 in section B. A Russian peat corer, 1 m in length and 4.5 cm in diameter, was used for sampling. Supplementary cores were also required in order to provide material for dating. Therefore, 15 parallel cores were retrieved as overlapping sequences in a 2-m² grid surrounding the master core. A Russian peat corer, 70 cm long and 7 cm in diameter, was used to retrieve this material. The cores were stored in a cold room with a constant temperature of 4–5°C. The stratigraphy of the master core was investigated in the field, combined with a more detailed description in the laboratory. The description was based on analysis of the material under a stereo microscope with ×25 magnification using Munsell color charts, combined with loss on ignition.

Figure 2.

The stratigraphy of Lina myr fen and coring locations.

Radiocarbon dating

In the laboratory, the 15 parallel cores were visually correlated by aligning stratigraphic boundaries and color variations. Subsamples were soaked overnight in 10% KOH and were then sieved using a 0.25-mm sieve. Macrofossils were identified using a stereo microscope under ×25 magnification and extracted with soft tweezers and radiocarbon dated. Eight dates were based on terrestrial macrofossils, which were combined with five bulk sediment samples (Supplemental Table 1, available online). Bulk dates were used where no macrofossils were available. To estimate the reservoir ages for the basin, a bulk sample was dated at the same depth as a terrestrial macrofossil sample (201 cm depth). Calibration of ¹⁴C ages and the construction of the age–depth model, which was a basic linear regression between neighboring levels, were carried out in Clam 2.2 and R Workspace using IntCal13 (Reimer et al., 2013). The reservoir age was subtracted from the bulk sediment ¹⁴C ages prior to calibration. Ages are reported as cal. yr BP (before 1950 AD).

Organic matter content

%LOI₅₅₀ (loss on ignition 550) and % carbonate content (%LOI₉₅₀) determinations (Dean, 1974) were made to estimate TIC and TOC (total inorganic and organic carbon, respectively). For sediments with low carbon values, a multiphase carbon analyzer Leco RC-512 was used.

Geochemistry

The cores were analyzed using an Itrax x-ray fluorescence (XRF) scanner (Croudace et al., 2006). The master core was scanned, using a molybdenum tube set at 30 kV voltage and current of 50 mA. Measurements were taken at 0.2 mm resolution for 15 s each. The obtained dataset includes a total of 15 elements (Br, Ca, Cu, Fe, K, Mn, Ni, Pb, Rb, S, Si, Sr, Ti, Zn, Zr) along with Mo incoherent and coherent scattering signals. Itrax data are uncalibrated (counts or count-rate) and therefore all the values were normalized by Ti which is considered to be a conservative element (Weltje and Tjallingii, 2008). Furthermore, normalization helps to compensate for the influence of the dilution effect of organic matter and physical defects (Löwemark et al., 2011). Average values were calculated for 1 cm intervals for smoothing the graphs.

Diatom analysis

Samples were treated according to the standard method compiled by Battarbee et al. (2002). Counting of diatoms was performed under light microscope using ×1000 magnification and immersion oil. Identifications of diatom valves were mainly based on Cleve-Euler (1951, 1952, 1953, 1955), Mölder and Tynni (1967–1973), Tynni (1975, 1976, 1978), Krammer and Lange-Bertalot (1986, 1988, 1991a, 1991b), Snoeijs (1993), Snoeijs and Balashova (1998), Snoeijs and Kasperoviciene (1996), Snoeijs and Popatova (1995), Krammer (2000, 2002, 2003), Lange-Bertalot (2001), Levkov (2009), Lange-Bertalot et al. (2011), and Levkov et al. (2013, 2016). The aim was to count 300 valves; however, lower basic sums were accepted where diatoms were sparse (lowest 85.5). Identified diatom species were grouped according to their salinity preferences (coastal brackish-marine taxa, planktonic brackish-marine taxa, halophilous taxa, lagoonal taxa, indifferent taxa, freshwater taxa), as well as water depth (coastal, planktonic, and benthic). The diatoms were grouped according to the following literature: Miller (1973), Lie et al. (1983), McQuoid and Hobson (1988), Zong (1997), Laffaille et al. (2002), and Hedenström and Risberg (2003). Since the Baltic Sea has not experienced fully marine conditions, that is, c. 3%, we use the term brackish-marine taxa (since some diatoms also live in marine conditions, but most prefer much lower salinities). We define coastal taxa as those which prefer shallow bays with brackish water. These bays are formed due to continued isostatic uplift in this formerly glaciated area. Lagoonal taxa (previously ‘clypeus flora’, named after Campylodiscus clypeus) are typical in shallow, low salinity bays (Alhonen, 1972; Eronen et al., 2001; Miller, 1986). Ancylus Lake taxa (previously ‘clear water’) have been noted in sediment accumulated within the Ancylus Lake stage. They are freshwater taxa and many are still common in large freshwater lakes such as Mälaren, Vänern, Vättern, Peipsi, and Ladoga (Alhonen, 1972; Eronen et al., 2001). Indifferent taxa are defined here as species with a broad tolerance to salinity and are therefore somewhat unhelpful for reconstructing shore displacement histories.

Tilia 2.0.41 software was used for plotting diagrams. CONISS cluster analysis was employed as an additional tool to eye-matching for defining diatom zones (Grimm, 1987).

Carbon and nitrogen determination

The C/N ratio is useful for indicating the type and source of organic matter (Rice and Hanson, 1984). Variations in C/N ratio can occur because algae are enriched in nitrogen and depleted in carbon compared with vascular plants (Tyson, 1995). C/N ratios of <8 typically indicate marine sediments (Bordovskiy, 1965) and according to García-Alix et al. (2012), a C/N ratio of <10 indicates aquatic organic matter. A C/N ratio of 10–20 indicates a mix of terrestrial and aquatic organic matter inputs and a C/N ratio of >20 signifies that terrestrial organic matter was predominant (Jones et al., 2013). Due to these variations in the C/N ratio of different sediments, C/N ratios can be used as proxy for relative sea level in isolation basins (Mackie et al., 2005). Forty subsamples spaced 7 cm apart at around 1 cm³ volume were subsampled from the gyttja and calcareous gyttja. No subsamples were taken from the fen peat as the source of the organic material was obviously terrestrial. No subsamples were analyzed from the clay as it can be assumed this formed in deep water and to have low organic matter content. The samples were then treated with 10% HCl and stirred in order to break up calcium carbonates, then dried overnight at 50°C in an oven and ground in a marble pestle and mortar. A microbalance was used to weigh 100 μgC–1 mgC into LUDI SWISS ϕ9.0-mm height 10-mm tin foil capsules and sealed with flat-edged tweezers. The C/N ratio measurement samples were made using a CHNS analyzer. The overall uncertainty in determining C/N (n = 41) was σ ± 1.19.

Pollen analysis

A total of 28 samples were prepared and counted with higher resolution in the upper part of the core. A standard procedure for pollen preparation was used following Berglund et al. (1986). Hydrofluoric acid was used on all samples in order to remove silicates. To calculate pollen concentration rates, tablets with a known number of Lycopodium clavatum spores were added to the samples and the exotic pollen were counted alongside the native fossil pollen (Stockmarr, 1971). The prepared slides were then mounted with glycerin. The pollen grains were identified according to Beug (2004), Fægri and Iversen (1989), Moore et al. (1991), and a reference collection. The aim for the minimum pollen count was set at 300; however, lower sums were accepted. The pollen diagram was created in Tilia 2.0.41. CONISS cluster analysis was applied to the data (Grimm, 1987). Only taxa with over 5% in abundance were included in the CONISS calculations. The CONISS output diagram was used to group the pollen data into different pollen assemblage zones; however, owing to the archeological importance of some indicator taxa such as Hordeum-type and Plantago lanceolata, zonation was also partly based on the occurrence of these taxa.

Results and interpretations

Stratigraphy

In general, the Lina myr basin is filled with glacial and postglacial clay. A thin sandy layer lies on top of the clay, covered by gyttja and fen peat (Figure 2). The gyttja was divided into three layers: two layers of calcareous gyttja separated by a layer of gyttja. In the master core (core 7), the following eight sedimentary facies were identified: clay and sand (0.15 %LOI₅₅₀ and 1.42 %LOI₉₅₀ 400–339 cm), calcareous gyttja (6.8 %LOI₅₅₀ and 28.28 %LOI₉₅₀ 339–307 cm), gyttja (26.58 %LOI₅₅₀ and 3.46 %LOI₉₅₀ 307–79 cm), calcareous gyttja (21.38 %LOI₅₅₀ and 30.95 %LOI₉₅₀ 79–57 cm) and fen peat (84.32 %LOI₅₅₀ and 2.33 %LOI₉₅₀ 57–40 cm). The upper 40 cm of the sequence was not used for analysis since the top surface of the fen was reworked due to plowing.

Chronology

Thirteen radiocarbon dates, ranging between 9102 and 2357 cal. yr BP (Supplemental Table 1, available online), were used to construct the age-depth model (Figure 3). At 201 cm depth, the date of a terrestrial macrofossil (5567 ± 51 ¹⁴C yr BP) was compared with the date of a bulk gyttja sample (6018 ± 32 ¹⁴C yr BP) and the 451-year age difference between the samples was used to estimate the reservoir effect. This reservoir age was subtracted from all bulk sediment dates prior to calibration. Two dates (Ua-54786 and Ua-54320) were rejected from the age-depth model since they were regarded as outliers (see Supplemental material, available online, for ages).

Figure 3.

Age-depth model Clam 2.2 and R Workspace IntCal 13.¹⁴C atmospheric curve for the Northern hemisphere (Reimer et al., 2013). The red points indicate outliers and the blue boxes indicate bulk dates which have had the reservoir age subtracted to them prior to calibration.

XRF geochemistry

When Br is compared with Ti, which is considered a relatively depleted element in the coarser-grained fractions of sediments, Br/Ti can be used as a proxy for marine organic matter (de Boer et al., 2014; Ziegler et al., 2008). Br/Ti has been used to indicate the ingress of marine water into coastal lakes caused by cyclone activity (Oliva et al., 2018). It could be argued that longer term changes, such as marine transgressions, could also be detected by fluctuations in the Br/Ti ratio. There were three apparent peaks in the Br/Ti for Lina myr – between 275 and 245 cm depth, culminating around 265 cm depth, between 165 and 155 cm and a sustained increase in the calcareous gyttja between 57 and 79 cm (Figure 4). The Ca/Ti ratio can be used to show relative changes in calcium carbonate content (Piva et al., 2008). The Ca/Ti ratio peaks in the upper and lower calcareous gyttja and mirrors LOI₉₅₀.

Figure 4.

The master core stratigraphy, loss on ignition, Br/Ti, Ca/Ti, C/N, brackish-marine planktonic diatom species share (including Paralia sulcata) as percentages and the diatom and pollen assemblage zones (DAZ and PAZ). L1-L3 represent Littorina Sea transgressive sub-phases.

Carbon-to-nitrogen ratios

The C/N was relatively high for the lower calcareous gyttja (C/N ratio between 11 and 17) indicating mixed sources of organic matter. The ratio started to shift toward lower values (C/N ratio of 9) after the transition to gyttja; however, the ratio did not fall below 7, which would indicate predominance of marine organic matter (Emerson and Hedges, 1988). After the initial shift to lower C/N ratios in the gyttja, the ratio shifted to a higher value of 12 and then again to a lower value of 9 a total of three times. These three shifts to lower values at 208, 173, and 117 cm depth in the stratigraphy may indicate increases in marine influence. The C/N ratio increased again to 25 at 89 cm depth with the boundary of the gyttja and upper calcareous gyttja deposits but then decreased again to 11 indicating a shift from terrigenous organic matter in the lake to a mixed source of organic matter.

Pollen record

The pollen diagram (Figure 5) was dominated by arboreal taxa such as Pinus, Betula, and Corylus. Since trees generally produce more pollen than herbs, it is common that tree pollen dominates the pollen records. However, when interpreting past plant communities and the effect of human activity, pollen from herbs and charcoal particles are important to consider, although they often only reach low percentages. Four pollen zones were identified based on CONISS and visual interpretation of indicator taxa such as Hordeum-type and Plantago lanceolata.

PAZ 1 (330–260 cm, 8850–7440 cal. yr BP, 6900−5490 BC)

Arboreal taxa of Pinus, Betula, and Corylus dominated the pollen record. Pinus reached relatively high values (45%) in the lowest part of the zone. In the middle of PAZ 1, Pinus decreased to about 28%, while Betula and Corylus peaked (Betula at 310 cm with 22.4% and Corylus at 290 cm with 28.1%). In the uppermost part of PAZ 1, Pinus recovered and Betula and Corylus values were lower. The percentages of the Quercetum Mixtum taxa (QM = Quercus, Tilia, and Ulmus) were low in the lower part of the zone but reached slightly higher values toward the upper part. Polypodiaceae appeared in the upper part of the zone, at 270 cm, reaching relatively high values (7.9%).

PAZ 2 (260–190 cm, 7440–6210 cal. yr BP, 5490–4260 BC)

Pinus, Corylus, and Betula were still the most dominant taxa in the pollen record; however, the occurrences of herbs and QM taxa increased. Poaceae grains <40 μm reach higher values than in PAZ 1 (2.2%); these are interpreted as pollen from wild grass, including reeds. In addition, Ericaceae, Cyperaceae, Artemisia, Chenopodiaceae, Ranunculaceae, and Rumex appeared more frequently than earlier. Artemisia and Chenopodiaceae can be described as indicators of ruderal communities while Ranunculaceae and Rumex are described as indicators of wet meadows and pastures (Behre, 1981). The general trend within PAZ 2 was that the tree pollen decreased in favor of pollen from herb taxa, of which some can indicate human disturbance. Since charcoal also displays higher values, it is possible that fire was used by humans to open up the landscape at this time.

PAZ 3 (190–97 cm, 6210–4680 cal. yr BP, 4260–2730 BC)

The most dominant taxa in this zone were Pinus, Betula, Corylus, and Alnus. Herbs such as Cyperaceae, Artemisia, and Rumex were all fairly consistent throughout the zone, whereas Chenopodiaceae occurred more sporadically. Equisetum spores also appeared relatively consistent throughout PAZ 3, whereas Polypodiaceae increased toward the end of zone 3. Charcoal values were relatively consistent throughout the zone.

PAZ 4 (97–40 cm, 4680–2350 cal. yr BP, 2730–400 BC)

The onset of PAZ 4 was defined by the introduction of the human impact indicator taxa Hordeum-type and Plantago lanceolata in the pollen record. The presence of cereal pollen of Hordeum-type indicates cultivation, while Plantago lanceolata commonly indicates grazing (Behre, 1981). Two pollen grains from the Poaceae family, >40 μm in diameter, and not identified as Hordeum-type, were classified as being Cerealia-type, since it was not possible to identify them to a lower taxonomic level. A marked change in the pollen composition of PAZ 4 is the large increase in Poaceae <40 μm. The grass curve influenced the pollen record to such a high degree that Poaceae, together with Pinus, Corylus, and Betula, were the most abundant taxa in the pollen record of PAZ 4. The increase in Poaceae <40 μm could be interpreted as an indication of either opening up of the landscape or wild grass and reeds growing in the Lina myr basin. Tree pollen percentage decreased while cultivation and grazing was underway in the area. This supports the idea that a large part of the increase in grass pollen during PAZ 4 was related to an opening up of the landscape. Pollen from trees such as Corylus, Ulmus, Tilia, and Fraxinus decreased and almost disappeared in the uppermost part of the zone, which could imply that areas were deforested and transformed into arable fields. Based on the stratigraphy and the diatom record, it could be concluded that the Lina myr area underwent a large transformation within PAZ 4. In the lowermost part of the zone where gyttja was deposited, the Lina myr basin was a bay connected to the Littorina Sea. Later the bay was isolated from the sea and a lake phase began, coinciding with the deposition of the calcareous gyttja layer. With time the lake became a fen with accumulation of fen peat. The increase of Cyperaceae during the fen phase indicated that sedges were growing on the fen surface. It is also possible that the Poaceae peak was partly related to wild grass or reeds growing in and around the fen. In the uppermost part of PAZ 4, spores of Equisetum and Polypodiaceae increased. The spores of Equisetum may represent several different species associated with wet or moist environments. It is therefore likely that the increase in Equisetum was related to the formation of the fen. This indicates that Equisetum grew together with Cyperaceae on the fen surface during this phase. The Polypodium spores are most likely from the species Polypodium vulgare, which is a fern that could be related to dry pastures and grazed forests (Behre, 1981), adding further indications of grazing to the record.

Figure 5.

Pollen percentage diagram for Lina myr.

Diatom record

A total of 163 diatom taxa were identified and seven diatom assemblage zones (DAZ 1–7) were established (Figure 6). No diatoms were observed between 400 and 338 cm where the sediment consisted of clay and sand.

DAZ 1 (339–307 cm, 9300–8500 cal. yr BP, 7350–6550 BC)

The diatom composition was characterized by freshwater benthic taxa. Dominant genera are Gomphonema spp. and Eunotia spp., constituting about 30–35% and 20–25%, respectively. Indifferent benthic taxa made up the second largest proportion and increase upward from 20% to 45%, mainly due to a rise of Epithemia spp. Ancylus Lake taxa, represented by Aulacoseira islandica and Cyclotella radiosa reached their highest values in this zone, comprising c.10%.

DAZ 2 (307–290 cm, 8500–8050 cal. yr BP, 6550–6100 BC)

This was based on one diatom assemblage and was dominated by taxa with a wide range of salinity preferences, for example, Epithemia adnata, Epithemia argus, and Epithemia goeppertiana. Halophilous and freshwater taxa constituted 15–20% each. Lagoonal taxa, that is, brackish water taxa, reached their peak at 7%, due to an increase of Campylodiscus bicostatus. Brackish-marine taxa remained low (<15%) and was composed of Diploneis smithii + v. rhombica. The Ancylus Lake taxa almost disappeared and planktonic species were negligible.

DAZ 3 (290–275 cm, 8050–7700 cal. yr BP, 6100–5750 BC)

This was characterized by a predominance of the coastal brackish-marine diatom Melosira westii. The second largest proportion comprised halophilous benthic species, especially Epithemia turgida. Indifferent and freshwater taxa decreased in comparison with DAZ 2 (<10%).

DAZ 4 (275–202 cm, 7700–6400 cal. yr BP, 5750–4450 BC)

This displayed even shares of brackish-marine and halophilous taxa. However, Melosira westii disappeared and was replaced by Paralia sulcata as a dominant coastal brackish-marine species. The halophilous Epithemia turgida constituted 40–50%. Other coastal brackish-marine species, for example, Hyalodiscus scoticus, Synedra crystalline, and Navicula peregrina, increase making up 5–7% each (refer to Supplemental Material, available online).

DAZ 5 (202–150 cm, 6400–5650 cal. yr BP, 4450–3700 BC)

This was dominated by coastal brackish-marine taxa reaching c. 90%. A predominant species was Paralia sulcata, culminating in the middle of the zone. Halophilous taxa showed a substantial decrease when compared with DAZ 4, but remained the second largest group. Other coastal brackish-marine taxa, for example, Diploneis didyma, rose sharply to 45% in the uppermost 10 cm of the zone.

DAZ 6 (150–79 cm, 5650–3800 cal. yr BP, 3700–1850 BC)

This was dominated by coastal brackish-marine taxa reaching c. 50%. Cocconeis scutellum, Navicula peregrine, and Mastogloia spp. were the most common taxa. Indifferent benthic taxa increased upward and constituted the second largest portion (30–40%) with 5–10% of Navicula rhyncocephala, Amphora ovalis, and Amphora pediculus.

DAZ 7 (79–53 cm, 3800–2600 cal. yr BP, 1850–650 BC)

This was characterized by low diatom abundance. Freshwater benthic species, for example, Cymbopleura inaequalis, Eunotia spp., and Gomphonema spp., were dominant. Occasional indifferent and brackish-marine species were also present.

Figure 6.

The diatom summary diagram. Dominating taxa for the various groups are: Aulacoseira islandica (Ancylus), Paralia sulcata and Pseudopodosira westii (coastal brackish-marine), Chaetoceros spp. resting spores (planktonic brackish-marine), Epithemia turgida (halophilous benthic), Campylodiscus bicostatus (lagoonal), Epithemia adnata (indifferent benthic), Stephanodiscus medius (indifferent planktonic), Gomphonema spp. (freshwater benthic), and Pantocsekiella ocellata (freshwater planktonic).

Discussion

Shore displacement reconstruction

The shore displacement curve shown in Figure 7 illustrates nine phases based on stratigraphy and diatoms combined with geochemical parameters. The aim of this curve is to tentatively reconstruct past relative sea-level fluctuations. The shore displacement curve is based on the elevation of the Lina myr threshold and the highest shoreline for the Littorina Sea (22 m). This highest shoreline is based on the occurrence of beach ridges and wave washed material found 21–22 m a.s.l. close to Lina myr.

Figure 7.

The tentative shore displacement curve constructed from investigations of the master core from Lina myr and comparisons for other studies around the Baltic Sea (modified from Barliaev, 2017). L1, L2, and L3 represent transgressional sub-phases of the Littorina Sea.

Stratigraphic data from phase I (prior to 8980 cal. yr BP) indicated that Lina myr was submerged beneath the Ancylus Lake. Thin layers of sand above the clay indicated reworking of glacial till through wave action and possible erosion. According to Svensson (1989), the Ancylus regression was rapid in the beginning (5–10 m) followed by a slowdown. Svensson (1989) also stated that the first isolation of the Lina myr basin, after the Ancylus Lake stage, occurred around 8600 radiocarbon yr BP, that is, c. 9550 cal. yr BP, which is c. 500 years older compared with our model for Lina myr. This difference in age could be attributed to differences between the age-depth model in this study and that used by Svensson (1989). In sediments low in organic matter, Svensson (1989) used correlations in regional pollen stratigraphies to date isolation levels. Organic carbon and carbonate content was low, as was the Br/Ti ratio, indicating sediment accumulation in a freshwater environment. Diatoms were not present perhaps due to poor preservation in these sediments.

Phase II (8980–8500 cal. yr BP) represented an isolation of the basin from the Ancylus Lake allowing the formation of calcareous gyttja, which was accumulated in a small freshwater lake, as indicated by Eunotia spp. and Gomphonema spp. (refer Supplemental Table, available online, for diatom groupings). Some occurrences of possibly reworked Ancylus Lake diatoms were also identified, for example, Aulacoseira islandica. The C/N ratio displayed values between 11 and 17 indicating mixed sources for terrestrial and aquatic organic matter. The Br/Ti ratio was relatively low indicating the presence of freshwater. Organic content was relatively low but carbonate content was high caused by the presence of shells. The freshwater lake phase persisted until the Littorina transgression (L1) inundated the area with marine water.

Diatom analysis from phase III (8500–8050 cal. yr BP) suggested that the lower portion of the gyttja represents a lagoonal stage with brackish conditions (Miller, 1986). This was supported by a decrease in the C/N ratio, which indicated a shift toward marine influence. The water depth and salinity were still low, as non-brackish and benthic diatoms dominated; this was also reflected in the relatively low Br/Ti ratio. This interpretation is supported by a peak in lagoonal taxa, for example, Campylodiscus clypeus and Campylodiscus bicostatus, which is characteristic for this type of environment (Alhonen, 1972; Eronen et al., 2001; Miller, 1986).

The timing of the onset of the Littorina transgression is in agreement with L1 in Blekinge (Berglund et al., 2005). It also coincides with the initiation of the transgression at Narva-Luga lowland and Pärnu area (Rosentau et al., 2013; Sandgren et al., 2004; Veski et al., 2005).

The initial period of phase IV (8050–7700 cal. yr BP) was characterized by the predominance of Melosira westii, indicating an environment with moderate salinity (Miller, 1973). Combined with the increase of the Br/Ti ratio, this represents evidence of a relative sea-level rise. The C/N ratio decreased sharply also indicating an increase in marine influence.

Throughout phase V (7700–6400 cal. yr BP), C/N ratios were relatively low, indicating a stronger marine influence. Melosira westii was replaced by another coastal brackish-marine diatom, Paralia sulcata, which tolerates higher salinity and has a competitive advantage in deep water (Lie et al., 1983; McQuoid and Hobson, 1988; Zong, 1997). Paralia sulcata peaked around 7500–7400 cal. yr BP coinciding with the highest Br/Ti ratios. This suggests the culmination of a transgressional sub-phase (L2). Salinity remained consistent during this period as indifferent diatom species still comprised a large share. This suggested sub-phase of the Littorina transgression is in agreement with studies from Blekinge, southern Sweden (Berglund et al., 2005; Miller and Robertsson, 1981), the eastern Gulf of Finland (Rosentau et al., 2013), the Karelian Isthmus (Miettinen et al., 2007), and the Kattegat region (Bjørnsen et al., 2008; Rasmussen et al., 2018).

The L2 sub-phase was followed by a decrease in planktonic diatoms and in the Br/Ti ratio. This corresponded to a possible regression as the sea-level rise was overtaken by postglacial isostatic uplift. We find that a regression took place between 7400 and 7000 cal. yr BP, which coincides with still stand conditions at the Karelian Isthmus (Miettinen et al., 2007) and a regression in Blekinge (Berglund et al., 2005) and the eastern Gulf of Finland (Rosentau et al., 2013).

After L2, low proportions of planktonic brackish-marine taxa indicate a low sea level. The relatively low Br/Ti ratios support a decrease in marine influences. It is, however, impossible to estimate the absolute water depth lowering.

The dominance of brackish-marine planktonic diatoms and a peak in the Br/Ti ratio in phase VI (6700–6200 cal. yr BP) may indicate an increase in water salinity and an increase in relative water depth around 6100 cal. yr BP. C/N ratios were relatively low at this time supporting the premise that aquatic organic matter input was relatively higher. This increase in salinity is in agreement with Westman et al. (1999), who suggested the highest salinity conditions (about 20‰) of the Littorina Sea between 6500 and 5000 cal. yr BP. This peak was caused by an increased volume of saline water inflow from the Atlantic as a result of the enlarged cross-section of the Danish straits (Gustafsson and Westman, 2002; Westman et al., 1999). As the study lacks any proxy for direct estimation of past water depths, it is difficult to assess whether the Littorina limit (i.e. 22 m a.s.l. at Väänä Lagoon NW Estonia, (Saarse et al., 2009a) was reached during the L2 or L3 sub-phase.

The L3 sub-phase, recorded in the older part of phase VI, was followed by a regression, which led to the eventual isolation of the basin. No evidence of minor transgressions after 6100 cal. yr BP was found, despite that indications of minor transgressions have been detected in eastern Svealand and Blekinge, Sweden (Berglund et al., 2005; Risberg et al., 2005). Planktonic brackish-marine diatoms were gradually replaced by coastal brackish-marine taxa indicating salinity remained relatively high, whereas water depth was falling. This supports the premise that the global eustatic sea level was stable, while isostatic land uplift became the main factor in shoreline displacement on Gotland.

Phase VII (5700–3800 cal. yr BP) was characterized by the dominance of coastal brackish-marine diatom taxa as a result of isostatic uplift and basin infilling. The final isolation from the Littorina Sea occurred 3800 cal. yr BP when the sea level dropped below the Lina myr basin threshold, which had probably been eroded and was lower than during the previous isolation event. The younger freshwater stage, phase VIII (3800–2700 cal. yr BP), was characterized by a lower water level due to erosion of the threshold and further sediment infilling of the basin. The calcareous gyttja deposits which accumulated during phase II were thicker and geographically more widespread than the more recent calcareous gyttja deposits. This indicated that the older isolated freshwater lake was larger than the younger isolated phase. The scarcity of diatoms frustules circumstantially indicates a high pH environment since diatoms are often poorly preserved in high pH environments (Flower and Ryves, 2009). An increase in the C/N ratio indicates a shift to higher inputs of terrestrial organic matter. The relatively high Br/Ti ratio may be caused by an increase in organic matter (Kalugin et al., 2013) rather than marine water input. As the basin shallowed over time, due to continuing input of sediments, the lake became overgrown. Accumulation of fen peat began 2700 cal. yr BP at the sampling site, which is located in the middle of the fen.

Phase IX (2700 cal. yr BP until present) is represented by the formation of a fen dominated by Cladium mariscus, a species of the sedge family typical for Gotland which thrives on soft ground with lime sledge and generally grows in fens and lakes less than 0.5 m deep. Comparisons between the pollen and diatom assemblage zones are shown in Figure 8.

Figure 8.

Summary of findings. L1, L2, and L3 represent transgressive sub-phases of the Littorina Sea.

Vegetation reconstruction and human impacts

Early humans and the environment

The first human settlements on Gotland are from about 9200 cal. yr BP (Apel et al., 2018). Population numbers are interpreted to be generally low during the Mesolithic, with periods of variation in settlement intensity and probably a marked decline during the Late Mesolithic. During part of this period, there is evidence of hunting for marine mammals (Apel and Storå, 2017). One of few Mesolithic sites found on Gotland is at Svalings, close to Lina myr. The site was dated to about 7000 cal. yr BP by radiocarbon dating of seal bones (Apel et al., 2018). Seal bones are prone to having an inbuilt marine reservoir age; the study by Apel et al. (2018) attempted to correct for this by subtracting 90 years from the uncalibrated ages. As seen from the C/N ratio and diatom data in the present study, part of the Mesolithic period is characterized by increased marine influence (diatom phase V: 7700−6400 cal. yr BP), and a sub-phase of the Littorina transgression. Hunter-gatherer populations are difficult to detect in pollen records since they generally have low impact on their surrounding vegetation. The first Early Neolithic finds on Gotland are dated to about 5800 cal. yr BP; however, the Early Neolithic culture could have been introduced about 200 years earlier, around 6000 cal. yr BP (Apel et al., 2018). The Early Neolithic finds belong to the Funnel Beaker Culture (FBC), which is the first farming society in Northern Europe and Scandinavia. The FBC complex, typical for the Early Neolithic period, lasts until about 5000 cal. yr. BP on Gotland (Apel et al., 2018). Studies from Scania, southern Sweden (Gron et al., 2015), and Mälardalen, eastern central Sweden (Isaksson and Hallgren, 2012), show that cattle were used for dairy farming during this period. The earliest evidence for cattle farming in Sweden comes from Almhov, Scania, and is dated to 5900−5450 cal. yr BP (3950−3500 cal. BC). It is believed that the Early Neolithic culture on Gotland had connections with Southern Scandinavia based on finds such as imported flint axes, amber, and possibly even pottery (Österholm, 1989). During the Early Neolithic, pigs were brought to Gotland; however, bones from sheep/goat and cattle are first found during the Middle Neolithic (Apel et al., 2018).

Gotland’s first collective burial site, from the late Early Neolithic is located at Ansarve in eastern Gotland. Analyses of bone material dated to 5300−4900 cal. yr BP indicates mainly terrestrial food intake for the studied individuals (Apel et al., 2018). Terrestrial diets may have been more important than marine at this time; however, impacts on vegetation are not evident in the pollen record.

Human impacts on vegetation were first seen clearly in the Lina myr pollen diagram at 4580 cal. yr BP (lower part of PAZ 4). This is after pigs are thought to have been introduced to Gotland and after the first identification of dairy farming in Scandinavia (Gron et al., 2015). Since the pollen record from Lina myr is interpreted to represent general regional vegetation change for large parts of Gotland, this indicates that humans began to change the general vegetation patterns on the island from the Middle Neolithic period (5000 cal. BP to 4200 cal. yr BP), associated with the Pitted Ware Culture (PWC). According to frequency distribution of dates for the PWC culture, this culture is linked to a marked population intensification on Gotland (Apel et al., 2018). The most important pollen taxa indicating human impact in PAZ 4 is Hordeum-type (barley), which first appeared in the pollen record at 4580 cal. yr BP and then occurs continuously throughout the zone. Hordeum itself can be associated with natural, wet habitats (Kuusk et al., 1979). However, since Hordeum pollen appeared at the same time as other signs of human impacts in the pollen record, and since it is not present in any previous zone, it is most likely that the Hordeum pollen grains seen in PAZ 4 are the first signs of cultivation introduced to Gotland by early farmers. This interpretation is supported by studies from around the Baltic Sea where Hordeum-type pollen first appear in pollen diagrams at about the same time, which is generally interpreted as signs of onset of cultivation. In studies of several Estonian sites, Hordeum was recorded as the earliest occurring cereal pollen, sometimes together with Avena. Some of the earliest identifications of Hordeum in Estonia are dated to 5161 cal. yr BP (Pirrus and Rõuk, 1998), 4700 ¹⁴C yr BP (Veski, 1998), 4650 cal. yr BP (Jaanits et al., 1982), and 4580 cal. yr BP (Poska et al., 2004). Similarly, Hordeum is also the earliest cereal seen in Latvia, where it has been found from the Late Neolithic from about 4000 to 3500 cal. yr BP and onward (Rasiņš and Taurina, 1983). The earliest signs of possible human impact in Uppland, eastern central Sweden, are two Hordeum-type pollen grains from the Late Mesolithic, dated to about 6250 BP/4300 BC (Karlsson, 2007). Although other microfossil counts show a vague support for this being true signs of early cultivation, the fact that no cereal grains are found in the area during the following Early Neolithic period renders the suspicion that these early Hordeum-type pollen grains from Uppland are from wild grass (Karlsson, 2007). The next occurrences of pollen linked to cultivation in Uppland are seen in the Middle Neolithic period, with pollen from cereals (Hordeum and Triticum-type) and Plantago lanceolata.

Within PAZ 4, Plantago lanceolata occurs at the same depths as the Hordeum-type pollen. Generally, Plantago lanceolata is interpreted as an indicator of grazing and cattle breeding linked to the environments of meadows and pastures (Behre, 1981; Karlsson, 2007); however, Plantago lanceolata is also common in rotational cultivation where it usually indicates undisturbed grasslands and ley farming (Burrichter, 1969). Other indicators of human impact in PAZ 4 are signs of deforestation and opening of the landscape seen as general decrease of pollen from trees and shrubs, for example, the decrease in Corylus (hazel) from 48% at the onset of PAZ 4 to 4% at the end. The increase of Poaceae from 2% to 21% at the same time supports the premise of an opening up of the landscape, since Poaceae is common on grazed and ruderal surfaces. However, part of the increase in Poaceae could also be caused by an increase in Phragmites related to the onset of fen conditions in the Lina myr basin. Within PAZ 4 herb taxa also increase in number, which can be used as an indicator of a rise in human impact since human-induced disturbance on the landscape is linked to floristic diversity and palynological richness (Berglund et al., 2008).

Lina myr was connected to the Littorina Sea until the Late Neolithic when it became a lake at about 3820 cal. yr BP. It is possible that Lina myr was once part of an inland water system; however, this must then have been before 3820 cal. yr BP and hence during the Mesolithic and Neolithic BP and hence during the Mesolithic and Neolithic periods. Gothemshammar (dated to around 2850−2450 cal. yr BP) and other Bronze Age finds in the area such as Tjelvar’s grave (dated to 3056−2941 cal. yr BP) and a location named ‘Killingborgen’ close by were likely used as ritual sites. The large cairn Majsterrojr just to the south of Tjelvar’s grave is a type of structure built in the Early Bronze Age (around 3500 cal. yr BP) but probably served as a ritual site for several thousand years (Martinsson-Wallin and Wehlin, 2017). These structures were built after the isolation of the lake from the Littorina Sea. After the isolation, the basin was a lake for about 1000 years, until the Late Bronze Age, when it became a fen. It is likely that the relatively large lake that filled the Lina myr basin was a good resource for fish and freshwater for the Bronze Age population around Lina myr and the Gothem River. The Gothem River connected the lake with the Sea to the east and the Hörsne River. During the early part of the Middle Bronze Age, at 2920 cal. yr BP, pollen grains of Cerealia-type first appear in the Lina myr pollen record. The lake in the Lina myr basin became a mire in the beginning of the Early Bronze Age. Lina myr was still connected to the Gothem and Hörsne Rivers and some areas of open water until the mire was drained in the 1940–1950s.

Conclusion

In the earliest part of the Lina myr record (9000−8980 cal. yr BP), the basin was submerged by the Ancylus Lake. At about 8980 cal. yr BP, the basin threshold located at the minimum altitude of 12 m a.s.l. emerged and the basin was isolated for the first time. Between 9500 and 8500 cal. yr BP, a freshwater lake with a maximum depth of 5 m and highly reducing conditions existed in the Lina myr basin. The basin became a bay of the Littorina Sea during a transgression, which initiated at about 8500 cal. yr BP. Two major sub-phases of the Littorina transgression could be identified. The first lasted from 8100 to 7500 cal. yr BP (L2), and the second from 6500 to 6000 cal. yr BP (L3). The highest Littorina shoreline was formed during one of these sub-phases. The highest salinity was reached during the second sub-phase (L3). This study agrees about the timing of the Littorina transgression with other studies with similar patterns of postglacial isostatic uplift. These findings suggest that these events were trans-Baltic. From 6000 cal. yr BP, the area experienced a regression that eventually led to the second isolation of the basin at about 3820 cal. yr BP. The following lake phase lasted for about 1000 years. During this time the lake slowly became shallower and at 2700 cal. yr BP also the central parts of the basin had transformed into a fen. Since the Lina myr basin is relatively large, the pollen record from the Lina myr core mainly represents the vegetation of most of Gotland. Prior to the Neolithic, most vegetation changes were due to ecological factors such as adaptation of plants to changing environmental conditions due to, for example, changing sea levels in connection with the Littorina transgressions. At 4580 cal. yr BP, in the Middle Neolithic, barley, a clear indicator of cultivation, is observed for the first time in the pollen record. Barley continues to be present in the pollen record together with Plantago lanceolata, a plant indicating grazing and/or undisturbed grasslands and ley farming. From the Middle Neolithic period to the uppermost part of the pollen record (2500 cal. yr BP), human transformation of the vegetation of Gotland is seen through a marked decrease in pollen from trees and shrubs (mainly hazel) and an increase in grass pollen and general floristic diversity. At 2920 cal. yr BP, in the early part of the Middle Bronze Age, a cereal other than barley was introduced in the pollen record.

It is clear from the archeological record that sites surrounding Lina myr were of great importance to Gotland’s prehistoric inhabitants. Perhaps this importance was due to its coastal location at this time which may explain why stone ship settings are orientated in a different way to stone ship settings on the rest of Gotland; the bow of many of these ships would have been oriented toward the modern day location of Lina myr. Although people have been in this area for around 7000 years, no obvious human impacts were seen in the vegetation record in the Mesolithic and Early Neolithic. Compared with the known ages of archeological sites, there was a lag time between the arrival of pigs to Gotland (Early Neolithic) and farming or animal husbandry to the opening up of the landscape at Lina myr. The intensification of land-use in the Middle and Late Neolithic may be due to increased populations and the pressure to produce food.

Supplemental Material

Diatom_diagram_supplementary_data-01-01 – Supplemental material for Landscape development at Lina myr fen, Eastern Gotland, 9000−2500 cal. yr BP

Supplemental material, Diatom_diagram_supplementary_data-01-01 for Landscape development at Lina myr fen, Eastern Gotland, 9000−2500 cal. yr BP by Nichola Ann Strandberg, Aleftin Barliaev, Helene Martinsson-Wallin, Jan Risberg, Martina Hättestrand, Ian Croudace, Malin Kylander and Yusuke Yokoyama in The Holocene

Supplemental Material

Diatom_groupings_with_reference_list – Supplemental material for Landscape development at Lina myr fen, Eastern Gotland, 9000−2500 cal. yr BP

Supplemental material, Diatom_groupings_with_reference_list for Landscape development at Lina myr fen, Eastern Gotland, 9000−2500 cal. yr BP by Nichola Ann Strandberg, Aleftin Barliaev, Helene Martinsson-Wallin, Jan Risberg, Martina Hättestrand, Ian Croudace, Malin Kylander and Yusuke Yokoyama in The Holocene

Supplemental Material

Supplemenary_documents_most_common_diatom_taxa-01 – Supplemental material for Landscape development at Lina myr fen, Eastern Gotland, 9000−2500 cal. yr BP

Supplemental material, Supplemenary_documents_most_common_diatom_taxa-01 for Landscape development at Lina myr fen, Eastern Gotland, 9000−2500 cal. yr BP by Nichola Ann Strandberg, Aleftin Barliaev, Helene Martinsson-Wallin, Jan Risberg, Martina Hättestrand, Ian Croudace, Malin Kylander and Yusuke Yokoyama in The Holocene

Supplemental Material

Table_1 – Supplemental material for Landscape development at Lina myr fen, Eastern Gotland, 9000−2500 cal. yr BP

Supplemental material, Table_1 for Landscape development at Lina myr fen, Eastern Gotland, 9000−2500 cal. yr BP by Nichola Ann Strandberg, Aleftin Barliaev, Helene Martinsson-Wallin, Jan Risberg, Martina Hättestrand, Ian Croudace, Malin Kylander and Yusuke Yokoyama in The Holocene

Footnotes

Acknowledgements

We would like to thank the landowners who allowed us to conduct fieldwork. Many thanks to Erik Wallin and Anton Uvelius for assistance with the coring. We are grateful to Stefan Wastegård for comments on our masters’ theses. A huge thanks to the late Sven Karlsson, to whom this publication is dedicated, for help with identification of pollen grains and assistance in the laboratory. We are also grateful to Paul Wallin for input concerning the archeology. Many thanks to Alexander Strandberg for proof reading the manuscript and to Nathaniel Baurley for support and help to produce the map. Thanks to the anonymous reviewers for their comments.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by Institute of Archeology and Ancient History, Uppsala University, the Berit Wallenberg Foundation, and by the Gerard De Geer Foundation 2016 stipend.

ORCID iD

Nichola Ann Strandberg

Supplemental material

Supplemental material for this article is available online.

References

Alhonen

(1972) The stages of the Baltic Sea as indicated by the diatom stratigraphy. Acta Botanica Fennica 92. Available at: https://helda.helsinki.fi/bitstream/handle/10138/36720/299944_092_1971.pdf?sequence=1&isAllowed=y.

Apel

Storå

(2017) The pioneer settlements of Gotland: A behavioural ecology approach. In: Persson

Riede

Skar

, et al. (eds) The Early Settlement of Northern Europe – An Ecological Approach. Sheffield: Equinox Publishing, pp. 277–309.

Apel

Wallin

Storå

, et al. (2018) Early Holocene human population events on the island of Gotland in the Baltic Sea (9200-3800 cal. BP). Quaternary International 465: 276–286.

Barliaev

(2017) Paleogeography and shore displacement of eastern Gotland between 9.5 and 2.8 ka cal BP. Master’s thesis, Department of Physical Geography, Stockholm University.

Battarbee

Jones

Flower

, et al. (2002) Diatoms. In: Last

Bricks

HJB

Smol

(eds) Tracking Environmental Change Using Lake Sediments, Volume 3: Terrestrial, Algal, and Siliceous Indicators. Dordrecht: Kluwer Academic Publishers, pp. 155–202.

Behre

K-E

(1981) The interpretation of anthropogenic indicators in pollen diagrams. Pollen et Spores 23: 225–245.

Berglund

(1971) Littorina transgressions in Blekinge, South Sweden a preliminary survey. Geologiska Föreningen i Stockholm Förhandlingar 93: 625–652.

Berglund

Ralska Jasiewiczowa

(1986) Handbook of Holocene Paleoecology and Paleohydrology. Chichester: John Wiley & Sons.

Berglund

Sandgren

(1996) The early Littorina Sea environment in Blekinge: Chronology, transgressions, salinity and shore vegetation. GFF 118: 64–65.

10.

Berglund

Gaillard

Bjorkman

, et al. (2008) Long-term changes in floristic diversity in southern Sweden: Palynological richness, vegetation dynamics and land-use. Vegetation History and Archaeobotany 17: 573–583.

11.

Berglund

Sandgren

Barnekow

, et al. (2005) Early Holocene history of the Baltic Sea, as reflected in coastal sediments in Blekinge, southeastern Sweden. Quaternary International 130: 111–139.

12.

Beug

H-J

(2004) Leitfaden der Pollenbestimmung: für Mitteleuropa und angrenzende Gebiete. Munich: Friedrich Pfeil.

13.

Bjørnsen

Clemmensen

Murray

, et al. (2008) New evidence of the Littorina transgressions in the Kattegat: Optically stimulated luminescence dating of a beach ridge system on Anholt, Denmark. Boreas 37: 157–168.

14.

Björck

Svensson

(1992) Climatic changes and uplift patterns-past, present and future. Swedish nuclear fuel and waste management company. Technical report 92-38. Available at: http://www.skb.com/publication/8673/TR92-38webb.pdf.

15.

Bordovskiy

(1965) Sources of organic matter in marine basins. Marine Geology 3: 5–31.

16.

Burrichter

(1969) Das Zwillbrocker Venn, Westmünsterland, in Moor-und vegetationskundlicher Sicht. Mit einem Beitrag zur Wald-und Siedlungsgeschichte seiner Umgebung. Münster: Landesmuseum für Naturkunde zu Münster, Abhandlungen aus dem Landesmuseum für Naturkunde zu Münster in Westfalen, 31: 1.

17.

Christensen

Nielsen

(2008) Dating Littorina Sea shore levels in Denmark on the basis of data from a Mesolithic coastal settlement on Skagens Odde, northern Jutland. Polish Geological Institute, Special Papers 23: 27–38.

18.

Cleve-Euler

(1951) Die Diatomeen von Schweden und Finnland (Kungliga Svenska Vetenskapsakademiens Handlingar). Stockholm, Almqvist & Wiksells Boktryckeri AB, 163 p.

19.

Cleve-Euler

(1952) Die Diatomeen von Schweden und Finnland. Teil 5. (Kungliga Svenska Vetenskapsakademiens Handlingar, Fjärde Serien, Band 3). Stockholm, Almqvist & Wiksells Boktryckeri AB, 154 p.

20.

Cleve-Euler

(1953) Die Diatomeen von Schweden und Finnland. Teil 3. Monoraphideae, Biraphideae 1 (Kungliga Svenska Vetenskapsakademiens Handlingar, Fjärde Serien, Band 4). Stockholm, Almqvist & Wiksells Boktryckeri AB, 255 p.

21.

Cleve-Euler

(1955) Die Diatomeen von Schweden und Finnland. Teil 4. Biraphideae 2 (Kungliga Svenska Vetenskapsakademiens Handlingar, Band 5). Stockholm, Almqvist & Wiksells Boktryckeri AB, 232 p.

22.

Croudace

Rindby

Rothwell

(2006) ITRAX: Description and evaluation of a new multi-function X-ray core scanner. Geological Society 267: 51.

23.

de Boer

Tjallingii

Velez

, et al. (2014) Climate variability in the SW Indian Ocean from an 8000-yr long multi-proxy record in the Mauritian lowlands shows a middle to late-Holocene shift from negative IOD-state to ENSO-state. Quaternary Science Reviews 86: 175–189.

24.

Dean

(1974) Determination of carbonate and organic matter in calcareous sediments and sedimentary rocks by loss on ignition: Comparison with other methods. Journal of Sedimentary Research 44: 242–248.

25.

Emerson

Hedges

(1988) Processes controlling the organic carbon content of open ocean sediments. Paleoceanography 3: 621–624.

26.

Eriksson

(1992) Natural History of Xerotherm Vegetation and Landscapes on Stora Karlsö, an Island in the Western Baltic Basin, Sweden. Uppsala: Societas Upsaliensis Pro Geologia Quaternaria.

27.

Eronen

(1974) The history of the Litorina Sea and associated Holocene events. Societas Scientarum Fennicae. Commentationes Physico-Mathematicae 44: 79–195.

28.

Eronen

Gluckert

Hatakka

, et al. (2001) Rates of Holocene isostatic uplift and relative sea-level lowering of the Baltic in SW Finland based on studies of isolation contacts. Boreas 30: 17–30.

29.

Fægri

Iversen

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

30.

Flower

Ryves

(2009) Diatom preservation: Differential preservation of sedimentary diatoms in two saline lakes. Acta Botanica Croatica 68: 381–399.

31.

García-Alix

Jiménez-Moreno

Anderson

, et al. (2012) Holocene environmental change in southern Spain deduced from the isotopic record of a high-elevation wetland in Sierra Nevada. Journal of Palaeolimnology 48(3): 471–484.

32.

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.

33.

Gron

Montgomery

Rowley-Conwy

(2015) Cattle management for dairying in Scandinavia’s earliest Neolithic. PLoS ONE 10: e0131267.

34.

Gustafsson

Westman

(2002) On the causes for salinity variations in the Baltic Sea during the last 8500 years. Paleooceanography 17: 12-1–12-14.

35.

Hallström

(1971) Boplatser och gravar på Nygårdsrum i Vallstena. Visby: Gotländskt Arkiv, pp. 114–115.

36.

Hansson

(1927) Gotlands bronsålder. Stockholm: Kungliga Vitterhetsakademien Historie och Antikvitets Akademien.

37.

Hede

(1925) Beskrivning av Gotlands silurlager. In: Munthe

Hede

Post

(eds) Gotlands geologi, en översikt. Lund: Sveriges Geologiska Undersökning, pp. 3–33.

38.

Hedenström

Risberg

(2003) Shore displacement in northern Uppland during the last 6500 calendar years. Technical report TR-03-17. Stockholm: Swedish Nuclear Fuel and Waste Management Co.

39.

Isaksson

Hallgren

(2012) Lipid residue analyses of early Neolithic Funnel-Beaker pottery from Skogsmosse, eastern central Sweden, and the earliest evidence of dairying in Sweden. Journal of Archaeological Science 39: 3600–3609.

40.

Jaanits

Laul

Lõugas

, et al. (1982) Estonian Prehistory. Tallinn: Eesti Raamat.

41.

Jones

Lawson

Reed

, et al. (2013) Diatom-inferred late Pleistocene and Holocene palaeolimnological changes in the Ioannina basin, northwest Greece. Journal of Paleolimnology 49: 185–204.

42.

Kalugin

Darin

Rogozin

, et al. (2013) Seasonal and centennial cycles of carbonate mineralisation during the past 2500 years from varved sediment in Lake Shira, South Siberia. Quaternary International 290: 245–252.

43.

Karlsson

(2007) Stone age vegetation in Uppland, southeastern Svealand, Sweden. GFF 129: 295–305.

44.

Krammer

(2000) Diatoms of Europe: The Genus Pinnularia. Königstein: A.R.G. Gantner Verlag Kommanditgesellschaft.

45.

Krammer

(2002) Diatoms of Europe: Diatoms of the European Inland Waters and comparable habitats, Volume 3: Cymbella (ed Lange-Bertalot

). Ruggel: A.R.G. Gantner Verlag K.G., 584 pp.

46.

Krammer

(2003) Diatoms of Europe: Diatoms of the European Inland Waters and Comparable Habitats, Volume 4: Cymbopleura, Delicata, Navicymbula, Gomphocymbellopsis, Afrocymbella (ed Lange-Bertalot

). Ruggel: A.R.G. Gantner Verlag K.G., 529 pp.

47.

Krammer

Lange-Bertalot

(1986) Bacillariophyceae. 1. Teil: Naviculaceae. In: Ettl

Gerloff

Heynig

, et al. (eds) Süsswasser flora von Mitteleuropa. Stuttgart; New York: Gustav Fischer Verlag, p. 876.

48.

Krammer

Lange-Bertalot

(1988) Bacillariophyceae. 2. Teil: Bacillariaceae, Epithemiaceae, Surirellaceae. In: Ettl

Gerloff

Heynig

, et al. (eds) Süsswasserflora von Mitteleuropa. Jena: VEB Gustav Fischer Verlag, p. 596.

49.

Krammer

Lange-Bertalot

(1991a) Bacillariophyceae. 3. Teil: Centrales, Fragilariaceae, Eunotiaceae. Süsswasserflora von Mitteleuropa. Stuttgart; Jena: Gustav Fischer Verlag.

50.

Krammer

Lange-Bertalot

(1991b) Bacillariophyceae. 4. Teil: Achnanthaceae, Kritische Ergänzungen zu Navicula (Lineolatae) und Gomphonema, Gesamtliteraturverzeichnis Teil 1-4. In: Ettl

Gärtner

Gerloff

, et al. (eds) Süsswasserflora von Mitteleuropa. Stuttgart; Jena: Gustav Fischer Verlag, p. 437.

51.

Kuusk

Talts

Viljasoo

(1979) Flora of Estonian SSSR. Tallinn: Valgus.

52.

Laffaille

Feunteun

Lefebvre

, et al. (2002) Can thin-lipped mullet directly exploit the primary and detritic production of European macrotidal salt marshes? Estuarine, Coastal and Shelf Science 54: 729–736.

53.

Lange-Bertalot

(2001) Diatoms of Europe: Diatoms of the European Inland Waters and Comparable Habitats, Volume 2: Navicula sensu stricto, 10 general separated from Navicula sensu lato, Frustulia. Ruggell: A.R.G. Gantner Verlag K.G., 526 pp.

54.

Lange-Bertalot

Bak

Witkowski

, et al. (2011) Diatoms of the European Inland Waters and Comparable Habitats: Eunotia and Some Related Genera. Ruggell: A.R.G. Gantner Verlag K.G.

55.

Laufeld

(1974) Silurian Chitinozoa from Gotland. Fossils and Strata 5: 1–130.

56.

Levkov

(2009) Diatoms of Europe: Diatoms of the European Inland Waters and Comparable Habitats, Volume 5: Amphora sensu lato (ed Lange-Bertalot

). Ruggell: A.R.G. Gantner Verlag K.G., 916 pp.

57.

Levkov

Metzeltin

Pavlov

(2013) Diatoms of Europe: Diatoms of the European Inland Waters and Comparable Habitats, Volume 7: Luticola and Luticopsis (ed Lange-Bertalot

). Königstein: Koeltz Scientific Books.

58.

Levkov

Mitić-Kopanja

Reichardt

(2016) Diatoms of Europe: Diatoms of the European Inland Waters and Comparable Habitats, Volume 8: The diatom genus Gomphonema from the Republic of Macedonia. Ruggell: A.R.G. Gantner Verlag K.G., 552 pp.

59.

Lie

Stabell

Mangerud

(1983) Diatom stratigraphy related to Late Weichselian sea-level changes in Sunnmøre, Western Norway. Norges geologiske undersøkelse 380: 203–219.

60.

Löwemark

Chen

Yang

, et al. (2011) Normalizing XRF-scanner data: A cautionary note on the interpretation of high-resolution records from organic-rich lakes. Journal of Asian Earth Sciences 40: 1250–1256.

61.

Lund

A-M

(1996) Stenåldersboplatser i linaområdet. En kartläggning och av boplatser runt Linaviken. Stockholm: Stockholms Universitet; Visby: Högskolan på Gotland.

62.

Lundqvist

(1928) Beskrivning till kartbladet Slite (ed Munthe

Hede

). Uppsala: Sveriges Geologisk Undersökning.

63.

Mackie

EAV

Leng

Lloyd

, et al. (2005) Bulk organic d13C and C/N ratios as palaeosalinity indicators within a Scottish isolation basin. Journal of Quaternary Science 20: 301–408.

64.

McQuoid

Hobson

(1988) Assessment of palaeoenvironmental conditions on Southern Vancouver Island, British Columbia, Canada, using the marine tychoplankton Paralia sulcata. Diatom Research 13: 311–321.

65.

Martinsson-Wallin

Wehlin

(2017) Stones in the South: Decoding Bronze Age ritual practices on Gotland. Current Swedish Archaeology 25: 227–256.

66.

Miettinen

Savelieva

Subetto

, et al. (2007) Palaeoenvironment of the Karelian Isthmus, the easternmost part of the Gulf of Finland, during the Litorina Sea stage of the Baltic Sea history. Boreas 36: 441–458.

67.

Miller

(1973) Belägg för en subboreal transgression i Stockholmstrakten. Lund: University of Lund, pp. 96–104.

68.

Miller

(1986) Ecology and palaeoecology of brackish water diatoms with special reference to the Baltic basin. In: Ricard

(ed.) Proceedings of the Eighth International Diatom Symposium. Paris: Koeltz Scientific Books, pp. 601–612.

69.

Miller

Robertsson

A-M

(1981) Current biostratigraphical studies connected with archaeological excavations in the Stockholm region. Striae 14: 167–173.

70.

Mölder

Tynni

(1967–1973) Uber Finnlands rezente und subfossile Diatomeen I-VII Bull. Bulletin of the Geological Society of Finland: 39–45.

71.

Moore

Webb

Collison

(1991) Pollen Analysis. Oxford: Blackwell.

72.

Munthe

Hede

Post

EJLv

(1927) Beskrivning till kartbladet Hemse. Stockholm: Kungl. boktryckeriet, P.A. Norstedt & söner.

73.

Ohlsson

(1984) Minnesbilder från en myr. Gutabygd: årsskrift för den gotländska hembygdsrörelsen. Visby: Gotlands hembygdsförbund, pp. 30–46.

74.

Oliva

Viau

Peros

, et al. (2018) Paleotempestology database for the western North Atlantic basin. The Holocene 28: 1664–1671.

75.

Österholm

(1989) Bosättningsmönstret på Gotland under stenåldern. En analys av fysisk miljö, ekonomi och social miljö. PhD Thesis, Stockholm University.

76.

Påhlsson

(1977) A standard pollen diagram from the Lojsta area of central Gotland. Uppsala: Societas Upsaliensis pro Geologia Quaternaria.

77.

Pettersson

(1958) Dynamik och konstans i Gotlands flora och vegetation (Acta Phytogeographica Suecica 40). Uppsala: Almqvist & Wiksells Boktryckeri AB.

78.

Pirrus

Rõuk

A-M

(1998) Human impact as revealed in lake and bog deposits of Saadjärv drumlin. In: Rõuk

A-M

Selirand

(eds) The Natural Scientific Methods in Estonian Archaeology. Tallinn: Estonian Academy of Sciences, pp. 39–53.

79.

Piva

Asioli

Schneider

, et al. (2008) Climatic cycles as expressed in sediments of the PROMESS1 borehole PRAD1-2, central Adriatic, for the last 370 ka: 1. Integrated stratigraphy. Geochemistry, Geophysics, Geosystems 9: 1–21.

80.

Poska

Saarse

Veski

(2004) Reflections of pre- and early agrarian human impact in the pollen diagrams of Estonia. Palaeogeography, Palaeoclimatology, Palaeoecology 209: 37–50.

81.

Rasiņš

Taurina

(1983) Übersicht über den Artenbestand der Kulturpflanzen und Unkräuter aus archäoloischen Ausgrabungen in der Lettischen SSR. Archaeology and Ethnography 14: 152–176.

82.

Rasmussen

Bondevik

Corner

(2018) Holocene relative sea level history and Storegga tsunami run-up in Lyngen, northern Norway. Journal of Quaternary Science 33: 393–408.

83.

Raukas

Moora

Karukäpp

(1995) About history of the Baltic Sea and early inhabitants in the Pärnu area. In: Meidla

Jõeleht

Kalm

, et al. (eds) Meidla, Liivimaa geoloogia. Tartu: Tartu University Press, pp. 119–123.

84.

Reimer

Bard

Bayliss

, et al. (2013) IntCal13 and Marine13 Radiocarbon Age Calibration Curves 0-50,000 years cal BP. Radiocarbon 55: 1869–1887.

85.

Rice

Hanson

(1984) A kinetic model for detritus nitrogen: Role of the associated bacteria in nitrogen accumulation. Bulletin of Marine Science 35: 326–340.

86.

Risberg

Alm

Goslar

(2005) Variable isostatic uplift patterns during the Holocene in southeast Sweden, based on high-resolution AMS radiocarbon datings of lake isolations. The Holocene 15: 847–857.

87.

Rosentau

Muru

Kriiska

, et al. (2013) Stone Age settlement and Holocene shore displacement in the Narva-Luga Klint Bay area, eastern Gulf of Finland. Boreas 42: 912–931.

88.

Saarse

Heinsalu

Veski

(2009a) Litorina Sea sediments of Vääna Lagoon, northwestern Estonia. Estonian Journal of Earth Sciences 58: 85–93.

89.

Saarse

Vassiljev

Rosentau

(2009b) Ancylus Lake and Litorina Sea transition on the Island of Saaremaa, Estonia: A pilot study. Baltica 22: 51–62.

90.

Sandgren

Subetto

Berglund

, et al. (2004) Mid-Holocene Littorina Sea transgressions based on stratigraphic studies in coastal lakes of NW Russia. GFF 126: 363–380.

91.

Sernander

(1894) Studier öfver den gotländska vegetationens utvecklingshistoria. Uppsala: Nya Tidnings Aktie-Bolags Tr., p. 112.

92.

Seving

(1986) Tre mesolitiska boplatser på Gotland. Ett försök till tolkning av relation och lokaliseringsmönster. Stockholm: Stockholms Universitet.

93.

SGU (2017). Available at: http://www.sgu.se/produkter/geologiska-data/.

94.

Snoeijs

(1993) Intercalibration and Distribution of Diatom Species in the Baltic Sea. Uppsala: Opulus Press.

95.

Snoeijs

Balashova

(1998) Intercalibration and Distribution of Diatom Species in the Baltic Sea. Uppsala: Opulus Press.

96.

Snoeijs

Kasperoviciene

(1996) Intercalibration and Distribution of Diatom Species in the Baltic Sea. Uppsala: Opulus Press.

97.

Snoeijs

Popatova

(1995) Intercalibration and Distribution of Diatom Species in the Baltic Sea:. Opulus Press.

98.

Stockmarr

(1971) Tablets with spores used in absolute pollen analysis. Pollen et Spores 13: 615–621.

99.

Sundberg

(2008) Västerbjers – En plats för ritualer? Visby: Högskolan på Gotland.

100.

Svensson

(1989) Late Weichselian and Early Holocene shore displacement in the Central Baltic, based on stratigraphical and morphological records from Eastern Småland and Gotland, Sweden. Lundqua Thesis 25: 195.

101.

Thomasson

(1927) Baltiska tidsbestämningar och baltisk tidsindelning vid Kalmarsund. Geologiska Föreningen iStockholm Förhandlingar 49: 21–76.

102.

Tynni

(1975) Über Finnlands rezente und subfossile Diatomeen VIII. Espoo: Geological Survey of Finland, 55 p.

103.

Tynni

(1976) Über Finnlands rezente und subfossile Diatomeen IX. Espoo: Geological Survey of Finland, 37 p.

104.

Tynni

(1978) Über Finnlands rezente und subfossile Diatomeen X. Espoo: Geological Survey of Finland, 55 p.

105.

Tyson

(1995) Sedimentary Organic Matter: Organic Facies and Palynofacies. London: Chapman and Hall.

106.

Uścinowicz

(2003) Relative Sea Level Changes, Glacio-Isostatic Rebound and Shoreline Displacement in the Southern Baltic. Warszawa: Polish Geological Institute.

107.

Veski

(1998) Vegetation History, Human Impact and Palaeogeography of West Estonia: Pollen Analytical Studies of Lake and Bog Sediments. Uppsala: Societas Upsaliensis pro Geologia Quaternaria.

108.

Veski

Heinsalu

Klassen

, et al. (2005) Early Holocene coastal settlements and palaeoenvironment on the shore of the Baltic Sea at Pärnu, southwestern Estonia. Quaternary International 130: 75–85.

109.

Wallin

(2010) Neolithic monuments on Gotland: Material expressions of the domestication process. Baltic Prehistoric Interactions and Transformations: The Neolithic to the Bronze Age 5: 39–61.

110.

Wallin

Martinsson-Wallin

(2018) Gothemshammar-a Late Bronze Age coastal rampart on Gotland. Fornvännen 113: 65–75.

111.

Wehlin

(2010) Approaching the Gotlandic Bronze Age from Sea: Future possibilities from a Maritime perspective. In: Martinsson-Wallin

(ed.) Baltic Prehistoric Interactions and transformations. The Neolithic to the Bronze Age. Gotland: Gotland University Press, pp. 89–110.

112.

Welinder

(1975) Part 1, Scandinavia: Agriculture, inland hunting, and sea hunting in the western and northern Region of the Baltic 6000 – 2000 BC. In: Fitzhugh

(ed.) Prehistoric Maritime Adaptations of the Circumpolar Zone. The Hague: Mouton Publishers, pp. 21–41.

113.

Weltje

Tjallingii

(2008) Calibration of XRF core scanners for quantitative geochemical logging of sediment cores: Theory and application. Earth and Planetary Science Letters 274: 423–438.

114.

Westman

Wastegård

Schoning

, et al. (1999) Salinity change in the Baltic Sea during the last 8,500 years: Evidence, causes and models. Technical report TR-99-38. Stockholm: Swedish Nuclear Fuel and Waste Management Co.

115.

Wohlfarth

Björck

Funder

, et al. (2008) Quaternary of Norden. Episodes 31: 73–81.

116.

Ziegler

Jilbert

de Lange

, et al. (2008) Bromine counts from XRF scanning as an estimate of the marine organic carbon content of sediment cores. Geochemistry, Geophysics, Geosystems 9: 1–6.

117.

Zong

(1997) Implications of Paralia sulcata abundance in Scottish isolation basins. Diatom Research 12: 125–150.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.31 MB

0.03 MB

1.31 MB

0.01 MB