An Arctic seal in temperate waters: History of the ringed seal ( Pusa hispida ) in the Baltic Sea and its adaptation to the changing environment

Abstract

The ringed seal (Pusa hispida) is an early immigrant in the Baltic Basin and has since its arrival experienced substantial changes in the climate, salinity and productivity of the Basin. In this paper, we discuss the dispersal and distribution of the ringed seal during different stages of the Baltic Sea in relation to past and ongoing environmental changes. Subfossil ringed seal remains around the Baltic Sea and the Danish Straits were radiocarbon dated in order to map the distribution of the species in different time periods. The δ¹³C data were used in evaluating the changes in the marine character of the Baltic Basin. The sequence of the dates indicates a continuous presence of the species in the Baltic Basin. The earliest ringed seal finds come from the Skagerrak/Kattegat area (Denmark, Swedish west coast) and date to the full glacial period and Baltic Ice Lake. In the Baltic Basin, the species appears in the subfossil record during the Ancylus period, but the main part of the remains date to the Littorina stage. During the Littorina stage, the distribution of the species was at least periodically wider than today, covering also southern parts of the Baltic. The presence of breeding populations in southern parts of the Baltic during the Holocene Thermal Maximum (HTM) indicates that the winters were at least periodically cold enough for winter ice. The changes in the marine influence in the Baltic Basin can be seen in the seal collagen δ¹³C values, which serve as a proxy for qualitative changes in water mass salinity.

Keywords

Baltic Sea dispersal distribution isotopes postglacial climate change ringed seal

Introduction

The ringed seal (Pusa hispida) is an Arctic species, dependent on ice for breeding, and capable of surviving in permanent ice and land fast winter ice. Based on radiocarbon-dated subfossil remains, the ringed seal was the first seal species to immigrate into the Baltic Basin. The dispersal was possible after the Scandinavian Ice Sheet retreated from the area and the basin was filled with glacial melting waters (e.g. Ukkonen, 2002 and references therein).

Since then, the Baltic ringed seal population has repeatedly experienced profound environmental changes in the form of varying water temperature, ice conditions, salinity and food resources, as well as competition with other seal species. In the midst of the ongoing climate change, the question has been raised whether or not this Arctic species can survive the expected drastic changes in its environment (BACC, 2008; Meier et al., 2004). Current climate models project a drastically altered ice climate for the end of this century, leading to reduced seal pup survival in the Baltic. In this scenario, only the northernmost part of the Bothnian Bay would have a relatively stable ice-winter (Meier et al., 2004; Sundqvist et al., 2012). Projected climate change is one of the factors behind the listing of the subspecies as ‘Vulnerable’ by the World Conservation Union Pinniped Specialist Group (Kovacs et al., 2012) and Baltic Marine Environment Protection Commission (Helcom, 2013).

To be able to assess the effects of the ongoing climatic change on Arctic seal populations in the Baltic Sea, much more has to be known about their history. This is made possible through the rich subfossil seal fauna found along the coasts of the Baltic Sea (e.g. Aaris-Sørensen, 1998, 2009; Daugnora, 2000; Lõugas, 1997; Schmölcke, 2008; Sommer and Benecke, 2003; Storå, 2001; Ukkonen, 2002; Zagorska, 2000).

In this paper, we present new radiocarbon dates for nearly 40 ringed seal remains along the coasts of the Baltic Sea, as well as the spatio-temporal distribution of the species during the Late Glacial and Holocene based on the new data. We discuss the history of the species in relation to changes in climate, marine character and productivity of the Baltic Sea. Furthermore, we try to assess if the Baltic ringed seal has indeed survived all environmental changes in history or if there are any signs of recolonization following local extinction.

Baltic Sea and its development after the glaciation, marine and freshwater stages

The present-day Baltic Sea (Figure 1) is one of the largest brackish water basins in the world. The mixing of freshwater mainly of riverine origin and saline North Sea water entering through the Danish Straits creates a SW–NE salinity gradient in the basin. Surface water salinity is greatest, >20 psu (practical salinity unit), in the Kattegat, 8–10 psu in the southern Baltic, 7–8 psu in the Baltic Proper and 3–5 psu in the Gulf of Finland, Bothnian Sea and the Bothnian Bay (e.g. Andersson, 2014).

Figure 1.

The Baltic Sea and its parts.

Following the final retreat of the Scandinavian Ice Sheet from the Baltic Basin, the Baltic Sea has evolved to its present state through a succession of freshwater and saline phases. The development of the basin was controlled by a complex interplay of geographically and temporally varying isostatic rebound, changes in the position and depth of sills and barriers, and the changing climate. The short review here follows Björck (2008) unless indicated otherwise. The first of the late glacial Baltic stages is the Baltic Ice Lake, a fully freshwater basin which formed c. 16,000 cal. BP in front of the receding ice sheet. Because of isostatic uplift, the level of the Baltic Ice Lake gradually rose above sea level, and an outlet for the freshwater to flow into the sea was created in the Öresund area. The freshwater stage terminated abruptly as the lake was drained through a channel in south-west Sweden within a few years, soon after the Younger Dryas cold stage c. 11,600 cal. BP.

Following the freshwater phase, starting at c. 11,400–11,300 cal. BP, the basin experienced a brief c. 150–200-year-period of more brackish water – the Yoldia Sea stage. Water salinities were highest in the shallows south of Stockholm (e.g. Schoning, 2001), with indications of brackish bottom waters also reaching the southern Baltic (Andrén et al., 2007). The beginning of the ensuing regressive freshwater stage, the Ancylus Lake, is placed around 10,700 cal. BP. The waters of the Ancylus Lake were well mixed with no permanent stratification. The Ancylus regression came to an end at c. 10,200–10,000 cal. BP with a fairly abrupt lowering of water level and a new connection to the Atlantic through the Danish Straits.

The transition from the freshwater stage to the brackish Littorina Sea conditions was multi-phased and time-transgressive: first possible signs of marine influence are recorded as early as 9800 cal. BP in the southern Baltic (Andrén et al., 2000; Berglund et al., 2005), while brackish water reached the Gulf of Finland and Gulf of Bothnia significantly later (Donner et al., 1999). The transition period, from c. 9800 to 8500 cal. BP (coastal areas; Berglund et al., 2005) or 8000 cal. BP (offshore; Andrén et al., 2000), also called the Early or Initial Littorina Sea, is characterized by very low salinities and relatively low water levels in the south (Berglund et al., 2005). The Littorina transgressions starting at c. 8000–8500 cal. BP depending on area resulted in the gradual rise of water depth and salinity (Berglund et al., 2005; Emeis et al., 2003; Lübke et al., 2011; Rössler et al., 2011; Schmölcke et al., 2006; Willumsen et al., 2013), and the establishment of brackish-marine Littorina Sea sensu stricto. The timing and magnitude of maximum salinity conditions in different parts of the Baltic Sea are still actively investigated and debated. In the coastal southern Baltic and the Baltic Proper, peak salinities and highest primary productivity were evidently reached c. 7000–5000 cal. BP (Berglund et al., 2005; Gustafsson and Westman, 2002). In the northern Baltic Proper, maximum sea surface salinities, with estimates ranging from 9 to 17 psu, were experienced c. 7000–4000 cal. BP (Gustafsson and Westman, 2002; Punning et al., 1988; Widerlund and Andersson, 2011; Willumsen et al., 2013). Further north, in the Gulf of Bothnia, peak salinities from 10 to 12 psu are observed 6000–5000 cal. BP and around 3000 cal. BP (Widerlund and Andersson, 2011). Donner et al. (1999) suggest an increase of 4 psu in surface water salinity during peak Littorina times for the Gulf of Finland and Gulf of Bothnia. It should be emphasized that ¹⁴C dates reported by various research groups from different Baltic sub-basins are not necessarily comparable because of the difficulties in constraining the spatially and temporally varying reservoir effect (see Lougheed et al., 2012, 2013 for discussion).

From the peak Littorina conditions, water salinities in the Baltic Sea have decreased (Widerlund and Andersson, 2011; Willumsen et al., 2013), most likely owing to increased freshwater input and decreased inflow from the North Sea (Gustafsson and Westman, 2002).

Behaviour and population size of the ringed seal

Ringed seals use ice for breeding, moulting (shedding skin) and resting. As the species is capable of maintaining a breathing-hole system in permanent and land fast ice, its distribution is very wide. It occurs in all seas of the Arctic Ocean. Separate populations of subspecies status exist in the Baltic Sea (Pusa hispida botnica), the Sea of Okhotsk (Pusa hispida ochotensis), Lake Ladoga (Pusa hispida ladogensis) and Lake Saimaa (Pusa hispida saimensis) (Rice, 1998).

Ringed seal populations are not found in areas without a genuine ice-winter (Dippner et al., 2008; Kelly et al., 2010; Kovacs et al., 2008; McLaren, 1958; Meier et al., 2004; Reeves, 1998; Smith and Stirling, 1975; Sundqvist et al., 2012). The ice cover typically lasts at least 2–3 months. Relatively stable ice is needed for breeding. Female seals build breeding lairs in snow on top of ice. A single pup is born and lactated for 5–6 weeks. The lair provides a warm microclimate for the pup and protects it from predation (polar bear, arctic fox in the arctic areas; wolf, red fox and birds in the Baltic). Even when the snow depth does not allow for the construction of a proper lair, the ringed seals try to find suitable structures in refrozen drift-ice. Also males and young seals build lairs in winter (Kelly et al., 2010; Reeves, 1998; Smith and Stirling, 1975).

Baltic ringed seals breed in February–April and moult on ice in April–May. In the Baltic, they can also continue moulting on land. Small rocks or shoals are the preferred land habitat. As successful breeding on land has not been documented, the ringed seal is essentially ice-dependent (Kelly et al., 2010; Kovacs et al., 2008; Reeves, 1998).

The Baltic ringed seal population exceeded 180,000 individuals in the beginning of the 20th century, but declined to about 5000 by the 1980s because of hunting and environmental pollutants (Härkönen et al., 1998). Limitation of hunting led to population recovery, and the latest estimate for the Baltic ringed seal population from 2011 is at least 10,000 individuals (Sundqvist et al., 2012). The breeding distribution now consists of four separate breeding areas in the Gulf of Bothnia, the Archipelago Sea, the Gulf of Finland and the Gulf of Riga. South of the Gulf of Riga, ringed seals are registered only occasionally.

Diet of the ringed seal

Even though the most common food item for the modern Baltic ringed seal is the Baltic herring (Clupea harengus membras), it feeds on other fish as well, for example, three-spined stickleback (Gasterosteus aculeatus), smelt (Osmerus eperlanus), sculpins (Myxocephalus), and so on (Stenman and Pöyhönen, 2005; Tormosov and Rezvov, 1978). The diet also includes crustaceans. In lakes Saimaa (Kunnasranta et al., 1999) and Ladoga (Sipilä and Hyvärinen, 1998), the diet of the ringed seal consists mainly of vendace (Coregonus albula), roach (Rutilus rutilus), smelt, perch (Perca fluviatilis) and whitefish (Coregonus lavaretus). Different Gadidae fish are the main food for the Arctic ringed seals (White Sea, Barents Sea, Arctic Ocean, etc.). Thus, its diet is depending on the availability of fish species in the water body it inhabits (Kelly et al., 2010; Kovacs et al., 2008; Reeves, 1998).

Genetic variability in the ringed seals worldwide and in the Baltic Sea

The ringed seals have a relatively continuous range in polar and temperate waters of the northern hemisphere. The analysis of large-scale patterns of their genetic variability at nuclear microsatellite loci showed little evidence for genetic differentiation (Davis et al., 2008). Of the eight local populations analysed (Alaska, five Canadian Arctic sites, Svalbard and the White Sea), only the White Sea showed significant, but moderate, differentiation from other locations (Davis et al., 2008). This implied extensive gene flow among different parts of the species range, but with significant isolation-by-distance (Davis et al., 2008). Although adult individuals show fidelity to established breeding territories (Smith and Hammill, 1981), young seals may travel substantial distances (Kapel et al., 1998; Smith, 1987; Teilmann et al., 1999), which may lead to admixture between individuals originating from different locations (Davis et al., 2008).

Palo et al. (2001) found significant differentiation at nuclear microsatellite loci between ringed seals from the Baltic Sea and from Svalbard, but comparable with that between Svalbard and the White Sea (Davis et al., 2008). This suggests restricted, but existing, gene flow between the Baltic and the North Atlantic throughout the Holocene. Consistent with these results, Martinez-Bakker et al. (2013) found low-to-moderate genetic differentiation between ringed seals from the Baltic Sea and from the Arctic (Chukchi and Beaufort Seas), in contrast to high differentiation between these populations and the land-locked ring seal population from Lake Saimaa in Finland. The number of immigrants per generation from the Arctic to the Baltic Sea was estimated at 9 by Palo et al. (2001) and at 11 (based on mtDNA) or 45 (based on microsatellite loci) by Martinez-Bakker et al. (2013). Martinez-Bakker et al. (2013) also observed ring seals’ movements using satellite telemetry and found that individuals can migrate over distances larger than 1000 km within several months. They concluded that it is plausible that individuals from the Arctic can traverse the Norwegian and North Sea and immigrate into the Baltic Sea (Martinez-Bakker et al. (2013).

Expected heterozygosity in the Baltic population was similar as in Svalbard (Palo et al., 2001) and Chukchi Sea and Beaufort Sea (Martinez-Bakker et al., 2013), suggesting that a large effective population size has been maintained in the Baltic ringed seals for most of their evolutionary history (Palo et al., 2001). The long-term effective population size of the Baltic population was estimated at about 20,000 individuals, assuming the complete isolation scenario. No genetic differentiation was detected within the Baltic Sea (Palo et al., 2001), as the contemporary existence of four breeding areas is of recent origin, and the areas are close enough to allow for gene flow.

Expected heterozygosity in the ringed seals is higher as compared with the grey seals and the harbour seals (Phoca vitulina) in European waters (Palo et al., 2001). This may be a consequence of larger population sizes of the ringed seals as compared with the other seal species (Palo et al., 2001). In the Baltic Sea, expected heterozygosity in ringed seals (Palo et al., 2001) was not substantially higher than that in grey seals (Graves et al., 2009), but the nucleotide diversity at mitochondrial DNA control region sequence was about three times higher in the ringed seals (Valtonen et al., 2012) as compared with the grey seals (Graves et al., 2009). This may suggest either high mtDNA diversity of the founder population of the Baltic ringed seals or subsequent immigration of individuals with diverse mtDNA haplotypes. The ringed seal population from Lake Saimaa, which originated from the Baltic population and became land-locked about 9500 years ago, carries mtDNA haplotypes belonging to only one of several clades occurring in the Baltic Sea (Valtonen et al., 2012). This may suggest that some of the mtDNA diversity of the Baltic ringed seals have been acquired after the Lake Saimaa was formed rather than at the initial founding event of the Baltic population.

Material and methods

The material consists of radiocarbon-dated geological and archaeological subfossil ringed seal remains found around the Baltic Sea and at Danish Straits (Table 1, Figure 2). The finds concentrate in two areas: the Gulf of Bothnia and the Danish Straits/Swedish West coast. The rapid shore displacement in the northern parts of the Baltic has exposed large areas of former sea floor, which then by cultivation or other land use have revealed seals sunk to the bottom of the sea, where the preservation conditions have been favourable because of the sediments of clay. At the Swedish west coast and in Denmark, many of the seal bones have been found in heaps of mussels or in other contexts favouring preservation of bones.

Table 1.

Radiocarbon dates from ringed seals (Pusa hispida) around the Baltic Sea.

Part of the Baltic Sea	Locality	Cat	Sample	Age BP	Lab	cal. BP^a	δ¹³C	Reference
Bothnian Bay	Rutviksberg 1, Norrbotten	NRM Z 1521	Costa	3635 ± 50	LuS-7516	3640	failed
Bothnian Bay	Rutviksberg 2, Norrbotten	NRM Z 1522	Fibula	3410 ± 50	LuS-7517	3380	−16.1
Bothnian Bay	Ähtävä	ZMH 6945	Radius	8195 ± 135	Hela-435	8820	−19.2	Ukkonen (2002)
Bothnian Bay	Oulainen	ZMH 6957	Costa	9270 ± 60	LuS-7308	10,200	−19.6
Bothnian Bay	Muhos	ZMH 6965	Cranium	5115 ± 75	Hela-436	5560	−15.9	Ukkonen (2002)
Bothnian Bay	Kovjoki	ZMH 6974	Fibula	8270 ± 80	Hela-434	8920	−19.1	Ukkonen (2002)
Bothnian Bay	Ruukki	ZMH 6977	Costa	6565 ± 50	LuS-7310	7180	−17.9
Bothnian Bay	Oulujoki	ZMH 6980	Pelvis	6450 ± 55	LuS-7311	7040	−18.1
Northern Quark	Lapua	ZMH 6953	Costa	8430 ± 55	LuS-7306	9150	−20.3
Northern Quark	Vähäkyrö	ZMH 6963	Costa	6480 ± 50	LuS-7309	7080	−16.7
Northern Quark	Ilmajoki	ZMH 6968	Costa	8345 ± 55	LuS-7305	9040	−20.3
Northern Quark	Ylistaro	ZMH 6972	Radius	8495 ± 80	Hela-439	9220	−19.7	Ukkonen (2002)
Northern Quark	Ylistaro	ZMH 6976	Pelvis	5275 ± 50	LuS-7307	5730	−16.9
Northern Quark	Nurmo	ZMH 6978	Radius	9505 ± 85	Hela-440	10,440	−19.8	Ukkonen (2002)
Bothnian Sea	Bjärtrå	NRM	Three ribs, vertebra	6085 ± 115	St-4404	6610	not given in ref.	Fredén (1975)
Bothnian Sea	Trönö, Hälsingland	NRM Z 1379	Fibula	7045 ± 60	LuS-7515	7600	−19.5
Bothnian Sea	Rogsta, Hälsingland	SHM 25265		5865 ± 50	LuS-7914	6360	−17.1
Sea of Åland	Skattmansö, Uppland	NRM P 5639a	Costa	8285 ± 55	LuS-7912	8950	−21.7
Sea of Åland	Skattmansö, Uppland	NRM P 7834a	Fibula	8450 ± 50	LuS-7911	9170	−21.6
Sea of Åland	Åloppe, Uppland	SHM 11729	Humerus	3940 ± 50	LuS-7915	4040	−16.3
Sea of Åland	Åloppe, Uppland	SHM 14389:12	Os temporale	4235 ± 50	LuS-7917	4450	−17.5
Sea of Åland	Åloppe, Uppland	SHM 14389:A	Humerus	4360 ± 50	LuS-7916	4620	failed
Gulf of Finland	Kudruküla		Os temporale	4750 ± 100	Ua-4826	5120	−17.1	Lõugas et al. (1996)
Baltic Proper M	Rojrhage Grötlingbo		Dens	4670 ± 60	Ua-17546	5020	−17	Rundkvist et al. (2004)
Baltic Proper M	Stora Förvar		Mandibula	4460 ± 80	Ua-2939	4730	−16.4	Rundkvist et al. (2004)
Baltic Proper M	Stora Förvar		Mandibula	4115 ± 75	Ua-17178	4280	−18	Rundkvist et al. (2004)
Baltic Proper, M/S	Šventoji 23	1	Radius	4550 ± 50	LuS-7692	4860	−17.7
Baltic Proper, M/S	Šventoji 4	2	Pelvis	4420 ± 50	LuS-7693	4690	−17.0
Baltic Proper, M/S	Šventoji 1	3	Pelvis	4470 ± 50	LuS-7694	4740	−17.0
Baltic Proper, M/S	Šventoji 6	4	Humerus	4370 ± 50	LuS-7695	4630	−17.7
Baltic Proper, M/S	Šventoji 3	5	Mandibula	4605 ± 50	LuS-7696	4930	−17.3
Danish Straits	Vedbæk	ZMK113/1945	Radius	6995 ± 55	LuS-7688	7510	−18.7
Danish Straits	Egegaard	ZMK 2/1874		42,300 ± 2000	LuS-7371	45,920
Danish Straits	Nivågård	ZMK 8/1960	Vertebra	14,110 ± 250	Ua-1023	16,820	not given in ref.	Lagerlund and Houmark-Nielsen (1993)
Danish Straits	Strøby Jærne	ZMK 1/1910	Humerus	3070 ± 50	LuS-7690	2880	failed
Kattegat	Uddevalla, Bräcke	GNM 10044	Femur	10,895 ± 65	LuS-7297	12,410	−12
Kattegat	Laholms tegelbruk	GNM 10103	Phal 2	13,710 ± 70^c	LuS-7301	16,370	−11.9
Kattegat	Skene	GNM 10156	Mc (juv.)	12,995 ± 70	LuS-7304	14,840	−12.8
Kattegat	Onsala, Buera	GNM 10546	Femur	10,730 ± 60	LuS-7302a	12,170	−14.8
Kattegat	Uddevalla, Kuröd	GNM 5479	Femur	10,555 ± 65	LuS-7298	11,810	−14.6
Kattegat	Mölndal	GNM 6290	Costa	11,350 ± 65^d	LuS-7302b	12,850	−13.1
Kattegat	Göteborg, Bäckebol	GNM 8921	Tibia	12,330 ± 65	LuS-7303	13,800	−13.3
Kattegat	Halland	NRM Z 0547	Costa	13,480 ± 80	LuS-7514	15,870	−12.7
Kattegat	Hørmested	ZMK4/1937	Fibula	14,140 ± 80	LuS-7687	16,880	−13.1
Kattegat	Dokkedal	ZMK13/1957	Radius	330 ± 45	LuS-7689	70	−17.8
Kattegat	Ammendrup	ZMK41/1962	Mc II	42,200 ± 1000^b	LuS-7691	45,370	−15.3
Skagerrak	Känsö	GNM 10150	Humerus	11,365 ± 65	LuS-7299	12,860	−14.2

Parts of the Baltic, see Figure 1; Cat: Museum/Institution catalogue number; GNM: Göteborg Natural History Museum, Gothenburg; NRM: Swedish Museum of Natural History, Stockholm; NRM P: Swedish Museum of Natural History, Department of Palaeozoology, Stockholm; NRM Z: Swedish Museum of Natural History, Department of Vertebrate Zoology, Stockholm; SHM: The Swedish History Museum, Stockholm; ZMH: Finnish Museum of Natural History, Helsinki; ZMK: University of Copenhagen, Zoological Museum, Copenhagen; Sample: part of the skeleton from which the sample was taken; Age BP: radiocarbon date without calibration; Lab: Abbreviation of the dating laboratory and number of the analyse; cal. BP: calibrated radiocarbon age; δ¹³C: stable carbon isotope value reported as % deviations with reference to the international standard VPDB; Reference: published dates.

Median value.

Apparent age.

A vertebra from the same material was dated earlier (13,425 ± 1601 BP, St-3606, Fredén, 1975).

A vertebra from the same material was dated earlier (10,760 ± 2701 BP, St-4102, Fredén, 1975).

Figure 2.

Radiocarbon-dated subfossil ringed seal (Pusa hispida) finds around the Baltic Sea.

Altogether, 47 ringed seal finds have been radiocarbon dated around the Baltic Sea and the Danish Straits (Table 1). Of the dates, 36 are new and 11 were published earlier by the authors or other researchers (Fredén, 1975; Lagerlund and Houmark-Nielsen, 1993; Lõugas et al., 1996; Rundkvist et al., 2004; Ukkonen, 2002). To obtain the new ¹⁴C results, the processes were essentially the same in both laboratories in Lund and Helsinki (¹⁴C lab codes LuS- and Hela-). A modified Longin method was used for the extraction of collagen (Berglund et al., 1976; Håkansson, 1976; Longin, 1971). The collagen samples were combusted to CO₂, measured for δ¹³C values by Isotope Ratio Mass Spectrometer (IRMS) and converted to pure graphite samples for ¹⁴C determination. Eventually, the Accelerator Mass Spectrometric (AMS) measurements for graphitized samples were made on the Single-Stage AMS machine at the GeoBiosphere Science Centre, Lund University (LuS-), or by Tandem Accelerator of the University of Uppsala (Hela-).

All dates were calibrated. Here, we consider two areas of interest – Skagerrak/Kattegat and Baltic Sea – separately concerning the reservoir age corrections for radiocarbon ages. There is a trend of the average reservoir ages to decrease from the coast of Norway to Skagerrak/Kattegat and further to Baltic Sea (see database at http://calib.qub.ac.uk/marine/). Particularly, average deviations (ΔR) from the ‘mean global ocean’ reservoir ages (Hughen et al., 2004; Reimer et al., 2009) can be calculated based on 24 and 8 individual measurements from Skagerrak/Kattegat and Baltic Sea, respectively, stored in the database. Based on this analysis, we have adopted ΔR values of −23 ± 75 years for Skagerrak/Kattegat and −82 ± 47 years for Baltic Sea. These corrections have been implemented into calibration procedure with OxCal 4.1 software (Bronk Ramsey, 2009) by using Marine09 calibration curve (Reimer et al., 2009). However, we also note that the complex history of the Baltic Sea induces time-dependent and spatial effects on reservoir ages that cannot be fully taken into account with information available. For simplicity, the datings are discussed according to median values (cal. BP).

The stable carbon isotope values, δ¹³C, are reported for 40 specimens altogether. A total of 31 specimens, those with a ¹⁴C lab code ‘LuS-’, were analysed at the Department of Geography and Geology of the University of Copenhagen on a Micromass Iso Prime IRMS. A subsample of the collagen extracted for ¹⁴C dating was combusted at the Lund University Radiocarbon Dating Laboratory and the gas sent in sealed glass ampoules to Copenhagen for analysis. The laboratory cites a precision of ±0.1‰ for the δ¹³C value in bone collagen. For the samples with a ¹⁴C lab code ‘Hela-’, the δ¹³C measurements were performed with IRMS (Finnigan MAT Delta-E) on combusted CO₂ sample of bone collagen. Typical analytical uncertainties were below 0.1‰ deduced from reference measurements on Oxalic Acid II standard (NIST SRM 4990C).

The collagen δ¹³C values of nine further specimens were taken from the literature (Lõugas et al., 1996; Rundkvist et al., 2004; Ukkonen, 2002). No precision estimates are given for the δ¹³C values in the original publications, but considering the usual level of analytical uncertainty reported by isotope laboratories for δ¹³C measurements, the stable carbon isotope data can be considered accurate within ±0.2‰ uncertainty. No correction for Suess-effect was done. The data are reported as permil deviations with reference to the international standard VPDB, with no adjustment for the recent change in δ¹³C values of atmospheric CO₂.

Results and discussion

Baltic ringed seal finds and their dates

Ringed seal finds included in the study, their radiocarbon dates and calibrated ages, and δ¹³C values are given in Table 1. The spatio-temporal distribution of the dated specimens is presented in Figure 2. Two of the specimens from Kattegat and Danish Straits (Ammendrup and Egegaard) gave apparent ages older than 45,000 cal. BP and are of glacial age. Several specimens found in the area of Kattegat gave ages between 17,000 and 15,000 cal. BP belonging to the deglaciation period before the onset of the Baltic Ice Lake stage. All specimens dated to Baltic Ice Lake stage (15,000–11,600 cal. BP) come from Skagerrak/Kattegat. No specimens were dated to Yoldia Sea stage (sensu lato, for example, the time period between Baltic Ice Lake and Ancylus 11,600–10,700 cal. BP). The earliest dated finds east of the Straits derive from the Ancylus stage (10,700–10,200 cal. BP) and were found at Bothnian Bay and Northern Quark. All other specimens are dated to the Ancylus/Littorina transitional stage (10,200–8500 cal. BP) or Littorina Sea (8500 cal. BP–today).

In this work, we have used a twofold division of reservoir ages according to find location and based on the present database (http://calib.qub.ac.uk/marine/, Hughen et al., 2004). A similar trend of smaller reservoir age within the Baltic Sea farther away from the marine water inflow of Skagerrak/Kattegat has been observed recently by Lougheed et al. (2013) in a study of mollusc shells (Macoma genus). Particularly, the northern parts of the Baltic Basin were observed to suffer from modest modern reservoir ages of c. 100–200 ¹⁴C years. As the Baltic Basin is presently shallower compared with the early Holocene timeframe, the marine influence was probably larger during the Littorina phase (Lougheed et al., 2012). Most of our Baltic samples were from the shores of the Gulf of Bothnia. Therefore – and in the absence of consistent proxy for time-dependency of the reservoir effect – we feel that our selection of the average reservoir ages of ~300–400 ¹⁴C years is well justified. However, we point out that (1) the samples from Gotland may, in particular, carry hardwater influence because of bedrock limestone and (2) role of ancient organic carbon from melting glaciers is unknown. Excessive amount of hardwater carbon should possibly be visible as less negative δ¹³C values of Gotland samples, reflecting the formation history of the limestone as calcium carbonate. We do not see such a deviation from the general trend of the data. Actually, during the Littorina phase, the samples from Bothnian Bay – probably carrying less hardwater effect – display even less negative δ¹³C values compared with the samples from Baltic proper (Gotland). Therefore, we believe the effect is hidden under the overall development of the δ¹³C values towards higher marine influence. As demonstrated by Hågvar and Ohlson (2013), old carbon input from melting glaciers may enter into glacier foreland food web. Being at the top of the food chain within a large basin of glacial melting waters, Baltic seals may contain such ancient carbon during the early phases of the Baltic Sea, particularly. However, as the path of carbon from low trophic levels to the top is still poorly understood, we only notify such a possible effect to make the measured ages older without addressing its magnitude.

Dispersal

The two full glacial Danish dates show that whenever there was a palaeo-Skagerrak/Kattegat during the Weichselian, the ringed seal was probably present. This means that the invasion into the Baltic could happen instantaneously as soon as the Baltic Ice Lake joined with the ocean. Ringed seals were present in the Skagerrak/Kattegat area during the late glacial at c. 16,800 at the latest and during the Baltic Ice Lake stage at 15,000–11,600 cal. BP near the initial outlet at Öresund and near Gothenburg, where the waters of the Baltic Ice Lake drained into the ocean c. 11,600 cal. BP and the basin entered its Yoldia stage (Figure 3). During this stage, there was a strait connection between the Atlantic and the Baltic Basin in central Sweden. However, the ringed seal record in the Baltic shows no finds from the Yoldia stage. The oldest two seal finds within the Baltic Basin date to c. 10,440–10,200 cal. BP and are thus from the Ancylus Lake stage. The Ancylus Lake had its outlet river first through its isolation threshold in central Sweden and later through the Dana River over the Danish Straits. It is not possible to say, at this stage, whether the species immigrated into the Baltic Basin already during the Yoldia stage or only later, during the Ancylus stage via its outlet rivers.

Figure 3.

Radiocarbon-dated subfossil ringed seal (Pusa hispida) finds during different Baltic stages. (a) Glacial period (white dots) and deglaciation (black dots) (older than 15,000 cal. BP); (b) Baltic Ice Lake Proper (15,000–11,600 cal. BP); (c) Yoldia stage (sensu lato, for example, the time period between Baltic Ice Lake and Ancylus 11,600–10,700 cal. BP); (d) Ancylus Lake 10,700–10,200 cal. BP; (e) Ancylus/Littorina transition phase (10,200–8500 cal. BP); (f) Littorina Sea, increasing salinity (8500–6000 cal. BP); (g) Littorina Sea, decreasing salinity (younger than 6000 cal. BP). Approximate configuration of ice, land and sea adopted from multiple sources.

One may argue that the dates from Ancylus stage could be affected by older carbon reservoirs than implemented within the reservoir age correction and therefore being from the Ancylus/Littorina transition phase with Danish Straits open. However, even with the reservoir correction applied, the time difference between the dates to the beginning of the transition period is of the order of 1000 years. That is significantly larger than the typical reservoir ages observed for Atlantic Ocean (http://calib.qub.ac.uk/marine/). In addition, the Bothnian Bay is not believed to suffer from hardwater effect because of lack of limestone bedrock within its surroundings (Lougheed et al., 2013) and the essential marine influence bringing in older carbon started only later (Donner et al., 1999). On the other hand, the recently observed ancient carbon release of melting glaciers (Hågvar and Ohlson, 2013) may play a role in supplying old carbon into the food chain of seals also within the Baltic Basin. We, therefore, consider our suggestion of the oldest Gulf of Bothnia seal finds dating to Ancylus period merely tentative.

The ringed seal dates form a continuous sequence from the Ancylus Lake stage to the Littorina and thus indicate a constant presence of the species in the Baltic Basin. After the initial immigration, the ringed seal population has probably gained new individuals from the Arctic population throughout its whole history in the Baltic Sea (Palo et al., 2001; Martinez-Bakker et al., 2013), but there is no indication of local extinction and recolonization of the basin. Stray ringed seals presumably of Arctic origin were found in the West-European continental coast almost yearly in 1972–1994 (Van Bree, 1996). The high genetic variability of the Baltic population may also result from a very diverse founder population and a continuously large population size through Holocene (Palo et al., 2001).

Distribution and ecology

Ringed seal was present in the Baltic Basin during the freshwater Ancylus stage as well as during the Littorina Sea stage initially with increasing salinity but later with decreasing salinity levels. The ringed seal was obviously not dependent on marine fish resources, but fed on freshwater fish, such as three-spined stickleback, whitefish, smelt, sculpins, and so on. These fish species are considered as first immigrants in the Baltic Basin after the retreat of glaciers. Their Late Pleistocene bone remains are rare, but the colonization of newly opened water bodies by cold-adapted fish is evident (Enghoff et al., 2007; Lõugas et al., 2013; Paaver and Lõugas, 2003). Later, at the Littorina Sea stage, the ringed seal met saline water fish such as the cod (Gadus morhua) and herring.

With the opening of the Danish Straits and the gradual inflow of marine waters into the Baltic Basin, the productivity of the sea increased (e.g. Lepiksaar, 1986; Lõugas, 1997, 1999), which was probably reflected in the population size of the ringed seal. At the same time, however, new seal species entered the basin, which could have meant increased competition. It seems, however, that the more pelagic harp seal (Pagophilus groenlandicus) and the littoral ringed seal managed to cohabit the Northern Baltic Basin for a long period of time (Bennike et al., 2008; Lõugas, 1997, 1999; Storå, 2001; Ukkonen, 2002). Of the grey seal (Halichoerus grypus), on the other hand, very little evidence exists in the northern part of the Baltic Basin (Ukkonen, 2002). Probably, the more pelagic grey and harp seals did not significantly compete with the littoral ringed seal for food resources and breeding areas (e.g. Suuronen and Lehtonen, 2012; Wathne et al., 2000).

During the Littorina Sea stage, the species had apparently a more southerly distribution than today. From the Ancylus/Littorina transition phase, there are no records from southern Baltic in our data, but a rich archaeological assemblage exists on Stora Karlsö off the coast of Gotland dating to approximately 9200–8000 cal. BP (Lindqvist and Possnert, 1997; Pira, 1926). Here, also bones of new born pups are present in the bone material (Lindqvist and Possnert, 1997; Pira, 1926). Bones of ringed seal occur also in archaeological refuse faunas in Denmark, Germany, southern Sweden and on the Island of Öland (Aaris-Sørensen, 1998: 186, 2009; Ericson, 1989; Königsson et al., 1971; Lindqvist and Possnert, 1997; Magnell, 2005; Møhl, 1971; Schmölcke, 2008; Sommer and Benecke, 2003).

The highest summer temperatures in the Baltic Sea region were reached about 8000–6000 cal. BP. This Holocene Thermal Maximum (HTM) can be seen both in the continental (Hammarlund et al., 2003; Latałowa et al., 2013; Seppä et al., 2005, 2009) and marine records (Zillén et al., 2008). The marine records show that this period was associated with persistent hypoxia and high organic carbon content (Zillén et al., 2008).

During the HTM, the dated ringed seal finds concentrate in the Gulf of Bothnia, where also the archaeological ringed seal record is abundant (Ukkonen, 2002). However, the archaeological record from the Kõnnu site, southern coast of Saaremaa (Oesel) Island, shows that c. 7000 cal. BP (6460 ± 40 BP; Poz-30039) the ringed seal was heavily exploited also in the Gulf of Riga. Also, other sites in Saaremaa and Hiiumaa Island (Võhma, Kõpu IV and Kõpu VIII) indicate intensive hunting activities at that time (Kriiska and Lõugas, 1999; Lõugas, 1999). The Kõnnu site includes also bones of very young hunted seals. Archaeological bone materials reveal further that bones from new born ringed seals are present as far south as Öland (Königsson et al., 1971) and Gotland (Lindqvist and Possnert, 1997; Storå, 2001) at c. 6100–4000 cal. BP, and even in Poland (Rzucewo: Jan Storå pers. obs.) at c. 5300–4000 cal. BP. Bones from subadult ringed seals are present in Scania (Jonsson, 1986) and Denmark, but the absence of metric data hinders closer age estimations.

The presence of ringed seal populations in the central parts of the Baltic Basin during the warm period is quite puzzling, since the most critical ecological requirement for ringed seals is a genuine ice-winter with at least 2–3 months ice cover. Models (Meier et al., 2004) and current variability between winters show that during mild winters, predominantly the Gulf of Bothnia has sufficient ice days for successful breeding and moult. There are only few studies that have tried to reconstruct Holocene winter temperatures in the Baltic or more generally in northern high latitudes. Iversen (1944) suggested, based on the presence of plant species well known not to survive current winters, that winter temperatures were also warmer in the Baltic Sea area. Recent studies (Brown et al., 2012; Giesecke et al., 2008) indicate that winter temperatures may have been up to about 2°C warmer than the present (1961–1990), possibly also because of changes in water circulation, which were contributing to a milder and more maritime climate (see Giesecke et al., 2008). Such an increase in winter temperature is well below the projected increase in the Intergovernmental Panel on Climate Change (IPCC), Special Report on Emissions Scenarios (SRES), Scenario families A1 and B1 (IPCC SRES A1 and B1 scenarios; Nakicenovic and Swart, 2000) used in the ice models of Meier et al. (2004) for assessing the effects of the projected climate change. These models project an increase in winter temperature in the Baltic Sea area of 4–6°C or 5–9°C, respectively.

A mid-Holocene winter temperature increase of 1–2°C in the Baltic Sea area would lead to a reduction in mean ice cover, but ice winters would persist. An increase of 2°C would, according to the ice-temperature model of Jylhä et al. (2008) result in ice winters classified as mild predominating in the Baltic of mid-Holocene. Mean maximum ice extent in such a climate would be less than 130,000 km² (approximately a third of the Baltic Sea area), and ice would be concentrated in the northern part of the sea in most years.

The presence of breeding populations of ringed seals in the central parts of the Baltic indicates either much colder winter conditions – at least periodically – during the mid-Holocene warm period as suggested above or that central Baltic ringed seals survived periods with no genuine winter ice by migration and possible emergency breeding on land. Emergency land breeding of Baltic ringed seals is in a handful cases documented in sheltered environments such as reed beds and juniper bushes (Nordström et al., 2011; Sundqvist et al., 2012).

It has been noted that the body size of the harp seal and ringed seal was smaller in Prehistoric times than at present (e.g. Ekman, 1922; Holmquist, 1912; Lönnberg, 1909; Lõugas, 1999; Storå, 2001; Storå and Ericson, 2004), which may indicate that the species experienced suboptimal living conditions. Studies of Canadian ringed seals have shown that a larger body size was related to the access to fast ice where lactation of pups could continue for up to (the normal) 6 or 8 weeks (McLaren, 1958; Smith, 1973). In areas of pack ice, lactation was more often prematurely disrupted resulting in suboptimal growth and, finally, smaller adult body size. However, since the small size is evident both during the colder and the warmer periods, the size of the Baltic ringed seal may not be related to ice conditions. A straightforward interpretation of the small body size is not possible, since it depends on several factors such as ecosystem productivity, density-dependent factors, Bergman’s rule and reduced predation (for harp seal, see Schmölcke, 2008).

The archaeological record includes also several southern Baltic sites with ringed seal bones (Aaris-Sørensen, 1998: 186, 2009; Møhl, 1971) covering the last 8000 years and geographically ranging from Bornholm to Vendsyssel (Northern Kattegat). These bones result most probably from occasional visits of single stragglers or starvelings in the southern parts of the Baltic and do not indicate the presence of true populations in the area.

After 5000 cal. BP, the climate started to cool (Seppä et al., 2009). A cluster of nine dates from 4700 to 4100 cal. BP includes seals both from the northern and from the southern parts of the Baltic. This fits well the general picture of cooling, but would be expected later as cooling is thought to have been generally linear from 5000 cal. BP to preindustrial times (Wanner et al., 2008).

Ringed seal finds from Baltic proper in this period might mean that the winter climate in the region was at least periodically cooler in 5000–4000 cal. BP, in spite of higher summer temperatures suggested by Seppä et al. (2009).

The δ¹³C record: Diet and salinity

The δ¹³C values, from −21.7‰ to −11.9‰, display a range of nearly 10‰, related both to geographic find locality and age of the specimen (Figure 4). The oldest specimens with finite ages 17,000–12,000 cal. BP display the least negative δ¹³C values from −11.9‰ to −14.2‰ as expected, considering the marine nature of their find localities in the area of Skagerrak/Kattegat. Specimens dated to 10,000 cal. BP and younger, originating from the various presently brackish sub-basins of the Baltic Sea, show distinctively more negative δ¹³C values ranging from −21.7‰ to −15.9‰.

Figure 4.

δ¹³C values of dated ringed seal specimens around the Baltic Sea plotted against specimens age.

The stable isotopic composition of carbon (δ¹³C, against the VPDB standard) in seal bone collagen reflects the diet of the animal and indirectly yields information about the salinity of the ambient waters (e.g. Lindqvist and Possnert, 1997). The δ¹³C values of seal bones are ultimately traced back to the isotopic composition of carbon lower in the food web, that is, the photosynthesizing organisms that fix and convert dissolved inorganic carbon (DIC) into organic carbon and the heterotroph (e.g. zooplankton, nekton) consumers (e.g. Hobson et al., 1995; Rolff and Elmgren, 2000).

In marginal seas and estuarine environments such as the Baltic Sea, the δ¹³C values of inorganic and organic carbon characteristically display a spatial pattern of decreasing values with increasing distance from the marine connection, being thus linked to salinity of the water. The pattern observed by, for example, Spiker and Schemel (1979), Mook and Tan (1991), Rolff and Elmgren (2000), Emeis et al. (2003), Angerbjörn et al. (2006), Kiljunen et al. (2008) and Helsingen (2011) is explained by the mixing of contrasting isotopic compositions of marine versus terrigenous (freshwater) carbon sources. In view of this well-documented relationship between salinity and food-web δ¹³C values, we interpret the ringed seal collagen δ¹³C data in terms of strength of marine influence in the Baltic Basin.

The mean δ¹³C of the samples with finite ages from Kattegat/Skagerrak, −13.3 ± 1.0‰, is similar to that of modern seal bones in archaeological contexts (−12.3 ± 1.3‰), as well as sea lions (−12.7 ± 0.8‰) from various locations around the Atlantic and Pacific Oceans (Richards and Hedges, 1999), suggesting that they subsisted on fully marine resources. Collagen δ¹³C values for modern harbour seals (Phoca vitulina, n = 16) and grey seals (Halichoerus grypus; n = 1) in Skagerrak, after adjustment for lipid removal (Lidén et al., 1995) and the Suess-effect (Keeling, 1979), have a mean at −14.3 ± 1.3‰ (Enhus et al., 2011). Interestingly, this implies that these modern seals rely on a slightly more brackish food web compared with the late glacial ringed seals. This may be related to the observation that modern grey seal (e.g. Enhus et al., 2011; Helsingen, 2011) and harbour porpoise (Angerbjörn et al., 2006) populations have connectedness between Skagerrak, Kattegat and the Baltic Basin, whereas during 17,000–12,000 cal. BP, the strait connection between Skagerrak/Kattegat area and the Baltic Basin had not yet been established.

As described above, the present-day Baltic Sea features prominent spatial gradients in salinity (see ‘Introduction’) and δ¹³C values. However, this appears not to be the case in the subfossil ringed seal data set. When considering samples from the Baltic Basin only, there is no correlation between the latitude of sample locality and δ¹³C values (R² ~ 0.03). Instead, it seems that the age of the find plays a more significant role (Figure 4). Indeed, there is a strong negative correlation (R² = 0.56; $p ≪ 0.01$ ) between the δ¹³C value and the age of the specimen, that is, the δ¹³C values get progressively more negative with increasing age. The trend is unchanged if only samples from the Gulf of Bothnia are considered and persists even when the different sub-basins of the Gulf of Bothnia (Bothnian Bay, Northern Quark, Bothnian Sea, Sea of Åland) are viewed separately. The apparent similarity in magnitude and synchronicity of change in δ¹³C values, and thus, salinity, in the Baltic sub-basins possibly reflects the averaging effect caused by seal mobility. The water masses of the Baltic Ice Lake because of melting of the Scandinavian Ice Sheet were essentially oligotrophic (see Blumenberg et al., 2013). The millennium-long Yoldia Sea (sensu lato) phase only slightly added the marine character of the Baltic Ice Lake. The low δ¹³C values of the seals inhabiting the Ancylus Lake and the Ancylus/Littorina transitional phase around 9000 cal. BP are typical for animals living in terrestrial/freshwater ecosystems and reflect the dominance of terrigenous carbon in the food web. From a minimum (−21.7‰) at 9000 cal. BP, the δ¹³C values increase 5.8‰ units to a maximum (−15.9‰) at 5500 cal. BP. The seal collagen δ¹³C data indicate a maximum marine influence from c. 6500 to 3500 cal. BP in the Baltic Sea. During this period, collagen displays a mean δ¹³C value of −17.1 ± 0.5‰, in agreement with δ¹³C data for ringed seal collagen ranging from −17.7 to −15.6‰ from Pitted ware culture sites (c. 5250–4750 cal. BP) in Gotland, Öland and Södertörn in central coastal Sweden (Eriksson, 2004; Eriksson et al., 2008; Fornander et al., 2008). The timing of the maximum salinity period inferred from ringed seal collagen agrees with other salinity reconstructions in the Baltic Basin (Berglund et al., 2005; Donner et al., 1999; Punning et al., 1988; Widerlund and Andersson, 2011; Willumsen et al., 2013). Modern ringed and harbour seal collagen δ¹³C values in the Northern and Southern Baltic are around −18‰ (Enhus et al., 2011). The 1‰ lower modern mean δ¹³C value is consistent with the reported recent freshening of the Baltic Sea waters (Widerlund and Andersson, 2011; Willumsen et al., 2013).

The seal bones bear another isotopic tracer linked to water salinity. ⁸⁷Sr/⁸⁶Sr values of marine mammals record the ⁸⁷Sr/⁸⁶Sr ratio of the ambient water, which in the Baltic Sea is a proxy for water mass salinity (Andersson et al., 1992; Widerlund and Andersson, 2006). As a part of earlier studies, two ringed seal specimens included here (ZMH 6978, ZMH 6972) and two harp seal samples from the Northern Baltic dated to 10,400–5200 cal. BP yielded a trend of ⁸⁷Sr/⁸⁶Sr values progressively decreasing with age from 0.750 to 0.709 (Arppe, unpublished), testifying to the evolving salinity conditions experienced by the Baltic seal population.

In addition to yielding information on the strength of marine influence in brackish environments, seal tissue δ¹³C values have been linked to foraging location (e.g. Hobson et al., 2002; Helsingen, 2011; Sinisalo et al., 2007; Young and Ferguson, 2014). Pelagic/off-shore food webs are characterized by distinctly lower δ¹³C values than benthic/in-shore ones because of lower diffusion resistance in higher-turbulence systems (France, 1995). Following this reasoning, the ringed seal δ¹³C record would imply benthic foraging (e.g. abundance of cod) for the late glacial Kattegat/Skagerrak seals, and a dramatic development from pelagic-dominated feeding to benthic food resources for the seals of the Baltic Basin between 9000 and 5500 cal. BP. However, we know that modern ringed seals, whose δ¹³C values are close to those of the Littorina stage seal specimens, rely primarily on herring, a pelagic species, for subsistence (Enhus et al., 2011). Furthermore, the rise of salinity of the Baltic waters during the Ancylus/Littorina transitional and early Littorina stage is independently well documented, and the relationships between δ¹³C values and salinity in marginal seas are well founded. Thus, we argue that the ringed seal δ¹³C variations chiefly reflect the strength of marine influence in the basin.

Conclusion

The ringed seal entered the Baltic Basin during the Yoldia or Ancylus stage of the Baltic Sea and has since been part of the Baltic fauna, during both the freshwater Ancylus stage and the Littorina stage with increasing and decreasing salinity.

The sequence of the dates indicate a continuous presence of the species in the Baltic Basin. The high genetic variability of the Baltic population may reflect continuous gene flow from the Atlantic Ocean and/or a very diverse founder population and a continuously large population size through Holocene.

The species had a wider distribution during the Littorina stage than today. The presence of breeding populations in central parts of the Baltic during the HTM indicates that during that time the winters were at least periodically cold enough for winter ice to occur even in the central part of the Baltic Sea basin and/or that ringed seals adapted to the iceless winter conditions of the HTM by migratory movements and possible attempts of land breeding. It is important to note, however, that successful adaptation to land breeding behaviour has never been recorded in any ringed seal population, probably because the pup and lactating female are vulnerable to predation.

Seal collagen δ¹³C values in the Baltic Sea primarily reflect the strength of marine influence in the basin and serve as a proxy for inferences on qualitative changes in water mass salinity.

Footnotes

Acknowledgements

We wish to thank the following museums and institutions for allowing us to use their collections and sample the specimens: Göteborg Natural History Museum, Gothenburg; Swedish Museum of Natural History, Stockholm; Finnish Museum of Natural History, Helsinki; University of Copenhagen, Zoological Museum, Copenhagen; National Museum of Lithuania, Vilnius; and Klaipėda University, Klaipėda. Heikki Seppä kindly read the manuscript and gave valuable comments to the discussion. We would also like to thank Robert Sommer and two anonymous referees for their constructive suggestions.

Funding

The radiocarbon dates were funded by a grant (20070665) from Crafoordska stiftelsen, Lund, and Lund University to PU. Malgorzata Pilot was supported by Marie Curie Intra-European Fellowship from the European Commission.

References

Aaris-Sørensen

(1998) Danmarks forhistoriske Dyreverden. København: Gyldendal.

Aaris-Sørensen

(2009) Diversity and dynamics of the mammalian fauna in Denmark throughout the last glacial–interglacial cycle, 115–0 kyr BP. Fossils and Strata 57: 1–59.

Andersson

(2014) Hydrography and oxygen in the deep basins. HELCOM Baltic Sea Environment Fact Sheets. Available at: http://www.helcom.fi/baltic-sea-trends/environment-fact-sheets/ (accessed 10 June 2014).

Andersson

Wasserburg

Ingri

(1992) The sources and transport of Sr and Nd isotopes in the Baltic Sea. Earth and Planetary Science Letters 113: 459–472.

Andrén

Sohlenius

(2000) The Holocene history of the southwestern Baltic Sea as reflected in a sediment core from the Bornholm Basin. Boreas 29: 233–250.

Andrén

Berglund

. (2007) New insights on the Yoldia Sea low stand in the Blekinge archipelago, southern Baltic Sea. GFF 129: 273–281.

Angerbjörn

Börjesson

Brandberg

(2006) Stable isotope analysis of harbour porpoises and their prey from the Baltic and Kattegat/Skagerrak Seas. Marine Biology Research 2: 411–419.

BACC (2008) Assessment of Climate Change for the Baltic Sea Basin (ed The BACC Author Team). Berlin: Springer-Verlag.

Bennike

Rasmussen

Aaris-Sørensen

(2008) The harp seal (Phoca groenlandica Erxleben) in Denmark, southern Scandinavia, during the Holocene. Boreas 37: 263–272.

10.

Berglund

Håkansson

Lagerlund

(1976) Radiocarbon dated mammoth (Mammuthus primigenius Blumenbach) finds in south Sweden. Boreas 5: 177–191.

11.

Berglund

Sandgren

Barnekow

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

12.

Björck

(2008) The late Quaternary development of the Baltic Sea basin. In: The BACC Author Team (eds) Assessment of Climate Change for the Baltic Sea Basin. Berlin: Springer-Verlag, pp. 398–407.

13.

Blumenberg

Berndmeyer

Moros

. (2013) Bacteriohopanepolyols record stratification, nitrogen fixation and other biogeochemical perturbations in Holocene sediments of the central Baltic Sea. Biogeosciences 10: 2725–2735.

14.

Bronk Ramsey

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

15.

Brown

Seppä

Schoups

. (2012) A spatio-temporal reconstruction of Holocene temperature change in southern Scandinavia. The Holocene 22: 165–177.

16.

Daugnora

(2000) Fish and seal osteological data at Sventoji sites. Lietuvos Archeologija 19: 85–101.

17.

Davis

Stirling

Strobeck

. (2008) Population structure of ice-breeding seals. Molecular Ecology 17: 3078–3094.

18.

Dippner

Vuorinen

Daunys

. (2008) Chapter 5: Climate-related marine ecosystem change. In: The BACC Author Team (eds) Assessment of Climate Change for the Baltic Sea Basin. Berlin: Springer-Verlag, pp. 309–377.

19.

Donner

Kankainen

Karhu

(1999) Radiocarbon ages and stable isotope composition of Holocene shells in Finland. Quaternaria A: 31–38.

20.

Ekman

(1922) Djurvärldens utbredningshistoria på skandinaviska halvön. Stockholm: Bonniers.

21.

Emeis

K-C

Struck

Blanz

. (2003) Salinity changes in the central Baltic Sea (NW Europe) over the last 10000 years. The Holocene 13: 413–423.

22.

Enghoff

Brian

MacKenzie

. (2007) The Danish fish fauna during the warm Atlantic period (ca. 7000–3900 bc): Forerunner of future changes? Fisheries Research 87: 167–180.

23.

Enhus

Boalt

Bignert

(2011) A retrospective study of metals and stable isotopes in seals from Swedish waters. Report No. 5: 2011. Stockholm: Swedish Museum of Natural History.

24.

Ericson

PGP

(1989) Säl och säljakt i Östersjöområdet under stenåldern. In: Iregren

Liljekvist

(eds) Faunahistoriska studier tillägnade Johannes Lepiksaar: Symposium 26 maj 1988 (Report series No. 3). Lund: Institute of Archaeology, University of Lund, pp. 57–63.

25.

Eriksson

(2004) Part-time farmers or hard-core sealers? Västerbjers studied by means of stable isotope analysis. Journal of Anthropological Archaeology 23: 135–162.

26.

Eriksson

Linderholm

Fornander

. (2008) Same island, different diet: Cultural evolution of food practice on Öland, Sweden, from the Mesolithic to the Roman Period. Journal of Anthropological Archaeology 27: 520–543.

27.

Fornander

Eriksson

Lidén

(2008) Wild at heart: Approaching Pitted Ware identity, economy and cosmology through stable isotopes in skeletal material from the Neolithic site Korsnäs in Eastern Central Sweden. Journal of Anthropological Archaeology 27: 281–297.

28.

France

(1995) Carbon-13 enrichment in benthic compared to planktonic algae: Foodweb implications. Marine Ecology Progress Series 124: 307–312.

29.

Fredén

(1975) Subfossil Finds of Arctic Whales and Seals in Sweden. Appendix: Radiocarbon Determination of Miscellaneous Subfossil Finds of the Swedish West Coast. Uppsala: Sveriges geologiska undersökning, C 710.

30.

Giesecke

Bjune

Chiverrell

. (2008) Exploring Holocene continentality changes in Fennoscandia using present and past tree distributions. Quaternary Science Reviews 27: 296–1308.

31.

Graves

Helyar

Biuw

. (2009) Microsatellite and mtDNA analysis of the population structure of grey seals (Halichoerus grypus) from three breeding areas in the Baltic Sea. Conservation Genetics 10: 59–68.

32.

Gustafsson

Westman

(2002) On the causes of salinity variations in the Baltic Sea during the last 8500 years. Paleoceanography 17: 1–14.

33.

Hågvar

Ohlson

(2013) Ancient carbon from a melting glacier gives high ¹⁴C age in living pioneer invertebrates. Scientific Reports 3: 2820.

34.

Håkansson

(1976) University of Lund radiocarbon dates. Radiocarbon 18: 290–320.

35.

Hammarlund

Björck

Buchardt

. (2003) Rapid hydrological changes during the Holocene revealed by stable isotope records of lacustrine carbonates from Lake Igelsjön, southern Sweden. Quaternary Science Reviews 22: 353–370.

36.

Härkönen

Stenman

Jüssi

. (1998) Population size and distribution of the Baltic ringed seal (Phoca hispida botnica). In: Heide-Jorgensen

Lydersen

(eds) Ringed Seals in the North Atlantic. Tromsø: North Atlantic Marine Mammal Commission (NAMMCO) Scientific Publications, vol. 1, pp. 167–180.

37.

Helcom (2013) HELCOM Red List of Baltic Sea species in danger of becoming extinct. Baltic Marine Environment Protection Commission Proceedings 140: 1–106.

38.

Helsingen

(2011) Bioaccumulation of chemical elements in grey seals (Halichoerus grypus) from the Baltic Sea. MSc Thesis, Norges teknisk-naturvitenskapelige universitet, 67 pp.

39.

Hobson

Ambrose

Jr Renaud

(1995) Sources of primary production, benthic-pelagic coupling, and trophic relationships within the Northeast Water Polynya: Insights from δ13C and δ15N analysis. Marine Ecology Progress Series 128: 1–10.

40.

Hobson

Fisk

Karnovsky

. (2002) A Stable isotope (δ13C, δ15N) model for the North Water food web: Implications for evaluating trophodynamics and the flow of energy and contaminants. Deep-Sea Research II 49: 5131–5150.

41.

Holmquist

(1912) Tierknochen an den steinzeitlichen Wohnplätzen in Visby und bei Hemmor sowie aus einem Öländischen Ganggrabe. Kungliga Svenska Vetenskapsakademiens Handlingar 49: 71–75.

42.

Hughen

Baillie

MGL

Bard

. (2004) Marine04 marine radiocarbon age calibration, 0–26 cal kyr BP. Radiocarbon 46: 1059–1086.

43.

Iversen

(1944) Viscum, Hedera and Ilex as climatic indicators. A contribution to the study of past-glacial temperature climate. Geologiska Föreningens i Stockholm Förhandlingar 66: 463–483.

44.

Jonsson

(1986) The vertebrate faunal remains from the Late Atlantic Settlement Skateholm in Scania, South Sweden. In: Larsson

(ed.) The Skateholm Project I. Man and Environment. Stockholm: Acta Regiae Societatis Humaniorum Litterarum Lundensis 79, pp. 56–88.

45.

Jylhä

Fronzek

Tuomenvirta

. (2008) Changes in frost, snow and Baltic sea ice by the end of the twenty-first century based on climate model projections for Europe. Climatic Change 86: 441–462.

46.

Kapel

Christiansen

Heide-Jorgensen

. (1998) Netting and conventional tagging used for studying movements in ringed seals (Phoca hispida) in Greenland. In: Heide-Jorgensen

Lydersen

(eds) Ringed Seals (Phoca hispida) in the North Atlantic. Tromsø: North Atlantic Marine Mammal Commission (NAMMCO) Scientific Publications, vol. 1, pp. 211–228.

47.

Keeling

(1979) The Suess effect: ¹³Carbon-¹⁴Carbon interrelations. Environmental International 2: 229–300.

48.

Kelly

Bengtson

Boveng

. (2010) Status review of the ringed seal (Phoca hispida) (NOAA Technical Memorandum: NMFS-AFSC-212). Washington, DC: US Department of Commerce.

49.

Kiljunen

Peltonen

Jones

. (2008) Coupling stable isotopes with bioenergetics to evaluate sources of variation in organochlorine concentrations in Baltic salmon ( Salmo salar). Canadian Journal of Fisheries and Aquatic Science 65: 2114–2126.

50.

Königsson

Lepiksaar

Königsson

L-K

(1971) Stenålderboplatsen i Alby på Öland. Preliminärt meddelande. Fornvännen 66: 34–46.

51.

Kovacs

Lowry

Härkönen

(IUCN SSC Pinniped Specialist Group) (2008) Pusa hispida. IUCN 2014. IUCN Red List of Threatened Species. Version 2014.1.

52.

Kovacs

Aguilar

Aurioles

. (2012) Global threats to pinnipeds. Marine Mammal Science 28: 414–436.

53.

Kriiska

Lõugas

(1999) Late Mesolithic and Early Neolithic Seasonal Settlement at Kõpu, Hiiumaa Island, Estonia. In: Miller

Hackens

Lang

. (eds) Environmental and Cultural History of the Eastern Baltic Region (PACT 57). Rixensart: PACT, pp. 157–172.

54.

Kunnasranta

Hyvärinen

Sipilä

. (1999) The diet of the Saimaa ringed seal Phoca hispida saimensis. Acta Theriologica 44: 443–450.

55.

Lagerlund

Houmark-Nielsen

(1993) Timing and pattern of the last deglaciation in the Kattegat region, southwest Scandinavia. Boreas 22: 337–347.

56.

Latałowa

Pędziszewska

Maciejewska

. (2013) Tilia forest dynamics, Kretzschmaria deusta attack, and mire hydrology as palaeoecological proxies for mid-Holocene climate reconstruction in the Kashubian Lake District (N Poland). The Holocene 23: 667–677.

57.

Lepiksaar

(1986) The Holocene history of Theriofauna in Fennoscandia and Baltic Countries. In: Königsson

L-K

(ed.) Nordic Late Quaternary Biology and Ecology (Striae, vol. 24). Uppsala: Societas Upsaliensis pro Geologia Quaternaria, pp. 51–70.

58.

Lidén

Takahashi

Nelson

(1995) The effects of lipids in stable carbon isotope analysis and the effects of NaOH treatment on the composition of extracted bone collagen. Journal of Archaeological Science 22: 321–326.

59.

Lindqvist

Possnert

(1997) The subsistence economy and diet at Jakob/Ajvide, Eksta Parish and other Prehistoric Dwelling and Burial sites on Gotland in long-term perspective. In: Burenhult

(ed.) Remote Sensing. Applied Techniques for the Study of Cultural Resources and the Localization, Identification and Documentation of Sub-Surface Prehistoric Remains in Swedish Archaeology. I. Thesis and Papers in North-European Archaeology 13:a. Hässleholm.

60.

Longin

(1971) New method of collagen extraction for radiocarbon dating. Nature 230: 241–242.

61.

Lönnberg

(1909) Några fynd av subfossila vertebrater. Arkiv för Zoologi 6: 1–28.

62.

Lõugas

(1997) Subfossil seal finds from archaeological coastal sites in Estonia, east part of the Baltic Sea. Anthropozoologica 25–26: 699–706.

63.

Lõugas

(1999) Postglacial development of fish and seal faunas in the Eastern Baltic water system. In: Benecke

(ed.) The Holocene History of the European Vertebrate Fauna. Modern Aspects of Research. Rahden/Westf: Marie Leidorf, pp. 185–200.

64.

Lõugas

Lidén

Erle

(1996) Resource utilisation along the Estonian Coast during the Stone Age. In: Hackens

Hicks

Lang

. (eds) Coastal Estonia: Recent Advances in Environmental and Cultural History (PACT 51). Rixensart: PACT, pp. 399–420.

65.

Lõugas

Wojtal

Wilczynski

. (2013) Palaeolithic fish from Southern Poland: A palaeozoogeographical approach. Archaeofauna 22: 123–131.

66.

Lougheed

Filipsson

Snowball

(2013) Large spatial variations in coastal 14C reservoir age – A case study from the Baltic Sea. Climate of the Past 9: 1015–1028.

67.

Lougheed

Snowball

Moros

. (2012) Multiple approach Baltic Sea geochronology using 14C dating, palaeomagnetic secular variation (PSV) and atmospheric Pb deposition; assessment of geochronology methods and inference of 14C reservoir age. Quaternary Science Reviews 42: 43–58.

68.

Lübke

Schmölcke

Tauber

(2011) Mesolithic hunter-fishers in a changing world: a case study of submerged sites on the Jäckelberg, Wismar Bay, northeastern Germany. In: Benjamin

Bonsall

Pickard

. (eds) Submerged Prehistory. Oxford: Oxbow Books, pp. 21–37.

69.

McLaren

(1958) The Biology of the Ringed Seal (Phoca hispida Schreber) in the Eastern Canadian Arctic (Bulletin of the Fisheries Research Board of Canada 118, 97 pp.). Ottawa: Fisheries Research Board of Canada.

70.

Magnell

(2005) Tracking Wild Boar and Hunters: Osteology of Wild Boar in Mesolithic South Scandinavia. Lund: Almqvist and Wiksell International, Lund University.

71.

Martinez-Bakker

Sell

Swanson

. (2013) Combined genetic and telemetry data reveal high rates of gene flow, migration, and long-distance dispersal potential in Arctic ringed seals (Pusa hispida). PLoS ONE 8(10): e77125. DOI: 10.1371/journal.pone.0077125.

72.

Meier

HEM

Döscher

Halkka

(2004) Simulated distributions of Baltic Sea-ice in warming climate and consequences for the Winter habitat of the Baltic Ringed Seal. Ambio 33: 249–256.

73.

Møhl

(1971) Oversigt over dyreknoglerne fra Ølby Lyng. En østsjællandsk kystboplads med Ertebøllekultur. Aarbøger for nordisk Oldkyndighed og Historie 1970: 43–77.

74.

Mook

Tan

(1991) Stable isotopes in rivers and estuaries. In: Degens

Kempe

Richey

(eds) Biogeochemistry of Major World Rivers. Chichester: John Wiley & Sons, pp. 245–264.

75.

Nakicenovic

Swart

(eds) (2000) Special Report on Emissions Scenarios. A Special Report of Working Group III of the Intergovernmental Panel on Climate Change. Cambridge and New York: Cambridge University Press, 570 pp. Available at: http://www.ipcc.ch/ipccreports/sres/emission/index.htm (accessed 21 May 2014).

76.

Nordström

Högmander

Halkka

A et al

. (2011) Itämerennorppa Saaristomerellä – unohdettu uhanalainen. Maailman luonnon säätiön WWF Suomen rahaston raportteja 28. Available at: http://wwf.fi/mediabank/930.pdf

77.

Paaver

Lõugas

(2003) Origin and history of the fish fauna in Estonia. In: Ojaveer

Pihu

Saat

(eds) Fishes of Estonia. Tallinn: Estonian Academy Publishers, pp. 28–46.

78.

Palo

Mäkinen

Helle

. (2001) Microsatellite variation in ringed seals (Phoca hispida): Genetic structure and history of the Baltic Sea population. Heredity 86: 609–617.

79.

Pira

(1926) On bone deposits in the cave ‘Stora Förvar’ on the Isle of Stora Karlsö, Sweden (A contribution to the knowledge of prehistoric domestic animals). Acta Zoologica 7: 123–217.

80.

Punning

J-M

Martma

Kessel

. (1988) The isotope composition of oxygen and carbon in the subfossil mollusc shells of the Baltic Sea as an indicator for paleosalinity. Boreas 17: 27–31.

81.

Reeves

(1998) Distribution, abundance and biology of ringed seals (Phoca hispida): An overview. In: Heide-Jorgensen

Lydersen

(eds) Ringed Seals (Phoca hispida) in the North Atlantic. Tromsø: North Atlantic Marine Mammal Commission (NAMMCO) Scientific Publications, vol. 1, pp. 9–45.

82.

Reimer

Baillie

MGL

Bard

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

83.

Rice

(1998) Marine Mammals of the World: Systematics and Distribution. Lawrence: Society for Marine Mammalogy.

84.

Richards

Hedges

REM

(1999) Stable isotope evidence for similarities in the types of marine foods used by Late Mesolithic Humans at sites along the Atlantic coast in Europe. Journal of Archaeological Science 26: 717–722.

85.

Rolff

Elmgren

(2000) Use of riverine organic matter in plankton food webs of the Baltic Sea. Marine Ecology Progress Series 197: 81–101.

86.

Rößler

Moros

Lemke

(2011) The Littorina transgression in the southwestern Baltic Sea: new insights based on proxy methods and radiocarbon dating of sediment cores. Boreas 40: 231–241.

87.

Rundkvist

Lindqvist

Thorsberg

(2004) Rojrhage in Grötlingbo. A multi-component Neolithic shore site on Gotland. Stockholm Archaeological Reports 41, Department of Archaeology, University of Stockholm.

88.

Schmölcke

(2008) Holocene environmental changes and the seal (Phocidae) fauna of the Baltic Sea: Coming, going and staying. Mammal Review 38: 231–246.

89.

Schmölcke

Endtmann

Klooss

. (2006) Changes of sea level, landscape and culture: A review of the south-western Baltic area between 8800 and 4000 BC. Palaeogeography, Palaeoclimatology, Palaeoecology 240: 423–438.

90.

Schoning

(2001) The brackish Baltic Sea Yoldia Stage – palaeoenvironmental implications from marine benthic fauna and stable oxygen isotopes. Boreas 30: 290–298.

91.

Seppä

Bjune

Telford

. (2009) Last nine-thousand years of temperature variability in Northern Europe. Climates of the Past 5: 523–535.

92.

Seppä

Hammarlund

Antonsson

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

93.

Sinisalo

Jones

Helle

. (2007) Changes in diets of individual Baltic ringed seals (Phoca hispida botnica) during their breeding season inferred from stable isotope analysis. Marine Mammal Science 24: 159–170.

94.

Sipilä

Hyvärinen

(1998) Status and biology of Saimaa (Phoca hispida saimensis) and Ladoga (Phoca hispida ladogensis) ringed seals. In: Heide-Jorgensen

Lydersen

(eds) Ringed Seals (Phoca hispida) in the North Atlantic. Tromsø: North Atlantic Marine Mammal Commission (NAMMCO) Scientific Publications, vol. 1, pp. 83–99.

95.

Smith

(1973) Population Dynamics of the Ringed Seal in the Canadian Eastern Arctic (Bulletin of the Fisheries Research Board of Canada 181). Ottawa: Fisheries Research Board of Canada.

96.

Smith

(1987) The ringed seal, Phoca hispida, of the Canadian western Arctic. Canadian Bulletin of Fisheries and Aquatic Sciences 216: 1–81.

97.

Smith

Hammill

(1981) Ecology of the ringed seal, Phoca hispida, in its fast ice breeding habitat. Canadian Journal of Zoology 59: 966–981.

98.

Smith

Stirling

(1975) The breeding habitat of the ringed seal (Phoca hispida): The birth lair and associated structures. Canadian Journal of Zoology 53: 1297–1305.

99.

Sommer

Benecke

(2003) Post-Glacial history of the European seal fauna on the basis of sub-fossil records. Beiträge zur Archäozoologischen und Prähistorischen Anthopologie VI: 16–28.

100.

Spiker

Schemel

(1979) Distribution and stable-isotope composition of carbon in San Francisco Bay. In: Conomos

(ed.) San Francisco Bay: The Urbanized Estuary. San Francisco, CA: American Association for the Advancement of Science, pp. 195–212.

101.

Stenman

Pöyhönen

(2005) Food remains in the alimentary tracts of the Baltic grey and ringed seals (Abstract). In: Helle

Stenman

Wikman

(eds) Symposium on biology and management of seals in the Baltic area, Helsinki, 15–18 February 2005, pp. 51–53. Helsinki: Finnish Game and Fisheries Research Institute.

102.

Storå

(2001) Reading bones – Stone age hunters and seals in the Baltic. Academic Dissertation, Stockholm University.

103.

Storå

Ericson

PGP

(2004) A prehistoric breeding population of Harp seals (Phoca groenlandica) in the Baltic Sea. Marine Mammal Science 20: 115–133.

104.

Sundqvist

Härkönen

Svensson

. (2012) Linking climate trends to population dynamics in the Baltic Ringed Seal: Impacts of historical and future winter temperatures. Ambio 41: 865–872.

105.

Suuronen

Lehtonen

(2012) The role of salmonids in the diet of grey and ringed seals in the Bothnian Bay, northern Baltic Sea. Fisheries Research 125–126: 283–288.

106.

Teilmann

Born

Acquarone

(1999) Behaviour of ringed seals tagged with satellite transmitters in the North Water polynya during fast-ice formation. Canadian Journal of Zoology 77: 1934–1946.

107.

Tormosov

Rezvov

(1978) Information on the distribution, number and feeding habits of ringed and grey seals in the Gulfs of Finland and Riga in the Baltic Sea. In: Lampio

Stenman

(eds) Proceedings of the symposium on the conservation of Baltic seals, 26–28 April 1977, pp. 14–17. Haikko, Finland: Finnish Game and Fisheries Research Institute.

108.

Ukkonen

(2002) The early history of seals in the northern Baltic. Annales Zoologici Fennici 39: 187–207.

109.

Valtonen

Palo

Ruokonen

. (2012) Spatial and temporal variation in genetic diversity of an endangered freshwater seal. Conservation Genetics 13: 1231–1245.

110.

Van Bree

PJH

(1996) Extralimital records of the Ringed Seal, Phoca hispida Schreber, 1775, on the West-European continental coast. Bonner Zoologische Beiträge 46: 377–383.

111.

Wanner

Beer

Bütikofer

. (2008) Mid- to Late Holocene climate change: An overview. Quaternary Science Reviews 27: 1791–1828.

112.

Wathne

Haug

Lydersen

(2000) Prey preference and niche overlap of ringed seals Phoca hispida and harp seals P. groenlandica in the Barents Sea. Marine Ecology Progress Series 194: 233–239.

113.

Widerlund

Andersson

(2006) Strontium isotopic composition of modern and Holocene mollusc shells as a palaeosalinity indicator for the Baltic Sea. Chemical Geology 232: 54–66.

114.

Widerlund

Andersson

(2011) Late Holocene freshening of the Baltic Sea derived from high-resolution strontium isotope analysis of mollusk shells. Geology 39: 187–190.

115.

Willumsen

Filipsson

Reinholdsson

. (2013) Surface salinity and nutrient variations during the Littorina Stage in the Fårö Deep, Baltic Sea. Boreas 42: 210–223.

116.

Young

Ferguson

(2014) Using stable isotopes to understand changes in ringed seal foraging ecology as a response to a warming environment. Marine Mammal Science 30: 706–725.

117.

Zagorska

(2000) Sea Mammal Hunting Strategy in Eastern Baltic. Lietuvos Archeologija 19: 275–285.

118.

Zillén

Conley

Andrén

. (2008) Past occurrences of hypoxia in the Baltic Sea and the role of climate variability, environmental change and human impact. Earth-Science Reviews 91: 77–92.

An Arctic seal in temperate waters: History of the ringed seal ( Pusa hispida ) in the Baltic Sea and its adaptation to the changing environment

Abstract

Keywords

Introduction

Baltic Sea and its development after the glaciation, marine and freshwater stages

Behaviour and population size of the ringed seal

Diet of the ringed seal

Genetic variability in the ringed seals worldwide and in the Baltic Sea

Material and methods

Results and discussion

Baltic ringed seal finds and their dates

Dispersal

Distribution and ecology

The δ13C record: Diet and salinity

Conclusion

Footnotes

Acknowledgements

Funding

References

The δ¹³C record: Diet and salinity