Geological Survey of Denmark and Greenland Bulletin 28, 2013, 17-20 17 A Baltic Ice Lake lowstand of latest Allerød age in the Arkona Basin, southern Baltic Sea Ole Bennike and Jørn Bo Jensen After the last deglaciation, the Baltic Sea underwent a com- plex salinity history and dynamic shore-level development with several lacustrine and marine stages: the Baltic Ice Lake, the Yoldia Sea, the Ancylus Lake and the Littorina Sea (Björck 1995). In connection with shallow seismic profiling in the south-western Baltic Sea, two marked and widespread erosional unconformities have been identified ( Jensen et al. 1997, 1999; Lemke et al. 1998; Larsen 2004). The older un- conformity occurs within sediments deposited in the Baltic Ice Lake, whereas the younger one separates Baltic Ice Lake sediments from Holocene lake and mire deposits. The latter unconformity is dated to the transition between the Young- er Dryas and the Holocene, corresponding to c. 11.7 cal. ka BP and formed due to a sudden drop in the level of the ice- dammed Baltic Ice Lake of around 25 m, caused by ice reces- sion from Mt. Billingen in south central Sweden. The age of the older erosional unconformity is poorly con- strained. However, it has been suggested that the level of the Baltic Ice Lake also experienced a sudden drop during the Al- lerød chronozone. The temperature increased during the Al- lerød, and it is possible that the margin of the Fennoscandian Ice Sheet also receded from the Mt. Billingen area at this time. If so, the older erosional unconformity may have formed in connection with an early drainage of the Baltic Ice Lake. The question about such an early lake-level fall was dis- cussed by Björck (1995), who listed a number of arguments for and against it. One of the arguments for drainage is that evidence of a rapid lake-level fall is seen in both south-east- ern Sweden and in other Baltic countries. A study in south- western Sweden indicates a high discharge of freshwater, and data from south central Sweden indicate a significant glacial recession west of Mt. Billingen during the Allerød, which could have led to drainage of the Baltic Ice Lake. It has been suggested that drainage happened around 13 cal. ka ago (Uścinowicz 2006; Andrén et al. 2011). In connection with sediment coring in the Arkona Basin, firm evidence of an early lowstand was for the first time iden- tified in a sediment core. An organic-rich sediment which can be referred to this lowstand was found, and material was submitted for 14C dating. The aim of this paper is to report on the age and its implications. The location of the coring site in the Arkona Basin is shown in Fig. 1. The basin is up to 49 m deep and the surface of the till is found at a depth of 30 to 70 m (Moros et al. 2002). Material and methods Sediment coring was carried out using a 10 cm diameter vibrocorer with a 6 m long steel tube. Normally, coring positions are selected from high-resolution, shallow seis- mic profiles, but in the deeper parts of the Arkona Basin, gas-bearing, organic-rich, Holocene marine sediments are widespread and hamper the use of seismic data. Hence the core described here was collected at a site with poor seismic data. The core was sampled at a water depth of 39.5 m, at 54°45.005́ N, 13°45.876́ E. The core was collected c. 5 km ENE of the northern end of a seismic profile published by Lemke et al. (1998), where the unconformity is clearly seen. The lower unconformity is also seen on seismic data collect- ed in the region during the Baltic Pipe survey (Larsen 2004). We collected two cores at the site, designated 258000-1 and 258000-2. The first core was collected in plastic foil, split lengthwise in the ship laboratory and described and subsam- pled for palaeoecological analysis. The second core was col- lected in a PVC tube and cut into 1 m sections for storage. However, the second core penetrated somewhat deeper than the first core, and in the core catcher, clay was found with abundant plant remains underlain by clay-rich, medium- grained sand. Ten 1 kg samples from core 258000-1 and from © 2013 GEUS. Geological Survey of Denmark and Greenland Bulletin 28, 17–20. Open access: www.geus.dk/publications/bull 14°E 14°E 55°N 55° 54°30´N 13°E 13°E 55–60 50–55 45–50 40–45 35–40 75–80 70–75 65–70 60–65 30–35 25–30 20–25 15–20 10–15 5–10 Depth (m) 20 km 0–5 54°30´N 258000 Skåne Bornholm Møn Rügen Fig. 1. Bathymetrical map of the Arkona Basin. The red dot shows the lo- cation of the studied sediment core. The core site is also shown on Fig. 5. 1818 the core catcher of 258000-2 were brought to the Geological Survey of Denmark and Greenland, where the samples were wet sieved shortly after the cruise. Fruits of Cladium mari- scus (a reed plant) were dried and shortly after submitted for accelerator mass spectrometry (AMS) radiocarbon age deter- mination at the Leibniz Laboratory for Radiometric Dating and Isotope Research in Kiel, Germany. Results and discussion Core 258000-1 consists of olive-grey mud from the core top down to a depth of 412 cm (Fig. 2). Some shells of Macoma balthica and Mytilus edulis were noted, which shows that the mud is marine and of Holocene age. From 412 to 473 cm, clay without carbonate is found; this succession was divided into three units according to variations in colour and tex- ture. Shells and head shields of Chydoridae (cladocerans, wa- ter f leas) are common, head capsules of larvae of Chironomi- dae (non-biting midges) are present but rare, egg cocoons of the fish leach Piscicola geometra were found in two samples, and statoblasts of the bryozoan Cristatella mucedo in four samples (Fig. 3; Table 1). These invertebrate remains show that the clay was deposited in a lake. The presence of rare remains of Betula sect. Albae (tree birch) and Pinus sylvetris (pine) indicates an early Holocene age. The lower part was rich in radicells (tiny roots) of reed plants, which show that the shoreline was not far away from the coring site. The clay was probably deposited in the Ancylus Lake and perhaps in the Yoldia Sea. D ep th b el ow c or e to p (c m ) Holocene marine mud indistinct layering Holocene marine mud homogeneous Fine- to medium-grained sand Late glacial clay Late glacial clay with plant remains Late glacial sand Holocene lacustrine clay 0 100 200 300 400 500 Fig. 2. Composite lithological log of core 258000 from the Arkona Basin. The two lower units were only found in core 258000-2. The core numbers refer to the system used at the Department of Marine Geolog y at the Insti- tute for Baltic Sea Research in Warnemünde. Fig. 3. Macrofossil diagram of the lower parts of cores 258000-1 and 258000-2 from the Arkona Basin. The hollow bars show remains not counted. 2 1 50 10 100 20 1 200 1 5 10 400 420 440 460 480 500 520 540 560 580 600 D ep th (c m ) Be tu la s ec t. Al ba e sp . Pi nu s s ylv es tri s Ph ra gm ite s a us tra lis Cl ad iu m m ar isc us Ca re x sp . Co m ar um p al us tre Ci cu ta v iro sa R ad ic el ls C ha ra ce ae in de t. A m bl ys te gi ac ea e in de t. M en ya nt he s t rif ol ia ta N ym ph ae a sp . Pi sc ico la g eo m et ra C la do ce ra C hi ro no m id ae in de t. Cr ist at el la m uc ed o Pi sc es in de t. Sc irp us la cu st ris rare common abundant Terrestrial Telmatic Lacustrine Li th ol og y 19 The clay is underlain by 2 cm of fine- to medium-grained grey sand. This thin sand layer may have been deposited at or after the final drainage of the Baltic Ice Lake. Below the sand layer and down to the bottom of the core at 570 cm, grey clay is found. This unit is rich in carbonate and barren of fossils and it is interpreted as late-glacial clay deposited in the Baltic Ice Lake during the Younger Dryas. From the deeper core 258000-2, a sample rich in plant remains was analysed for macrofossils. The plant remains were dominated by fruits and seeds of the telmatic plants Me- nyanthes trifoliata and fruits of Carex, mainly Carex vesicaria, Cladium mariscus, Scirpus lacustris, Comarum palustris and Cicuta virosa (Fig. 3). Limnic plants and animals were rep- resented by a seed of Nymphaea sp., common stems of Amb- lystegiaceae (mosses, not shown), rare shells and head shields of Chydoridae and rare head capsules of larvae of Chironomi- dae. The fossil assemblage and in particular the occurrence of abundant remains of telmatic plants show that the sediment was deposited in shallow water near the shore of a lake, prob- ably just outside the reed belt. The presence of fruits of Betula sect. Albae indicates that the land was covered by birch forests. The sample of Cladium mariscus fruits yielded an age of 10  980 ± 55 14C years BP (KIA-21680). This is calibrated to 12.674–13.069 cal. ka BP, according to the INTCAL09 dataset, which corresponds to the youngest part of the Al- lerød chronozone or the oldest part of the Younger Dryas. An age corresponding to the warm Allerød Chronozone was expected from the fossil assemblage, because Cladium mariscus and Scirpus lacustris are thermophilous plants. Cladium mariscus was recorded from late-glacial deposits in south-eastern Denmark by Bennike & Jensen (1995), but its presence was probably due to down-core contamination, and there are no secure records of it from late-glacial deposits in Denmark (Iversen 1954; Jensen et al. 1997; Bennike et al. 2004). Its northern geographical limit during the Allerød may thus have been located near the coring site. As the core was collected at a water depth of 39.5 m and the sample comes from a core depth of c. 5.8 m, the dated sample comes from a depth of c. 45.3 m below present sea lev- el. The sediment is fine-grained and was probably deposited at a water depth of several metres, and we suggest that the shore level during deposition was around 40 m lower than at present. Both before and after this lowstand episode, the relative shore level was around 20 m below the present sea level according to Jensen et al. (1997). This implies that the shore-level fall towards the end of the Allerød chronozone was of the same magnitude as the fall at the Younger Dryas – Holocene transition, i.e. considerably more than 5–10 m as suggested by Björck (1995). The new data allow us to modify the shore-level model proposed by Bennike & Jensen (1998) and extend it back in time. Figure 4 shows a new model for relative shore-level changes in the Arkona Basin from the last deglaciation to the present. It is seen that transgressions were interrupted by sudden regressions. South of the Arkona Basin, late-glacial sediments reach elevations lower than 20 m (Lampe 2005). However, the rela- tionship between these sediments and the regional shore level of the southern Baltic Basin is uncertain. Some of the sedi- ments are glaciof luvial and were deposited above shore level, other late-glacial sediments may have been deposited in local basins, perhaps in part dammed by bodies of stagnant ice. At present, the most enigmatic stage in the history of the Baltic Basin is that of the early Holocene Ancylus Lake; the shore-level curve for this stage has been drawn as a dashed line on Fig. 4. It has been suggested that the Ancylus regres- Fig. 4. Tentative curve showing relative shore-level changes in the Arkona Basin during the late-glacial and the Holocene. ka: kilo-annum (1000 years), BIL: Baltic Ice Lake, YS: Yoldia Sea, All: Allerød, YD: Younger Dryas. Modified from Bennike & Jensen (1998). Table 1. Macrofossils in core 258000 r: rare, c: common. 412–420 – – – – – – – – – – – c r – 420–430 – – – – – – – – – – – c r – 430–440 3 – – – – – – – – – – c r 4 440–449 2 – – – – – – – – – – c – 10 449–454 – – – – – – – – – – – c – 1 454–459 – – – – – – – – – – 5 c r 8 459–463 2 r – – – – – c – – 1 c r – 463–468 – – – – – – – c – – – c r – 468–473 1 r – – – – – c – 1 – c r – c. 580 3 – 50 15 150 20 1 c 200 1 – r r – Be tu la s ec t. Al ba e sp . D ep th (c m ) Pi nu s sy lve st ris Cl ad iu m m ar isc us Sc irp us la cu st ris Ca re x sp p. Co m ar um p al us tre Ci cu ta v iro sa Ra di ce lls M en ya nt he s tr ifo lia ta N ym ph ae a sp . Pi sc ico la g eo m et ra C hy do ri da e in de t. C hi ro no m id ae in de t. Cr ist at el la m uc ed o D ep th (m b .s. l.) 12 10 8 6 4 2 0 Age (cal. ka BP) 0 10 20 30 40 Littorina SeaAncylus LakeYSBIL Focus of this paper HoloceneYDAll 2020 sion was around 20 m (Björck 1995), but more recently fig- ures of 5 m and 10 m were also proposed (Björck et al. 2008; Rosentau et al. 2013). No erosional unconformity has been reported from the south-western Baltic Basin that formed during this regression ( Jensen et al. 1999). Figure 5 shows a model of the palaeogeography of the Baltic Basin after the late Allerød drainage. Although the Baltic Basin was now at the same level as the sea, it probably remained a freshwater lake. The connection to the sea was narrow, and we suggest that the outf low of huge amounts of fresh water coming from rivers and from the melting ice sheet hindered seawater from entering the Baltic Basin. Concluding remarks The rich occurrence of remains of reed plants at a depth of 45  m below sea level in the Arkona Basin provides firm evidence of a lowstand. A radiocarbon age shows that it oc- curred at the end of the Allerød Chronozone. We suggest that the shore level fell about 20 m, similar to the shore-level fall at the Younger Dryas – Holocene boundary. Acknowledgements The captain and crew of the former R/V Alexander von Humboldt from the Institute for Baltic Sea Research in Warnemünde are thanked for their help during the marine cruise. This paper is dedicated to the memory of Wolfram Lemke, who invited us to take part in the cruise during which cores 258000-1 and 258000-2 were collected. References Andrén, T., Björck, S., Andrén, E., Conley, D., Zillén, L. & Anjar, J. 2011: The development of the Baltic Sea during the last 130 ka. In: Harff, J. et al. (eds): The Baltic Sea Basin, 75–97. Berlin: Springer Verlag. Bennike, O. & Jensen, J.B. 1995: Near shore Baltic Ice Lake deposits in Fakse Bugt, southeast Denmark. Boreas 24, 185–195. Bennike, O. & Jensen, J.B. 1998: Late- and postglacial shore level changes in the southwestern Baltic Sea. Bulletin of the Geological Society of Denmark 45, 27–38. Bennike, O., Jensen, J.B., Lemke, W., Kuijpers, A. & Lomholt, S. 2004: Late- and postglacial history of the Great Belt, Denmark. Boreas 33, 18–33. Björck, S. 1995: A review of the history of the Baltic Sea, 13.0–8.0 ka BP. Quaternary International 27, 19–40. Björck, S., Andrén, T. & Jensen, J.B. 2008: An attempt to resolve the partly conf licting data and ideas on the Ancylus–Littorina transition. Polish Geological Institute Special Papers 23, 21–26. Iversen, J. 1954: The Late-Glacial f lora of Denmark and its relation to climate and soil. Danmarks Geologiske Undersøgelse II. Række 80, 87–119. Jensen, J.B., Bennike, O., Witkowski, A., Lemke, W. & Kuijpers, A. 1997: The Baltic Ice Lake in the southwestern Baltic: sequence-, chrono- and biostratigraphy. Boreas 26, 217–236. Jensen, J.B., Bennike, O., Witkowski, A., Lemke, W. & Kuijpers, A. 1999: Early Holocene history of the southwestern Baltic Sea: the Ancylus Lake stage. Boreas 29, 437–453. Lampe, R. 2005: Lateglacial and Holocene water-level variations along the NE German Baltic Sea coast: review and new results. Quaternary Inter- national 133–134, 121–136. Larsen, C.S. 2004: Sequence stratigraphy based on vibrocore description and shallow seismic data from the south-western Baltic Sea. Danmarks og Grønlands Geologiske Undersøgelse Rapport 2004/53, 28 pp. Lemke, W., Endler, R., Tauber, F., Jensen, J.B. & Bennike, O. 1998: Late- and postglacial sedimentation in the Tromper Wiek (western Baltic). Meyniana 50, 155–173. Moros, M., Lemke, W., Kuijpers, A., Endler, R., Jensen, J.B., Bennike, O. & Gingele, F. 2002: Regression and transgressions of the Baltic basin re- f lected by a new high-resolution deglacial and postglacial lithostratigra- phy for Arkona Basin sediments (western Baltic Sea). Boreas 31, 151–162. Rosentau, A. et al. 2013: Stone Age settlement and Holocene shore dis- placement in the Narva-Luga Klint Bay area, eastern Gulf of Finland. Boreas. http://dx.doi.org/10.1111/bor.12004 Uścinowicz, S. 2006: A relative sea-level curve for the Polish southern Bal- tic Sea. Quaternary International 145–146, 86–105. Wohlfarth, B., Björck, S., Funder, S., Houmark-Nielsen, M., Ingólfsson, Ó., Lunkka, J.-P., Mangerud, J., Saarnisto, M. & Vorren, T. 2008: Qua- ternary of Norden. Episodes 31, 73–81. Authors’ address Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark. E-mail: obe@geus.dk Fig. 5. Generalised palaeogeographical map of the Baltic region after the drainage at the end of the Allerød. Modified from Björck (1995) and Wohlfahrt et al. (2008). B: Billingen. 250 km Ice Sea Lake Land B 258000 http://dx.doi.org/10.1111/bor.12004 mailto:prj@geus.dk