A reconstruction of the effects of Post- Glacial Rebound on the lake system dynamics in the Baltic Basin since the Last Glacial Maximum to explain the high biodiversity anomaly.

(1)

Alex Nap

Universiteit van Amsterdam 04-07-2017, Amsterdam

Supervisor: Dr. K.F. Rijsdijk

A reconstruction of the effects of

Post-Glacial Rebound on the lake system

dynamics in the Baltic Basin since the

Last Glacial Maximum to explain the high

biodiversity anomaly.

(2)

1

Abstract

The Baltic Basin has had a dynamic history since the Last Glacial Maximum 22 000 years ago. The post-glacial rebound following the retreating of the ice sheet changed the landscape of the peri-Baltic region significantly. This study aims at identifying the effects of these landscape changes on freshwater lakes surrounding the Baltic Basin, to help clarify the hypothesis that the connectivity between lakes and the Baltic Basin made an exchange of freshwater lake species possible. With digital elevation and ice sheet models in ArcGIS the connectivity and ice coverage of selected major lakes in the area were analysed. The results show when the selected major lakes became ice-free, if and how they were connected to the Baltic Basin and when they disconnected. The main conclusion drawn from this study is that some lakes in the peri-Baltic have been connected with the Basin for thousands of years and have been part in the dynamic changes of the Baltic Basin, suggesting that an exchange of species would have been possible until recently.

Keywords: Baltic Basin, post-glacial rebound, Last Glacial Maximum, biodiversity, Fennoscandia, freshwater lakes.

(3)

2

Introduction

The Baltic Basin has had a dynamic geographical history during the last glacial periods. During the Quaternary glaciation, the area around the Baltic Basin has been covered by ice sheets several times for thousands of years and this came with a continuous changing landscape (Lehman & Jones, 1991). According to Björck (1995) there are four main reasons why the scientific institutes, mostly from countries around the Baltic Sea - also peri-Baltic Region - are interested in the geographical past of the Baltic Basin since the Last Glacial Maximum – LGM – 22 000 BP:

1) The salinity records of since the Last Glacial Maximum – LGM - 22 000 BP, have varied widely for a relative short period of time and the subsequent species changes have varied as well. 2) There had been land bridges connecting and disconnecting different parts of the peri-Baltic

Region and the waterbodies of the Baltic Basin, creating a way for flora and fauna to migrate to other areas.

3) The deglaciation events from the LGM until the forming of the current Baltic Sea as it is known today, it is the most recent large-scale glaciation of the Fennoscandian peninsula.

4) The Baltic Basin gives an extensive amount of varve chronology, allowing for detailed paleo-geographic conditions prevailing since the LGM.

(5)

4 The Baltic Basin is part of the Baltic Shield and is surrounded by the Fennoscandian peninsula. The depression that is today known as the Baltic Sea was formed by tectonically active horst and graben system that created the current low lying Finnish and Swedish lowlands (Sub-Cambrian peneplain) and the Baltic Basin (Gaál & Gorbatschev, 1987; Pira, Flodén & Mokrik, 2003). This basin was later filled with sediment and in the Miocene the Eridanos river system flowed through the Basin until the Middle Pleistocene, this could explain the slightly dendritic shape of the Baltic Basin, together with former tectonic activity (Pira, Flodén & Mokrik, 2003). During more recent, Quaternary, history the Baltic Basin has been covered three to seven times with an ice sheet, during the Elsterian, Saalian and Weichselian (Marks, 2004; Stewart & Lonegan, 2011). The focus in this study will be on the history of the Baltic Basin from the LGM until present.

During the LGM the Fennoscandian ice sheet covered extensive parts of Northern Europe. An area from Northern Germany, Denmark, Norway, Sweden, Finland, The Baltic States and parts of North Western Russia were occupied by an ice sheet, measuring 3 – 4 kilometres in thickness in some parts (Svendsen et al., 2004). The weight of this ice sheet was large enough to press down the Earth’s crust considerable amounts. In the central part of the ice sheet this meant that the crust was pressed down more than 400 metres compared to today (Peltier, 1996; Fjeldskaar et al., 2000). This effect is known as isostatic depression.

When around 18 000 BP the ice sheet started melting and retreating, the decline of the weight of the ice pressing down on the Earth’s crust began as well. Due to the elasticity of the crust, a vertical upward crustal motion started. This effect is called post-glacial rebound, this is a part of the isostasy theory (Peltier, 1996).

This post-glacial rebound and declining size of the ice sheet led to some sudden and severe changes in the Baltic landscape. Not only through the post-glacial rebound but the glaciers also left behind huge end moraines in Central Europe (Marks, 2002; Kalm, 2012). Lobes on the end of the ice sheet carved out areas and left depression in the landscape, later filled up with lakes and sediments (Kalm, 2012). Underneath the ice sheet several forms of erosion and sedimentation occurred, examples are drumlins and subglacial meltwater riverbeds. All these effects of glacial erosion and sedimentation are still visible and significant in the landscape.

This thesis will focus mainly on the forming of freshwater lakes and their relation to the retreating ice sheet and the newly formed waters of the Baltic Basin. In academic literature, there is consensus about a few stages of the Baltic Basin since the LGM, the system of Björck (1995; 2008) is considered a standard (table 1).

Table 1: Stages of the Baltic Basin according to Björck (2008):

Name of waterbody Salinity Age

The Baltic Ice Lake Freshwater 15.0 – 11.6 ka BP

Yoldia Sea Brackish 11.6 – 10.7 ka BP

Ancylus Lake Freshwater 10.7 – 8.5 ka BP

Littorina Sea / Baltic Sea Brackish/Salt 8.5 – present

It is important to notice that the shift from Littorina Sea to the current Baltic Sea does not have a clear distinct border but rather a gradual transition from a larger waterbody containing about twice the amount of water and approximately 1¼ the size of the current Baltic Sea (Björck, Andrén & Jensen, 2008). The salinity of these stages has been researched through molluscs in sediments, hence the names of the last three lakes, these are all gastropod species.

(6)

5 As Björck (1995) postulated in

the reasons why the Baltic Basin research is so extensive, floral and faunal migration is an important one. Recent research showed that there is an unexpected “hump” when one looks at the species richness graphs of freshwater lake species around the Baltic Sea between the 50th_{and 60}th

latitude (Hof et al., 2008; Dehling et al., 2010). This hump in species richness is in contrast with the general biogeographic rule that species richness decreases with increasing latitudes (Huggett, 2004). The researchers in these studies give the hypothesis that the connectivity between lakes around the Baltic Basin due to post-glacial rebound could be a reason that there is a species abundance there (Hof et al., 2008).

The area between the 50th_{and 60}th

latitude around the Baltic Sea coincides with the most dynamic part of this waterbody; the ice retreated here the first and all the regression and transgression events involving the Baltic Basin have included this part of the Basin (Björck, 2008). Therefore, this study will try to identify how the lakes and Baltic Basin have been connected and what their history was considering ice coverage, salinity stages and interconnectivity. The ice sheet eroded out basins and left behind other remnants where lakes could form (Kalm, 2012). It is necessary to include the effects of post-glacial rebound on the change of the peri-Baltic landscape and essential to combine this with the information that is available on the retreating Fennoscandian ice sheet to provide a basis for further research on the biodiversity “hump”.

The research for this thesis will include a model of the retreating ice sheet combined with the uplifting landscape and it will analyse these models.

The following research questions will be addressed on the given models and the read literature: ▪ How did the retreating ice sheet influence the forming of lakes around the current Baltic

Sea since the LGM?

▪ How did the post-glacial rebound influence the forming of lakes around the current Baltic Sea since the LGM?

Figure 2: The relation between latitude and species richness around the Baltic Basin (Hof et al., 2008)

Figure 3: Map of the total uplift rate of the Baltic Basin. Dark blue is the most uplift and yellow is no uplift or depression.

(7)

6 ▪ How did the connectivity of the freshwater lakes around the Baltic Basin change over

different stages in time since the LGM?

▪ How did the post-glacial rebound and retreating ice sheet affect the salinity and water quality of the current Baltic Sea and major lakes in the peri-Baltic region?

Methods and Data

The data used for this thesis consists of three parts. The first one is a set of Digital Elevation Models (DEM), the second one reconstructions of the Fennoscandian ice sheet and the last one polygons of lakes in the peri-Baltic region, for use in ArcGIS.

The DEMs were obtained from Erik Koene, a PhD student from ETH Zürich. These models were made using the current DEM of Europe and by using an algorithm a new DEM was created for every 1000 years until 22 000 years BP. This DEM shows the elevation of Northern Europe per square kilometre every time step of 1000 years. This also includes the sea level at that time for every time frame of the DEM. This is important because it shows the connectivity between the large waterbodies of the North Sea and Baltic Sea and surrounding lakes, bays and inlets. This means that 0 metres of altitude on the DEM means 0 metres of altitude at that given time.

From a research group of the University of Bergen in Norway a reconstruction of the Fennoscandian ice sheet was obtained. This reconstruction was made by the team by using all the existing literature of the Fennoscandian ice sheet and eventually putting this information in a reconstruction model (Hughes et al., 2016). The reconstruction shows the minimum, maximum and most probable extent of the ice sheet. Here also time steps of 1000 years are used, going back slightly further in time than the DEMs of Koene, about 25 000 years.

The polygons of lakes were used before in a biodiversity research by Georgopoulou et al. (2016). The polygons show the shape and location of lakes around the Baltic Sea. This is necessary to see, using the aforementioned model and reconstruction, where a lake is positioned geographically.

The main software used for this thesis is ArcGIS. This is a Geographical Information System. All the parts mentioned above were imported to ArcGIS for further use.

When all the DEMs, ice sheet reconstruction maps and lake polygons were imported into ArcGIS the research process could start. The first calculation that was carried out was to measure what the post-glacial rebound rates were according to the DEMs. The first DEM of 0 ka BP was subtracted from the last DEM of 21 ka BP. This created a map where it was visible where the post-glacial rebound was on its maximum and where the least rebound had taken place or where there even was a post-glacial depression.

The next step was creating maps that were more convenient to use than the DEMs of Koene, because these showed just the elevation without thresholds of certain altitudes or sea level. It was necessary for the convenience of the analysis to show especially the different sea levels and connectivity changes. The new maps were made by reclassifying Koene’s DEMs in ArcMap and used different colours to show different elevations. These newly created maps also provided a more aesthetic view.

(8)

7 A selection of lakes was made for the analysis of connectivity and ice free-ness. This selection was based upon several criteria. The reasons of choice and list of lakes is visible in the appendix.

Figure 4: The selection of lakes. 1. Holstein and Pomeranian Lakeland 2. Masurian Lakeland 3. Vänern, Vättern and Hjalmaren (left to right) 4. Peipus, Ladoga and Onega (left to right) 5. Saimaa

Results

The used methods in ArcGIS eventually led to some interesting maps of the Baltic Basin. The results of these mapping processes will be shown below. Most of the maps show how lakes are connected to the Baltic Basin and when they became ice free.

(9)

8

Overlook of the history of the Baltic Basin from 21 ka BP until 1 ka BP

Figure 5a: Reconstruction of Baltic Basin 21 ka BP, 1 = Pomeranian lake district, 2 = Masurian lake district, 3 (from left to right): Vättern, Vänern and Hjälmaren, 4 = Peipus, 5 = Saimaa, 6 = Ladoga, 7 = Onega.

21 ka BP

The Fennoscandian ice sheet is still covering the Baltic Basin in this period. All selected lakes are still beneath the ice except for the moraine lake districts in Central Europe, some of these are still half covered with ice and some are in front of the ice sheet. Other lake groups, such as the Mecklenburg Lake Plateau were also formed at the edge of the ice sheet in an area with moraines and tunnel lakes but probably due to the post-glacial rebound and shape of moraines these lakes debouche in the Elbe and eventually the North Sea instead of the Baltic Sea and have therefore been excluded from this study (figure 5a).

(10)

9 Figure 5b: Reconstruction of Baltic Basin 17 ka BP

17 ka BP

The ice sheet is retreating in this period. The moraine lake districts are not covered by the ice sheet anymore. The first ice lobes can be seen stretched from lakes Ladoga and Onega. The coastline can be already seen transgressing towards the center of the Baltic Basin compared to 21 ka BP, for instance the patch of land between Peipus and Ladoga which used to be an island and is now connected (figure 5b)

(11)

10 13 ka BP

The lakes Ladoga, Peipus and Onega are ice free in this period. The ice sheet has retreated from the southern part of the Baltic Basin and on the map a narrow strait is visible between southern Sweden and Denmark. This could mean that saltwater could have flowed through the strait into the Baltic Basin, creating a saltwater environment in some places and on other place brackish or freshwater due to the large runoff of the melting ice sheet (figure 5c).

(12)

11 Figure 5d: Reconstruction of Baltic Basin 11 ka BP

11 ka BP

An extra map was added between the 4 year steps because between 13 ka BP and 9 ka BP most of the changes happened in the size of the ice sheet. Some spall patches of land can be seen on present day Finland forming a small archipelago. North of Onega is a large area visible of the White Sea which still is connected to the Baltic Basin. It is not sure whether this connection went further east towards the Arctic Ocean or if there was a land bridge connecting the Scandinavian peninsula and Eurasia. The strait between Denmark and Sweden is still open as well (figure 5d).

(13)

12 9 ka BP

The ice sheet completely vanished in this period. The shape and contour of Finland are visible as an archipelago, still containing the contours of present day Lake Saimaa between these islands. Lake Vättern and Hjälmaren are still connected as a strait between the Baltic Basin and Skagerrak. Between Denmark and Sweden straits connecting the Baltic Basin and Skagerrak are visible as well. Peipus and Vänern are disconnected from the Basin and are probably freshwater lakes connected only through rivers to the Basin. Ladoga is a large bay south of Finland and Onega is still connected most likely to the White Sea (figure 5e).

(14)

13 5 ka BP

The only lakes still connected to the Baltic Basin are Ladoga and Hjälmaren. Finland is no longer an archipelago but proper land. The straits between Denmark and Sweden have widened and the straits that contained Vättern and Hjälmaren are no longer present. Hjälmaren is now an inlet and Ladoga is rather an inland sea connected via a small strait to the Baltic Basin than a bay.

(15)

14 1 ka BP

Ladoga is still connected with the Finnish gulf via a very narrow strait according to the map. The rest of the lakes have all been disconnected from the main Baltic Basin.

Closer look at some lakes’past

Figure 6: Map of the maximum extent of the ice sheet 21 ka BP. The purple highlighted lakes are the Masurian and Pomeranian Lake districts, formed on the edge of the ice sheet.

(16)

15 The maximum extent of the ice sheet 22 ka BP and the present location of the lake districts can be seen at the edge of the ice sheet (figure 6). The lakes are known to have formed on the end moraine areas, leaving many small basins and valleys behind.

The left part of figure 8 shows how the lobes of the retreating ice sheet are roughly the same shape as the present lakes Ladoga and Onega, the sheet has already retreated from lake Peipus. The right part shows how Lake Ladoga according to the model is still connected to the Finnish Gulf about 1000 years BP.

Figure 8: Stages of Lake Peipus, Ladoga and Onega, 12 ka BP (left) and 01 ka BP (right)

Figure 7: The Baltic Ice Lake Stage, with the ice sheet damming the area over South Sweden. But in these results a narrow strait between Denmark and Sweden is visible.

(17)

16 Figure 9: the connection of the South Swedish lakes (left to right): Vänern, Vättern and Hjälmaren, 10 ka BP. As can be seen in figure 4 the lake system of Southern Sweden formed a connection between the Baltic Basin and the Skagerrak 10 ka BP. The area of Hjälmaren here is still surrounded by an extensive amount of water and will be disconnected from the Baltic Basin 5 ka BP.

Table 2 shows the most important data from this thesis in one figure. The maps from figures 5-8 were part of the larger model that was made of every 1000 years. The analysis shows when the lakes have become ice-free, so when the ice sheet was retreated from the area of the present lake and how they were connected to the Baltic Basin and how their interconnectivity was. To help visualize these results a figure has been made, this demonstrates how lakes were connected in the Baltic Basin at the same time. The blue area within the black line between the Yoldia and Baltic stage is the Ancylus Lake stage, when the Baltic Basin was a freshwater basin.

The data from table 2 can be visualized to give a more clear picture of when which lake became ice free, when a lake was disconnected from the Baltic Basin and what the rough salinity was of that period (figure 9a & 9b).

(18)

17 Table 2: Lake ice coverage and connectivity, visualized in figure 9:

Lake or Lakeland Ice Free Connection Baltic Basin Interconnectivity Holsteinische Schweiz

(Lake District)

20 ka BP None except for rivers. Yes, swamps and rivers. Pomeranian Lake

District

20 ka BP None except for rivers. Yes, swamps and rivers. Masurian Lake

District

17 ka BP None except for rivers. Yes, swamps and rivers. Lake Peipus 14 ka BP Yes, bay of Baltic Ice Lake and

Yoldia Sea until 10 ka BP.

-

Lake Vänern 11 ka BP Yes, either a bay of the Baltic Ice Lake or part of a strait during Yoldia Sea and Ancylus Lake. Disconnected 8 ka BP.

Yes, with Vättern and Hjälmaren, until 2 ka BP.

Hjälmaren 11 ka BP Yes, a bay or strait during all the stages of the Baltic Basin, including Littorina Sea. Disconnected 5 ka BP.

Yes, with Vättern and Vänern, until 2 ka BP.

Lake Vättern 12 - 11 ka BP

Yes, either a bay of the Baltic Ice Lake or part of a strait during Yoldia Sea and Ancylus Lake. Disconnected 9 ka BP.

Yes, with Vänern and Hjälmaren, until 2 ka BP.

Lake Ladoga 12 ka BP Yes, as a bay in every stage of the Baltic Basin until 1 ka BP.

Yes, from 7 ka BP with Onega and from 5 ka BP with Saimaa.

Lake Onega 12 ka BP Yes, as a bay of the Yoldia Sea and Ancylus Lake, then disconnected from Baltic Basin.

First connected to White Sea and then changed watershed around 5 ka BP to Lake Ladoga.

Saimaa 11 ka BP Yes, was part of a larger area of lakes and islands during every stage from Yoldia Sea to Littorina Sea.

Yes, in the same watershed as Ladoga.

(19)

18 Both the results for a reconstruction of lake stages can be made from existing literature (figure 9a) and from the results from table 2 (figure 9b). These results show a difference between the salinity of the lakes still part of the Baltic Basin from 14 ka BP until 9 ka BP. This reconstruction was based on the models created by using the DEMs of Koene, where a strait between Denmark and Sweden is constantly visible in that period suggesting, together with the large freshwater outflow from the retreating glacier, a brackish climate.

Figure 10a: Stages for every lake per ka BP according to literature (Davydoval, 1995; Björck, 2008; Bendixen et al. 2017) : White = ice sheet coverage; blue = freshwater; light green = brackish or saltwater/ Yoldia Sea; Green = Saltwater/Baltic Sea. The area within the black line is when lakes have been part of the Baltic Basin waterbody according to the model. The dotted line is the Ladoga catchment area.

(20)

19

Discussion

The effects of the ice sheet and its retreat have had different effects on the forming of the selected lakes around the Baltic Sea. It is convenient to put them in four different groups, based on common paleogeographic evolution:

1. Lake Districts: Holstein, Pomerania and Masuria 2. Lake Peipus, Onega and Ladoga

3. Vänern, Vättern and Hjälmaren 4. Saimaa

1. Lake Districts

The Lakelands of Holstein, Pomerania and Masuria have some landscape features in common. If the map is observed of 21 ka BP (figure 7) it is visible that the lakes are on the edge of the maximum extent of the ice sheet. The ice sheet pushed forward all sediments and shaped end moraines on the outer edges (Marks, 2002). Just behind the end moraines the glacier carved out depressions known as tunnel lakes in the landscape (Marks, 2002). When the ice sheet retreated the plains were part of glaciofluvial systems for a while and then partly covered by the Baltic Ice Lake. This area, covering the coastline and land deeper inwards from Jutland in Denmark towards Estonia is called the Baltic Uplands (Houmark-Nielsen, 2011). This is a stretched chain with end moraines of about 200 metres in altitude and right before this elevated chain lay several lake districts. The borders between these lake districts are mostly shaped through common catchment areas (Raukas & Gaigalas, 1993). The lake districts of Holstein, Pomerania and Masuria are also a part of this chain. The lakes were never part of the Baltic Basin’s large waterbody but are connected through streams and swamps to the Baltic Basin.

The Lakelands were also affected by the post-glacial rebound and this could have influenced watershed forming. Most of the lakes are well interconnected through streams flowing through old glacier eroded valleys or through vast wetlands, this is not visible in the analysis (Bajkiewicz-Grabowska & Borowiak, 2008; Hillbricht-Ilkowska, Rybak & Rzepecki, 2000).

2. Peipus, Onega and Ladoga

These three large lakes all share the same origin. As is visible (figure 8) the ice sheet lobes covered the areas of the lakes during different stages of the ice sheet decline (Saarnisto & Saarinen, 2001). The lobes eroded and carved out some depressions approximately the current shape of these lakes. During the Baltic Ice Lake stage, the meltwater from the ice sheet filled up these depressions left by the retreating glaciers. Onega and Peipus became ice free between 15 and 14 ka BP and became glacier lakes subsequently. The ice lobe covering Ladoga retreated about 1000 years later and the lakes became part of the larger Baltic Ice Lake. A large waterbody that is no longer extant northeast of Onega is visibly connected to Ladoga and Peipus and via another waterway to the larger Ice Lake in 13 ka BP. After these events, the ice sheet did not directly affect the three lakes through erosion or sedimentation anymore.

The post-glacial rebound largely affected the lakes Peipus, Ladoga and Onega. Due to the uplift, the first lake to lose its connection to the Baltic Basin was lake Peipus. According to the analysis this would have happened between 12 ka BP and 11 ka BP. The lake was however still connected through an outlet with the Baltic Basin. Lake Onega lost its connection with the Baltic Basin and Lake Ladoga around 10 ka BP. Due to post-glacial rebound the lake eventually changed its watershed to the White Sea (Saarnisto & Saarinen, 2001). Later, due to erosion and post-glacial rebound the Onega basin changed from catchment area again and started flowing to Lake Ladoga and the Baltic Basin again, this happened around 7 ka B. This change of watershed however is hardly visible on the maps because the analysis maps just show if there is connectivity and not with which waterbody, White Sea or Baltic Basin.

(21)

20 Lake Ladoga was the last lake to lose its connection to the Baltic Basin. According to the analysis Ladoga became disconnected from the Baltic Sea just 1 ka BP due to the post-glacial rebound. This could mean that the river Neva was a sea arm just until recently. It is postulated however, as the Finnish Gulf where Ladoga has its outflow is currently on the threshold of freshwater, this could also be true for Lake Ladoga at that time. The outflow channel would have been very narrow and the inflow of freshwater would have been quite abundant with the connections of Lake Saimaa and Lake Onega flowing towards Ladoga. Davydoval et al. (1996) on the other hand conclude that Ladoga got disconnected with Ancylus Lake around 8 ka BP and that it was until the Neva River started draining Ladoga into the Baltic Basin around 3 ka BP when Ladoga was connected with the Baltic Sea again, together with the other lakes in the Ladoga catchment area, Onega and Saimaa.

3. Lake Saimaa

Lake Saimaa’s fundaments lay on an extensive drumlin field in Southern Finland (Nenonen & Portaankova, 2009). For most of the time since the LGM this area was an archipelago shaped by small islands and waterways eroded away by the retreating glacier. During the post-glacial rebound stage, the archipelago’s land area grew while the waterways in between the islands slowly decreased in size. The area of the lake lost its connection with the Baltic Basin between 7 ka BP and 6 ka BP according to the model. After it was disconnected Lake Saimaa started to drain towards Lake Ladoga and currently shares its catchment area with Lake Ladoga and Onega through the Vuoksi River (Nenonen & Portaankova, 2009).

4. Lake Vänern, Vättern and Hjälmaren

Vänern, Vättern and Hjälmaren are lakes that all have a tectonic history before being covered by ice sheets during the last glacial. There were already depressions before the LGM started and covered them with ice. The ice however did carve out sediments from the lakes and probably deepened them during the LGM (Kvarnäs, 2001). All the lakes were ice free between 12 ka BP and 11 ka BP according to the analysis.

When the ice sheet left the area of the Mid-Swedish lakes they formed a connection between the North Sea and the newly formed Yoldia Sea 12 ka BP, after the Baltic Ice Lake drained. This connection lasted until the end of the Ancylus Lake stage 9 ka BP, where the lake had been one of the outflows, together with the waterways in the present Danish Archipelago, visible on figure 8. Then the rebound lifted the land beneath Vänern and Vättern so much that they lost the connection with the Baltic Basin at sea level. Lake Hjälmaren however was an inlet for the Baltic Basin until 5 ka BP. Hjälmaren is still well connected via waterways and Lake Mälären with the Baltic Sea.

The post-glacial rebound and retreating ice sheet and stages of the Baltic Basin.

The retreating ice sheet and the post-glacial rebound both had large effects on the shape and landscape of the Baltic Basin. As can be seen in the model, visualized in figure 6, the ice sheet dammed the outlet of the Baltic Ice Lake, covering the depression where currently the South-Swedish lakes are located. But when this dam melted away the Lake drained and water from the North Sea could flow into the Basin, probably turning it into brackish water, combining the saltwater from the sea with the fresh- meltwater from the glaciers. In the model (figure) there is still a much broader strait connecting the BIL and the Skagerrak than would be expected, explained further in the relations to the literature paragraph. The post-glacial rebound changed the shape of the Baltic Basin since the LGM. The Baltic Basin became narrower and less deep due to the uplift and many inlets got cut off and became either lakes or land; such as Lake Ladoga - inlet to lake - or the areas in Southeast Sweden - inlet to land. During the Ancylus Lake stage, 10.7 – 8.5 ka BP, it may have been possible for freshwater species to spread out over vast areas of the Baltic Basin and exchange biodiversity. Later when the Littorina Sea substituted the Ancylus Lake the area of the waterbody was still much larger than it was today and connections with present day

(22)

21 lakes were still abundant. This makes it clear how vivid the history of the Baltic Basin was over a relative short amount of time.

Improvements and the models’ relation to literature

Some of the outcomes of the model show similarities with existing literature, as is explained in the prior part of the discussion. This is for instance the case with the Lake Districts, Peipus and Lake Saimaa. However, there are some parts of the model that seem to contradict with the academic literature on the geography of the Baltic Basin. The created maps of the Lake districts in the results are coherent with existing literature about these areas (Marks, 2002; Raukas & Gaigalas, 1993). The lake districts could be worth for biodiversity research through their interconnected wetlands and streams and further research if these streams and wetlands could have been well connected to the Baltic Basin’s freshwater stages as well could be a meaningful addition to the biodiversity anomaly research.

The difficulty with using the DEMs from Koene was that all the maps created using the algorithm have the current DEM of Europe as a basis. This means for instance that the Dutch Flevopolder is still visible in 8 ka BP and that some geographical features that are visible on the maps might not have been there at that given period of time.

This proved to be an obstacle when reconstructing the time frame for the Baltic Ice Lake stage an Ancylus Lake stage of the Baltic Basin in particular. According to Bjöck (2008) the BIL had no connection with the Skagerrak and was dammed by a land bridge between Denmark and Sweden and the ice sheet towards the north. On the maps of the analysis of the period between 15 – 12 ka BP – a period of which general agreement exists in literature that the Baltic Basin was an ice lake - there is a strait connecting the Skagerrak and the Baltic Basin (Björck, 1995; 2008). This suggests an inflow of saltwater into the ice lake.

In literature, there is also a discussion about where the Ancylus lake drained in the North Sea, either via the Hjälmaren – Vänern - Göta system (HVG) through Central Sweden or through channels between the Belt islands of Denmark. A sudden drop of water level is observed in the HVG outlet system around 10.5 ka BP and the Ancylus Lake started draining through the straits between Denmark and Germany, also known as Belt (Björck, 2008). On the maps of the analysis there is both a connection through the HVG and the Belt systems. Feldens & Schwarzer (2012) conclude that there have been deeply incised valleys through the seafloor of the Belt islands where the Ancylus Lake would have drained into the Skagerrak and where later the saltwater of the Skagerrak entered the Baltic Basin, creating the Littorina Sea, 8 ka BP. The maximum width of this incised valley according to Feldens & Schwarzer (2012) is about 1 kilometre, while the models made for this thesis show a strait of about 8 and 6 kilometres wide at different places between Denmark and Sweden between 11 and 9 ka BP. There is a plausible explanation for this. During the transition to the Littorina Sea, 8 ka BP, the waters of the Great Belt and Oresund suddenly deepened, in some places more than 10 metres (Berglund et al., 2005). The event that caused this was a rapid eustatic sea level rise after the Ancylus stage of the Baltic Basin around 9 ka BP; the sediments left by the Ancylus drainage system were eroded away due to rapid flooding of the area (Bendixen et al., 2017) This sudden deepening could explain why the models show a strait during the BIL – 15 ka BP until 12 ka BP – and Ancylus lake period – 10 ka BP until 9 ka BP, because they use the DEM where the straits are already deepened out, calculating them back to other periods where they did not exist yet. This is also why in according to the models a brackish environment is chosen instead of freshwater in the visualization of results, using the DEMs of the model (figure 9a & 9b).

This flaw in the model could be responsible for more contradictions with existing literature. For instance the connection of Ladoga with the Baltic Basin until very recently. It could be that the deepening of the Neva 3 ka BP (Davydoval et al. 1996) is also visible on the older maps when the Neva did not drain Ladoga yet. The valley of the Neva is then calculated back to other periods longer ago where the Neva valley did not exist yet, creating a connection in the analysis between Ladoga and Baltic Basin on the maps.

(23)

22

Consequences for biodiversity

The initial reason to research the dynamics of the Baltic Basin since the LGM is the unusual high biodiversity rates of the area. The results of the model show that there indeed was a strong connectivity between lakes and the Baltic Basin for significant periods since the LGM. Especially the connections during the BIL, 15.0 – 11.6 ka BP and Ancylus Lake stage, 10.7 – 8.5 ka BP, could help explain why there is the anomaly of freshwater species in lakes in the peri-Baltic region. However this model suggestss a more brackish environment, contradicting to the freshwater environment that is suggested in other literature such as Davydoval et al. (1995), Björck (2008), Feldens & Schwarzer (2012) and Bendixen et al. (2017). Some lakes – Hjälmaren and Ladoga - have been connected for to the brackish/saltwater stages of the Littorina and Baltic Sea as well, suggesting that in these lakes the salinity had been higher for a longer period. This study has tried to bring some insight into how the Baltic Basin changed shape and how waterbodies were connected with each other after the ice sheet melted away. A new hypothesis could be stated that the interconnectivity in the Baltic Basin made an exchange of species possible, explaining why the biodiversity is so high. This could be an interesting topic for future research.

Conclusion

This study aimed to reconstruct and analyse the changes of the Baltic Basin since the LGM. The Baltic Basin has had a myriad of changing landscapes since the LGM. This is mostly due to retreating ice sheet and the subsequent post-glacial rebound. This also meant that freshwater lakes around the Baltic Basin were part of these extreme landscape changes. Some lakes have been part of the large waterbody of the Baltic Basin for thousands of years and others have been formed by the ice sheet’s erosion and sedimentation. The connectivity of the Baltic Basin is remarkable, where very large waterbodies such as Ladoga have been part of the Baltic Sea until very recently. This connectivity together with the often-changing salinity of the Baltic Basin could have influenced the exchange of species in the peri-Baltic region and help explain the “hump” of freshwater lake species. There is however a large discrepancy between the connectivity found in literature and connectivity of the model used, especially of the period between 15 ka BP and 8 ka BP, where the models suggest a brackish or salt aquatic environment whereas literature describes two freshwater stages; the Baltic Ice Lake, 15.0 – 11.7 ka BP, and the Ancylus Lake stage, 10.7 – 8.6 ka BP. This is an interesting starting point for further research into the aquatic environment of the Baltic Basin and the anomaly in biodiversity on the crossroads of biology and geography.

(24)

23

Literature

1. Bajkiewicz-Grabowska, E., & Borowiak, D. (Eds.). (2008).Anthropogenic and natural transformations of lakes. Department of Limnology, University of Gdańsk.

2. Bendixen, C., Jensen, J. B., Boldreel, L. O., Clausen, O. R., Bennike, O., Seidenkrantz, M. S., & Hübscher, C. (2017). The Holocene Great Belt connection to the southern Kattegat, Scandinavia: Ancylus Lake drainage and Early Littorina Sea transgression. Boreas, 46(1), 53-68.

3. Berglund, B. E., Sandgren, P., Barnekow, L., Hannon, G., Jiang, H., Skog, G., & Yu, S. Y. (2005). Early Holocene history of the Baltic Sea, as reflected in coastal sediments in Blekinge, southeastern Sweden.Quaternary International,130(1), 111-139.

4. Björck, S. (1995). A review of the history of the Baltic Sea, 13.0-8.0 ka BP.Quaternary international,27, 19-40.

5. Björck, S. (2008). The late Quaternary development of the Baltic Sea basin.Assessment of climate change for the Baltic Sea Basin, 398-407.

6. Björck, S., Andrén, T., & Jensen, J. B. (2008). An attempt to resolve the partly conflicting data and ideas on the Ancylus-Littorina transition.Polish Geological Institute Special Papers,23,

21-26.

7. Darracq, A., Greffe, F., Hannerz, F., Destouni, G., & Cvetkovic, V. (2005). Nutrient transport scenarios in a changing Stockholm and Mälaren valley region, Sweden. Water Science and

Technology, 51(3-4), 31-38.

8. Davydoval, N. N., Arslanov, K. A., Khomutova, V. I., Krasnov, I. I., Malakhovsky, D. B., Saarnisto, M., ... & Subetto, D. A. (1996). Late-and postglacial history of lakes of the Karelian Isthmus. Hydrobiologia, 322(1), 199-204.

9. Dehling, D. M., Hof, C., Brändle, M., & Brandl, R. (2010). Habitat availability does not explain the species richness patterns of European lentic and lotic freshwater animals.Journal of biogeography,37(10), 1919-1926.

10. Feldens, P., & Schwarzer, K. (2012). The Ancylus Lake stage of the Baltic Sea in Fehmarn Belt: Indications of a new threshold?.Continental Shelf Research,35, 43-52.

11. Fjeldskaar, W., Lindholm, C., Dehls, J. F., & Fjeldskaar, I. (2000). Postglacial uplift, neotectonics and seismicity in Fennoscandia.Quaternary Science Reviews,19(14), 1413-1422

12. Gaál, G., & Gorbatschev, R. (1987). An outline of the Precambrian evolution of the Baltic Shield.Precambrian Research,35, 15-52.

13. Georgopoulou, E., Neubauer, T. A., Harzhauser, M., Kroh, A., & Mandic, O. (2016). Distribution patterns of European lacustrine gastropods: a result of environmental factors and deglaciation history.Hydrobiologia,775(1), 69-82.

14. Hillbricht-Ilkowska, A., Rybak, J., & Rzepecki, M. (2000). Ecohydrological research of lake– watershed relations in diversified landscape (Masurian Lakeland, Poland).Ecological Engineering,16(1), 91-98.

15. Hof, C., Brändle, M., & Brandl, R. (2008). Latitudinal variation of diversity in European freshwater animals is not concordant across habitat types.Global Ecology and Biogeography,17(4), 539-546.

16. Houmark-Nielsen, M. (2011). Pleistocene glaciations in Denmark: a closer look at chronology, ice dynamics and landforms.Developments in Quaternary Science,15, 47-58.

17. Hugget, R. J. (2004). Fundamentals of biogeography. Routledge Fundamentals of Physical

Geography, New York.

18. Hughes, A. L., Gyllencreutz, R., Lohne, Ø. S., Mangerud, J., & Svendsen, J. I. (2016). The last Eurasian ice sheets–a chronological database and time‐slice reconstruction, DATED‐ 1.Boreas,45(1), 1-45.

19. Kalm, V. (2012). Ice-flow pattern and extent of the last Scandinavian Ice Sheet southeast of the Baltic Sea.Quaternary Science Reviews,44, 51-59.

(25)

24 20. Kvarnäs, H. (2001). Morphometry and hydrology of the four large lakes of Sweden.AMBIO: A

Journal of the Human Environment,30(8), 467-474.

21. Lehman, S., & Jones, G. A. (1991). Initiation of Fennoscandian ice-sheet retreat during the last deglaciation.Nature,349(6309), 513.

22. Marks, L. (2002). Last glacial maximum in Poland.Quaternary Science Reviews,21(1),

103-110.

23. Marks, L. (2004). Pleistocene glacial limits in Poland.Developments in Quaternary Sciences,2,

295-300.

24. Nenonen, J., & Portaankorva, A. (2009). The geology of the lakeland Finland area.Geological Survey of Finland–Northern Periphery Program.

25. Peltier, W. R. (1996). Mantle viscosity and ice-age ice sheet topography.Science,273(5280),

1359.

26. Pira, V., Flodén, T., & Mokrik, R. (2003). The Baltic Sea basin in the geology of Fennoscandia and Baltic region.

27. Raukas, A., & Gaigalas, A. (1993). Pleistocene glacial deposits along the eastern periphery of the Scandinavian ice sheets‐an overview.Boreas,22(3), 214-222.

28. Saarnisto, M., & Saarinen, T. (2001). Deglaciation chronology of the Scandinavian ice sheet from the Lake Onega basin to the Salpausselkä end moraines.Global and Planetary Change,31(1), 387-405.

29. Stewart, M. A., & Lonergan, L. (2011). Seven glacial cycles in the middle-late Pleistocene of northwest Europe: Geomorphic evidence from buried tunnel valleys.Geology,39(3), 283-286.

30. Svendsen, J. I., Alexanderson, H., Astakhov, V. I., Demidov, I., Dowdeswell, J. A., Funder, S., & Hubberten, H. W. (2004). Late Quaternary ice sheet history of northern Eurasia.Quaternary Science Reviews,23(11), 1229-1271.

Acknowledgements

I would like to thank my supervisor Kenneth Rijsdijk for his help and feedback on this thesis, his academical feedback was indispensable. I want to thank Elisavet Georgopoulou for her help with the biodiversity explanations, the help at the start of this project and the lake models. Thanks to Erik Koene for his DEMs provided for this project, they formed the fundament for this study. And at last I would like to thank the team of the University of Bergen that created the reconstructions of the ice sheet since the LGM.

(26)

25

Appendix 1: Selection of lakes

Lake Background information on lakes

Holsteinische Schweiz (Lake District)

Lakeland on the edge of the maximum extent of the ice sheet. Lies in a land of end moraines and glacial eroded valleys. Strong interconnectivity between the lakes in the area, rivers flow towards Baltic Sea.

Pomeranian Lake District Lakeland on the edge of the maximum extent of the ice sheet. Lies in a land of end moraines and glacial eroded valleys. Strong interconnectivity between the lakes in the area, rivers flow towards Baltic Sea

Masurian Lake District An extensive area with interconnected lakes situated on the place where the ice sheet extended the farthest. The lakes are formed near the end moraines on eroded basins.

Lake Peipus A large lake on the border of Estonia and Russia. Has a large catchment area in a lowland with an outflow in the Baltic Sea. Lake Vänern The largest lake of Sweden, formed by a depression in a now

eroded mountain chain. Then the Fennoscandian ice sheet eroded bedrock away from this depression further deepening the lake. Hjälmaren Lake near the coast of Sweden. Formed by tectonic horst and

graben activity. Has been connected via Malmaren Lake to the Baltic Sea until now (Darracq et al., 2005).

Lake Vättern Second largest lake of Sweden, formed in an old graben fault. Has an interesting history together with Vänern, both have been part of the connection Skagerrak – Baltic Basin.

Lake Ladoga Largest lake in Europe. Has an abundance of species (Georgopoulou et al., 2016).

Lake Onega Second largest lake in Europe. Changed watershed two times from Ladoga to White Sea and back again to Ladoga.

Saimaa A very young lake in South Finland. Is rather a collection of different interconnected basins than a proper lake. Is very much affected by post-glacial rebound. Lies in the catchment area of Ladoga and Onega