Palaeogeographic reconstruction of proglacial lakes in Estonia ¨ RI VASSILJEV, LEILI SAARSE AND AVO MIIDEL ALAR ROSENTAU, JU
BOREAS
Rosentau, A., Vassiljev, J., Saarse, L. & Miidel, A. 2007 (April): Palaeogeographic reconstruction of proglacial lakes in Estonia. Boreas, Vol. 36, pp. 211 221. Oslo. ISSN 0300-9483. This paper describes a Geographical Information System (GIS)-based palaeogeographic reconstruction of the development of proglacial lakes formed during deglaciation in Estonia, and examines their common features and relations with the Baltic Ice Lake. Ice marginal positions, interpolated proglacial lake water levels and a digital terrain model were used to reconstruct the spatial distribution and bathymetry of the proglacial lakes. Our results suggest that the proglacial lakes formed a bay of the Baltic Ice Lake after the halt at the Pandivere Neva ice margin about 13.3 cal. kyr BP. Shoreline reconstruction suggests that two major proglacial lake systems, one in eastern and the other in western Estonia, were connected via a strait and thus had identical water levels. The water budget calculations show that the strait was able to transfer a water volume several times greater than the melting glacier could produce. As this strait compensated for the water level difference between the two lake parts, the subsequent further merging in north Estonia did not result in catastrophic drainage, as has been proposed. Alar Rosentau (e-mail:
[email protected]), Institute of Geology, University of Tartu, Vanemuise 46, 51014 Tartu, Estonia; Ju¨ri Vassiljev, Leili Saarse and Avo Miidel, Institute of Geology at Tallinn University of Technology, Ehitajate tee 5, 19086 Tallinn, Estonia; received 27th January 2006, accepted 1st August 2006.
During deglaciation of the Late Weichselian ice sheet, large proglacial lakes were formed in front of the retreating ice (Kvasov 1979; Bjo¨rck 1995; Mangerud et al. 2004). Their size and water level varied considerably depending on the position of the retreating ice sheet, isostatic rebound and the location of outlets. Two major proglacial lake systems developed in Estonia during the last deglaciation, one in the Lake Peipsi basin in eastern Estonia (Raukas & Ra¨hni 1969; Hang 2001) and the other in western Estonia (Lo˜okene 1959; Pa¨rna 1960; Raukas et al. 1971; Kessel & Raukas 1979). The proglacial lake in the Lake Peipsi basin formed when ice retreated from the HaanjaLuga marginal formations (Fig. 1) about 14.7 cal. kyr BP (corrected varve years of Kalm 2006 after Sandgren et al. 1997) and gradually enlarged to the north as ice retreated further. The level of Glacial Lake Peipsi lowered significantly during deglaciation of the lake basin as a result of shifting outlets and isostatic rebound (Fig. 2). A large proglacial lake in western Estonia emerged during the PandivereNeva stade about 13.3 cal. kyr BP (Fig. 1B, 3), when western Estonia became ice-free, and it is traced in relief by the two highest coastlines, termed local ice-lake stages A1 and A2 (Lo˜okene 1959; Pa¨rna 1960, 1962; Raukas et al. 1971). The proglacial lake at stage A1, marked by the highest shoreline in western Estonia, formed between the ice margin (Pandivere Neva) and the northwest slope of the Pandivere and Sakala Uplands and in the Lake Vo˜rtsja¨rv basin (Fig. 3). The proglacial lake at stage A2 probably formed c. 120150 years later, after the retreat of the ice to the north from the Pandivere Neva ice marginal formations (Fig. 3). On the basis of
the location of glaciofluvial deltas and eskers and several maps compiled earlier (Pa¨rna 1960; Ra¨hni 1966; Raukas et al. 1971; Karuka¨pp et al. 1996; Kajak 1999), the ice margin position during stage A2 could have been located between the PandivereNeva and Palivere ice margins (Fig. 3). The coastal landforms of the proglacial lake at stage A2 are located at lower altitudes than A1 because of the isostatic land uplift. Different hypotheses have been proposed concerning the relations between Glacial Lake Peipsi and the proglacial lake in western Estonia (stages A1 and A2) and their association with the Baltic Ice Lake (BIL). First, Pa¨rna (1960) showed that Glacial Lake Peipsi had a higher water level than proglacial lake stage A1, but later he proposed that Glacial Lake Peipsi was part of the proglacial lake stage A1 (Pa¨rna 1962). Most Estonian researchers have considered the shorelines of stages A1 and A2 in western Estonia to represent shores of local (isolated) ice lakes (Lo˜okene 1959; Pa¨rna 1960; Kessel & Raukas 1979) and BIL occurred only when the ice retreated to the Salpausselka¨ moraines, as defined in Finland (Fig. 1B; Glu¨ckert 1995). Kvasov & Krasnov (1967) and later Kvasov & Raukas (1970) suggested that the ice retreat at Ma¨nnikva¨lja kamefield at the end of the Pandivere Neva stage (Fig. 3) caused catastrophic drainage of Glacial Lake Peipsi into the proglacial lake at stage A2 in western Estonia and designated it as the beginning of the BIL. They compared the impact of this event with the final drainage at Billingen in Sweden. However, this proposal was challenged by the lack of drainage varves in the Lake Peipsi basin (Hang 2001). Furthermore, recent reexamination of available data on lateglacial coastal landforms of Estonia showed that the water levels of DOI 10.1080/03009480600991938 # 2007 Taylor & Francis
212
Alar Rosentau et al.
BOREAS 36 (2007)
Fig. 1. A. Overview map showing the study area in relation to the Baltic Ice Lake (BIL) and Glacial Lake Peipsi (GLP) in front of the Scandinavian Ice Sheet during the initial damming of the BIL (Bjo¨rck 1995). B. Main Lateglacial ice marginal positions around the Baltic with ages (cal. kyr BP) according to Lundqvist & Wohlfarth (2001), Saarnisto & Saarinen (2001) and Kalm (2006). The study area is marked by the square. The location of the height distance diagram (dotted line) for the GLP (Fig. 2) is shown.
proglacial lake stages A1 and A2 in western Estonia were similar to the water level of Glacial Lake Peipsi (Vassiljev et al. 2005). During the last few decades, knowledge of the early development of the BIL in western and southern parts of the Baltic has improved significantly (Svensson 1991; Bjo¨rck 1995; Us´cinowicz 1999, 2006; Moros et al. 2002; Lampe 2005). However, the extension of the BIL to the east and its relation with the neighbouring Glacial Lake Peipsi has remained unclear because of the lack of recent palaeogeographic research. A regional study by Bjo¨rck (1995) indicated that the eastern coast of the initial dammed BIL extended to western Estonia and did not reach Glacial Lake Peipsi
(Fig. 1A). Bjo¨rck’s (1995) study acknowledges that the reconstruction of the eastern part of the BIL is hypothetical as it is mainly based on extrapolation of shore displacement data from the western coast. Our main objective was to explore the history of the proglacial lake development using shoreline proxies, a digital terrain model (DTM) and Geographical Information System (GIS) analysis. The proglacial lake shorelines and bathymetry were reconstructed in order to understand how proglacial lakes in the Peipsi basin and western Estonia were distributed and related. Palaeoreconstructions helped to clarify the controversial hypothesis about the history of the beginning of the BIL in Estonia.
Palaeogeographic reconstruction of proglacial lakes in Estonia
BOREAS 36 (2007)
SE
Fig. 2. Generalized height distance diagram (modified from Raukas & Ra¨hni 1969) showing the water level changes in Glacial Lake Peipsi since deglaciation up to 13.3 cal. kyr BP. The location of the profile is shown in Fig. 1B.
water level (m a.s.l.)
100
213 NW 100
~ 14.7 cal. kyr BP
90
90
80
80 70
70 r BP cal. ky ~ 14.0
60
60 r BP cal. ky ~ 13.3
50
50 Recent lake level
40
40 30
30 0
20
40
60
80
100
120
140
160
180
distance (km)
Fig. 3. Map of the study area (indicated by the square) showing the spatial distribution of shorelines of stages A1 and A2 in Estonia, after Kessel & Raukas (1979). Black solid lines mark the ice marginal positions after Kalm (2006) and dashed lines mark two hypothetical positions of the ice margin during the proglacial lake stage A2. Critical threshold areas covered with detailed altitudinal data are marked on the map: I, in the Ma¨nnikva¨lja kame field; II, in the Navesti River valley; III, in the Emajo˜gi River valley. The profiles A A? and B B? are shown in Fig. 7.
214
Alar Rosentau et al.
Material and methods Reconstruction of proglacial lake shorelines and bathymetry was based on GIS analysis, by which interpolated water-level surfaces and average thicknesses of Holocene peat deposits were systematically removed from the modern DTM (Rosentau et al. 2004). Interpolated water level surfaces of the proglacial lake stages were retrieved from the Lateglacial shoreline database for Estonia, containing 149 sites, 83 of which represent the proglacial lake at stage A1 and 66 the proglacial lake at stage A2, as reported by different authors (Vassiljev et al. 2005). Spatial checks showed that some shoreline sites were either too high or too low compared with their neighbouring sites and they were removed from the simulations. Then point kriging interpolation with linear trend was used to reconstruct the water level surfaces of proglacial lakes. Kriging is advantageous because it interpolates accurate surfaces from irregularly spaced data and it is easy to identify outliers in the data set. The interpolated water level surfaces were smoothed using residuals (the difference between the actual site altitude and the interpolated surface) and sites with residuals more than 9/0.7 m were discarded. The final interpolated water level surfaces of proglacial lake stages A1 and A2 were based on 40 and 42 sites, respectively (for a complete description see Vassiljev et al. 2005). A modern DTM with a grid size of 200 /200 m (Fig. 3) was generated using the linear solution of the natural neighbour interpolation of the altitudinal data from the Digital Base Map of Estonia at a scale of 1:50 000 (Estonian Land Board 1996). The DTM was complemented with more detailed altitude data at critical lake threshold areas: in the Emajo˜gi valley (Fig. 3III) from Soviet military topographic maps (scale 1:10 000, 0.5-m contour interval), and in the Navesti River valley and Ma¨nnikva¨lja kamefield (Fig. 3I, II) from Soviet military topographic maps (scale 1:25 000, 2.5-m contour interval). The Holocene peat deposits were removed from the DTM, using constant mean thicknesses for three main types of mires: 4 m for raised bogs, 2 m for transitional mires and 1 m for fens (Orru 1995). Leverington et al. (2002) pointed out limitations in DTM-based palaeoreconstructions because of the impact of deposition subsequent to the time being modelled. The influence of this deposition on our modelling relates to Holocene peat deposits, which cover 22% of the territory in Estonia (Orru 1995). After removal of the Holocene peat and the interpolated water level surface from the DTM, the shoreline and bathymetry were derived for stages A1 and A2 and related to the ice marginal positions. The geometry of the water level surfaces of the proglacial lakes were analysed using spatial statistics
BOREAS 36 (2007)
such as surface profiling and terrain aspect. Terrain aspect was used to characterize the azimuth of the fastest uplift. All the spatial statistic calculations were performed within the same study area (showed by the square on Fig. 3) using the Conformal Transverse Mercator projection: TM-Baltic (Estonian Land Board 1996).
Results Proglacial lake stage A1 The proglacial lake at stage A1 developed in front of the Pandivere Neva ice margin (Fig. 4A), which formed the ice-proximal coast of the lake, marked by several ice-contact slopes, flat-topped glaciolacustrine and glaciofluvial landforms. Accumulative and abrasional coastal landforms developed in more ice-distal positions. Because of differential land uplift, the altitude of the shoreline was about 40 m a.s.l. in southern Estonia and up to 93 m a.s.l. in northern Estonia. The lake covered western Estonia and the Lake Vo˜rtsja¨rv and Lake Peipsi basins during this stage (Fig. 4A). The water depth was up to 50 m in western Estonia and up to 30 m in the Lake Peipsi depression. These lakes were joined in central Estonia through a strait in the Navesti and Emajo˜gi river valleys (Fig. 4A). A narrow strait was also present in the Raudna and Ta¨nnassilma river valleys (Fig. 4A). The reconstruction suggests that a narrow strait also existed in northern Estonia immediately in front of the PandivereNeva ice margin. However, this is dependent on the exact position of the ice margin. Proglacial lake stage A2 The proglacial lake at stage A2 developed after the ice margin retreated northwards from the Pandivere Neva ice marginal position (Figs 3, 4B). Because of isostatic land uplift, the shoreline was c. 1015 m lower in the northern part of the study area but only 12 m lower in the southern area compared with the proglacial lake stage A1 shoreline (Fig. 4A, B). The lower shoreline position narrowed the connection between the waterbodies in western Estonia and the Lake Peipsi basin in central Estonia. The proximal strait in northern Estonia opened and widened (Fig. 4B) and became the main connection between western Estonia and the Lake Peipsi basin. The lake outline is well defined by accumulative and abrasional coastal landforms (Fig. 4B). The reconstruction indicates that the lake occupied large areas of western Estonia and the Lake Vo˜rtsja¨rv and Peipsi basins (Fig. 4B). The water depth reached up to 45 m in western Estonia and up to 25 m in the Lake Peipsi depression (Fig. 4B).
BOREAS 36 (2007)
Palaeogeographic reconstruction of proglacial lakes in Estonia
215
Fig. 4. Palaeogeographic reconstructions of the shorelines and bathymetry of proglacial lake stages A1 (A) and A2 (B), with indications of the water level surface isobases. Proximal glaciolacustrine (squares), distal accumulative (circles) and abrasional (triangles) formations are shown.
Fig. 5. Generalized cross-section of the threshold area in the Emajo˜gi River valley at Tartu. For location see Fig. 3. Compiled after Ritsberg et al. (2005).
Threshold in Emajo˜gi River valley During stage A1 the Emajo˜gi River valley at Tartu was the critical threshold between the waterbodies in western Estonia and the Lake Peipsi basin (Figs 4A, 5). The threshold is cut into the Devonian sandstones and its bottom lies below 25 m a.s.l. (Ritsberg et al. 2005), covered by glaciolacustrine sediments (Fig. 5). The exact age of the formation of glaciolacustrine sediments is unknown, so it could have been deposited
before the proglacial lake A1 stage. Therefore we assumed that the threshold bottom during proglacial lake stage A1 was at least 28 m a.s.l. (Fig. 5, upper surface of glaciolacustrine sediments) and the strait in Tartu was c. 1 km wide and 13 m deep. The crosssectional area of the strait (below water level) was at least 0.0011 km2. Assuming low water flow velocities (0.2 m/s), the strait could have transferred an annual water volume of at least 67 km3 (Table 1). If the ice sheet at 13.3 cal. kyr BP had been 1000 m thick and
216
Alar Rosentau et al.
BOREAS 36 (2007)
Table 1. Calculated yearly water volume passing the threshold at Tartu by different water flow velocities. Flow velocity (m/s) 0.1 0.2 0.3 0.4 0.5 1
Water volume (km3/yr) 33 67 100 134 167 334
Table 3. Direction of the fastest uplift for A1, A2 and B3 water level surfaces. Proglacial lakes
Mean (degrees)
Median (degrees)
SD
Azimuth, after Pa¨rna (1962)
A1 A2 B3
340.7 336.3 323.8
337.6 337.6 323.1
18.8 19.8 7.0
c. 335 c. 335 c. 326
3200 Table 2. Predicted yearly water volume produced by melting glaciers by different ice sheet thicknesses. Meltwater volume (km3/yr)
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
3.41 6.83 10.24 13.65 17.06 20.48 23.89 27.30 30.71 34.13
2400 area (km2)
Ice sheet thickness (km)
1600
800
B3
had withdrawn 200 m/yr (Hang 2001), meltwater production would have totalled about 35 km3/yr (Table 2). According to Lambeck et al. (1998), the maximum ice thickness of the ice sheet in Estonia during the Last Glacial Maximum was about 800 1000 m. Thus the ice thickness was considerably less during the A1 stage. Our calculations do not consider the amount of water contributed by rivers and precipitation. However, the current annual river runoff and precipitation are about 9.4 km3 and 2.3 km3, respectively (Kullus 1973) and it is likely that these contributions were within the same magnitude in the past. Thus the strait in the Emajo˜gi valley was able to transfer a water volume well in excess of the amount of meltwater the glacier could have generated. Therefore, there is no reason to suggest that the water level in Glacial Lake Peipsi was higher than in its counterpart in western Estonia during stage A1 even if the proximal strait at Ma¨nnikva¨lja (Fig. 3) had been completely closed by ice (Fig. 4A). During the proglacial lake A2 stage, a strait north of the Pandivere Upland opened or widened (Fig. 4B) and the water level in Glacial Lake Peipsi could not have been higher than in western Estonia.
300
330
A2 360
A1 30
60
direction of the fastest uplift (degrees) Fig. 6. Direction of the fastest uplift for stages A1, A2 and B3 water level surfaces given as a frequency distribution diagram, where the y -axis represents the sum of the grid cell area (in km2) for an appropriate fastest uplift direction, given in the x-axis in degrees.
BIL (stage B3), dated to 11.6 cal. kyr BP (derived from data by Saarse et al. 2003). The results indicate that the direction of the fastest uplift during proglacial lake A1 and A2 stages was similar, but westwards during the B3 stage. Two lake-level tilting profiles, one for central (Fig. 7A) and another for western (Fig. 7B) Estonia, show a concave water surface for stage A1 and a slightly concaveconvex water surface for stage A2. A comparison of these surfaces with BIL stage B3 shows that water-surface tilting gradients for stage A1 (0.5 0.6 m/km) are higher than for stage B3 (0.3 m/km) in northern Estonia but lower in southern Estonia (Fig. 7). The water surface of stage A2 shows tilting gradients similar to stage B3 in the central part of the study area but lower gradients in southern Estonia and in front of the assumed ice margin (Fig. 7).
Characteristics of the reconstructed water level surfaces The direction of the fastest uplift during proglacial lake A1 and A2 stages, as derived from the interpolated water surfaces, is shown in Table 3 and Fig. 6. These were compared with the well-developed shoreline of the
Discussion The reliability of reconstructions of shorelines and bathymetry of the proglacial lake stages A1 and A2
Palaeogeographic reconstruction of proglacial lakes in Estonia
A
wat er lev el (m a.s.l.)
100
A`
A1
B`
B
A
217
100
B A1
A2
A2
80
80
60
B3
60
B3
40
40
20
20 0
25
50
75
100
125
distance (km)
150
0
25
50
75
100
w at er level (m a.s. l.)
BOREAS 36 (2007)
125
distance (km)
Fig. 7. Tilting profiles of water level surfaces for proglacial lakes at stages A1, A2 and B3. The location of the profiles is shown on Fig. 3.
depends on the accuracy of shoreline proxy data. Proximal shoreline data of stage A1 match well with the reconstructed PandivereNeva ice margin, dated by varve chronology at 13.3 cal. kyr BP (Hang 2001; Kalm 2006), representing the age of this shore. The reconstructed A1 shoreline position coincides well with the distribution of glaciolacustrine deposits in western Estonia (Fig. 8). This supports the view that the shoreline of the proglacial lake at stage A1 is the highest and oldest in this area. However, in the southwest part of the study area, the A1 stage shore lies slightly inside (at lower altitude) the outline of glaciolacustrine sediments (Fig. 8). Previous studies (Lo˜okene 1959; Pa¨rna 1960; Kessel & Raukas 1979) have shown that small local ice lakes existed during the Sakala stade between the ice margin and the western slope of the Sakala Upland (Fig. 8). Their coastal landforms and corresponding sediments are up to 10 m higher than the shore at stage A1. Our results show that the direction of the fastest uplift changed from more northerly (c. 3378 during stages A1 and A2) to more westerly (3238 during stage B3) after deglaciation (Table 3). The calculated directions of the fastest uplift agree with those previously reported: for proglacial lake stages A1 and A2, 3358, and for stage B3, 3268 (Pa¨rna 1962). This westerly trend continued in Estonia from the late Holocene, from the post-Littorina, at c. 3208 (Kessel & Raukas 1979; Saarse et al. 2003), to the present, currently c. 3108 (Torim 1998). This systematic trend in the changes of direction of the fastest uplift is probably related to the pattern of deglaciation of the Late Weichselian ice sheet. The similar direction of the fastest uplift (Fig. 6) of stage A1 and A2 shorelines and relatively small regression in southern Estonia (Fig. 7A, B) suggest a small age difference in these shorelines, inferred to be 120 150 years (Vassiljev et al. 2005). However, the water level drop of c. 15 m in northern Estonia (Fig. 7A, B)
during this time seems to be excessive as it requires c. 3 times higher land uplift values than reported by various studies (Raukas et al. 1971; Svensson 1991). The inferred ice front position during proglacial lake stage A2 (Fig. 3) indicates that it could have been near the Palivere ice marginal zone (12.8 12.7 cal. kyr BP; Kalm 2006), so that the age difference between stage A1 and A2 shorelines could be as great as c. 600500 years. A similar difference in regression was noted by comparing the shore displacement curves from eastern Sma˚land and Blekinge (southeast Sweden) between 14 and 12.8 cal. kyr BP (Fig. 9B; Svensson 1991; Bjo¨rck 1995). Thus the regression at Blekinge was only 2.5 m (from 44.5 to 42 m a.s.l.) at the time when a 13-m regression (from 89 to 74 m a.s.l.) took place at eastern Sma˚land (Fig. 9B; Svensson 1991). The low regression at Blekinge has been attributed to the rising water level of the BIL as a ¨ resund strait (Bjo¨rck 1995) result of the closing of the O or changes in uplift (Svensson 1991). Our shoreline reconstructions (Fig. 4A, B) and the present knowledge of the ice retreat pattern (Fig. 1B) indicate that proglacial lakes at stages A1 and A2 were open to the west, where the BIL was located. Moreover, the BIL shoreline during stage B3 is located at similar altitudes in eastern Sma˚land and northwestern Estonia (about 59 m a.s.l.) and in Blekinge and southwestern Estonia (about 30 m a.s.l.), as shown in Fig. 7B and 9B. Prior to the first drainage at 12.8 cal. kyr BP, the water levels at the same locations in eastern Sma˚land were 74 m a.s.l. and in Blekinge 42 m a.s.l., and correspondingly in northwestern Estonia 72 m a.s.l. and in southeastern Estonia 42 m a.s.l. during stage A2 (Fig. 7B, 9B). Thus the similar behaviour of the water levels in Blekinge and eastern Sma˚land suggests that the formation of proglacial lake at stage A2 took place prior to the first drainage of the BIL, probably about 12.8 cal. kyr BP, which is equivalent to the age of the Palivere stade.
218
Alar Rosentau et al.
BOREAS 36 (2007)
Pand iver
Glaciolacustrine deposits Holocene peat
e- Ne va
Others ediments and water
Ice marginal zones Shoreline of stage A1 0
Pandivere Upland
20km
LAKE PEIPSI
Sa kal a
LAKE VÕRTSJÄRV
Sakala Upland
Otepää Heights
Fig. 8. Comparison of the reconstructed stage A1 shoreline with mapped glaciolacustrine deposits in the study area (compiled after Kajak 1999).
Our results show that the tilting gradients of stages A1 and A2 decrease towards the southeast (Fig. 7), significantly further south of the current zero uplift isobase (Fig. 9A). Low tilting gradients of lateglacial shores in zero uplift and subsiding areas have been noted by Pa¨rna (1960, 1962) and Orviku (1960), and later by correlation of overdeepened river mouths of the Emajo˜gi, Ahja and Obdekh rivers (Figs 3, 9A; Miidel et al. 1995), lateglacial river terraces (Hang et al. 1964, 1995; Hang 2001) and raised shores on the eastern coast of Lake Vo˜rtsja¨rv (Moora et al. 2002; Miidel et al. 2004). However, explanations for this anomaly in southeast Estonia remain elusive. Recent vertical movements indicate that the southern margin of the Baltic Shield is enclosed by a belt of subsidence,
which could be associated with a collapsed structure of a circum-Fennoscandian ring bulge of the upper mantle after melting of the Weichselian ice sheet (Fig. 9A; Bylinski 1990; Fjeldskaar 1994; Harff et al. 2001). Measurements of recent vertical movement in Estonia place the subsidence belt in southeast Estonia, with the lowest values in the southern part of the Peipsi basin, whereas uplift occurs north of it (Fig. 9A). This pattern of crustal tilting coincides well with the Late Weichselian shoreline tilting pattern (Fig. 4A, B) and could reflect the ongoing forebulge collapse in southeast Estonia. The development of rivers and lakes supports the forebulge hypothesis in southeast Estonia and the neighbouring area in Russia (Grachev & Dolukhanov 1970; Bylinski 1990). The forebulge in
Palaeogeographic reconstruction of proglacial lakes in Estonia
BOREAS 36 (2007)
219
110
A
B
100
9
90 80 70
m.a.s.l.
8
7
60 50 40 30 20
6
10
5
0 uncal. years BP
Narva
4
Fig. 7B
3
2
B`
12000
10000
3 mm/y Forebulge zones: 18.0-12.5 uncal. kyr BP 12.5-10.3 uncal. kyr BP
Fig. 9. A. The study area in relation to Fennoscandian land uplift (Ekman 1996) and peripheral subsiding region associated with collapsing forebulge (Harff et al. 2001). This subsiding area concurs approximately with two possible forebulge zones, identified according to the increased incision and high sedimentation rates of rivers by Bylinski (1990). Note that the higher incision of river valleys discharging into Lake Peipsi (dots with incision depth; after Miidel et al. 1995 and Hang & Miidel 1999) is in areas of higher rates of recent subsidence. B. BIL shore displacement curves for eastern Sma˚land (I; Svensson 1991) and Blekinge (II; Bjo¨rck 1995), southeast Sweden. Major events presented in the curve and discussed in the text are given in calibrated years according to Bjo¨rck (1995).
northwest Russia and eastern Europe was studied in detail by Bylinski (1990). He identified two major forebulge zones, at 18.0 12.5 and 12.5 10.3 uncal. kyr BP (Fig. 9A), and argued that the forebulge caused the increased incision and high sedimentation rates of rivers in these areas. It is probable that the great depth of the lateglacial valleys (up to 70 m) in southeast Estonia and the northern Pskov region (Russia),
discharging water into the Peipsi basin, is related to Bylinski’s (1990) forebulge (Fig. 9A). Thus the uplift caused by the forebulge during the deglaciation could account for the deep incision of rivers. This uplift might also have produced a regression and formation of several isolated waterbodies in the southern part of the Lake Peipsi basin and adjoining areas (Hang & Miidel 1999). At the end of the Late Weichselian, the
220
Alar Rosentau et al.
rapid uplift included the northern part of the lake basin and relative subsidence became prevalent in the south (Hang & Miidel 1999). In addition to the effect of the forebulge, the tectonically active fault zone between Pa¨rnu and Narva (Fig. 9A) could have affected the uplift of shorelines (Orviku 1960; Pa¨rna 1960, 1962; Vallner et al. 1988). It should be noted that seismic events had also occurred in the Pa¨rnuNarva fault zone (Sildvee & Vaher 1995). The reactivation of old fault zones, accompanied by strong earthquakes in the Late Weichselian and Holocene, has been detected in many parts of Fennoscandia (Lundqvist & Lagerba¨ck 1976; Mo¨rner 1978; Ojala et al. 2004). Our reconstructions do not support Kvasov & Krasnov’s (1967) proposal that the BIL started with a catastrophic drainage of Glacial Lake Peipsi into the proglacial lake at stage A2. Reconstructions show that the water levels in Glacial Lake Peipsi and western Estonia were similar at that time. Water-balance calculations also indicate that the critical threshold near Tartu was unable to dam the Glacial Lake Peipsi since the proglacial lake A1 stage, so there is no physical reason to support catastrophic drainage.
Conclusions . Shoreline reconstruction shows that Glacial Lake Peipsi had a strait-like connection with a proglacial lake in western Estonia during stages A1 and A2 via the Emajo˜gi Navesti river valleys. . The strait north of the Pandivere Upland could have existed since the proglacial lake stage A1 (13.3 cal. kyr BP), depending on the exact ice margin position. . The joining of lakes east and west of the Pandivere Upland did not cause a catastrophic drainage of Glacial Lake Peipsi, as proposed earlier, because the water levels east and west were the same. . Reconstructions do not support the opinion that the BIL started by the joining of Estonian proglacial lakes north of the Pandivere Upland. Proglacial lakes at stages A1 and A2 were both already open to the BIL, as indicated by their inferred ages and iceretreat pattern. These were formed during the initial damming of the BIL, as revealed by water-level similarities with eastern Sma˚land and Blekinge, southeast Sweden. . The direction of the fastest uplift during the Late Weichselian changed from northerly to more westerly and was probably related to the pattern of the deglaciation. . Effects of a forebulge or/and the influence of the Pa¨rnuNarva tectonic fault zone could account for the low tilting and relatively higher than expected altitude of Late Weichselian shorelines in southeastern Estonia.
BOREAS 36 (2007)
Acknowledgements. We thank Prof. J. Mangerud and Prof. E. Larsen for critical remarks and suggestions to improve the manuscript. Many thanks to Dr. Robert Szava-Kovats and Helle Kukk checking the language. This study was funded by Estonian target funding projects 0182530s03 and 0332710s06 and Estonian Science Foundation Grants no. 5370, 5342, 5923 and 6736.
References Bjo¨rck, S. 1995: A review of the history of the Baltic Sea, 13.0 8.0 ka BP. Quaternary International 27, 19 40 (Calibrated ages from http://www.geol.lu.se/personal/seb/). Accessed October 2006. Bylinski, E. N. 1990: Swell-like lithospheric uplifts due to glacioisostasy and their probable influence on oil and gas deposits location in the north of Europe. Geomorphology 4, 3 13 (in Russian with English summary). Ekman, M. 1996: A consistent map of the postglacial uplift of Fennoscandia. Terra Nova 8, 158 165. Estonian Land Board. 1996: Digital Base Map of Estonia in scale 1:50 000. Estonian Land Board, Tallinn. Fjeldskaar, W. 1994: The amplitude and decay of glacial forebulge in Fennoscandia. Norsk Geologisk Tidsskrift 74, 2 8. Glu¨ckert, G. 1995: The Baltic Ice Lake in south Finland and its outlets. Quaternary International 27, 57 51. Grachev, A. F. & Dolukhanov, P. M. 1970: Postglacial crustal movements of Canada and Fennoscandia. Baltica 4, 297 312 (in Russian with English summary). Gudelis, V. 1976: History of the Baltic Sea. In Gudelis, V. & Emelyanov, E. (eds.): Geology of the Baltic Sea, 95 116. Mokslas, Vilnius (in Russian with English summary). Guobyte, R. 2004: A brief outline of the Quaternary of Lithuania and the history of its investigation. In Ehlers, J. & Gibbard, P. L. (eds.): Quaternary Glaciations: Extent and Chronology. Part I. Europe, 245 250. Elsevier, Amsterdam. Hang, T. 2001: Proglacial sedimentary environment, varve chronology and Late Weichselian development of Lake Peipsi, eastern Estonia. Quaternaria A11, 1 44. Hang, T. & Miidel, A. 1999: Main features of the morphology and formation of the river valleys. In Miidel, A. & Raukas, A. (eds.): Lake Peipsi. Geology, 110 119. Sulemees Publishers, Tallinn. Hang, E., Liblik, T. & Linkrus, E. 1964: On the relations between Estonian valley terraces and lake and sea levels in the Late-Glacial and Holocene periods. Transactions of the Tartu State University 156. Publications on Geography IV, 29 42. Hang, T., Miidel, A. & Pirrus, R. 1995: Late Weichselian and Holocene water-level changes of Lake Peipsi, eastern Estonia. PACT 50, 121 131. Harff, J., Frischbutter, A., Lampee, R. & Meyer, M. 2001: Sea-level changes in the Baltic Sea: interrelation of climate and geological processes. In Gerhard L. C., Harrison, W. E. & Hanson, B. M. (eds.): Geological Perspectives of Global Climate Change, AAPG Studies in Geology 47, 231 250. AAPG, Tulsa. Kajak, K. 1999: Quaternary Deposits of Estonia, 1:400 000. 1 sheet Geological Survey of Estonia, Tallinn. Kalm, V. 2006: Pleistocene chronostratigraphy in Estonia, southeastern sector of the Scandinavian glaciation. Quaternary Science Reviews 8, 960 975. Karuka¨pp, R., Moora, T. & Pirrus, R. 1996: Geological events determining the Stone Age environment of Kunda. PACT 51, 219 229. Kessel, H. & Raukas, A. 1979: The Quaternary history of the Baltic, Estonia. In Gudelis, V. & Ko¨nigsson, L.-K. (eds.): The Quaternary History of the Baltic, 127 146. Acta Universitatis Upsaliensis, Uppsala. Kullus, L. 1973: Water balance of Lake Peipsi-Pihkva. IAHS-AISH Publication 109, 159 163.
BOREAS 36 (2007)
Palaeogeographic reconstruction of proglacial lakes in Estonia
Kvasov, D. D. 1979: The late-quaternary history of large lakes and inland seas of eastern Europe. Annales Academie Scientiarum Fennicae, Ser. A III, Geologica-Geographica 127, 1 71. Kvasov, D. D. & Krasnov, I. I. 1967: Main problems in the history of proglacial lakes in northeast. In Kvasov D. D. (ed.): Northwest Lakes History, 7 17. Geographical Society of the Soviet Union, Leningrad (in Russian). Kvasov, D. D. & Raukas, A. 1970: Postglacial history of the Gulf of Finland. Proceedings of the Geographical Society of the Soviet Union 102, 432 438 (in Russian). Lambeck, K., Smither, C. & Ekman, M. 1998: Tests of glacial rebound models for Fennoscandinavia based on instrumented sea- and lake-level records. Geophysical Journal International 135, 375 387. Lampe, R. 2005: Lateglacial and Holocene water-level variations along the German Baltic Sea coast: review and new results. Quaternary International 133 134, 121 136. Leverington, D. W., Teller, J. T. & Mann, J. D. 2002: A GIS method for reconstruction of late Quaternary landscapes from isobase data and modern topography. Computers & Geosciences 28, 631 639. Lundqvist, J. & Lagerba¨ck, R. 1976: The Pa¨rve Fault: a late-glacial fault in Precambrian of Swedish Lapland. Geologiska Fo ¨ reningens i Stockholm Fo ¨ rhandlingar 98, 45 51. Lundqvist, J. & Wohlfarth, B. 2001: Timing and east west correlation of south Swedish ice marginal lines during the Late Weichselian. Quaternary Science Reviews 20, 1127 1148. Lo˜okene, E. 1959: Geomorphologie des no¨rdlichen Teils des Ho¨hengebiets von Sakala. Transactions of the Tartu State University 75. Publications on Geology I, 119 154 (in Russian with German summary). Mangerud, J., Jakobsson, M., Alexanderson, H., Astakhov, V., Clarke, G. K. C., Henriksen, M., Hjort, C., Krinner, G., Lunkka, J. P., Mo¨ller, P., Murray, A., Nikolskaya, O., Saarnisto, M. & Svendsen, J. I. 2004: Ice-dammed lakes and rerouting of the drainage of northern Eurasia during the Last Glaciation. Quaternary Science Reviews 23, 1313 1332. Miidel, A., Hang, T., Pirrus, R. & Liiva, A. 1995: On the development of the southern part of Lake Peipsi in the Holocene. Proceedings of the Estonian Academy of Sciences, Geology 44, 33 44. Miidel, A., Raukas, A. & Vaher, R. 2004: Geology of the lake basin. In Haberman, J., Pihu, E. & Raukas, A. (eds.): Lake Vo˜rtsja ¨ rv, 33 46. Estonian Encyclopaedia Publishers, Tallinn. Moora, T., Raukas, A. & Tavast, E. 2002: Geological history of Lake Vo˜rtsja¨rv. Proceedings of the Estonian Academy of Sciences, Geology 51, 157 179. Mo¨rner, N.-A. 1978: Faulting, fracturing, and seismicity as functions of glacio-isostasy in Fennoscandia. Geology 6, 41 45. Moros, M., Lemke, W., Kuijpers, A., Endler, R., Jensen, J.B., Bennike, O. & Gingele, F. 2002: Regressions and transgressions of the Baltic basin reflected by a new high-resolution deglacial and postglacial lithostratigraphy for Arkona Basin sediments (western Baltic Sea). Boreas 31, 151 162. Noormets, R. & Floden, T. 2002: Glacial deposits and ice-sheet dynamics in the north-central Baltic Sea during the last glaciation. Boreas 31, 362 377. Ojala, V. J., Kuivama¨ki, A. & Vuorela, P. 2004: Postglacial Deformation of Bedrock in Finland. Nuclear Waste Disposal Research. Geological Survey of Finland, Report YST-120, 23 pp. Orru, M. 1995: Handbook. Estonian Mires. 240 pp. Geological Survey of Estonia, Tallinn (in Estonian with English summary). Orviku, K. 1960: Geological data on neotectonic movements of the ˇ elnin G. A., Earth’s crust in the Estonian SSR. In Orviku K. K., Z Gudelis V. K., Springis K. J. & Villmann C. I. (eds.): Materials on
221
the Meeting of the Neotectonic Movements in Peribaltic, 120 143. Academy of Sciences of Estonian SSR, Tartu (in Russian with English summary). Pa¨rna, K. 1960: Zur Geologie des Baltischen Eisstausees sowie der lokalen grossen Eisstauseen auf dem Territorium der Estnischen SSR. Eesti NSV Teaduste Akadeemia Geoloogia Instituudi uurimused V, 269 278 (in Russian with German summary). Pa¨rna, K. 1962: On the geology of the Baltic Ice Lake and large local proglacial lakes on the territory of Estonia. Ph.D. dissertation, Academy of Sciences of ESSR, 163 pp. Persson, C. 1983: Glacial deposits and the Central Swedish end moraine zone in eastern Sweden. In Ehlers, J. (ed.): Glacial Deposits in North-West Europe, 131 139. Balkema, Rotterdam. Rattas, M. & Kalm, V. 2004: Glaciotectonic deformation patterns in Estonia. Geological Quarterly 48, 15 22. Ra¨hni, E. 1966: Eskers and ice marginal formation of the last glaciation on the Pandivere Upland. Ph.D. dissertation, Academy of Sciences of ESSR, 265 pp. (in Estonian). Raukas, A. & Ra¨hni, E. 1969: On the geological development of the Peipsi-Pihkva depression and the basins distributed in that region. Proceedings of the Estonian Academy of Sciences, Chemistry, Geology 18, 113 125 (in Russian with English summary). Raukas, A., Ra¨hni, E. & Miidel, A. 1971: Marginal Glacial Formations in North Estonia. 226 pp. Valgus, Tallinn (in Russian with English summary). Ritsberg, K., Hang, T. & Miidel, A. 2005: On the geology and geomorphology of the Emajo˜gi River valley. In Raukas, A. (ed.): Eesti Geograafia Seltsi Aastaraamat 35, 130 145. Entsu¨klopeediakirjastus, Tallinn. Rosentau, A., Hang, T. & Miidel, A. 2004: Simulation of the shorelines of glacial Lake Peipsi in eastern Estonia during the Late Weichselian. Geological Quarterly 48, 299 307. Saarnisto, M. & Saarinen, T. 2001: Deglaciation chronology of the Scandinavian ice sheet from east of Lake Onega basin to the Salpausselka¨ end moraines. Global and Planetary Change 31, 387 405. Saarse, L., Vassiljev, J. & Miidel, A. 2003: Simulation of the Baltic Sea shorelines in Estonia and neighbouring areas. Journal of Coastal Research 19, 261 268. Sandgren, P., Hang, T. & Snowball, I. F. 1997: A Late Weichselian geomagnetic record from Lake Tamula, SE Estonia. GFF 119, 279 284. Sildvee, H. & Vaher, R. 1995: Geologic structure and seismicity of Estonia. Proceedings of the Estonian Academy of Sciences, Geology 44, 15 25. So¨derberg, P. 1988: Notes on continuation of the Salpausselka¨ ice marginal zone in the northern Baltic Proper. Geological Survey of Finland Special Paper 6, 69 72. Svensson, N. O. 1991: Late Weichselian and Early Holocene shore displacement in the central Baltic Sea. Quaternary International 9, 7 26. Torim, A. 1998: Renovation of the Estonian Levelling Network, 55 pp. Manuscript in Estonian Land Board Development Centre. Tartu. Uscinowicz, S. 1999: Southern Baltic area during the last deglaciation. Geological Quarterly 43, 137 148. Uscinowicz, S. 2006: A relative sea-level curve for the Polish Southern Baltic Sea. Quaternary International 145 146, 85 105. Vallner, L., Sildvee, H. & Torim, A. 1988: Recent crustal movements in Estonia. Journal of Geodynamics 9, 215 233. Vassiljev, J., Saarse, L. & Miidel, A. 2005: Simulation of the proglacial lake shore displacement in Estonia. Geological Quarterly 49, 253 262.
