This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
Author's personal copy
Quaternary International 207 (2009) 58–68
Contents lists available at ScienceDirect
Quaternary International journal homepage: www.elsevier.com/locate/quaint
Reconstruction of postglacial palaeoenvironmental changes in eastern Lithuania: Evidence from lacustrine sediment data V. Sˇeiriene_ a, *, M. Kabailiene_ b, J. Kasperovicˇiene_ c, J. Ma zeika a, R. Petrosˇius a, R. Pasˇkauskas c Institute of Geology and Geography, T. Sˇevcˇenkos 13, LT 03223 Vilnius, Lithuania Department of Geology and Mineralogy, M.K. Ciurlionio 21/27, LT 2600 Vilnius, Lithuania c Institute of Botany, Zˇaliuj ˛ u˛ e zeru˛ 49, LT 08406 Vilnius, Lithuania a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Available online 31 December 2008
Postglacial climatic conditions were inferred from the cores taken from two lakes in eastern Lithuania. Pollen, diatom, pigment, chemical analysis and 14C dating were performed. Vegetation with open herb/ grass predominating flourished on the unstable soils, surrounding oligotrophic palaeobasins with high water levels and low pigment concentrations that existed at about 13,200 cal. The succeeding stage (13,200–12,600 cal BP) indicates a formation of pine–birch woodland, rising humidity as well as biogenic productivity, and points to climatic warming. At about 12,600 cal BP the re-establishment of open herb– grass–shrub communities was registered. Gradual improvement of climatic conditions started after 11,200 cal BP. The period from w8700 to w6000 cal BP shows an expansion of broad-leaved trees, significant increase in diatom diversity, diatom and pigment concentration, and drops in C/N ratio indicating the most favorable climatic conditions during the postglacial. After w4800 cal BP the broadleaved trees markedly decreased and the significance of eutrophication and human activity increased. Ó 2009 Elsevier Ltd and INQUA. All rights reserved.
1. Introduction Lakes are excellent sensors of environmental changes and lake sediments can provide a means of environmental reconstruction at the local and regional levels across continents. However, climate change influences lakes in many different ways and the direct and indirect linkages between climate and lake water column need to be understood. For this reason a multi-proxy, whole-lake approach is recommended as the best way forward, not only as a means of reconstructing past environment but also as a mean of assessing the impact of climate change on lake systems (Battarbee, 2000). In Lithuania broad-scale patterns of climatic variation during the lateglacial and Holocene at the regional to subcontinental level have been identified primarily from palynological studies _ 1993, 1998, 2006a; Bla (Kabailiene, zauskas et al., 1998; Stancˇikaite_ nas et al., 2005). However, most of the studies et al., 1998; Sˇinku were concentrated on the general development of vegetation and were infrequently 14C dated. Only during the last decade has more attention been paid to reliability of the age determinations of palaeoevents, based on 14C dating (Stancˇikaite_ et al., 2002, 2003, 2004, 2008; Sˇeiriene_ et al., 2006).
* Corresponding author. Institute of Geology and Geography, T. Sˇevcˇenkos Str. 13, LT 03223 Vilnius, Lithuania. Tel.: þ370 52104707; fax: þ370 52104695. _ E-mail address:
[email protected] (V. Sˇeiriene). 1040-6182/$ – see front matter Ó 2009 Elsevier Ltd and INQUA. All rights reserved. doi:10.1016/j.quaint.2008.12.005
Pollen, diatom, pigment, chemical analysis and 14C dating were applied in recent investigations. Pollen analysis is the main proxy for reconstruction of past vegetation and climate changes. Diatoms are sensitive ecological indicators and have been widely used to quantify past changes in pH, salinity, nutrients, water level changes (e.g. Fritz, 1989; Battarbee and Renberg, 1990; Bennion, 1994; Battarbee, 2000). Since many of the factors that control trophic state and lacustrine primary productivity are climatically related, plant pigments are useful in obtaining information on palaeoenvironment and palaeoclimate (Sanger,1988; Leavitt et al., 1997; Guilizzoni et al., 2002). 14C dating provides the sedimentation time control and enables age determination of the main palaeoevents. This multidisciplinary study was undertaken to better understand the longterm environmental changes and to supplement the existing postglacial environmental change models with new important data. 2. Regional setting Two lakes of different size, depth, and anthropogenic influence situated in the eastern part of Lithuania (Fig. 1) were chosen for _ investigations. Varenis Lake is located in southeastern Lithuania (54170 N; 24 330 E) on the glaciofluvial plain of the Last (Weichselian) Glaciation at 120 m a.s.l. The lake covers an area of 23.4 ha with the maximal water depth reaching 8.65 m in the centre of the lake, and a mean depth of 3 m. The sedimentary basin has two
Author's personal copy
V. Sˇeiriene_ et al. / Quaternary International 207 (2009) 58–68
59
Fig. 1. Location of investigated lakes and sections, bathymetry of lakes.
_ e_ River, meandering through an outflows and a tributary, the Varen agricultural landscape. Telmatic vegetation forms a wide belt along _ the lake shore. Varenis Lake is of glaciokarst origin and is surrounded by pine forest growing on sandy soils. Baltys Lake is situated in northeastern Lithuania (55 350 N; 25 310 E) on the area of the marginal formations of the Last Glaciation at 100 m a.s.l. The lake is of glaciokarst origin with an area of 9.2 ha. Maximal water depth reaches 7 m in the western part of the lake, and the mean depth is 3–4 m. The lake is surrounded by coniferous forest, has no inflow or outflow streams, and anthropogenic influence is insignificant.
particles. Additionally, the acetolysis method defined by Erdtman (1936) was used. More than 500 terrestrial pollen grains were counted for each sample. Pollen and spore identification was based on Faegri and Iversen (1989), Moore et al. (1991), and Reille (1992, 1998). The pollen percentages of arboreal (AP) and non-arboreal (NAP) pollen were calculated from the basic sum of terrestrial P P P pollen grains ( AP þ NAP ¼ P). The sum of Quercetum mixtum (QM) includes Ulmus, Quercus, Tilia, Carpinus, Fraxinus and Fagus. Pollen concentrations (grain cm3) were determined by adding Lycopodium tablets (Stockmarr, 1971) to a known volume of the sediment.
3. Materials and methods 3.3. Diatom analysis 3.1. Coring and sampling To select sampling places, prospect holes to establish the lithology and the depths of the sediments were drilled in both _ lakes. Varenis Lake was cored from a rubber boat in June 2000 using a Russian corer (chamber length 1 m, inside diameter 5 cm). Samples were taken from the eastern part of the lake (Fig. 1), where water depth reaches 3.50 m. Baltys Lake was cored from ice in March 2001 with the same Russian corer and the coring site was situated in the eastern part of the lake. Water depth was about 1.70 m. Sediment cores from the lakes later were subsampled every 2–10 cm for pollen, diatom, pigment and chemical investigations. Bulk samples of 20–25 cm were taken for 14C dating. 3.2. Pollen analysis The samples were prepared according to standard procedure described by Grichiuk (1940) which includes treating the sediments with a heavy liquid (CdI2 þ KI) to remove minerogenic
Diatom frustules were extracted from the sediments in the conventional manner described by Battarbee (1986) and Miller and Florin (1989). HCl was added to remove the carbonates and 30% H2O2 to oxidise organic material. Decanting and flotation in heavy liquids (CdI2 þ KI) were applied to remove clay particles and mineral material. Afterwards, the residue was mounted in NBS Naphrax (R.I. ¼ 1.74) and examined under a light microscope with an oil immersion objective at a magnification of 1000. The identification of species mainly followed Krammer and Lange-Bertalot (1988, 1991a,b, 1997), counting at least 500 diatom valves along the central part of horizontal transects per slide. Diatoms were subdivided into groups according to ecological requirements (Hustedt, 1937–1939; Van Dam et al., 1994). The succession of the most frequent and ecologically important taxa is presented in the percentage diagrams based on the total sum of the identified items. All spreadsheets as well as the percentage diagrams for pollen and diatom were plotted using the programs TILIA and TILIA-GRAPH (Grimm, 1992).
Author's personal copy
V. Sˇeiriene_ et al. / Quaternary International 207 (2009) 58–68
60
3.4. Chemical and pigment analysis
Table 1 Lithological description of the studied sections.
Sediment samples collected for chemical and pigment analyses were frozen (20 C) before analyzed. Estimation of total organic carbon (TOC) and nutrients (phosphorus and nitrogen) were performed according to Potapova et al. (1980) and Merkiene_ and Ceponyt e_ (1994). Total phosphorous (TP) was analyzed by acid digestion followed by the molybdate ascorbic acid method, and a persulfate–H2SO4 digestion. Afterwards total nitrogen (TN) was analyzed by the Kjeldahl method. Pigment analyses were conducted using 90% acetone as an extracting agent over a 24-h period at 4 C, filtered on membrane filter Vladipor No. 9, and quantified spectrophotometrically according to Jeffrey and Humphrey (1975). Pigments quantity was calculated according to the methodology proposed by SCOR UNESCO (1966). The concentrations of pigments are expressed as mg g1 of dry sediment.
Depth, cm
Description
_ Varenis Lake 0–350 170–220 220–320 320–390 390–720 720–975 975–1085
Water column Dark grey, silty gyttja Black, calcareous silty gyttja with mollusc shells Grey clayey lake marl Dark grey, calcareous gyttja with plant remains No data Dark grey calcareous gyttja
Baltys Lake 0–170 350–500 500–600 600–750 750–780 780–1000 1000–1250
Water column Light brown gyttja with dark brown gyttja interlayers Clayey lake marl Brown gyttja with mollusc shells Peat Brown, compact gyttja with plant remains Peat
3.5.
14
C dating
14
C dating was performed on 9 bulk (organic or carbonate) samples that were more or less evenly spaced along the sediment sequences. Samples were analyzed in the Radioisotope Research Laboratory of the Institute of Geology and Geography, Vilnius, Lithuania. After drying the samples were treated by acid–alkali–acid washing. The specific activity of 14C was measured by liquid scintillation counting (LSC) method as described in Gupta and Polach (1985), Arslanov (1985), Kovaliukh and Skripkin (1994), and Bowman (1995). The radiocarbon calibration program OxCal v3.10 (Bronk Ramsey, 2001) with the IntCal04 data set (Reimer et al., 2004) was used for the calibration of radiocarbon dates and estimation of sediment accumulation rate. Time scales were constructed on the basis of the linear interpolation between available dates and all ages are given as calibrated years before 1950 AD (cal BP).
However, the rise of TOC towards the top of the core (Fig. 6) demonstrates the increasing eutrophication as well. Fluctuation in the total phosphorous is relatively small within the investigated sediment section (0.2–0.26%) and decreases more than two times (till 0.1%) in the uppermost sediments. Total nitrogen (Nsum) exceeds 1%, varying along the sequence. Decreasing amounts of pigments are recorded approaching the uppermost part of the core. Baltys Lake sediments are rich in TOC, and this amount fluctuates within inconsiderable pales (19.7–39.4%) (Fig. 7). The trend of total phosphorus (Psum) increases (0.027–0.055%) upwards, indicating an increasing eutrophication level in the lake. Total nitrogen (Nsum) exceeds 3% and has an uneven distribution along the core. Table 2 _ Description of local pollen assemblage zones distinguished at Varenis Lake. LPAZ Depth, cm
Description
VP9 170–250
Herb pollen reaches 12–18% and Poaceae together with Artemisia predominate among them. Slight increase in Alnus pollen (up to 18%) is noticed. Picea shows 10–18%, Pinus – 50–60%. Single pollen of cultural grains and aquatic plants are present Picea compounds about 39.5%, Pinus – 70–80%. Broad-leaved trees decrease, only Tilia somewhere reaches 1%. Corylus makes 1–2%, Alnus – 2–3%, Betula – 1–6%. Single Calluna pollen and Pteridium spores are registered Broad-leaved trees culminate and Tilia (1–1.5%) dominates among them. Ulmus shows about 1%, Quercus and Fraxinus – >1%, Corylus about 1–2%. Alnus varies from 1 to 16%, Betula – 1–15%, Pinus – 50– 75%. Some increase of herb pollen (mainly Poaceae and Artemisia) is noticed. Poaceae has an even continuous presentation, while the highest number of Artemisia is in the lowermost part of the zone (about 9%) and in few samples in the upper part of the zone (up to 6%). Slight increase of Polypodiaceae spores is registered Alnus increase up to 18%, Betula – up to 15% and Corylus reaches 1– 3%. Single grains of Ulmus, Tilia and Quercus are present. Pinus ranges in interval of 60–85%. Total sum of herbs is about 0.5–3% Pinus culminates in this zone and reaches 98%. Pollen grains of other trees are very sparse. Total sum of herbs makes only 0.5–2.5% Betula increase up to 30%, Pinus shows about 60% and Picea decrease to 10%. Alnus comprise about 1% and Juniperus (>1%) is noticed Herb pollen rise to 5% and Poaceae (about 2%) predominate among them, Artemisia shows about 1.5% and Cyperaceae shows 1%. Pinus (70–80%) is best represented among trees. Picea has a peak of about 25% Decreasing number of NAP and rising of AP pollen. Pinus prevail comprising about 90%, Picea shows 2–4%, Betula and Juniperus – about 1% NAP pollen compounds about 10–12%. Poaceae and Artemisia are the best represented among them. Chenopodiaceae shows 1%. Pinus reaches up to 78–81%, Picea – 5–6%, Betula and Juniperus about 1%. Single spores of Bryales, Polypodiaceae and Botrychium are noticed
4. Results 4.1. Lithology The sediment sequences were subdivided into lithological units according to the visual description. Later, the lithological composition of sediments was specified at the laboratory (Table 1) as described in Bengtsson and Enell (1986). The relative amount of organic matter was determined as loss-on-ignition (LOI).
VP8 250–320
VP7 320–465
4.2. Palynological results The pollen data was visually subdivided into pollen assemblage _ zones (LPAZ). Nine LPAZ were recognized in the Varenis Lake pollen diagram (Table 2, Fig. 2), and 13 for Baltys Lake (Table 3, Fig. 3).
VP6 465–545
4.3. Diatom survey
VP4 605–620
Based on diatom investigations, local diatom zones (LDAZ) were _ determined: seven in Varenis Lake (Table 4) and seven in Baltys Lake (Table 5) sediment sections. The diatom taxa were grouped according to their ecological conditions (Figs. 4 and 5).
VP5 545–605
VP3 620–680
VP2 680–705
4.4. Chemical and pigment investigations VP1 705–720
_ Sediments of Varenis Lake are relatively rich in total organic carbon (TOC). The amount of TOC fluctuates between 13.2 and 17.1% (Fig. 6). This could be related with short water turnover time and prevailing terrigenic sedimentation due to inflow streams.
Author's personal copy
V. Sˇeiriene_ et al. / Quaternary International 207 (2009) 58–68
61
_ Fig. 2. Percentage pollen diagram for Varenis Lake.
Nevertheless, rising values are seen in the uppermost part of the sediments. Pigment investigations in Baltys Lake sediments show _ higher representation in comparison with Lake Varenis and a striking tendency of increase in concentrations of chlorophyll a, carotenoids, and chemically stable pigments ubiquitous among algae towards the top of the core. These results confirm the increasing productivity of the lake. Whereas variability of the total phosphorous (Psum) is relatively small in the investigated lakes, the eutrophication trend is visible according to TOC, nitrogen and pigments. The C/N ratio has been used for the evaluation of the relative influence of terrestrial and lake organic matter in several ecosystems (Tyson, 1995). Saito et al. (1989) suggested a ratio exceeding 20 could be related with the terrestrial origin of the material and that varying between 5 and 7 – as a pelagic one. In addition, Stein (1991) reported that a C/N ratio lower than 10 indicates a limnic origin of the sediments and values around 10 represent both limnic and terrestrial organic components in the sediment. Taking into account these facts, the C/ N ratio observed in the sediments of the investigated lakes seems to reflect a mixed origin of the organic matter in both lakes.
calibration curve. The sedimentation rates based on calibrated 14C ages with 68.2% confidence limits for Baltys Lake are: between depths 538 and 713 cm – 1.0–1.2 mm/year; between depths 713 and 813 cm – 0.6–0.7 mm/year; and between depths 813 and 1110 cm – 0.9–1.0 mm/year. 5. Environmental phases Based on the age–depth curves (Figs. 8 and 9), tentative ages (cal BP) were assigned to the lithostratigraphic units and to pollen and Table 3 Description of local pollen assemblage zones distinguished at Baltys Lake. LPAZ Depth, cm Description BP13 350–430
BP12 430–460
BP11 460–570 BP10 570–630
4.5. Chronology Radiocarbon dates from the investigated sequences are presented in Table 6. An age–depth model (Figs. 8 and 9) was compiled in order to decrease the impact of large standard errors which appear after the calibration of the dates due to the presence of radiocarbon plateaus (Ammann and Lotter, 1989; Stuiver et al., 1998). _ The sediment sequence of Varenis Lake (170–720 cm) comprises a calibrated 14C age interval of 3870–13,280 cal BP. For some dated intervals (170–190 cm and 500–520 cm), the calibrated age has larger error terms compared to uncalibrated age errors due to plateaus and wiggles on the calibration curve. Two dates of sediment (intervals 370–390 cm and 500–520 cm) have no significant plateaus and wiggles on calibration curve. The sedimentation rates based on _ Lake are: calibrated 14C ages with 68.2% confidence limits for Varenis interval from depth 180 (centre of sampled interval) to 380 cm – 0.6– 0.7 mm/year; interval from depth 380 to 510 cm – 0.6–0.9 mm/year; and interval from depth 510 to 710 cm – 0.5–0.6 mm/year. The sediment sequence of Baltys Lake (525–1150 cm) comprises a calibrated 14C age interval of 4440–11,200 cal BP. Most dated intervals (except 800–825 cm) have plateaus and wiggles on the
BP9 630–800
BP8 800–880 BP7 880–970 BP6 970–1040
BP5 1040–1080 BP4 1080–1110 BP3 1110–1180
BP2 1180–1210 BP1 1210–1250
Pinus reaches 88%, Picea 26%, Betula forms a low continuous curve. Small peaks of Poaceae, Cerealia, Rumex and other herbs are present Pinus shows 65%, Picea – 20%. Alnus gradually increases and reaches 8% in the middle of the zone. Cyperaceae, Poaceae, Cerealia, Rumex and Ranunculaceae forms a low curves High representation of Pinus (82–90%), Picea shows 12–20%, Betula – 1–1.5% Pinus gradually increases to the top of the zone from 40% to 82%. Total amount of broad-leaved trees decreases and Picea increases reaching 36%. Alnus shows 3.2% Culmination of Quercetum mixtum and drop in Pinus (25–55%), Ulmus, Tilia and Corylus are registered. NAP pollen increases up to 5% Rise of Quercetum mixtum, a decrease in Pinus (up to 33–45%) and increase in Alnus and Corylus (4–12%). Betula rise up to 10–22% Total amount of AP pollen reaches its maximum in the diagram and comprises about 99%. Pinus curve rises up to 91% Total amount of NAP markedly decreased from 10% to 1.5%. Pinus curve rises up to 90%. The Picea pollen is also significant and varies from 5 to 10% The NAP pollen increased up to 26%, Pinus drops down to 73%. Poaceae curve culminate at 14% and Artemisia increases up to 2.5% Pinus curve increases up to 94% and arboreal pollen are better represented than NAP Pinus curve rises up to 82–85%. Cyperaceae (3.5%), Poaceae (3–5%) and Artemisia (2–3.6%) are represented continuously while the rest taxa occur sporadically Predomination of the AP pollen (89–92%). Pinus is most abundant reaching 50–82%. Picea has a peak of 38%. Betula shows 1.6% High representation of AP pollen (82–90%). Trees are dominated by Pinus pollen and Picea is present as well. Herb pollen makes 18%. Poaceae prevails among herbs. Artemisia and Helianthemum are registered
Author's personal copy
62
V. Sˇeiriene_ et al. / Quaternary International 207 (2009) 58–68
Fig. 3. Percentage pollen diagram for Baltys Lake.
diatom zones (Tables 2–5, Figs. 6 and 7). Seven major phases are recognized in the environmental and climatic development of the region: >13,200 cal BP; 13,200–12,600 cal BP; 12,600–11,200 cal BP; 11,200–8700 cal BP; 8700–6000 cal BP; 6000–3800 cal BP; and
