High-resolution lacustrine record of the late glacial/holocene transition in central Europe

High-resolution lacustrine record of the late glacial/holocene transition in central Europe
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Quaternary Science Reviews, Vol. 12, pp. 287-294, 1993. 0277-3791/93 $24.00 Printed n Great Britain. All fights eserved. © 1993 Pergamon Press Ltd HIGH-RESOLUTION LACUSTRINE RECORD OF THE LATE GLACIAL/HOLOCENE TRANSITION IN CENTRAL EUROPE Tomasz Goslar, 1 Tadeusz Kuc, 2 Magdalena Ralska-Jasiewiczowa, 3 Kazimierz R6z~tnski, 4 Maurice Arnold, 5 Edouard Bard, 6 Bas van Geel, 7 Mieczys4aw E Pazdur, 1 Krystyna Szeroczyriska, 8 Bogumil( Wicik, 9 Kazimierz Wi~ckowski 1° and Adam Walanus 1 1Radiocarbon Laboratory, Institute of Physics, Silesian Technical University, Krzywoustego 2, 44-100 Gliwice, Poland 2Institute of Physics and Nuclear Techniques, Academy of Mining and Metallurgy, Mickiewicza 30, 30-059 Krak6w, Poland 3Szafer Institute of Botany, Polish Academy of Sciences, Lubicz 46, 31-512 Krak6w, Poland 4International Atomic Energy Agency, Section of Isotope Hydrology (RIPC) Wagramerstrasse 5, POB 100, 1400 Vienna, Austria 5Centre des Faibles Radioactivitds, CNRS-CEA, 91198 Gif-sur-Yvette, France 6Universite d Aix-Marseille III, Geosciences de l environnement, URA CNRS 132, CASE 431, Faculte des Sciences et Techniques de St-Jerome, Avenue Escadrille Normandie-Niemen, 13397 Marseille Cedex 13, France 7Hugo de Vries-Laboratorium, University of Amsterdam, Kruislaan 318, 1098 SM Amsterdam, The Netherlands Slnstitute of Geological Sciences, Polish Academy of Sciences, 2~wirki i Wigury 93, 02-089 Warszawa, Poland 91nstitute of Geography, Warsaw University, Krakowskie Przedmielcie 26/28, 02-110 Warszawa, Poland ]°Institute of Geography and Spatial Organization, Polish Academy of Sciences, Krakowskie Przedmie~cie 30, 02-110 Warszawa, Poland In this paper we present the high-resolution record of proxy climatic data in central Europe during the final stages of the last deglaciation, derived from the annually laminated sediments of Lake Gos'ci,a~. central Poland). The isotopic, palynological and other microfossil data confirm sudden changes of climate at the onset and termination of the Younger Dryas (completed within 150 and 70 years, respectively), in close agreement with the previous estimates derived from the polar ice cores and marine sediments. In the upper YD some amelioration of climate took place already about 600 years before the main YD/Preboreal transition. Counting of annual varves in the lake sediments allows a direct estimate of the duration of the Younger Dryas in central Europe; it lasted approximately 1640 years, substantially longer than suggested by previous estimates derived from laminated lake sediments and glacial varves, but agreeing with the radiocarbon calibration data obtained for Barbados corals. The calendar ages of the boundaries of the YD, 12,920 and 11,280 cal BP, are tentatively set. INTRODUCTION The precise and accurate chronology of the last deglaciation is of crucial interest for deciphering the mechanisms responsible for the complex nonlinear response of the global climate to the extraterrestrial radiative forcing (Milankovitch, 1941; Broecker and Denton, 1989; Liu,  1992). In particular, a better understanding of the Younger Dryas cold espidose between 11-10 ka BP (Wright, 1989; Kennett, 1990; Peteet, 1992), superimposed on the global improvement of climate, is urgently needed in view of possible analogues with the predicted global warming. The study of dynamic processes affecting global climate, however, requires the time scale expressed in calendar years ('cal BP) but not in radiocarbon years (14C BP). Here we present the high-resolution record of proxy climatic data in central Europe during the final stages of the last deglaciation, derived from the annually laminated sediments of Lake Go~ci~ (central Poland). SITE DESCRIPTION Lake Go~ci~Z (52 30'N, 10020'E, alt. 64.4 m, a.s.1.) is situated in the area reached by the southern lobe of the ice- sheet during the maximum extent of the Vistulian glaciation (Fig. 1A). The deepest place is situated in the central part of R L3 1 t~. FIG. 1. (A) A Map of Central Europe with he drill sites mentioned n the text. 287  64.4 N 288 T. Goslar etal. 0 lkm i i i B WoslMrn profile [GI,~JOI Contnd s~eT 18 ~ PB lo4o~-~ YD ....... ~ o ,,\~-- 21 o1 ~,..~ oo At. I H C 0 1© li5 dl m] quality of lamination sand peat excellent high medium poor layers FIG. 1. (B) Bathymetric map of Lake Go,cir. The positions of central and western profiles are indicated by crosses. (C) Scheme of the varve chronology of Lake Go~ci~ sediments (at present). The numbers of varves in continuous segments of chronology are printed vertically. The former (Ralska-Jasiewiczowa t al., 1992) number of the oldest varve n western profile s given n parenthesis. The accuracy of varve counting s expressed n terms of quality of lamination: < 0.1%-excellent, < 1%-high, ~5%-medium, > 10%-poor. the lake; another deepening occurs in its western part, about 300 maway (Fig. 1B). The northern shallow bay is drained by a small stream that passes through four lakes and leads to the Vistula river. More details about the lake and its surroundings have been presented elsewhere (Ralska- Jasiewiczowa et al., 1987; Goslar et al., 1992). VARVECHRONOLOGY The sediment was analyzed on five cores, four of them from the lake centre and one from its western part (Fig. 1B). The most important sediment feature is a distinct lamination (couplets of the light mostly calcitic layer and the dark layer  The Late Glacial/Holocene Transition 289 GI~O 13.53 1070 1040 o 1 . lolo 980 13.5~ FIG. 2. The fragment of core G1/90 at the YD/PB boundary (varve number 1040). rich in organic detritus) encompassing the whole sediment profile in the central part of the lake and the lower section of the sediment in its western part (Fig. 2). The annual character of lamination was confirmed (Goslar et al., 1992; Ralska- Jasiewiczowa et al., 1992) by microscope examination of thin sections from the lower parts and tape-peel slides from the top part of sediment according to the methods introduced by Merkt (1971) and Simola (1977), respectively. The very good distinctiveness of the lamination in the central profile permitted the correlation of individual varves among the cores. In varve correlation the characteristic patterns of variations of thickness and brightness of calcite and organic layers along the profiles were compared (cf. Fig. 2 in Ralska-Jasiewiczowa et al., 1992). The occurrences of additional features, e.g. diatom layers, were also taken into account. The confidence of correlation found by such a method, basically similar to that used in dendrochronology, rises with the length of compared sequences, and depends also on the preliminary information about the relative position of both sequences. In most cases, the errors in varve counting disable the correlation, though sometimes the comparison of sequences may be helpful in pointing to the levels where such errors probably occurred. Using such a method it was possible to construct almost continuous, well-replicated varve chronology (Fig. 1C), at least for the sediment below a depth of 8 m ( Goslar et al., 1989). The only breaks are a 0.6 m sand layer and two thin massive silt layers (ca. 1 cm each). The accuracy of varve counting, which was limited by the occurrence of couplets difficult to identify as annual varves, varies with depth and is expressed in terms of the quality of the lamination. Because of the large counting error in the uppermost 8 m, the chronology of the lower part of the sediment should be regarded as floating. The core G1/90 (western profile) is exactly correlative with the central profile in its part corresponding with the laminated sequence above the sandy layer. In the basal part of G1/90 the lamination is continuous and not interrupted by a sand layer. In our previous papers (Goslar et al., 1992; Ralska-Jasiewiczowa et al., 1992) the varves in this part were numbered from -640 to 0 and, according to the preliminary recognition of AL/YD boundary, should be correlative with the varves occurring below the sand in central profile. However, the visual comparison of varved sequences gave two relative positions equally good, i.e. the exemplary 290th varve in the lowermost sequence of central profiles could correspond to varve number -505 as well as varve number -536 in G1/90. Both correlations needed similar serious revision of varve counting in the very basal part of G1/90, i.e. expanding the segment between -500 and -640 in G1/90 by approximately 30 varves. Such an expansion was not excluded since the poor quality of the lamination in this segment (much poorer than in the corresponding segment of central profiles) caused serious doubts in recognizing annual varves. Having no certain correlation we decided to regard the bottom most sequences of both profiles as not being correlative (Goslar et al., 1992; Ralska-Jasiewiczowa et al., 1992). The recently obtained 613C and 5180 curves from the lowest section of G1/90 fit the appropriate curves from central profile reasonably well if we expand the chronology of G1/90 and link the sequences from both profiles at the second of relative positions pointed by visual correlation of varves. In addition, the pollen data correspond better when linked at the second position. For those reasons we have decided to regard the expanded chronology of G1/90 as true and have chosen the second relative position as the most probable. According to the correlation chosen, the number of the oldest varve in G1/90 is -675, and the lowest laminated section in central profiles comprises varves between -536 and -825. Ten radiocarbon dates of samples of terrestrial macrofossils selected from sections encompassing 10-20 varves were obtained by the AMS method using the Gif-sur- Yvette Tandetron facility. The macrofossils are rather rare and small in Lake Go~ci~Z sediments and we were generally obliged to mix samples from adjacent varves in order to gather enough matter for AMS dating (> 150 of C). In order to evaluate the 14C chemistry blank for such material, we prepared blank samples with small pieces of wood collected in the Eemian section of the Grande Pile peat bog. This study showed that the blank ranges between 0.2 pmC and 0.8 pmC for samples decreasing in size between 1.5 mgC  290 T. Goslar et al. TABLE 1. The radiocarbon ages of terrestrial macrofossils from the lateglacial and early Holocene parts of laminated sediments of lake Go~ci02 Sample Varves Mass ktgC ~C age BP Material dated G201M 345-362 170 10,030 + 250 G202M 363-371 760 10,450 + 140 G223M + G224M + G225M 637-666 145 9600 _+ 280 G233M 754-763 590 9950 + 150 G236M + G237M 794-813 355 9870 + 150 G238M 824-833 174 9760 + 330 G244M + G245M + G245AM 886-910 233 9970 + 240 G252M + G255M 973-1020 330 10,360 + 160 G264M 1121-1130 233 9750 + 210 G268M 1171-1180 830 10,050 + 120 Pinus seed wing, wood Pinus needle, Betula nutlet Betula nutlet, seed, scales Bark Betula nutlets, scales Betula nutlets, scales Bud scales Pinus seed wings, Betula nutlets, bud scales Plant detritus Pinus needles and 200 I.tgC. This information was used to correct the 14C ages of the Go~ci~Z microfossils which are listed in Table 1. THE SAMPLES FOR ISOTOPE AND MICROFOSSIL ANALYSES The 13C/12C and ~so/Jro isotope ratios in carbonates (average > 50 of dry mass) in the lower fragments of sediment were measured for samples encompassing 3-10 varves in continuous sequence from the lower sections of central and western profiles. Oxygen isotopic composition of authigenic lake carbonates is controlled by the 180 content and the temperature of the upper region of the lake (epilimnion) during summer months, when the precipitation of calcite takes place (Siegenthaler, Eicher, 1986; Schelske, Hodel, 1991). In open lakes with small evaporation to inflow ratio (fast water turnover) the isotopic composition of lake water reflects that of precipitation over the lake catchment area which, at mid-latitude regions, is mainly temperature dependent (R6~arlski, 1992). Our studies imply that present water turnover in the Lake Go~ci~Z is rather fast, with a mean residence time presumably not exceeding several years, with only minor 180 evaporative enrichment of the lake water when compared to the average isotopic composition of inflow (T. Kuc, unpublished data). Thus, the 180 content of calcite deposited under such conditions may be used as a proxy indicator of climatic episodes in the lake history. The short-term changes of 5~80 and ~3C in the central and western profiles (Fig. 3) are correlative indicating the same parameters of the transition periods, but a systematic shift of absolute values suggests a constant difference in calcite precipitation which could occur in a case of two basins with a 'weak' connection between them. A remarkable warmer (shallower) western basin with relatively high biological activity could create conditions favorable to the observed differences in the isotope composition. This problem will be further investigated. The analyses of plant microfossils and Cladocera analyses were performed on samples encompassing 6-10 varves. THE ALLEROD/YOUNGER DRYAS AND YOUNGER DRYAS/PREBOREAL BOUNDARIES Both the transitions Aller6d/Younger Dryas (AL/YD) and Younger Dryas/Preboreal (YD/PB) were recognized in the central and western profiles from the isotope and palynological data. The most significant pollen and other microfossil curves are shown together with the stable isotope records in Fig. 3. The rec0naissance pollen spectra from the G1/90 profile are superimposed on the pollen curves from G 1/87 core by a precise varve correlation. Some individual differences between the two pollen records are apparent, but the general consistency of changes in the basic pollen assemblages is good. The G1/90 record starts somewhat earlier, with the mire and shallow-water gyttja (high Gramineae and Cyperaceae pollen values) preceding the formation of varved sediment. The basal parts of both profiles represent the late phase of Aller6d with the open Pinus-Betula woods being the dominant regional vegetation type. The Allerrd/Younger Dryas transition is expressed by a decline of Betula (tree pollen-type) and Populus tremula type) frequencies followed by rises ofJuniperus, Salixpolaris type, Artemisia, Chenopodiaceae, Gramineae, Cyperaceae and other indicative herb pollen types (e.g. Gypsophila fastigiata, Rumex acetosa etc.), indicating the reduction of woods and the progressive development of open shrub-herb communities. The sharp drop in the alga Tetraedron minimum coincides exactly with the decrease of 8180, both changes being the reaction to declining summer temperature. Similarly, the decrease of Betula is synchronous with a decline of 8180 by ca. 2 per million in ca. 150 years and an increase in 813C by ca. 4 per million. Both pollen and isotope records reveal some minor fluctuations during the Younger Dryas, which, however, will not be discussed in terms of vegetation changes until the full pollen sequences from the core G1/90 and from cores collected across the northern lake bay (D. Demske, in preparation) are reconstructed. The YD/PB transition appears in pollen records as a sharp boundary between varves 990 and 1040, preceded, however, by a phase of 300-600 years that could be called a 'descending part of the Younger Dryas'. This phase is characterized by the increased though fluctuating Betula pollen values, including Betula nan type, and by consistently declining Juniperus, Salix polaris type and Chenopodiaceae pollen curves, followed by a decrease in Artemisia. All these changes suggest a gradual reduction of open shrub-herb communities of rather xeric character by developing birch copses. The mean values of ~51so increase by ca. 0.5 per million. The boundary itself is indicated by a definite reduction in Juniperus, Salix polaris type, Artemisia and Chenopodiaceae pollen percentages, a distinct rise of  The Late Glacial/Holocene Transition 291 .- ..... -~ .................... f--~ ............ 1 .... r ..................... 7. .................. 11 I 8~I Im oo~ o 3~ ~ 0 O~ 9 0 eu~u01 e~cl ~l ~ m~f N '(,,,/') em 0 0 I,I < _J aAJ c~ I : iji±i 3tL. tt EO> • 0 0 0 = 0
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks