in il

Stable and studies of Late Weichselian lake sediments in southern Sweden and northern Poland, with palaeoclimatic implications

Dan Hammarlund

Loud 1994 Volume 31 Loud University, Department of Quaternary LUNDQUA Thesis 31

Stable carbon and oxygen isotope studies of Late Weichselian lake sediments in southern Sweden and northern Poland, with palaeoclimatic implications

Dan Hammarlund

Avhandling att med tillstånd från Matematisk-Naturvetenskapliga Fakulteten vid Lunds Universitet för avläggande av filosofie doktorsexamen, offentligen försvaras i Naturgeografiska Institutionens föreläsningssal, Sölvegatan 13, Lund, fredagen den 6 maj 1994 kl. 10.15

Lund 1994 Lund University, Department of Quaternary Geology Organization LUNDUNIVERSnY DOCTORAL DISSERTATION

Date of issue April ,994

Department of Quaternary Geology CODEN: SE-LUNBDS/NBGK-94/31+30p

Authors) Sponsoring organization Dan Hammarlund

Title and subtitle Stable carbon and oxygen isotope studies of Late Weichselian lake sediments in southern Sweden and northern Poland, with palaeoclimalic implications.

Abstract Late Weichselian lacustrine sediment sequences from southern Sweden and northern Poland were studied by means of stable isotope in order to reconstruct the climatic development and climatically induced environmental changes in the respective regions. The methods used include analyses of the stable carbon isotope composition (813C) of bulk organic material, and the stable carbon and oxygen isotope compositions (S'-'C, 8'8O) of bulk carbonates and carbonate shells of aquatic organisms. These results were complemented and supported by lithological, chemical and biostratigraphic data (plant macrofossils, insects, molluscs). Chronological data were obtained by AMS radiocarbon dates and correlations based on pollen analysis. The carbon isotope data are mainly interpreted in terms of aquatic vegetation production, decomposition, and soil development, although a number of other processes may also be of importance, e.g. dissolution of silicates. In southern Sweden the oxygen isotope data primarily reflect changes in lake- temperature and hydrological parameters, such as evaporation, whereas in Poland variations in oxygen isotope composition are mainly related to changes in mean annual air temperature. At c. 12,400 BP a climatic change from arctic, dry, and continental, to subarctic and more humid and maritime conditions occurred in southern Sweden. The Older Dryas stadial (c. 12,200-12,000 BP) is characterized by a temporary return to generally colder, drier, and more continental conditions, followed by generally favourable (subarctic), although unstable, climatic conditions. At c. 11,300 BP a gradual transition towards a colder and more continental climate was initiated, followed by total absence of limnic carbonates during the Younger Dryas stadial (c. 11,000-10,200 BP), indicating arctic and continental conditions. The transition to the Holocene is characterized by a rapid and strong climatic wanning. The results from northern Poland point to some important differences compared to this development. A climatic wanning around 13,000 BP was followed by generally favourable climatic conditions enabling continuous sedimentation of limnic carbonates during the Late Weichselian. The oscillations towards stadial climatic conditions were less pronounced, although clearly discernaWe and synchronous with corresponding features in southern Sweden. Distinct dep'M >ns of 13C in lacustrine organic material at the transition to the Holocene were recorded in southern Sweden, also demonstr .< I y decreasing mean values obtained from an extensive compilation of 813C data. A number of processes that may influent- é i of organic lake sediments are discussed. Other methodological results and problems of general importance, related to stf • •«» ope analysis of lake sediments are also addressed.

Keyword! Stable isot-ip s Jarbon , Oxygen isotopes, Biostratigraphy, Lake sediments, Palaeoclimate, Late Weichselian, Southern Sweden, Nor vrn Poland

Classifies t i system and/or index terms (if any)

Supplementary bibliographical information l**W»V English

400 copies ISBN ISSN and t y tide O28I.3O33 LUNDQUA THESIS

Recipient's aotes Number of pages 30+4 app. **• 60 SEK

Security classification

Distribution by (name and address) Department of Quaternary Geology, Tornavägen 13, S-223 63 Lund, Sweden I, the undersigned, being the copyright owner of the abstract of the above-mentioned dissertation, hereby grant to all reference sources permission to publish and disseminate (he abstract of Che above-mentioned dissertation.

Signature C%0 GWl4é*V/mi& Date '994-03-18 To Elisabet and Agnes Contents

1 Introduction 2

2 Investigation area 3

3 Methods 5

4 Summaries of the papers 6 4.1 Paper I 6 4.2 Paper U 6 4.3 Paper ffl 8 4.4 Paper IV 8

5 Discussion and conclusions 10 Methodological results and problems 10 .1 Sediment chemistry data 10 .2 Organic carbon isotope data 11 .3 Carbonate carbon isotope data 15 .4 Carbonate oxygen isotope data 16 5.2 Palacoclimatic implications for southern Sweden 17 5.2.1 Before c. 12,400 BP 17 5.2.2 c. 12,400 to c. 12,200 BP 18 5.2.3 c. 12,200 to c. 12,000 BP 19 5.2.4 c 12,000 to c. 11,000 BP 19 5.2.5 c. 11,000 to c. 10,200 BP 20 5.2.6 c. 10,200 to c. 9,500 BP 20 5.3 Comparisons with the Polish study 22 5.4 Potential of the method in southern Sweden 23

Acknowledgements 24

References 25

Appendices I-IV: Papers included in the thesis Stable carbon and oxygen isotope studies of Late Weichselian lake sediments in southern Sweden and northern Poland, with palaeoclimatic implications

by Dan Hammarlund

Department of Quaternary Geology, Lund University, Tornavägen 13, S-223 63 Lund, Sweden

This thesis is based on the following four papers and the present summary and extension:

I. Hammarlund, D. 1993: Evidence of a dis- tinct 5"C decline in organic lake sediments at the Pleistocene-Holocene transition in southern Sweden. Boreas 22, 236-243.

II. Hammarlund, D. & Lemdahl, G. 1994: A Late Weichselian stable isotope strati- graphy compared with biostratigraphical data: a case study from southern Sweden. Journal of Quaternary Science 9, 13-31.

III. Hammarlund, D. & Keen, D. H. (manu- script): A Late Weichselian stable isotope and molluscan stratigraphy from southern Sweden. Submitted to Boreas.

IV. Hammarlund, D. 1994 (preprint): Late Weichselian lake sediments analysed for stable carbon and oxygen isotopes: a study at Imiotki, Lake Lednica, Great Poland. Accepted for publication in Biblu/teka Studiow Lednickich, Poznari.

Reference to the papers is made by using their roman numerals, which also apply to their respective locations as appendices in the thesis. Paper I is reprinted with the permission of Boreas and paper II is reprinted with the permission of Journal of Fig. I. Core section of faintly laminated, clayey, Quaternary Science. An english version of paper IV calcareous gyttja with mollusc-sl.'ll fragments from is presented with the permission of Biblioteket Vanstads mosse, southernmost Sweden (scale bar in Studiow Lednickich. cm).

1 1 Introduction

The present investigation, which is a contribution to archives for detailed reconstructions of Late Weich- IGCP 253 (Termination of the Pleistocene), was selian and Holocene climate . initiated in 1989. The main aim of the study is The oxygen isotope composition of limnic carbo- twofold. Firstly to reconstruct the climatic and nates can, at favourable conditions, provide useful environmental development during the Late Weich- information on past regional climate, as it reflects selian Substage within the southern part of the large-scale atmospheric circulation, lake-water tem- Scandinavian giaciation area, secondly to evaluate perature and hydrological conditions. However, the the potential of stable isotope studies of lacustrine complex relation to changes in temperature calls for sediments in the region (i.e. southern Sweden). great caution at the interpretation of oxygen isotope Oxygen isotope studies have played an important data. Results from carbon , on the role for the knowledge about Quaternary climate his- other hand, may reflect local limnic conditions rela- tory. Records of 5[*O variations in carbonate micro- ted to a number of biological and chemical pro- fossils from deep-sea sediments have provided a cesses, most of which in turn depend on climate. stratigraphic framework of isotope stages repre- Carbon isotope data obtained on carbonates and or- senting glacials and interglacials. Most of the ganic material respectively, provide complementary research carried out today, concerned with Quater- information of great value for reconstructions of nary stratigraphy and climate history, is related to local environmental changes. these isotope stages. Furthermore, the last glacial This thesis presents a summary of stable carbon cycle has been subject of detailed studies based on and oxygen isotope studies of Late Weichselian 5'*O variations in glacier ice-cores from Greenland lacustrine sediments from southern Sweden and and Antarctica. Apart from these well established northern Poland, supported by biostratigraphic data. fields of research, the last decades have experienced The results are discussed in terms of climatic an increasing number of stable isotope studies changes and climatically induced environmental focused on lake sediments, (see App. Ill for a brief development. Methodological results and problems of historical review). The high resolution, rapid re- general importance at the interpretation of stable sponse to climatic changes, and relative accessibility, isotope data are also presented. The emphasis is means that lacustrine deposits are very suitable placed on processes influencing the carbon isotope composition of organic material in lake sediments. 2 Investigation area

The principal part of the studies within this project Weichselian Substage. This can be tested by was, for several reasons, located to southernmost stable isotope analysis. Sweden: The deglaciation of southern Sweden is 1. With the exception of a study on the island of relatively well known through a number of strati- Gotland (Morner & Wallin 1977), mainly concen- graphic studies. As a synthesis of this work, a model trating on Holocene climatic development, and with ice recession lines was presented by Björck et two short sequences deposited around 10,000 BP al. (1988). The solid lines of Fig. 2 roughly corre- from the Billingen area in the Middle Swedish spond to this model. The Kullaberg Peninsula in the end-moraine zone (Björck & Digerfeldt 1984), no northwestern part of Skåne (Fig. 1 of App. II) has environmental isotope studies of Late Weichselian been considered to be the first part of Sweden that lake sediments have up to now been carried out in became ice-free, around 14,000 BP according to Sweden. Lagerlund (1980; 1987). However, the deglaciation chronology of Skåne is not known in detail and is 2. The province of Skåne was the earliest deglaciated still under debate. An alternative model was presen- part of Sweden, which means that the oldest lacu- ted by Lagerlund & Houmark-Nielsen (1993), pro- strine sediments can be expected to be found here. posing a somewhat earlier deglaciation, initiated along the western coast of Skåne, well before 14,000 3. The region is very well studied by means of a BP (dashed lines of Fig. 2). The present study in- variety of stratigraphic methods. Several recon- cludes two investigated sites in Skåne, Körslätta- structions of Late Weichselian climatic changes mossen (App. II) and Vanstads mosse (App. III). The based on biostratigraphy have been presented (e.g. locations of the sites are shown in Fig. 2. In addition Berglund 1971; Berglund et al. 1984; Lemdahl to these studies, stable carbon isotope data from 1988a; Liedberg Jönsson 1988; Bergsten & c. 90 sites in southern Sweden, south of approxima- Nordberg 1992). Other studies have been based tely 00°N, were compiled and presented separately on lithostratigraphy and glacial geology (e.g. (App. I). Berglund 1979; Björck 1981; Berglund &Mörner As a complement to the studies in southern 1984; Björck & Möller 1987). This gives ex- Sweden, a lacustrine sequence at Imioiki in North cellent opportunities to compare the first results of Central Poland (Fig. 2) was investigated (App. IV). stable isotope studies with existing models. The ice recession in northern Poland can be traced by a number of distinct moraine ridges, mainly 4. The geographical position of southern Scandinavia oriented in an east-westerly direction (Kasprzak & at present involves frequent shifts of air-masses of Kozarski 1988). However, insufficient dating makes continental and maritime origin respectively. It the deglaciation chronology uncertain. The southern- can be assumed that the region was influenced by most ridges, corresponding to the r.aximum extent of similar patterns of change in different -per- the Weichselian glaciation, are believed to have been spectives during the climatically dynamic Late formed at c. 20,000-18,000 BP (Kozarski 1980). NORWAY

100 km i i 52°-

10°

Fig. 2. Map of the southern Baltic Sea Region showing the geographical positions of the investigated sites, K = Körslättamossen (App. II), V = Vanstads mosse (App. III), I = Imiolki (App. IV). The deglaciation chronology of southern Sweden is illustrated by approximate Late Weichselian ice-marginal positions (/4C years BP), based mainly on the model presented by Björck et al. (1988; solid lines). The dashed lines refer to an alternative deglaciation chronology model (Lagerlund & Houmark-Nielsen 1993; Lagerlund unpubl. data). The heavy line represents the maximum extent of the Weichselian glaciation (20,000-18,000 MC years BP). 3 Methods

The general principles of environmental isotope analyses have been described in a number of text- books. A theoretical background to studies of lake sediments was presented by Pearson & Coplen (1978). Siegenthaler & Eicher (1986) discuss the same problems from a more practical point of view. For more detailed literature reviews and considera- tions on fractionation processes, see Deines (1980) for organic material, and Buchardt & Fritz (1980) for carbonates. The laboratory procedures used at the different physical, cnemical and stable isotope analyses vary slightly between the different studies. These, as well as the fieldwork methods (Fig. 3) are described in detail for each site in Appendices I-IV. Reproduci- bility of the analytical results is better than ±3% for carbon analysis, and better than ±4% for humus ex- traction (measured on a standard). Humic content of the sediments is related to organic carbon content (relative spectro-photometric absorbance). In the following all isotope notations are expressed as per mille deviations from the international PDB standard (Cretaceous belemnite from the Peedee formation, South Carolina; Craig 1957);

oC'C, " I03 %c (1)

Reproducibility of the stable isotope analyses is Fig. 3. Coring at Vanslads mosse. May 1989 (J. within ±0.07%c. Ising and G. Lemdahl). 4 Summaries of the papers

4.1 Paper I An important factor proposed as explanation for the depletion of "C is a more extensive dissolution of U silicates in the catchments and water bodies of the Evidence of a distinct 5 C decline in lakes during the Late Weichselian than during the organic lake sediments at the Pleistocene- Holocene. An abundant supply of fine-graineo Holocene transition in southern Sweden, silicate particles and high concentrations of by Hammarlund, D.; Boreas 22, 236-243 aggressive in precipitation and surface (1993). water, would enhance the chemical weathering and to a relative enrichment of "C in lake water DIC during the Late Weichselian. Other contributing This paper presents a compilation of 8"C values factors may be an increased supply of "C-depleted obtained from conventionally radiocarbon dated bulk carbon dioxide from soil processes, and a general sediment samples from southern Sweden (S. of change in the limnic vegetation towards increased 60°N). Dates in the range 13,000-8,000 BP obtained production of phytoplankton, as a response to the on pure lacustrine deposits in areas of non-calcareous I4 climatic warming at the transition to the Holocene. bedrock were used, rejecting samples with C ages On the other hand a global enrichment of "C in significantly deviating from expected ages according atmospheric carbon dioxide of c. \%e around 10,000 to pollen data and/or deglaciation chronology. Thus, BP probably counteracted the observed trend. The 286 samples from 91 sites were included in the com- lake sediment data of the present study are closely pilation. resembled by decreasing values of 8"C at this Running mean values calculated for the entire in North American wood cellulose. This may indicate population reveal a distinct depletion of "C in bulk a mutual relation to changes in the global carbon organic material, centred around 10,000 BP and cycle at the termination of the last glacial. amounting to c. 5%c. The mean levels before and after 10,000 BP are -23.2%c and -28.2%» respectively. Similar trends had earlier been recorded across the Pleistocene-Holocene boundary in the continuous 4.2 Paper II carbon isotope records of Körslättamossen and Van- stads mosse in southernmost Sweden (App. II and A Late Weichselian stable isotope strati» III). This study was thereby conducted in order to prove the occurrence of secular changes in the graphy compared with biostratigraphical region, and to evaluate possible causes of 6"C data: a case study from southern Sweden, variations in lake sediments in general. by Hammarlund, D. & Lemdahl, G.; Jour- A comparison of the compilation results with 8|JC nal of Quaternary Science 9, 13-31 (1994). data obtained on bulk sediment organic material in the two individual lacustrine sequences, which are This paper is based on studies of a partly carbonate- partly rich in carbonates, shows that the rapid rich lake sediment sequence at Körslättamossen in depletion of "C at c. 10,000 BP is mainly in- the northwestern part of the province of Skåne, dependent of endogenic carbonate production. The southern Sweden (Fig. 2). The following stable iso- change from complete absence of carbonates during tope parameters were analysed: 8"C of bulk organic the Younger Dryas stadial (c. 11,000 to c. 10,200 material (S"C,,r,). 8I3C and 8'"O of bulk carbonates

BP) to abundant precipitation of carbonate shells (8"Ccllfc, 8"OCJrt,), and 8"C and 8'"O of carbonate during the Younger Dryas-Preboreal transition zone shells, mainly pelecypods of the family Sphaeriidae IJ (c. 10,200 to c. 10,000 BP) could to a minor extent (8 Cm,,n, 8'"Om(l)1). The interpretations of these results have led to a depletion of I3C in dissolved inorganic were supported by plant macrofossil and insect re- carbon (DIC) by withdrawal of I3C for carbonate cords (G. Lemdahl), as well as carbon analysis, production. However, the main part of the observed extraction of humic compounds, and relative quanti- depletion must be attributed to other processes. The fication of shell carbonate. The chronology was carbon isotope composition of the organic component established by a combination of correlations based on of lake sediments is most probably affected by a pollen analysis, and five radiocarbon dates obtained combination of processes related to local environ- by accelerator mass-spectrometry. The 4m thick se- mental conditions, which in turn depend on regional quence analysed, consisting mainly of calcareous climate. gyttja clay and clayey gyttja, was deposited between c. 13,000 and c. 10,000 BP. Comparisons with the changing plant macrofossil A successive depletion of I3C in bulk carbonates at assemblages and the content of shell carbonate the transition to the Holocene may be related to the l3 throughout the sequence, made it possible to demon- same processes as those affecting 5 Cl)rt data. strate the relation between carbon isotope compo- Within the lower parts of the sequence, characterized sition of bulk organic material and limnic environ- by low carbonate contents, relatively high values of l8 mental conditions. Values of 5'"'ClÄ( around -2b%c in 5 OtlA (c. -5%c) were recorded. A rapid shift to the lower part of the sequence reflect dominance of values ;.i the range -I0%e to -9%c was recorded at bryophytes and assimilation of atmospheric carbon c. 12,400 BP, a level which was retained until dioxide. A shift to values close to -24%c may be c. 11,000 BP, with the exception of the period related to increased re-deposition of terrestrial between c. 12,200 and c. 12,000 BP. In the sedi- organic material, followed by even higher and some- ments deposited during the latter period, the Older what fluctuating values reflecting the immigration of Dryas stadial, values around -7%c were recorded. The macrophytes. At this stage (c. 12,500 to c. 11,800 strong enrichment of '"O in bulk carbonates during BP) a relatively sparse production of carbonate shells the above-mentioned periods is interpreted to reflect and insignificant recycling of carbon dioxide from a substantial evaporation from the lake-water surface. decomposition, led to a relative enrichment of L'C in This may indicate a continental climate characterized organic material. The opposite situation within the by dry summer conditions. At c. 12,400 BP a distinct upper part of the interstadial (c. 11,800 to c. 11,000 shift towards a more humid and maritime climate is BP) led to successively lower values of 8"Cwg, assumed to have occurred. The lack of data represen- reaching c. -217,,. A rapid and total cessation of the ting the Younger Dryas stadial is followed by a de- carbonate production during the Younger Dryas creasing trend within the Younger Dryas-Preboreal stadial resulted in a shift towards high values transition zone (c. 10,200 to c. 10,000 BP). The (c. -22%c), followed by an even more pronounced de- depletion of "O in bulk carbonates at this stage was pletion at c. 10,200 BP, giving rise to values of probably caused by an increase in lake-water tem- 5"C,,r| around -29%e (see App. 1 and section 5.1.2). perature. A few values of 5'"O, obtained on limnic 5"Ctart) data principally reflects the rate of aquatic ostracods were recorded in the lowermost part of the vegetation production. Relatively high values sequence. These data, mainly reflecting winter con- (c. +2%o) in the lowermost, moss-rich part of the se- ditions, are depleted in "O by c. 2%c compared to quence, are followed by a decreasing trend to corresponding values of S'^O,^, a difference that c. -2%c, reflecting decreasing abundance of bryo- supports the theory of a substantial summer evapo- phytes. At c. 12,400 BP a rapid increase in vegeta- ration during the early lake history. A relatively tion production led to an enrichment of "C in bulk constant level of 8'"OmoH data from c. 12,400 to l3 carbonates. However, the high values of 5 Ccart, c. 11,000 BP around -8.5%o, with no considerable subsequently recorded (c. +O.5%o) were temporary re- change shortly before 12,000 BP, in contrast to placed by lower values (c. -\%c) in the sediments S"Ocart,, may indicate that the Older Dryas climatic representing the Older Dryas stadial (c. 12,200 to oscillation was mainly related to summer conditions. c. 12,000 BP). This depletion of "C in limnic carbo- Mollusc shell carbonates are assumed to be formed nates, interpreted to reflect a climatically induced during most of the year, whereas limnic carbonates decrease in productivity, was accompanied by a simi- produced by mainly precipitate during lar decrease in values of 5"Cmol| due to a general the summer. depletion of "C in DIC. At c. 11,800 BP an abrupt The interpretations based on biostratigraphic data decline in values of S'3C amounting to c. 2%e was cart) are generally well in accordance with the isotope recorded, preceded and followed by relatively con- results. At c. 12,400 BP an increase in insect stant levels. The possibility that this shift was caused diversity and vegetation density was recorded. by an adaptation of the limnic vegetation to high Quantifications of mean summer and winter tempera- contents of bicarbonate in the lake water is dis- tures based on beetle analysis clearly identifies the cussed. The occurrence of amorphous carbonate Younger Dryas stadial, showing a distinct cooling aggregates at this stage indicates the introduction of around 11,000 BP and a strong warming at c. 10,200 a type of photosynthetic assimilation known as BP. However, the Older Dryas oscillation (c. 12,200 ion pumping. A complete absence of car- BP to c. 12,000 BP) was not recognized in biostrati- bonates in the sediments deposited during the graphic data. Instead a mean July temperature op- Younger Dryas stadial results in lack of data. timum, based on insect data, was recorded around 12,000 BP. 4.3 Paper III In the lowermost part of the sequence values of 8"Otltb around -5%o were most probably inherited by detrital carbonates. Shortly before c. 12.40QJJP even A Late Weichselian stable isotope and higher values (close to -3%o) indicate evaporative molluscan stratigraphy from southern enrichment of "O in limnic carbonates, corre- Sweden, by Hammarlund, D. & Keen, D.H. sponding to the development at Körslättamossen at (manuscript submitted to Boreas). this stage. A large and rapid shift at c. 12,400 BP to values around -10%o was preceded by a c. 2%e en- This paper is based on studies of a partly carbonate- richment of "O in mollusc shell carbonate. This rich lake sediment sequence at Vanstads mosse in the indicates a substantial climatic change towards more southern part of the province of Skåne, southern maritime conditions, involving increased humidity Sweden (Fig. 2). The following stable isotope para- and subsequently decreased evaporative enrichment meters were analysed: 5"C of bulk organic material during the summer. A relative enrichment of '*O in 13 IS (8 Cort). 5"C and 5 O of bulk carbonates (8"Ccrt> precipitation as a year-around average, indicated by 18 8'"Ocart,), and 5"C and 5 O of carbonate shells, increasing values of 8"O in mollusc shell carbonate mainly ostracods of the species Candona Candida probably indicates a climatic wanning at 12,500- 1J (8 C0Mr, 8'*Ollstf). The interpretations of these results 12,400 BP. A slight enrichment of "O in bulk car- were supported by a study of the mollusc fauna bonates may indicate a return to a more continental (D.H. Keen), as well as by carbon analysis, ex- climate during the Older Dryas stadial (c. 12,200 to traction of humic compounds, and relative quanti- c. 12,000 BP). At the transition to the Holocene a fication of shell carbonate. The chronology was decreasing trend is in good agreement with the re- established by a combination of correlations based on sults from Körslättamossen, indicating increasing pollen analysis, and four AMS radiocarbon dates. lake-water temperature. The record of 8'"OMtt data The 7m thick sequence analysed, consisting mainly exhibits relatively stable values around -6%o from of calcareous gyttja clay and clayey gyttja (Fig. 1), c. 12,400 to c. 11,300 BP, followed by a decreasing was deposited between c. 13,000 and c. 9,500 BP. trend to c. -8%o shortly before 11,000 BP. This is interpreted to reflect a successive climatic cooling The variations in 5^Cn,t data throughout the and an increase in continentality well before the sequence principally coincide with corresponding initiation of the Younger Dryas stadial. This assump- results from Körslättamossen (App. II). The onset of, tion is supported by the introduction of two cold and increase in limnic productivity at c. 12,400 BP demanding mollusc species at this stage. is reflected by an enrichment of "C in bulk organic material. Minor fluctuations shortly before 12,000 BP Deviating values of 8"C and S'*O obtained on and between c. 11,800 and c. 11,000 BP are closely bulk carbonates at the beginning of the Younger related to the content of shell carbonate, lowered Dryas stadial were found to be related to the occur- shell production around 11,500 BP being accompa- rence of siderite (FeCO,) in the sediments. These nied by a slight enrichment of I3C in organic isotope results cannot be interpreted in terms of l3 material. Relatively low values of 8 Corj before climatic changes. c. 11,000 BP are followed by a rising trend during Important information on the limnic environment the Younger Dryas stadia! and a very rapid decline facilitating the interpretation of stable isotope data at c. 10,200 BP (see App. I and section 5.1.2). can be derived from mollusc analysis. Based on a l3 Also the interpretations of the 8 Ctart, data agree synthesis of results from the different methods in- with corresponding results from Korslp' nossen. A cluded in the study, variations in different climatic productivity increase at c. 12,400 Bh is iterrupted parameters as well as general climatic changes were by a distinct drop at c. 12,200 BP and a second estimated throughout the Late Weichselian and at the increase at c. 12,000 BP, the latter oscillation transition to the Holocene. probably being the result of a climatic cooling during the Older Dryas stadial. In the lowermost part of the sequence very stable values of 5"Ctjrb are attributed 4.4 Paper IV to detrital input of Palaeozoic limestone particles. Between c. 12,400 and c. 11,000 BP variations in Late Weichselian lake sediments analysed values of S"C0,lr of similar character and magnitude IJ for stable carbon and oxygen isotopes: a as those of 8 C,)r(, data were recorded. A slight enrichment of "C in DIC at c. 11,500 BP, accompa- study at Imiotki, Lake Lednica, Great nied by a very sparse mollusc fauna may be inter- Poland, by Hammarlund, D.; Biblioteket preted as a climatic cooling involving lowered lake- Studiöw Lednickich, Poznati (1994). water temperature during this period. This paper is based on studies of a highly carbonate- corded at this stage, resembling the decreasing trends rich lake sediment sequence at Imiolki, c. 35km observed in southern Sweden at this stage. northeast of Poznari in North Central Poland (Fig. 2). A distinct enrichment of I3C in bulk carbonates to The following stable isotope parameters were ana- c. +2%c in the lower part of the sequence supports lysed: 8"C of bulk organic material (8I3C ), and the theory of an increase in aquatic vegetation I8 8"C and 8 O of bulk carbonates (8"CclIb, 8™Oclrt>). production at an early stage. The following fluctu- The interpretations of these results were supported by ating values at a lower level (c. +0.5%c) are probably comparisons with separate biostratigraphic studies of caused by increased recycling of "C-depleted carbon the same sequence, mainly based on plant macro- dioxide from decomposition of organic material. fossil analysis (L. Kubiak-Martens unpubl. data). The During the Younger Dryas stadial a distinct decrease l3 physical composition of the sediments was analysed in values of 8 Cclrtl to c. -0.5%o, followed by stable by loss on ignition. The chronology of the sequence values, point to a generally lowered limnic produc- was established by means of biostratigraphic corre- tivity, which seems to increase again at the transition lations with radiocarbon dated lacustrine sequences to the Holocene. in the close vicinity of Imiolki. Pollen analysis Stable values of 8"OCill() close to -5%e in the results (A. Wawrzyniak-Gluszak unpubl. data) in- lowermost part of the sequence, in combination with 13 dicate that the 2.5m thick section analysed was stable values of 8 Cclrtl around +0.5%c, indicates deposited between c. 13,500 and c. 9,500 BP. erosional input of detrital carbonates. Subsequently, 1 The record of 8"COT| data differs considerably minor fluctuations in values of S ^^ in the range from corresponding results of the studies in southern -%%c to -6%« until c. 11,000 BP, closely resemble re- Sweden. Initially high values, c. -20%c, are followed sults from similar studies in Germany, Poland and by a distinctly decreasing trend to c. -27%c, in the Baltic Region. It may be possible to correlate two combination with substantial fluctuations during the oscillations in 8"Ocart data towards slightly lower Allerod interstadial. This development is accompa- values with similar features recorded in Central nied by an increase in carbonate content to con- Europe. These oscillations have been dated to shortly tinuously high values. This probably indicates a before 12,000 BP and shortly before 11,000 BP re- comparatively early development towards high litnnic spectively, and are interpreted as due to depletions of productivity, although considerable oscillations can !IO in precipitation during cold episodes. During the be assumed to have Owurred before c. 11,000 BP. In Younger Dryas stadial a distinct depletion of 18O in sharp contrast to this period, the Younger Dryas bulk carbonates was recorded, possibly distinguished stadial seems to have been characterized by a some- as two phases separated by a period of slightly what lowered productivity in combination with more higher values. This depletion of "O most probably stable conditions. With respect to productivity con- corresponds to similar data obtained from several ditions, the transition to the Holocene indicates a lacustrine sequences in Continental Europe, and return to at least interstadial conditions, although suggests that the climatically induced change in only restricted data are available. It should be noted atmospheric circulation patterns recorded during the that no depletion of I3C in organic material was re- Younger Dryas stadial, affected also northern Poland.

Fig. 4. View of the present Lake Lednica close to the sampling point at Imiolki, Great Poland, September 1990.

9 5 Discussion and conclusions

In the following sections selected results and tal sediment dry weight, the siderite predominantly problems, obtained from the different studies within occurring in the sediments deposited during the the project, are discussed in a broader sense. More Younger Dryas stadial and within the Younger comprehensive and detailed discussions are presented Dryas-Preborcal transition zone. The siderite was in the respective papers (App. I-IV). probably authigenically precipitated in the pore-water after deposition of the sediments as a result of microbial methane production (Engström & Wright 5.1 Methodological results and 1984; Curtis et al. 1986). Authigenic formation requires anaerobic conditions, low ratios of dissolved problems sulphates, and Fe/Ca ratios greater than 0.05 (Berner 1971). The precipitation of sidcrite shortly after Below principles of general importance for the 11,000 BP at Vanstads mosse (Fig 5 of App. Ill) interpretation of stable isotope data, as well as probably reflects a climatically induced environ- lithological and chemical data, are addressed. mental change with cessation of carbonate production (see section 5.2.5). However, the significance of siderite as a climatic indicator seems to be of 5.1.1 Sediment chemistry data uncertain value, as the siderite content increases shortly before 10,000 BP at Körslättamos en (Fig. 5 Without specifying a lower limit, Siegenthaler & of App. II). The carbon and oxygen isotope composi- Eicher (1986) state thnt the sediments should contain tions of the pure siderite samples from the Vanstads a sufficient amount of carbonates to meet the mosse sequence (Fig. 5 of App. Ill) are in good demands of stable isotope analysis. Most earlier agreement with data presented by Mozley & Wersin s udies have been conducted on sequences of pure (1992). Stable isotope data obtained on bulk carbo- lake-marl or sediments with very high contents of nate samples contaminated by siderite should be calcium carbonate, generally more than 70% of total regarded with caution, since a "C fractionation effect dry weight, based on loss on ignition (e.g. Bengtsson has been recorded relative to calcite (Curtis et al. & Enell 1986). Most of the sequence at Imiotki ex- 1986). These considerations accentuate the impor- hibits such high carbonate contents (Fig. 2 of App. tance of chemical analysis accompanying stable iso- IV), whereas the two sequences studied in southern tope studies of sediment carbonates, at least con- Sweden are partly characterized by only slightly cerning sequences with low carbonate contents. calcareous clay gyttjas. In these studies, however, the Based on analysis of the humic content of the carbon content of the sediments was determined by sediments according to Bahnson (1968), information the use of a high precision carbon determinator, en- on build-up and degradation of permafrost in the abling organic material and carbonates of different region has been obtained. Generally relatively high chemical composition to be distinguished and quanti- humic values at, or shortly after climatic shifts, fied as elemental carbon. At Körs lättamossen parts of inferred from stable isotope and biostratigraphic data, the isotope data were obtained on sediments with less are assumed mainly to represent changes in perma- than 1 % carbonate carbon content (corresponding to frost conditions. During the earlier part of the less than c. 8% t.d.w. calcium carbonate; Fig. 9 of Younger Dryas stadial increasing contents of humic App. II). These results could be affected by pre- substances were recorded (Fig. 5 of App. II; Fig. 3 paration errors (B. Buchardt, pers. comm.) and of App. Ill), probably reflecting presence of perma- should be regarded with caution. However, the close frost around the lakes. Thawing of the active layer resemblance with corresponding data from Vanstads during the summer seasons may have induced ero- mosse (Fig. 5 of App. Ill) indicates that the results sion of cryoturbated organic soils. Repeating of this are most probably reliable. There is no reason to process during several hundreds of years probably assume that sediments with low carbonate contents caused a successive depletion of organic material in would be less suitable for stable isotope analysis, the soils, and lower input of humus to the lakes provided that the chemical composition of the car- during the later part of the Younger Dryas stadial. bonates is properly confirmed. The occurrence of permafrost during the Younger By means of carbon analysis and x-ray diffraction Dryas stadial is clearly demonstrated by finds of analysis restricted amounts of siderife (FeCO3) were fossil ice-wedges and polygonal patterns in south- identified in parts of the South Swedish sequences. western Sweden (Svensson 1988; 1990), as well as The concentrations are generally less than 1 % of to- other permafrost features in southern Scandinavia

10 (e.g. Kolstrup 1985), dated to this period. The cli- A. Lake-water temperature matic warming at c. 10,200 BP most probably re- sulted in degradation of permafrost, major soil dis- Dissolved bicarbonate in equilibrium with atmo- turbances, and increased erosive input of coarse spheric carbon dioxide or with carbon dioxide minerogenic material as well as humic compounds to dissolved in lake water is increasingly enriched in the lakes (ef. Berglund & Malmer 1971). Relatively 13C with decreasing temperature (Deuser & Degens high values of humic substances were recorded at 1967; Mook et al. 1974). The fractionation effect is this stage. Slightly increasing humic values shortly relatively modest, 0.08%o/°C, although it could to after 12,400 BP and around 12,000 BP at Vanstads some extent have contributed to the recorded de- mosse imy also indicate breakdown of permafrost pletion of "C, on assumption of a general increase in that probably existed before c. 12,400 BP (Björck lake-water temperature at the transition to the 1979) and during the Older Dryas stadial (Björck & Holocene. Möller 1987) respectively. In contrast, relatively low values were recorded during the interstadial parts of the sequences (c. 12,400 to c. 12,200 BP and B. Latitude c. 12,000 to c. 11,000 BP), indicating absence of permafrost and stabilized soils in the region Evidence of decreasing values of 8"C with higher (ef. H.J.B. Birks 1986). These results are well in latitude have been presented, both concerning lake accordance with stable isotope and biostratigraphic sediments (Stuiver 1975), and wood (Håkansson data, and constitute parts of the basis for the model (1985). This is probably an effect of increasing of climatic change in terms of continental versus supply of carbon dioxide in areas of colder climate maritime conditions proposed by Kammarlund & according to point G below. However, the relation to Keen (Fig. 7 of App. III). latitude can be neglected in the present study.

5.1.2 Organic carbon isotope data C. Marine influence

The most significant feature of the stable isotope Due to kinetic fractionation associated with several data compiled in the present study is probably the reaction steps within photosynthesis, plant material is pronounced depletion of "C in bulk organic material depleted in '3C with respect to the carbon reservoir across the Pleistocene-Holocene boundary (Fig. 2 of utilized (see Deines 1980 for a review). Marine App. I). This trend was recorded at both the studied plants are generally enriched in I3C compared to la- sites in southernmost Sweden (App. II and III), and custrine plants. However, this factor can be ignored was verified and quantified by means of an extensive in the present study since the sediments of all the compilation of 8'3C data from southern Sweden. A 13 sequences included were deposited in pure lacustrine distinct depletion of C in lake sediment organic basins. material at c. 10,000 BP has also been recorded in Denmark (Noe-Nygaard 1994) and in Britain (Harkness & Walker 1991). This study (Hammarlund D. Aquatic vegetation 1993; App. I) includes a discussion of possible causes of the trend. Furthermore, Håkansson (1985) A change from production of mainly submerged proposed a number of processes as responsible for macrophytes during the Younger Dryas stadial to I3 the C depletion, which has earlier been recorded at mainly plankton during the early Holocene, was eight localities in southern Sweden (Håkansson proposed by Håkansson (1985) as a major cause of 1986). However, the complexity of the problem, the depletion of I3C. This hypothesis was based on probably involving several contributing processes comparisons of carbon isotope data obtained on plant mainly as a result of the general climatic wanning material and net plankton respectively. However, the around 10,000 BP, qualifies for a more detailed relative enrichment of I3C recorded for macrophytes analysis of the topic. The following compilation of is most important in hard-wa'.er lakes (Stuiver 1975; processes is more comprehensive and the conclusions Håkansson unpubl. data). The lakes included in the have been partly modified compared to the corre- compilation by Hammarlund (1993) had mainly rela- sponding discussion by Hammarlund (1993). Most of tively soft water, since they were all situated in non- the known factors affecting the carbon isotope calcareous areas. Therefore, this factor can probably composition of DIC and plant material are referred not be attributed to as a main cause of the depletion to, and their respective importance as possible causes of )3C in bulk organic material, although it may to of the observed trend is discussed. some extent have contributed.

11 E. Telmatic and terrestrial vegetation G. Supply of carbon dioxide

As a result of differences in photosynthetic mecha- Decreasing availability of carbon dioxide results in a nisms, considerable differences in carbon isotope decreasing kinetic fractionation during photosynthesis composition have been demonstrated between plants from c. 20%c at excess supply, to a minimum of 6%o following the C, and C4 photosynthetic pathways at limiting conditions (Deines 1980; Hollander & respectively (Bender 1971,O'Leary 1981). C4 plants McKenzie 1991). As a result of temperature varia- are enriched in I3C by 12-14%o as an average com- tions, significant changes in the amount of carbon pared to C, plants, the latter group exhibiting a mean dioxide dissolved in lake water can be assumed to of about -26%o (Deines 1980). Talbot & Livingstone have taken place during the climatically dynamic (1989) in a study of East African lakes, demonstrated Late Weichselian (cf. Kelts & Hsu 1978). At low that sediment sequences characterized by relative concentrations of carbon dioxide several macrophytes enrichment of "C in bulk organic material could be are able to assimilate bicarbonate during photo- correlated with periods of lowered lake levels. The synthesis (Allen & Spence 1981; Fry & Sherr 1984; 5I3C anomalies were caused by expansions of tropi- Raven et al. 1985). The uptake of bicarbonate cal grasses using C4 photosynthesis, on exposed lake usually occurs as a relatively slow and energy- shores during periods of increased dryness. See also demanding diffusion through the cell membrane, Talbot & Johannesen (1992) and Nordt et al. (1994) followed by a conversion to carbon dioxide within and references therein. C4 plants are confined to the cell. This process may result in a diminished areas of tropical and warm temperate climate, pre- carbon isotope fractionation between DIC and or- dominantly arid regions (Foyer 1984), which means ganic material compared to direct assimilation of that this factor can be neglected in northern Europe. dissolved carbon dioxide (cf. Aravena et al. 1992). However, changes in the relative amount of terrest- Deines (1980) states that the 7-9%« enrichment of 13C rial and aquatic plant material may in some cases in bicarbonate compared to carbon dioxide at equi- influence 513C of bulk sediment organic material. librium conditions (Deuser & Degens 1967) is com- LaZerte & Szalados (1982) demonstrated a difference pensated for by a corresponding increase in frac- of c. 8%o between submerged aquatic macrophytes tionation, which means that the carbon isotope and terrestrial C, plant, the macrophytes being composition of the organic material produced is enriched in °C. A similar relation was stated by generally independent of carbon sources. However, Michel et al. (1989) as a cause of changes in SI3C of this assumption probably ignores complicating factors lake sediment organic material. This effect is pro- such as diffusion (see point I. below). Variations in bably partly related to diffusion rates (see point I. the concentrations of inorganic carbon species as a below), and may be more important in hard-water response to changes in lake-water temperature and lakes. pH during the Late Weichselian may to some extent I3 have influenced 5 COTJ data (Talbot 1990; Talbot & Johannesen 1992). F. h'3C of atmospheric carbon dioxide Another aspect is provided by the global varia- tions in carbon dioxide concentration of the atmo- Based on decreasing values of 5I3C in cellulose sphere related to glacial cycles (e.g. Siegenthaler nitrate from ancient wood (Crayton & Epstein 1977; 1991). The increased partial pressure of carbon similar trends were later recorded by Krishnamurthy dioxide at the transition to the Holocene may have & Epstein 1990 and Becker et al. 1991), a depletion caused an increased supply of dissolved carbon of "C in atmospheric carbon dioxide has been pro- dioxide in the lake water, enabling the maintenance posed (Stuiver & Braziunas 1987), accompanying the of a large fractionation during photosynthesis. This 30-40% increase in carbon dioxide concentration re- process may to some extent have contributed to the 13 corded at the last glacial-interglacial transition depletion of C at the transition to the Holocene (cf. (Siegenthaler 1991). However, this hypothesis has Fontugne & Calvert 1992). recently been rejected through studies of plant ma- terial from packrat middens, and of carbon dioxide enclosed in glacier ice. Evidence of a global en- H. Hydrogen ion pumping richment of "C amounting to c. 1 %o around 10,000 BP, was independently presented by Marino et al. At the assimilation of bicarbonate by hydrogen ion (1992) and Leuenberger et al. (1992). Thus, the pumping (McConnaughy 1991), only 50% of the '.-ar- depletion of "C recorded in lake sediment organic bon dioxide produced at the dehydration according to material, probably was instead counteracted by this the following reaction is used for photosynthesis (see process. App. II);

12 2HCO3 + 2H* -> 2CO2 + 2H2O (2) in DIC, and subsequently in new aquatic plant ma- terial (Rau 1978). As a result of increased limnic The remaining 50% of the carbon dioxide is vegetation production and increased accumulation of consumed at the precipitation of calcium carbonate; decaying organic material on the lake bottoms, a substantial increase in this kind of recycling of 2 Ca * + CO2 + CaCOj + 2W (3) carbon most probably occurred at the transition to the Holocene. This process may partly explain the rapid This process may thereby facilitate selective uptake depletion of I3C in organic material at c. 10,000 BP. of I2C compared to assimilation of dissolved carbon dioxide, and hence induce a relative depletion of "C in organic material (Fronval & Jensen 1992). Hydro- L Fermentation and methane production gen ion pumping was probably not an important pro- cess in the soft-water lakes included in the com- Anaerobic degradation of organic material in lake pilation by Hammarlund (1993), but it could have sediments commonly generates methane and carbon contributed to the changes in values of S^C,,,, dioxide, the latter compound being relatively recorded in the two individual sequences from enriched in "C, exhibiting AI3C values around -5%o southernmost Sweden (App. II and III). according to Oana & Deevey (1960). However, bac- terial oxidation of the simultaneously produced methane may result in production of carbon dioxide /. Turbidity with 8I3C values as low as -70%o to -8O%o (Shoell 1980). This complex process, which has been At aquatic photosynthesis, diffusion of carbon discussed by Håkansson (1985), predominantly dioxide through a stagnant boundary layer sur- occurs during periods of high organic productivity, rounding the plant tissues causes a relative and could possibly to some extent have contributed enrichment of I3C in DIC of the boundary layer. This to the decreasing trend around 10,000 BP. effect, which is more pronounced in of stag- nant lakes than in rapidly flowing streams, may result in a relative enrichment of I5C of aquatic plant M. Terrestrial soils material (Osmond et al. 1981). The general im- portance of this process in the present study is As a result of the early Holocene climatic warming, difficult to evaluate, although it probably had no the establishment and increased density of terrestrial l3 major influence on the average values of 5 Cor| vegetation led to the development of stable soils in across the Pleistocene-Holocene transition. the region (H.J.B. Birks 1986; Liedberg Jönsson 1988). Subsequently, increased root respiration and decomposition of humic substances in the soils most J. Organic productivity probably generated an increased supply of carbon dioxide with 5>3C values around -25%o that reached Preferential consumption of >2C at aquatic photo- and lakes (Mook 1980). This process >3 synthesis to a relative enrichment of "C in the may have been important as a source of C-depleted 3 remaining DIC, as demonstrated by enrichments of carbon influencing 5' C of DIC and aquatic vegeta- >3C in limnic carbonates (McKenzie 1985; App. II tion at c. 10,000 BP. and III). This may also result in a relative enrichment of I3C in organic material (Nakai 1972; Stuiver 1975). However, on the assumption of a general in- N. Endogenic carbonate production crease in aquatic organic productivity at the transition to the Holocene, the results of the present study, as At the precipitation of carbonate shells (molluscs, well as the results of Harkness & Walker (1991), ostracods), a withdrawal of I3C may result in a re- clearly contradict with this relationship. lative depletion of "C in lake-water DIC. However, l3 comparisons with the individual records of 5 Cor, included in the present study (App. II and DI), K. Decomposition of organic material indicate that the re-established production of car- bonate shells at the transition to the Holocene is not At the degradation and oxidation of organic material, the main cause of the depletion of "C in organic carbon dioxide with a carbon isotope composition material, although it may have contributed. Within resembling that of reduced organic carbon is released the earlier parts of the Late Weichselian sequences, (Deines 1980), which may lead to a depletion of I3C variations in the amount of carbonate shells produced

13 was probably a more important factor, clearly in- all the bicarbonate is derived from the carbon dioxide fluencing the carbon isotope composition of DIC and responsible for the dissolution process. On assump- aquatic vegetation. tion of a relatively greater importance of this kind of silicate dissolution by reaction with atmospheric carbon dioxide during a few thousand years after the O. Dissolution of carbonates deglaciation (Locke 1986), higher concentrations of bicarbonate with 6"C values around +!%« (see At the dissolution of carbonates according to the App. I) would be expected in the lake waters. This following equilibrium; process may have resulted in a relative enrichment of "C in organic material due to bicarbonate assimi- 2 lation (cf. Hollander & McKenzie 1991) during the CaCO, + CO2 + H2O «=? Ca * + 2HCO, (4) Late Weichselian compared to early Holocene condi- half of the bicarbonate carbon originates from carbon tions (see point G. above). Thereby, this process can be attributed to as a possible contributing cause of dioxide and the other half derives from the dissolved B carbonate. Assuming carbonate 5"C values around the depletion of C in lake sediment organic material ±0%o (Keith & Weber 1964), this process may result at the Pleistocene-Holocene transition. The general in a carbon isotope composition of bicarbonate com- rate of dissolution of silicates was most probably parable to corresponding values at equilibration with considerably higher during the early Holocene. How- atmospheric carbon dioxide via diffusion through the ever, at this stage "C-depleted carbon dioxide from lake-water surface (Deuser & Degens 1967). Atmo- soil processes (see point M. above) can be assumed to have been mainly responsible for the dissolution, spheric carbon dioxide currently exhibits values in L the range -7%c to -8%e (Keeling et al. 1979; Levin et giving rise to a depletion of 'C in DIC and aquatic al. 1987). On the other hand, the carbon dioxide vegetation. These considerations were not clearly responsible for dissolution of carbonates contained in expressed by Hammarlund (1993; App. I). bedrock and soils is usually depleted in I3C due to soil processes, such as root respiration and decay of humic substances (Vogel & Ehhalt 1963). This pro- Q. Diagenesis cess may thus give rise to a depletion of "C in DIC. However, the supply of carbonates must have been In general, diagenetic processes are accompanied by very restricted in the catchments of the lakes in- a release of "C-depleted methane from the organic cluded in the compilation by Hammarlund (1993), material of the sediments, leading to a relative en- since these were all situated in areas of crystalline richment of "C in the residual carbon compounds bedrock. Old carbon from dissolution of carbonates, (Shoell 1980). However, diagenesis is probably a incorporated by submerged macrophytes, may give process of minor importance in this context. rise to erroneous results at (hard- water effect). This process and its possible bearing on the carbon isotope composition of plant material Conclusions has been studied by Aravena et al (1992). To summarize, the observed depletion of "C in lake sediment organic material across the Pleistocene- P. Dissolution of silicates Holocene boundary, amounting to c. 5%e (Fig. 2 of App. I), is probably related to a substantial environ- A substantial etching of silicate minerals contained in mental change, which in tum depends on the climatic glacial deposits, predominantly plagioclase feldspars, wanning. The observed trend can unambiguously be hornblende and epidote, with subsequent production attributed to a general depletion of "C in DIC, as of ultra-fine silicate particles, can be assumed to indicated by synchronous depletions of "C in bulk have occurred soon after the deglaciation (Locke carbonates (Fig. 9 of App. II; Fig 5 of App. III). The 1986; Sverdrup & Warfvinge 1991). This process most important processes contributing to the distinct probably induced a considerable input of silicate shift were probably decomposition of organic ma- particles to the lake waters during the Late Weich- terial and soil formation (points K. and M. above), selian. At the dissolution of silicates, represented by leading to an increased supply of l3C-depleted carbon the following equation (Vogel & Ehhalt 1963); dioxide to the lake water at the transition to the Holocene. Possibly, dissolution of fine-grained sili-

CaAl2(Si04)2 + 2CO2 + H2O -> cate particles (point P. above) could also have con- 2 Ca * + 2HC0, + AI2O3 + 2SiO2 (5) tributed by raising the bicarbonate concentration du-

14 Table 1. Compilation of different factors that may affect the stable carbon isotope composition of bulk organic material in lake sediments, and their respective influence on the observed decline in 8"C values across the Pleistocene-Holocene boundary in southern Sweden.

Process Influence on declining trend of S13^ values at c. 10,000 BP: (see text) Major neg. Minor neg. Irrelevant Minor pos. Major pos.

A. Lake-water temperature B. Latitude X C. Marine deposition X D. Aquatic vegetation X? E. Cj/C4 plant vegetation 13 F. 8 C of atmospheric CO2 G. CO2 supply X? H. Hydrogen ion pumping X? I. Turbidity X J. Vegetation production K. Decomposition (oxidation) X L. Fermentation/CH4 production X? X? M. Soil development X N. Carbonate production X O. Carbonate dissolution X P. Silicate dissolution X? Q Diagenesis X

ring the Late Weichselian. Variations in 5I3C of relations may indicate a mutual relation to changes in atmospheric carbon dioxide and changes in limnic the global carbon cycle, probably involving large- vegetation production (points F. and J. above) are scale changes in complex parameters such as ocean two processes that may have counteracted the re- circulation patterns and the concentration of atmo- corded trend. Though this is a highly complex spheric carbon dioxide through time (Broecker & problem, most other parameters mentioned above Peng 1986; Heinze & Hasselmann 1993; Archer & probably had either none or only restricted influence Maier-Reimer 1994). on the carbon isotope composition of bulk sediment organic material at c. 10,000 BP (Table 1). Interestingly, rapid depletions of I3C of similar 5.1.3 Carbonate carbon isotope data character as in the present study have been lecordcd in organic material extracted from totally different The carbon isotope composition of limnic carbonates media of both terrestrial and marine origin. Leavitt & is mainly a function of the general rate of limnic Danzer (1992), in a study of North American wood photosynthesis (McKenzie 1985). By preferential I3 I2 samples, found a depletion of C amounting to photosynthetic assimilation of CO2, iquatic vege- c. 3%o around 10,000 BP. Furthermore, bulk organic tation production induces a relative enrichment of I3C material in marine sediments from the eastern in DIC, and subsequently in the carbonates produced Mediterranean exhibit cyclic variations in SI3C as a result of displacement of the carbonate equi- related to glacials and interglacials, sapropel librium (Eq. 4). This general assumption seems to be deposited during interglacials being consistently valid only at the first stage of colonization and depleted in >3C (Fontugne & Calvert 1992). The increased blossoming of aquatic vegetation (cf. H.H. difference between the last glacial maximum and the Birks 1980), which is accompanied by distinct en- Holocene reaches 3-4%o. The changes are interpreted richments of I3C in bulk carbonates at all the studied as due to variations in freshwater input, supplying sites. Later on, as the limnic ecosystems evolve and l3C-depleted carbon dioxide produced by terrestrial the vegetation production respond to different en- biological activity to a greater extent during inter- vironmental changes, more complex relations arise, glacial periods. Furthermore, increased partial pres- and other processes affecting SI3C of DIC must be sure of carbon dioxide during interglacials is believed taken into account. These processes, which are to have contributed to the observed pattern. These mainly related to soil formation and decomposition I 15 of organic material, may lead to a relative depletion depends on lake-water temperature. At equilibrium of "C in DIC even in situations with an increasing or conditions and at constant lake-water 5>SO, a tem- generally high vegetation production. Such relations perature coefficient of c. -0.25%o/°C was empirically l3 are demonstrated by decreasing values of 8 CCilrt in demonstrated by Epstein etal. (1951; 1953) resulting combination with increasing productivity as indicated in the following relation, which has been slightly in biostratigraphic data (see section 5.2.6). The modified by Craig (1965); different contributing processes are also reflected by 2 the carbon isotope composition of bulk organic ma- T = 16.9 - 4.2(8C - 8J + O.13(8c - 8W) (6) terial (see App. I and section 5.1.2). 13 At the two Swedish sites distinct depletions of C where T is the water temperature in °C, Se is the in bulk carbonates were recorded shortly after 12,000 5"O of carbon dioxide released at dissolution of the

BP, preceded and followed by relatively constant carbonate with phosphoric acid at 25°C, and 5W is the levels. The biostratigraphic studies provide no evi- 6'8O of carbon dioxide in equilibrium with the sur- dence of decreasing vegetation production at this rounding water. Both isotope notations are related to stage. If anything, the opposite is indicated. The the PDB standard. 13 lowered values of 8 CCMb may be caused, or at least Secondly, the oxygen isotope composition of the l3 affected by, increased recycling of C-depleted surrounding lake water affects 518O of the carbonates carbon dioxide derived from decaying plant remains. precipitated. This factor is controlled by a number of This would be in analogy with the continuous deple- processes related to the hydrological regime of the I3 tions of C in organic material recorded in the range lake, such as sources of surface and groundwater in- c. 12,000 to c. 11,000 BP. However, the occurrence flow and evaporation (cf. von Grafenstein el at. of amorphous carbonate aggregates (Fig 6C of 1992). However, the oxygen isotope composition of App. II) in the sediments deposited after the decrease precipitation water has been considered the most l3 in values of 8 Ccart, at Körslättamossen, may be of important factor, and has been the subject of a importance at the interpretation of the carbon isotope number of detailed studies (Dansgaard 1964; data. The carbonate aggregates were probably formed Yurtsever 1975; Siegenthaler & Oeschger 1980; van during photosynthetic assimilation and subsequent der Straaten & Mook 1983; Rozanski et al. 1992). dehydration of bicarbonate, described as hydrogen The oxygen isotope composition of meteoric water ion pumping (see section 5.1.2; point H). The che- depends on the source of water vapour (5"O of mical reactions involved are described in App. II. ocean water), and the temperature gradient between This type of photosynthesis may be accompanied by the points of evaporation and condensation respec- kinetic fractionation effects responsible for the tively. According to Dansgaard (1964), the fractiona- observed depletion of "C in bulk carbonates. tion caused by the atmospheric circulation, known as At Vanstads mosse a similar relationship between a Rayleigh process, results in a correlation between carbon isotope composition and precipitation of 5"O of precipitation water and mean annual air tem- carbonates may be available. At this site, however, perature (t) of c. M)J0%c/°C: tubular encrustations originating from Chara coral- Una (Kelts & Hsu 1978) were frequently recorded 8"O = 0.7t- 13.' (7) prior to the decrease in values of 5"CCirt, described >3 above. Thus, in this case the depletion of C in bulk Several attempts (Stuiver 1968; Hendy & Wilson carbonates was accompanied by the disappearance of 1968; Dansgaard & Tauber 1969; Lemeille et a!. carbonate encrustations. On the assumption that the 1983; Siegenthaler & Eicher 1986) have been made two types of carbonate particles were formed by to combine the two fractionation processes involved, similar processes, these contradicting results probably which are oppositely correlated to temperature and reject the hypothesis of hydrogen ion pumping as thereby mutually counteracting. The combinations being a source of carbon isotope fractionation. result in a positive correlation between S'*O of Further stable isotope studies of well identified limnic carbonates and mean annual air temperature limnic carbonates are required to solve this problem. in the range 0.25-0.46%e/°C. However, this assump- tion ignores that the fractionation during carbonate precipitation is related to lake-water temperature, and 5.1.4 Carbonate oxygen isotope data not to mean annual air temperature. Taking this im- portant difference into consideration, it becomes Two main factors influencing the oxygen isotope obvious that the two fractionation processes may well composition of limnic carbonates can be dis- operate independently. An increase in lake-water tinguished, both of which are related to climate by temperature could thereby be accompanied by a de- means of fractionation processes. Firstly, at the pletion of "O in limnic carbonates, without the precipitation of calcium carbonate the fractionation influence of a contemporary enrichment of '*O in

16 precipitation and lake water induced by atmospheric HI) constitute the most important parts of the project, circulation changes or other factors that may in- presenting partly new aspects of Late Weichselian fluence Eq. 7. This is explicitly illustrated by the palaeoclimatology in the region. In the following depletions of I8O in bulk carbonates recorded around chronological sections the Late Weichselian climate or shortly before 10,003 BP (Fig 9 of App. II; Fig. 5 history of southernmost Sweden is summarized in of App. DI), which most probably reflect a climatic terms of general climatic development, as well as wanning. These results, which are supported by bio- changes in individual climatic parameters. The con- stratigraphic data, clearly contradict with most other clusions are based mainly on results from stable studies (e.g. Eicher & Siegenthaler 1976; Punning isotope analysis, although they are supported by et al. 1984; Ralska-Jasiewiczowa et al. 1992), in- comparisons with biostratigraphic data and recon- cluding the results from Imiotki (Fig. 2 of App. IV), structions based on other methods. The main palaeo- where an enrichment of "*O in bulk carbonates was climatic implications are summarized in Table 2 (cf. recorded at the Pleistocene-Holocene transition. The Fig. 7 of App. Dl). discrepancy may be caused by a variety of factors, As one of the major objectives of this study is to e.g. seasonal differences in the general distribution of reconstruct climatic development and compare cli- precipitation, and further emphasizes the complexity matic events, some of which were most probably of oxygen isotope studies. time-transgressive between different regions, the ll! The great variations in 8 OCilIb data recorded application of as proposed by Mangerud around 12,400 BP and during the Older Dry as stadial et al. (1974) may lead to misunderstandings. Instead (Fig. 9 of App. II; Fig. 5 of App. Ill), which are the stratigraphic terminology as far as possible is assumed to reflect changes in water balance (i.e. rate related to ages in conventional and uncorrected of evaporation), indicate that hydrological factors of radiocarbon years (BP). The terms local significance may sometimes be more important Older Dryas, Aller0d, and Younger Dryas are only than large-scale changes in atmospheric circulation used in a climatostratigraphic sense to define stadials and mean annual air temperature. As some of the and interstadials respectively (Fig. 5) main factors influencing the rate of evaporation arc air temperature and temperature of the water (Ahrens 1991), the rate of evaporation in an arctic or sub- 5.2.1 Before c. 12,400 BP arctic climate can be assumed to be restricted. How- ever, substantial rates of evaporation have been During the earliest stage of lacustrine deposition near recorded in polar regions (Orvig 1970). Furthermore, or total absence of mollusc-shell fragments (Fig. 6 of a mass-balance study of an exposed ice surface in App. Ill) and a very sparse plant macrofossil assem- Greenland (O. Humlum unpubl. data) shows that blage (Fig. 8 of App. II) indicate unfavourable con- evaporative weight-loss may amount to 2mm water ditions for most higher aquatic organisms. It may be equivalent per day at -3°C, probably as a result of argued that suitable floral and faunal elements were strong winds and a low relative humidity. Evapora- absent due to slow migration rates. However, rapid tion from an open water surface can be assumed to dispersal of aquatic plants at favourable climatic reach even higher values. These results strengthens conditions have been demonstrated by Iversen the hypothesis of high evaporation from the lake- (1973), and soon after the establishment of limnic water surfaces in southern Sweden during parts of vegetation the first molluscs commonly appear the Late Weichselian (see section 5.2), as explanation (Ökland 1990). A generally high content of coarse of the substantial enrichment of '"O recorded in bulk minerogenic material and input of detrital carbonates, carbonates. The annual evaporation/precipitation ratio also points to unfavourable conditions for limnic may have been very high, especially when taking production, characterized by extensive erosion and into account the impact of strong winds (Berglund & high sediment input. At Körslättamossen limnic Rapp 1988; Schlyter 1991), and the much higher in- ostracods were probably established in the lake as solation that prevailed at mid latitudes during the early as around 13,000 BP (Fig. 5 of App. II). Most Late Weichselian compared to present conditions in ostracod species are bottom-dwelling detritus-feeders the Arctic (Broecker & Denton 1989). or filtraters (Enckell 1980), and may thus be less dependent upon the presence of higher aquatic vege- tation. High frequencies of moss remains, probably of limnic origin, were also recorded. The occurrence 5.2 Palaeoclimatic implications of bryophytes is the most probable cause of the for southern Sweden relative enrichment of "C in bulk carbonates re- corded at Körslättamossen (Fig. 9 of App. II), the The two studies carried out in southern Sweden photosynthetic activity leading to an enrichment of I3 (Körslättamossen and Vanstads mosse; App. II and C in DIC (McKenzie 1985). At Vanstads mosse

17 CL1MATOSTRAT1GRAPHY (Mangenid et al. 1974) "C years BP Southern Sweden Northern Poland Preborcal

Chronozone Preboreal Preboreal 10.000 - Y.DVPreboreal transition 10.200 - Younger Dryas 10,400 - Younger Dryas Younger Drya 10,600 - stadia] stadial 10.800 - 11,000 - 11,200 - Allensd Allentd Allentd 11,400 - Chronozone interstadial interstadial 11,600 - 11.800 - Older Dryas Chronozone 12.000 - Older Dryas siadial Older Dryas stadial 12,200 - Boiling Belling interstadial 12.400 - Chronozone Belling 12,600 - interstadial 12,800 siadial 13,000 - conditions

stadial conditions

Fig. 5. Ciimatostratigraphic subdivisions of the Late Weichselian in southern Sweden and northern Poland respectively, inferred from the results of the present study and comparisons with independent data. The subdivisions are matched against the chronostratigraphic scheme proposed by Mangerud et al. (1974). Shaded zones represent insufficiently dated climatic changes.

(Fig. 5 of App. Ill), and possibly also at Kör- 5.2.2 c. 12,400 to c. 12,200 BP slättamossen, input of detrital carbonates is evident from the carbon and oxygen isotope compositions of Considerable changes in most stable isotope para- bulk carbonates in the lowermost parts of the se- meters as well as lithological and biostratigraphic quences (cf. Keith & Weber 1964; Buchardt & data indicate a major environmental shift at c. 12,400 Nielsen 1985). BP. A rapidly increasing limnic productivity is evi- Shortly before 12,400 BP macrophytes and dent from distinct enrichments of "C in bulk car- molluscs established in the lakes and the bonates, and increasing abundances of plant macro- subsequently deposited carbonates can be assumed to fossils originating from aquatic macrophytes. Simul- be exclusively of endogenic origin. However, a taneously the contents of organic and carbonate relative depletion of "C in bulk carbonates indicate carbon of the sediments increase significantly. At this a still restricted photosynthetic activity. Very high stage also the mollusc faunas become more dense values of 5"OCilrt, recorded before c. 12,400 BP are and diverse. These changes are at both sites accom- interpreted to be caused by strong evaporation from panied by a very pronounced depletion of "O in bulk the lake-water surfaces, subsequently leading to carbonates. This is interpreted as a result of a general enrichments of "O in lake water and limnic change towards a more maritime climate, giving rise carbonates (see section 5.1.4). These results suggest to moister summer conditions and a considerable de- very dry summer conditions. An arctic and highly crease in evaporation, probably corresponding to the continental climate with short summers was probably change from dry to more humid conditions proposed prevailing (cf. van Geel & Kolstrup 1978). by Björck & Möller (1987). Compilations of vegeta- Reconstructions based on fossil insects suggest mean tional data (Huntley & Birks 1983) indicate steep July temperatures below 10°C (Lemdahl 1988a). gradients between continental and oceanic climatic

18 regimes during the Late Weichselian (H.J.B. B irks However, the climatically induced environmental 1986). The rapid and substantial climatic change changes, identified by means of stable isotopes, are indicated by oxygen isotope data, may be a response hardly recognized in plant macrofossil and insect to a shift of the geographical position of the Polar data from Körslättamossen (Hammarlund & Lemdahl Front. Shortly before this climatic change an en- 1994; App. II), since these records are relatively richment of '*O in mollusc shell carbonates was poor. Temperature reconstructions based on beetle recorded at Vanstads mosse, which may indicate an analysis exhibit a mean July temperature optimum increase in mean annual air temperature. The estab- (12-14°C) shortly before 12,000 BP (Lemdahl lishment of aquatic vegetation and mollusc faunas 1988a). This is probably not in contradiction with the probably constitutes a reaction to changes in several stable isotope results of the present study, since a different climatic parameters, such as longer vegeta- continental climate with short summers may well tion periods and a general increase in mean summer support relatively high temperatures during a re- temperature. The latter factor is difficult to evaluate stricted period of the year. However, substantial on the basis of insect data from Körslättamossen, but depletions of I3C in bulk carbonates and lowered an increase in mean July temperature from below carbonate contents indicate a shortening of the 10°C to c. 12°C around 12,500 BP has been demon- growth season and probably a decrease in mean strated by studies of other Late Weichselian insect summer water-temperature. records from Skåne (Lemdahl 1988b). Discrepancies of this kind between palaeoclimatic data obtained by different methods can be further illustrated by Late Devensian records from the 5.2.3 c. 12,200 to c. 12,000 BP British Isles. Based on insect data (Atkinson el al. 1987), a Late Devensian climatic optimum was re- This restricted period, the Older Dryas stadial corded at c. 12,700 BP, with a still relatively warm (Björck 1984; Björck & Möller 1987), seems to con- climate prevailing shortly before 12,000 BP. In stitute a return to approximately the same climatic contrast to these results, vegetational and stable conditions as those prevailing shortly before c. carbon isotope data may indicate a restricted period 12,400 BP. The event is clearly delimited by con- of relatively cold climate shortly before 12,000 BP siderable climatic shifts. A distinct decrease in (Pennington 1975; Coope & Pennington 1977; aquatic vegetation production is indicated by pro- Walker & Lowe 1990; Harkness & Walker 1991). nounced depletions of "C in bulk and mollusc-shell These considerations in conjunction with the results carbonates. The productivity decrease is also re- of the present study and other independent data from flected by decreasing contents of carbonate carbon in Sweden (e.g. Björck & Möller 1987), and from the sediments. The mollusc assemblage at Vanstads northwestern as well as Central Europe (van Geel et mosse is characterized by small and thin walled al. 1989; Ponel & Coope 1990; Lotter et al. 1992), shells and a generally lowered species diversity. indicate that the Older Dryas stadial with certainty These conditions were probably caused by a climatic was a period of climatic cooling of great regional change involving shortening of the summer seasons significance. Furthermore, increased wind activity and, correspondingly, longer periods of ice cover. resulting in deposition of eolian sand is evident from Increased soil erosion is indicated by coarse minero- lithostratigraphic studies in Poland, Denmark and genic material (ef. Berglund et al. 1984). A general The Netherlands (Nowaczyk 1986; Kolstrup et al. change towards a more continental climate with 1990; van der Hammen et al. 1967). However, the summer drought is indicated by a substantial en- imprint on different biostratigraphic records is often richment of ISO in bulk carbonates at Körslätta- barely discernable or difficult to interpret (e.g. mossen. On the basis of pollen and insect analysis, Bohnckec/a/. 1987). the Older Dryas stadial in other studies in ths region, has been recognized as a period of increased dryness (van Geel & Kolstrup 1978; Kolstrup 1982; Lemdahl 5.2.4 c. 12,000 to c. 11,000 BP 1988a). Based on geomorphoiogy, lithostratigraphy and biosfratigraphy Björck & Möller (1987) conclude The sediments deposited during this relatively long that this period was characterized by a continental period of time generally reflect interstadial conditions climate with relatively warm summers and cold, re- with a predominant climate characterized by mari- latively snow-free winters. Permafrost probably time influence. An increase in length of the vege- developed in southern Sweden (see section S.I.I). tation period is indicated by high carbonate contents Increased wind activity (Berglund &. Rapp 1988) of the sediments. Relatively rich and diverse assem- could also have been a factor contributing to lake- blages of aquatic plant remains and mollusc shells water evaporation and subsequent enrichment of "O indicate favourable conditions for limnic production. in limnic carbonates. This hypothesis is supported by enrichments of "C

19 in bulk carbonates according to the principles defined cated by the insect data from Köislättamossen (Fig. 7 by McKenzie (1985). Shortly after 12,000 BP, slight of App. II), as well as by other insect records from depletions of "C in bulk carbonates were recorded at Skåne (Lemdahl 1988a; 1991a). An arctic and conti- both sites, although these changes were probably nental climate probably prevailed, although in- related to other processes than productivity (see dependent data suggest a considerable amount of section 5.1.3). General depletions of "C in bulk winter precipitation, at least in the western parts of organic material probably reflect mainly increasing Scandinavia, as indicated by glacier advances in importance of "C-depleted carbon dioxide from de- Norway (Mangerud 1980; 1991) and western Sweden composition of plant remains as a carbon source at (Björck & Digerfeldt 1984; 1991). In southeastern aquatic photosynthesis (increased recycling). At Sweden, however, the climate at this stage may have c. 11,500 BP a minor enrichment of "C in bulk been drier, resulting in a delayed glacier retreat only organic material at Vanstads mosse is accompanied (Kristiansson 1986, Björck et al. 1988). Fossil ice- by an impoverished mollusc fauna. This may reflect wedges, dated to the Younger Dryas stadial, are re- a minor and temporary climatic cooling. latively rare in southernmost Sweden (Johnsson

The 5'*Ocarb data exhibit relatively stable values in 1981), although they do occur (Svensson 1988; 1990) the range -IO%c to -9%c, a slightly lower level than which indicates an arctic climate and the presence of corresponding results from Grange mose in south- permafrost (see section 5.1.1). However, the for- eastern Denmark (Kolstrup & Buchardt 1982). Also mation of periglacial features may to some extent ls the values of 8 Omou (Fig. 9 of App. II) are rela- have been prohibited by winter snow-cover. tively constant during this period, whereas 5'*OoW Prolonged ice-cover seasons and cool summers data (Fig. 5 of App. Ill) exhibit a decreasing trend probably effectively prevented precipitation of limnic from c. 11,300 to c. 11,000 BP. On the basis of the carbonates. This means that carbon isotope compo- appearance of two cold-demanding mollusc species sition of organic material is the only isotope para- at this stage, this is interpreted as a result of a meter available from this period. As a response to gradual change towards a generally colder and more the ceased production of carbonate shells a consider- continental climate well before the onset of the able enrichment of >3C in DIC took place, giving rise n Younger Dryas stadial, involving a successive de- to successively higher values of ?> Ca,t. Low con- pletion of "0 in precipitation. Vegetational data from centrations of siderite (FeCO3) were identified in the southeastern Sweden (Björck & Möller 1987) also sediments deposited during the Younger Dryas indicate a progressive climatic cooling initiated at stadial, giving rise to anomalous isotope results (see c. 11,300 BP (ef. Lemdahl 1988a). The slightly oscil- section 5.1.1). lating appearance of most isotope data-sets during this period may be related to similar features re- corded in Greenland ice cores (Johnsen et al. 1992; 5.2.6 c. 10,200 to c. 9,500 BP Taylor el al. 1993), as a response to slightly unstable climatic conditions (cf. Lemdahl 1988a). Climatic The Younger Dryas-Preboreal transition zone fluctuations are also indicated in biostratigraphic data (c. 10,200 to c. 10,000 BP), represents a substantial from other parts of Europe (e.g. Atkinson et al. environmental change in the limnic ecosystem. Bio- 1987; Wohlfarth et al. 1994). stratigraphic data indicate a very rapid and sub- stantial climatic warming, involving the introduction of an insect fauna adopted to temperate conditions 5.2.5 c. 11,000 to c. 10,200 BP (Fig. 7 of App. II). A diverse aquatic vegetation including thermophilous species is evident from the The Younger Dryas stadial is usually very clearly plant macrofossil record (Fig. 8 of App. II), as well reflected as a lithological anomaly in South Swedish as from the establishment of a diverse mollusc fauna lacustrine sequences. At the two sites investigated the adopted to weed-rich conditions (Fig. 6 of App. III). sediments deposited during this period are charac- This development is also identified in the litho- terized by total absence of calcium carbonate. The stratigraphy of the studied sequences, as a shift shift from clayey calcareous gyttja to silty gyttja clay towards deposition of carbonate-rich sediments. represents a considerable environmental change, Terrestrial plant macrofossil data indicate the involving increased soil erosion (Berglund et al. development of an open woodland dominated by 1984) and unfavourable conditions for precipitation birch, and stabilized soils. Similar climatically of limnic carbonates. The mollusc and ostracod induced environmental changes have been demon- faunas were probably eliminated, a more sparse strated in a number of studies throughout the region aquatic plant macrofossil assemblage was recorded, (e.g. Liedberg Jönsson 1988). A marked and rapid and obligate arctic species characterize the insect climatic amelioration at c. 10,200 BP, involving an record. Mean July temperatures below 10°C are indi- increase in mean July temperature of 3-6°C, is

20 indicated by insect studies in southernmost Sweden combination with an additional process may be (Lemdahl 1991a). The rapidity of the climatic considered, related to the limnic development out- wanning around 10,200 BP, most probably also lined above. Increased recycling of carbon dioxide involving a shift towards generally maritime from decomposition of organic material, as a re- conditions, may well be compared to the abrupt sponse to an increased limnic productivity and an termination of stadia! conditions proposed by studies increased organic carbon content of the sediments at of Greenland ice cores (Dansgaard et al. 1989; this stage, could have contributed to the depletion of Johnsen et al. 1992; Alley et al. 1993), and insect "C (see section 5.1.2). The re-established precipita- studies from the British Isles (Ashwcrth 1972; 1973). tion of limnic carbonates provides additional stable In contrast to the relatively gradual mode of climatic isotope data. The records of S'^C,^ data exhibit change characterizing the initiation of the Younger decreasing trends, probably responding to the same Dryas stadial at c. 11,000 BP, a more abrupt change conditions as 6I3C of organic material, since both from a cold and continental climate to more maritime parameters reflect the carbon isotope composition of conditions probably took place at the Younger Dryas- DIC. A considerable depletion of "O in bulk carbo- Preboreal transition (cf. Goslar et at. 1993). nates was recorded within the Younger Dryas-Pre- Centred around, or shortly before 10,000 BP, very boreal transition zone at Körslättamossen (Fig. 9 ot l3 rapidly declining values of C1>ri were recorded, the App. II), whereas relatively constant values were shifts amounting to 6%o and 8%c respectively at the recorded at Vanstads mosse (Fig. S of App. III). At two sites. The causes behind this important strati- the latter site, however, a decreasing trend follow graphic event have been discussed by Hammarlund shortly after 10,000 BP. These results are not in (1993; App. I). A relative decrease in the supply of accordance with corresponding 6"O data from bicarbonate derived *rom dissolution of ultra-fine Imiotki in Poland (Fig. 2 of App. IV) or from other silicate particles at the transition to the Holocene studies in Central Europe (e.g. Eicher 1987), al- may partly explain the depletion of "C in organic though the depletion of "O in bulk carbonates was material. Furthermore, an increased supply of "C- most probably a result of the general climatic depleted carbon dioxide from soil processes, in wanning outlined above (see section S.I.4).

Table 2. General climatic and environmental conditions in southern Sweden during different parts of the Late Weichselian, inferred from stable isotope, chemical and lithological data supported by biostratigraphy.

Appr. period General climate Local environment (I4C years BP)

-12,400 Arctic, continental, dry. Sparse vegetation. Unstable soils, probably permafrost.

12,400-12,200 Subarctic, maritime, humid. Increased production of aquatic vegetation. Soil formation.

12,200-12,000 Subarctic, continental, dry, short Decreased production of aquatic vegetation. summers. Soil disturbance, probably permafrost.

12,000-11,000 Subarctic, maritime. Possible oscillation Increased production of aquatic vegetation. ate. 11,500 BP. Soil formation. Gradual transition from c. 11,300 BP.

11,000-10,200 Arctic-subarctic, continental, short Soil disturbance, permafrost. summers. Rapid transition at c. 10,200 BP.

10,200-9,500 Subarctic-temperate. Rapid vegetational development. Possible oscillation at c. 10,000 BP. Formation of stable soils.

21 The climatic amelioration at the transition to the culation changes in Continental Europe during the Holocene may have been interrupted by a short phase Late Weichselian (cf. Siegenthaler et al. 1984). of stagnation shortly after 10,000 BP (cf. van Geel & In contrast to the continuous lake-marl sequences Kolstrup 1978), indicated by the presence of siderite recorded in Continental Europe, distinct lithostrati- and a cold demanding mollusc species at this stage graphic variations characterize Late Weichselian at Vanstads mosse (App. III). lacustrine sequences from southern Sweden, the sedi- ments deposited during the Younger Dryas stadial being completely devoid of calcium carbonate. These 5.3 Comparisons with the Polish important differences suggest that the climatic changes responsible for the stratigraphic anomalies study on the continent, were more pronounced in Scandi- navia. According to a general-circulation-model study The lacustrine sequence investigated at Imiolki in presented by Rind etal. (1986), the decrease in mean Poland (Fig. 2 of App. IV) differs from the sequen- annual air temperature during the Younger Dryas ces studied in southern Sweden by exhibiting a con- stadial was significantly greater in the North Atlantic tinuous deposition of limnic carbonates throughout Region than in Middle Europe, partly due to the in- the Luie Weichselian. However, the Younger Dry as fluence of lowered sea-surface temperatures. Further- stadial is clearly identified in the Iithostratigraphy as more, the relative proximity to the retreating inland a slight lowering of the carbonate content, probably ice probably induced persistent anti-cyclonal circu- reflecting a climatic cooling involving less favourable lation patterns (cf. Kutzbach & Webb 1991), which conditions for precipitation of limnic carbonates. A may have contributed to the severity of the Younger climatically induced environmental change is also Dryas cooling in southern Sweden. The climatic indicated by pollen data (A. Wawrzyniak-Gtuszak change probably involved prolonged ice-cover sea- unpubl. data). A slight increase in the non-arboreal sons cool summers and the total cessation of limnic pollen (NAP) frequency is accompanied by lowered carbonate production. According to oxygen isotope frequencies of Betula and a corresponding increase data, the phases of increased evaporation during parts in Pinus. A similar decrease in carbonate content of the early Late Weichselian in southern Sweden, shortly before 12,000 BP most probably represents have no counterparts in Poland, which may illustrate the Older Dryas stadial. The pollen assemblage from the difference between a genuine arctic climate in this part of the sequence does not, however, point to Scandinavia and more subarctic conditions in the a restricted event of stadial conditions, as it is still northern parts of the European Continent (see section characterized by continuously high NAP values. 5.1.4). This difference is also evident from recon- The two parts of the sequence characterized by structions of mean July temperatures based on lowered carbonate contents, both exhibit relative palaeoentomological studies (Lemdahl 1988a; depletions of '"0 in bulk carbonates, most probably 1991b), and may correspond to the climatic gradient as a response to climatically induced changes in maintained by the position of the Polar Front, as atmospheric circulation. These oscillations may be discussed by H.J.B. Birks (1986). correlated with similar features in Switzerland After an initial enrichment of 13C in bulk carbo- presented by Lotter et al. (1992). Lowered tempera- nates as a result of the establishment of aquatic tures at the condensation of precipitation probably vegetation, lower values were recorded in the Imiolki I8 l3 resulted in a depletion of O in lake water and sub- sequence. Also the 5 Cor( data exhibit a persistent sequently in limnic carbonates (see section 5.1.4). A decreasing trend at this stage, probably mainly re- decrease in mean July temperature of at least 2-3°C flecting increasing recycling of "C-depleted carbon during the Younger Dryas stadial is evident from dioxide. This development is principally the same as insect studies carried out at two sites in Central in southern Sweden, although it probably took place Poland (Lemdahl 1991b). However, the Older Dryas as early as around 13,000 BP, whereas the corre- stadia) is not identified in insect data (see section sponding depletions of "C at the two Swedish sites 5.2.3). Similar depletions of I8O in bulk carbonates were initiated after 12,000 BP. This means that the during the Younger Dryas stadial have been recorded development of aquatic vegetation in southern in lake-sediment sequences from other parts of Sweden substantially lagged behind the conditions in Poland (Rozanski 1987; Rozanski et al. 1988; Continental Europe (Fig. 5), most probably as a Ralska-Jasiewiczowa et al. 1992), as well as from result of climatic differences (cf. Lemdahl 1991b; northeastern Germany (Pachur & Röper 1984), Cen- 1991c). Striking fluctuations characterize all isotope tral Europe (e.g. Eicher 1987), and the Baltic Region parameters recorded in the sediments deposited (Punning el al. 1984). The parallel trends demon- during the Aller0d interstadial (c. 12,000 to c. 11,000 l3 strate the regional significance of atmospheric cir- BP). Especially the S Cnr( data exhibits considerable

22 oscillations which may be caused by carbon isotope cies are also good indicators of past temperature variations in DIC, related to dissolution of carbonates conditions. Temperature quantifications based on and changes in aquatic vegetation (L. Kubiak- beetle analysis are useful at the interpretation of Martens unpubl. data). This points to a relatively oxygen isotope data, although they partly reflect unstable climate (cf. Wohlfarth et al. 1994), which is climatic parameters of different character than those also indicated by stable isotope and biostratigraphic responsible for the stable isotope variations. data from southern Sweden (App. II and ID). Knowledge about the preferred season of formation In contrary to the studies carried out in southern of limnic carbonates and shells of different aquatic Sweden, a relatively clear correlation between stable organisms is of vital importance when comparing carbon and oxygen isotope data obtained on bulk different data-sets. carbonates was recorded at Imioiki in Poland. Accor- On the other hand, the stable carbon isotope ding to Talbot (1990) this may indicate that the composition of bulk organic material, as well as carbonates were precipitated in a hydrologically individual plant taxa in lacustrine sequences may closed lake, or at least in a basin characterized by become an important stratigraphic tool in the future. relatively long water residence-. The present Additional continuous records from the Late Weich- Lake Lednica (Fig. 4) occupies an area which is selian can easily be obtained from different parts of 15-20 times as large as the former lakes at Körslätta- Scandinavia, to evaluate the significance of any mossen and Vanstads mosse. The differences in total secular changes. The decline in 8'3C of bulk sedi- volume were probably even more pronounced. ment organic material at the Pleistocene-Holocene transition is most probably a feature of great regional importance. 5.4 Potential of the method in Future studies should be accompanied by detailed southern Sweden analyses of mineralogy and petrography of the sedi- ments in order to obtain additional information about Continuous Late Weichselian lake-marl or carbonate- sediment composition and origin of the different shell sequences probably cannot be found in Sweden, constituents. This type of data, as well as stable which means that studies based on stable isotope isotope records obtained on well identified carbonates analysis of limnic carbonates will be seriously and macroscopic plant remains, would facilitate handicapped by discontinuous data. However, close- interpretation and provide more precise palaeo- interval sampling of selected carbonate-rich se- climatic information. quences, deposited during parts of the Late Weich- On a global scale, stable isotope analysis of lake selian and the early Holocene, may provide important sediments will probably withhold its position in the information on climatically induced environmental future, as an important and widely used method for changes. reconstructions of environmental changes related to Stable isotope data should always be accompanied the glacial cycles. Detailed studies may provide and supported by biostratigraphic investigations, as additional information on how different complex there are sometimes several possible interpretations parameters, such as the terrestrial biosphere and the available. The present study shows that plant macro- atmospheric circulation, react upon global climatic fossil data greatly facilitates the interpretation of changes in the past, as well as in the future. This carbon isotope results. Indirect information on aqua- type of data is probably of crucial importance e.g. tic vegetation can also be provided by studies of the for the understanding of the rapid glacial termi- mollusc assemblage. Furthermore, some mollusc spe- nations (cf. Broecker & Demon 1989).

23 Acknowledgements

I would like to express my sincere gratitude to the of my studies. I am also indebted to Ulrich following persons, who all in some way have contri- Siegenthaler, Bern, Magdalena Ralska-Jasiewiczowa, buted to the realization of this work; Krakow, and Antje Schwalb, Neuchätel, for stimu- Björn E. Berglund, head of the department of lating and constructive discussions. Kazimierz Quaternary Geology, has provided excellent working Tobolski, Poznari, provided generous fieldwork facili- facilities and a very qualified and up-to-date ties in Poland and contributed to the realization of research-education programme during my years of the Imiotki study. studies. He also promoted a stimulating and friendly Siv Olsson kindly assisted at the chemical ana- atmosphere at the department. He critically read parts lyses and helped me with various problems related to of the manuscript. sediment chemistry. Torben Fronval, Niels Bo Svante Björck, my main supervisor, has always Jensen, and Ole Humlum, Copenhagen, Ingrid U. been an important source of inspiration and has given Olsson, Uppsala, Lars Olof Björn, Gunnar Digerfeldt, me valuable scientific support. He has shown great Inger Ericsson, and Göran Skog are all acknow- interest in my work and he critically read a large ledged for interesting discussions and literature number of manuscript versions. references. Geoffrey Lemdahl, my other supervisor and close Britt Nyberg completed some of the drawings, coopcrator, provided important help and support, as and Tomas Nihlén made the photographs and copy well as valuable comments on various manuscripts. proofs. Jonas Ising, Thomas Persson, Jonas Ekström, Bj0rn Buchardt, Copenhagen, has been a person and Per Möller assisted with various computer prob- of invaluable importance for my thesis work. He in- lems and editorial handling of parts of the thesis. troduced me to the world of stable isotopes by nearly Mats Regnell was my sympathetic room mate for adopting me as a graduate student of his own. He most of the time. provided laboratory facilities for my analytical work I would also like to thank all other friends and and helped me by means of stimulating discussions colleagues in Lund and Copenhagen for surrounding on various problems related to stable isotope theory. my studies with a pleasant and enjoyable atmosphere. He also suggested important improvements of the Finally I want to thank my family, Elisabet and manuscript. Agnes, for great patience, moral support, and Birte Wanning and Inge Juul, Copenhagen, kindly excellent service at home during times of hard work. and carefully introduced me to the different ana- Especially the last months involved many lonely lytical techniques. Owe Gustafsson, Gothenburg and nights and week-ends for them. I am also greatly Rune S0raas, Bergen carried out the stable isotope indebted to my parents, Olle and Britt, and my analyses which I was unable to do on my own. parents-in-law, Lars and Inger, for invaluable support Göran Possnert, Uppsala, and Maurice Arnold, Gif- in many ways. sur-Yvette, carried out the radiocarbon datings. This project was financially supported by the Sören Håkansson has shown great interest in my Swedish Natural Science Research Council (grant no. work, especially concerning the many problems rela- G-GU 4637-305/310 to S. Björck), Knut & Alice ted to the stable carbon isotope composition of Wallenbergs Stiftelse, Kungliga Fysiografiska Säll- organic material. He also critically read an earlier skapet, and Lunds Geologiska Fältklubb. I held a version of the manuscript, and suggested valuable Faculty research scholarship from July 1989 to improvements. January 1993, and a graduate position at the Depart- Ulrich Eicher, Bern, generously introduced me to ment from February 1993. This thesis is a contri- the basics of stable isotope analysis at an early stage bution to IGCP 253 (Termination of the Pleistocene).

24 References

Ahrens, CD. 1991: Meteorology today. West, St Berglund, BE., Lemdahl, G., Liedberg Jönsson, B. Paul. 575pp. & Persson, T. 1984: Biotic response to climatic Allen, E.D. & Spence, D.H.N. 1981: The differential changes during the time span 13,000-10,000 BP - ability of aquatic plants to utilize the inorganic A case study from SW Sweden. In Mörner, N.-A. carbon supply in fresh waters. The New Phytolo- & Karlen, W. (eds), Climatic Changes on a Year- gist 87, 269-283. ly to Millenial Basis, 25-35. Reidel, Dordrecht. Alley, R.B., Meese, D.A., Shuman, C.A., Gow, A.J., Berglund, B.E. & Malmer, N. 1971: Soil conditions Taylor, K.C., Grootes, P.M., White, J.W.C., Ram, and Late Glacial stratigraphy. Geologiska För- M., Waddington E.D., Mayewski P.A. & Zielin- eningens i Stockholm Förhandlingar 93,575-586. ski, G.A. 1993: Abrupt increase in Greenland Berglund, B.E. & Mörner, N.-A. 1984: Late Weich- snow accumulation at the end of the Younger selian deglaciation and chronostratigraphy of Dryas event. Nature 362, 527-529. southern Scandinavia: problems and present "state Aravena, R., Warner, B.G., MacDonald, G.M. & of the art". In Mörner, N.-A. & Karlen, W. (eds), Hanf, K.I. 1992: Carbon Isotope Composition of Climatic Changes on a Yearly to Millenial Basis, Lake Sediments in Relation to Lake Productivity 17-24. Reidel, Dordrecht. and Radiocarbon Dating. Quaternary Research Berglund, B.E. & Rapp, A. 1988: Geomorphology, 37, 333-345. climate and vegetation in NW Scania, Sweden, Archer, D. & Maier-Reimer, E. 1994: Effect of deep- during the Late Weichselian. Geographia Polo- sea sedimentary calcite preservation on atmo- nica 55, 13-35.

spheric CO2 concentration. Nature 367, 260-263. Bergsten, H. & Nordberg, K. 1992: Late Weichselian Ashworth, A.C. 1972: A Late-Glacial insect fauna marine stratigraphy of the southern Kattegat, from Red Moss, Lancashire, England. Entomolo- Scandinavia: evidence for drainage of the Baltic gica Scandinavica 5, 211-224. Ice Lake between 12,700 and 10,300 BP. Boreas Ashworth, A.C. 1973: The climatic significance of a 21, 223-252. Late Quaternary insect fauna from Rodbaston Berner, R.A. 1971: Chemical . Hall, Staffordshire, England. Entomologica Scan- McGraw-Hill, New York. 240pp. dinavica 4, 191-205. Birks, H.H. 1980: Plant macrofossils in Quaternary Atkinson, T.C., Briffa, K.R. & Coope, G.R. 1987: lake sediments. Ergebnisse der Limnologie 15, Seasonal temperatures in Britain during the past 1-60. 22,000 years, reconstructed using beetle remains. Birks, H.J.B. 1986: Late-Quaternary biotic changes Nature 325, 587-593. in terrestrial and lacustrine environments, with Bahnson, H. 1968: Kolorimetriske bestemmelser af particular reference to north-west Europe. In humificeringstal i h0jmoset0rv fra Fugls0 mose på Berglund, B.E. (ed.), Handbook of Holocene Djursland. Meddelelser fra Dansk Geologisk För- Palaeoecology and Palaeohydrology, 3-65. Wiley, ening 18, 55-63. Chichester. Becker, B., Kromer, B. & Trimborn, P. 1991: A Björck, S. 1979: Late Weichselian stratigraphy of stable-isotope tree-ring timescale of the Late Blekinge, SE Sweden, and water level changes in Glacial/Holocene boundary. Nature 353,647-649. the Baltic Ice Lake. Thesis 7, Department of Bender, M.M. 1971: Variations in the "C/"C ratios Quaternary Geology, Lund University. 248pp. of plants in relation to the pathway of photosyn- Björck, S. 1981: A stratigraphic study of Late thetic carbon dioxide fixation. Phytochemistry 10, Weichselian deglaciation, shore displacement and 1239-1243. vegetation history in south-eastern Sweden. Bengteson, L. & Enell, M. 1986: Chemical analysis. Fossils and Strata 14, 1-93. In Berglund, B.E. (ed.), Handbook of Holocene Björck, S. 1984: Bio- and chronostratigraphic sig- Palaeoecology and Palaeohydrology, 423-451. nificance of the Older Drys Chronozone- on the Wiley, Chichester. basis of new radiocarbon dates. Geologiska För- Berglund, B.E. 1971: Late-Glacial stratigraphy and eningens i Stockholm Förhandlingar 106, 81-91. chronology in South Sweden in the light of bio- Björck, S., Berglund, B.E. & Digerfeldt, G. 1988: stratigraphic studies on Mt Kullen, Scania. Geo- New aspects on the deglaciation chronology of logiska Föreningens i Stockholm Förhandlingar South Sweden. Geographia Polonica 55, 37-49. 93, 11-45. Berglund, B.E. 1979: The deglaciation of southern Sweden 13,500-10,000 BP. Boreas 8, 89-118.

25 Björck, S. & Digerfeldt, G. 1984: Climatic changes Crayton, J.Y. & Epstein, S. 1977: Climatic implica- at Pleistocene/Holocene boundary in the Middle tions of D/H ratios of meteoric water over North Swedish end moraine zone, mainly inferred from America (9,500 - 22,000 B.P.) as inferred from stratigraphic indications. In Mörner, N.-A. & ancient wood cellulose C-H hydrogen. Earth and Karlen, W. (eds), Climatic Changes on a Yearly Planetary Science Letters 34, 333-350. to Millenial Basis, 37-56. Reidel, Dordrecht. Curtis, CD., Coleman, ML. & Love, L.G. 1986: 667pp. Pore water evolution during sediment burial from Björck, S. & Digerfeldt, G. 1991: Aller0d/Younger isotopic and mineral chemistry of calcite, dolo- Dryas sea level changes in southwestern Sweden mite and siderite concretions. Geochimica et and their relation to the Baltic Ice Lake deve- Cosmochimica Acta 50, 2321-2334. lopment. Boreas 20, 115-133. Dansgaard, W. 1964: Stable isotopes in precipitation. Björck, S. & Möller, P. 1987: Late Weichselian Tellus 16, 436-468. environmental history in southeastern Sweden Dansgaard, W. & Tauber, H. 1969: Glacier oxygen- during the deglaciation of the Scandinavian Ice 18 content and Pleistocene ocean temperatures. Sheet. Quaternary Research 28, 1-37. Science 166, 499-502. Bohncke, S.J.P., Vandenberghe, J., Coope, G.R. & Dansgaard, W., White, J.C. & Johnsen, S.J. 1989: Reiling, R. 1987: Geomorphology and palaeo- The abrupt termination of the Younger Dryas ecology of the Mark Valley (southern Nether- climate event. Nature 339, 532-534. lands): palaeoecology, palaeohydrology and Deines, P. 1980: The isotopic composition of re- climate during the Weichselian Late Glacial. duced organic carbon. In Fritz, P. and Fontes, J.C. Boreas 16, 69-85. (eds), Handbook of environmental isotope geo- Broecker, W.S. & Denton, G.H. 1989: The role of chemistry, Vol. 1A: The terrestrial environment, ocean-atmosphere reorganizations in glacial 329-406. Elsevier, Amsterdam. cycles. Geochimica et Cosmochimica Acta 53, Deuser, W.G. & Degens, E.T. 1967: Carbon isotope

2645-2501. fractionation in the system CO2(gas)-CO,(aque- Broecker, W.S. & Peng, T.-H. 1986: Carbon cycle: ous)-HCO3 (aqueous). Nature 215, 1033-1035. 1985, Glacial to interglacial changes in the Eicher, U. 1987: Die Spätglazialen sowie die friih- operation of the global carbon cycle. Radiocarbon postglazialen Klimaverhältnisse im Bereiche der 28, 309-327. Alpen: Sauerstoffisotopenkurven kalkhaltiger Buchardt, B. & Fritz, P. 1980: Environmental iso- Sedimente. Geographica Helvetica 2, 99-104. topes as environmental and climatological indi- Eicher, U. & Siegenthaler, U. 1976: Palynological cators. In Fritz, P. & Fontes, J.C. (eds). Handbook and oxygen isotope investigations on Late-Glacial of environmental isotope , Vol, 1A: sediment cores from Swiss lakes. Boreas 5, The terrestrial environment, 473-504. Elsevier, 109-117. Amsterdam. Enckell, P.H. 1980: Kräftdjur. Signum, Lund. 685pp. Buchardt, B. & Nielsen, AT. 1985: Carbon and oxy- Engström, D.R. & Wright, Jr HE. 1984: Chemical gen isotope composition of Cambro-Silurian stratigraphy of lake sediments as a record of limestone and anthraconite from Bornholm: Evi- environmental change. In Haworth, E.Y. & Lund, dence for deep burial Diagenesis. Bulletin of the J.W.G. (eds), Lake sediments and environmental Geological Society of Denmark 33, 415-435. history, 11-68. Leicester University Press. Coope, G.R. & Pennington, W. 1977: The Winder- Epstein, S., Buchsbaum, R, Lowenstam, H.A. & mere Interstadial of the Late Devensian. Philo- Urey, H.C., 1951: Carbonate-water isotopic tem- sophical Transactions of the Royal Society of perature scale. Geological Society of America London B 280, 337-339. Bulletin 62, 417-425. Craig, H. 1957: Isotopic standards for carbon and Epstein, S., Buchsbaum, R, Lowenstam, H.A. & oxygen and correction factors for mass-spectro- Urey, H.C., 1953: Revised carbonate/water iso- metric analysis of carbon dioxide. Geochimica et topic temperature scale. Geological Society of Cosmochimica Acta 12, 133-149. America Bulletin 64, 1315-1326. Craig, H. 1965: The measurement of oxygen isotope Fontugne, M.R. & Calvert, S.E. 1992: Late Pleisto- palaeotemperatures. In Tongiorgi, E. (ed.), Stable cene variability of the carbon isotopic compo- isotopes in oceanographic studies and palaeo- sition of organic matter in the eastern Mediter- temperatures, 161-182. Consiglio Nazionale delle ranean: monitor of changes in carbon sources and

Ricerche, Pisa. atmospheric CO2 composition. Paleoceano- graphy 7, 1-20.

26 Foyer, C.H. 1984: Photosynthesis, Wiley, New York. Harkness, D.D. & Walker, M.J.C. 1991: The Deven- 219pp. sian Lateglacial Carbon Isotope Record from Fronval, T. & Jensen, N.B. 1992: En stabilisotopisk Llanilid, South Wales. Quaternary Proceedings 1, unders0gelse af recente og Flandriske sedimenter 35-43. fra Arres0. M.Sc. Thesis, Geological Institute, Heinze, C. & Hasselmann, K. 1993: Inverse multi- University of Copenhagen. 141pp. parameter modeling of palaeoclimate cartn n cycle Fry, B. & Sherr, E.B. 1984: 6"C measurements as indices. Quaternary Research 40, 281-2%. indicators of carbon flow in marine and fresh- Hendy, C.H. & Wilson, AT. 1968: Paleoclimatic water ecosystems. Contributions in Marine data from speleothems. Nature 219, 48-51.

Sciences 27, 13-47. Hollander, D.J. & McKenzie, J.A. 1991: CO2 control van Geel, B., Coope, G.R. & van der Hammen, T. on carbon-isotope fractionation during aqueous 1989: Palaeoecology and stratigraphy of the photosynthesis: A paleo-pCO, barometer. Geology Lateglacial type section at Usselo (The 19, 929-932. Netherlands). Review of Palaeobotany and Huntley, B. & Birks, H.J.B. 1983: An Atlas of Past Palynology 60, 25-129. and Present Pollen Maps for Europe: 0-13000 van Geel, B. & Kolstrup, E. 1978: Tentative years ago. Cambridge University Press, Cam- explanation of the Late Glacial and early bridge. Holocene climatic changes in North-Western 'versen, J. 1973: The Development of Denmark's Europe. Geologie en Mijnbouw 57, 87-89. Nature since the Last Glacial. Geological Survey Goslar, T., Kuc, T., Ralska-Jasiewiczowa, M, of Denmark V 7C, 1-126. Rozanski, K., Arnold, M., Bard, E., van Geel, B., Johnsen, S.J., Clausen, H.B., Dansgaard, W., Fuhrer. Pazdur, M.F., Szeroczynska, K., Wicik, B., K., Gundestrup, N., Hammer, C.U., Iversen, P., Wieckowski, K. & Walanus, A. 1993: High- Jouzel, J., Stauffer, B. & Steffensen, J.P. 1992: resolution lacustrine record of the Late Irregular glacial interstadials recorded in a new Glacial/Holocene transition in Central Europe. Greenland . Nature 359, 311-313. Quaternary Science Reviews 12, 287-294. Johnsson, G. 1981: Fossil patterned ground in von Grafenstein, U., Erlenkeuser, H., Miiller, J. & southern Sweden. Geologiska Föreningens i Kleinmann-Eisenmann, A. 1992: Oxygen isotope Stockholm Förhandlingar 103, 79-89. records of benthic ostracods in Bavarian lake Kasprzak, S. & Kozarski, S. 1988: Ice-lobe contact sediments. Naturwissenschaften 79, 145-152. sedimentary scarps in marginal zones of the major Håkansson, S. 1985: A review of various factors Vistulian ice-sheet positions, West-central Poland. influencing the stable carbon isotope ratio of 1989 Quaestiones Geographicae, Special issue 2, organic lake sediments by the change from 69-81. Glacial to Post-Glacial environmental conditions. Keeling, CD., Mook, W.G., & Tans, P.P. 1979: Re- Quaternary Science Reviews 4, 135-146. cent trends in the 13C/I2C ratio of atmospheric Håkansson, S. 1986: A marked change in the stable carbon dioxide. Nature 277, 121-123. carbon isotope ratio at the Pleistocene-Holocene Keith, M.L. & Weber, J.N. 1964: Carbon and oxy- boundary in southern Sweden. Geologiska För- gen isotopic composition of selected limestones eningens i Stockholm Förhandlingar 108, and fossils. Geochimica et Cosmochimica Ada 155-158. 28, 1787-1816. Hammarlund, D. 1993: Evidence of a distinct 813C Kelts, K. & Hsu, K.J. 1978: Freshwater carbonate decline in organic lake sediments at the sedimentation. In Lerman, A. (ed.), Lakes - Pleistocene-Holocene transition in southern Chemistry, Geology, Physics, 295-323. Springer, Sweden. Boreas 22, 236-243. New York. Hammarlund, D. & Lemdahl, G. 1994: A late Kolstrup, E. 1982: Late-Glacial pollen diagrams from Weichselian stable isotope stratigraphy compared Hjelm and Draved mose (Denmark) with a sug- with biostratigraphical data: a case study from gestion of the possibility of drought during the southern Sweden. Journal of Quaternary Earlier Dryas. Review of Palaeobotany and Paly- Science 9, 13-31. nology 36, 35-63. van der Hammen, T., Maarleveld, G.C., Vogel, J.C. Kolstrup, E. 1985: A fossil frost mound of Late & Zagwijn, W.H. 1967: Stratigraphy, climatic Dryas age in middle Jutland (Denmark) Boreas succession and radiocarbon dating of the Last 14, 217-223. Glacial in The Netherlands. Geologie en Mijn- Kolstrup, E. & Buchardt, B. 1982: A pollen ana- bouw 46, 79-95. lytical investigation supported by an "O-record of a Late-Glacial lake deposit at Grange (Denmark). Review of Palaeobotany and Palynology 36, 205-230. Kolstrup, E., Griin, R., Mejdahl, V., Packman, S.C. Lemdahl, G. 1991b: (Fossil insects and palaeo- & Wintle, A.G. 1990: Stratigraphy and thermo- ecology - an example from a Lateglacial site in of Late Glacial cover sand in the Lednica region.) In Tobolski, K. (ed.), Wstep Denmark. Journal of Quaternary Science 5, do paleoekologii Lednickiego Parku Krajobrazo- 207-224. wego (Introduction to Palaeoecology of the Kozarski, S. 1980: An outline of the Vistulian Lednica Landscape Park). Biblioteka Studiéw stratigraphy and chronology of the Great Poland Lednickich, Poznan, 129-141. Lowland. Quaternary Studies in Poland 2,21-35. Lemdahl, G. 1991c: Late Vistulian insect assem- Krishnamurthy, R.V. & Epstein, S. 1990: Glacial- blages from Zabinko, western Poland. Boreas 20, interglacial excursion in the concentration of 1X-11. I2 atmospheric CO2: effect in the "C/ C ratio in Lemeille, E., Letolle, R., Meliere, F. & Olive, P. wood cellulose. Tellus 42B, 423-434. 1983: Isotope and other physico-chemical para- Kristiansson, J. 1986: The ice-recession in the meters of palaeolake carbonates. In Palaeocli- southeastern part of Sweden. A varve-chrono- mates and palaeowaters: A collection of environ- logical time scale for the latest part of the Late mental isotope studies, 135-150. International Weichselian. Report 7, Department of Quaternary Atomic Energy Agency, Vienna. Research, University of Stockholm. 132pp. Leuenberger, M., Siegenthaler, U. & Langway, C.C. Kutzbach, J.E. & Webb ffl, T. 1991: Late Qua- 1992: Carbon isotope composition of atmospheric

ternary climatic and vegetational change in CO2 during the last ice age from an Antarctic ice eastern North America: Concepts, models and core. Nature 357, 488-490. data. In Shane, L.C.K. & Cushing, EJ. (eds), Levin, I., Kramer, B., Wagenback, D. & Munnich, Quaternary Landscapes, 175-217. University of K.O. 1987: Carbon isotope measurements of

Minnesota Press. atmospheric CO2 at a coastal station in Antarctica. Lagerlund, E. 1980: Litostratigrafisk indelning av Tellus 39B, 89-95. Västskånes Pleistocen och en ny glaciationsmodcll Liedberg Jönsson, B. 1988: The Late Weichselian för Weichsel. Report 21, Department of Quater- macrofossil flora in western Skåne, southern nary Geology, Lund University. 120pp. Sweden. Thesis 24, Department of Quaternary Lagerlund, E. 1987: An alternative Weichselian Geology, Lund University. 12pp. glaciation model, with special reference to the Locke, W.W. 1986: Rates of Hornblende Etching in glacial history of Skåne, South Sweden. Boreas Soils on Glacial Deposits, Baffin Island, Canada. 16, 433-459. In Colman, S.M. & Dethier, D.P. (eds), Rates of Lagerlund, E. & Houmark-Nielsen, M. 1993: Timing Chemical Weathering of Rocks and Minerals, and pattern of the last deglaciation in the Kattegat 129-145. Academic Press, Orlando. region, southwest Scandinavia. Boreas 22, Lotter, A.F., Eicher, U., Siegenthaler, U. & Birks, 337-347. H.J.B. 1992: Late-Glacial climatic oscillations as LaZerte, B.D. & Szalados, J.E. 1982: Stable carbon recorded in Swiss lake sediments. Journal of isotope ratio of submerged freshwater macro- Quaternary Science 7, 187-204. phytes. Limnology and Oceanography 27, Mangerud, J. 1980: Ice-front Variations of Different 413-418 Parts of the Scandinavian Ice Sheet, 13,000- Leavitt, S.W. & Danzer, S.R. 1992: 8"C variations 10,000 BP. In Lowe, J.J., Gray, J.M. & Robinson, in C, plants over the past 50,000 years. Radio- J.E. (eds), Studies in the Lateglacial of North- carbon 54,783-791. West Europe, 23-30. Pergamon Press, Oxford. Lemdahl, G. 1988a: Palaeodimatic and palaeoeco- Mangerud, J. 1991: The Last Interglacial/Glacial logical studies based on subfossil insects from Cycle in Northern Europe. In Shane, L.C.K. & Late Weichselian sediments in southern Sweden. Cushing, E.J. (eds), Quaternary Landscapes, Thesis 22, Department of Quaternary Geology, 38-75. University of Minnesota Press, Minne- Lund University. 12pp. apolis. Lemdahl, G. 1988b: Late Weichselian insect Mangerud, J., Andersen, ST., Berglund, B.E. & assemblages from the Kullen Peninsula, South Donner, J.J. 1974: Quaternary stratigraphy of Sweden. Report 30, Department of Quaternary Norden, a proposal for terminology and classifi- Geology, Lund University. 55pp. cation. Boreas 3, 109-J 28. Lemdahl, G. 1991a: A rapid climatic change at the Marino, B.D., McElroy, M.B., Salawitch, R.J. & end of the Younger Dryas in south Sweden - Spaulding, W.G. 1992: Glacial to interglacial palaeoclimatic and palaeoenvironmental recon- variations in the carbon isotope composition of

structions based on fossil insect assemblages. atmospheric CO2. Nature 357, 461-465. Palaeogeography, Palaeoclimatology, Palaeo- ecology 83, 313-331.

28 McConnaughy, T. 1991: Calcification in Chara O'Leary, M.H. 1981: Carbon isotope fractionation in corallina: CO, hydroxylation generates protons plants. Phytochemistry 20, 553-567. for bicarbonate assimilation. Limnology and Orvig, S. (ed.) 1970: Climates of the Polar Regions, Oceanography 36, 619-628. World Survey of Climatology, Vol. 14. Elsevier McKenzie, J.A. 1985: Carbon isotopes and produc- 1970. tivity in the lacustrine and marine environment. In Osmond, C.B., Valaane, N., Haslam, S.M., Uotila, P. Stutnm, W. (ed.), Chemical processes in lakes, & Roksandic, Z. 1981: Comparisons of 5'3C 99-118. Wiley, New York. Values in Leaves of Aquatic Macrophytes from Michel, F.A., Fritz, P. & Drimmie, R.J., 1989: Evi- Different Habitats in Britain and Finland; some dence of climatic change from oxygen and carbon Implications for Photosynthetic Processes in isotope variations in sediments of a small arctic Aquatic Plants. Oecologia (Berlin) 50, 117-124. lake, Canada. Journal of Quaternary Science 4, Pachur, H.-J. & Röper, H.-P. 1984: Geolimnolo- 201-209. gische Befunde des Berliner Raumes. Berliner Mook, W.G. 1980: Carbon-14 in hydrological stu- Geographische Abhandlungen 36, 37-49. dies. In Fritz, P. & Fontes, J.C. (eds), Handbook Pearson, F.J. & Coplen, T.B. 1978: Stable isotope of environmental isotope geochemistry, Vol. I A: studies of lakes. In Lerman, A. (ed.), Lakes • The terrestrial environment, 49-74. Elsevier, Chemistry, Geology, Physics, 325-339. Springer, Amsterdam. New York. Mook, G.W., Bommerson, J.C. & Staverman, W.H. Pennington, W. 1975: A chronostratigraphic compa- 1974: Carbon isotope fractionation between dis- rison of Late-Weichselian and Late-Devensian solved bicarbonate and gaseous carbon dioxide. subdivisions, illustrated by two radiocarbon-dated Earth and Planetary Science Letters 22, 169-176. profiles from western Britain. Boreas 4, 157-171. Mörner, N.-A. & Wallin B. 1977: A 10,000-year Ponel, P. & Coope, G.R. 1990: Lateglacial and early temperature record from Gotland, Sweden. Flandrian Coleoptera from La Taphanel, Massif Palaeogeography, Palaeoclimatology, Palaeo- Central, France: Climatic and Ecological Implica- ecology 21, 113-138. tions. Journal of Quaternary Science 5, 235-249. Mozley, P.S. & Wersin, P. 1992: Isotopic compo- Punning, J.-M., Martma, T. & Vaikmäe, R. 1984: sition of siderite as an indicator of depositional Light isotope variations in carbonate •xdiments environment. Geology 20, 817-820. and their palaeogeographical value. Zeitschriftftir Nakai, N. 1972: Carbon isotopic variation and the Internationale Mitteilungen 84, 329-336. paleoclimate of sediments from Lake Biwa. Pro- Ralska-Jasiewiczowa, M., van Geel, B., Goslar, T. & ceedings of the Japan Academy 48, 5*6-521. Kuc, T. 1992: The record of the Late Glacial/ Noe-Nygaard, N. 1994: Sedimentary, geochemical Holocene transition in the varved sediments of and ecological evolution of a Lateglacial- Lake Gosciaz, central Poland. Geological Survey Postglacial lacustrine basin: lake-level and of Sweden Ca 81, 257-268. climatic influence on flora, fauna and human Rau, G. 1978: Carbon-13 Depletion in a Subalpine population (Aamosen, Denmark). Fossils and Lake: Carbon Flow Implications. Science 201, Strata (in press). 901-902. Nordt, L.C., Boutton, T.W., Hallmark, C.T. & Raven, J.A., Osborne, B.A. & Johnston, A.M. 1985:

Waters, MR. 1994: Late Quaternary Vegetation Uptake of CO2 by aquatic vegetation. Plant, cell and Climate Changes in Central Texas Based on and environment 8, 417-425. the Isotopic Composition of Organic Carbon. Rind, D., Peteet, D., Broecker, W., Mclntyre, A. & Quaternary Research 41, 109-120. Ruddiman, W. 1986: The impact of cold North Nowaczyk, B. 1986: The age of dunes, their textural Atlantic sea surface temperatures on climate: and structural properties against atmospheric implications for the Younger Dryas cooling. circulation pattern of Poland during the Late Climate Dynamics 1, 3-33. Vistulian and Holocene. Thesis 28, Department of Rozanski, K. 1987: The "O and 13C isotope investi- Geography, University of Poznari. 245pp. gations of carbonate sediments from the Lake Oana, S. & Deevey, E.S. 1960: Carbon 13 in lake Strazym (Brodnica Lake District). Ada Palaeo- waters, and its possible bearing on paleolim- botanica 27, 277-282. nology. American Journal of Science, Bradley Rozanski, K., Araguås-Araguås, L. & Gonfiantini, R. Volume 258-A, 253-272. 1992: Relation Between Long-Term Trends of Ökland, J. 1990: Lakes and Snails: Environment and Oxygen-18 Isotope Composition of Precipitation Gastropoda in 15000 Norwegian lakes, ponds and and Climate Science 258, 981-985. rivers. Universal Book Services/Dr W. Backhuys, Oegstgeest.

29 Rozanski, K., Wcislo, D., Harmata, K., Nory- Svensson, H. 1990: Relict periglacial structures. skiewicz, B. & Ralska-Jasiewiczowa, M. 1988: Occurrences, age and development in different Palynological and isotope studies on carbonate matrices on a coastal plain of southwestern sediments from some Polish lakes - Preliminary Sweden. Geografiska Annaler 72 A, 79-91. results. In Lang, G. & Schliichter, C. (eds), Lake, Sverdrup, H. & Warfvinge, P. 1991: On the geoche- Mire and River Environments during the last mistry of chemical weathering. In Rosén, K. (ed.), 15,000 years, 41-49. Balkema, Rotterdam. Chemical weathering under field conditions, Schlyter, P. 1991: Recent and periglacial wind action 79-118. Report 63, Department of forest soils, in Scania and adjacent areas of S Sweden. Zeit- Swedish University of Agricultural Sciences. schrift fur Geomorphologie N.F. 90, 143-153. Talbot, MR. 1990: A review of the palaeohydro- Shoell, M. 1980: The hydrogen and carbon isotopic logical interpretation of carbon and oxygen composition of methane from natural gases of isotopic ratios in primary lacustrine carbonates. various origin. Geochimica et Cosmochimica Acta Chemical Geology (Isotope Geoscience Section) 44, 649-661. 80, 261-279. Siegenthaler, U. 1991: Glacial-interglacial Talbot, M.R. & Johannesen, T. 1992: A high resolu-

atmospheric CO2 variations. In Bradley, R.S. tion palaeoclimatic record for the last 27,500 (ed.), Global changes of the past, 245-260. years in tropical West Africa frotr. the carbon and UCAR/Office for Interdisciplinary Earth Studies, isotopic composition of lacustrine orga- Boulder. nic matter. Earth and Planetary Science Letters Siegenthaler, U. & Eicher, U. 1986: Stable oxygen 110, 23-37. and carbon isotope analyses. In Berglund, B.E. Talbot, M.R. & Livingstone, D.A. 1989: Hydrogen (ed.), Handbook ofHolocene Palaeoecology and index and carbon isotopes of lacustrine organic Palaeohydrology, 407-422. Wiley, Chichester. matter as lake level indicators. Palaeogeography, Siegenthaler, U., Eicher, U., Oeschger, H. & Palaeoclimatology, Palaeoecology 70, 121-137. Dansgaard, W. 1984: Lake sediments as conti- Taylor, K.C., Lamorey, G.W., Doyle, G.A., Alley, nental 8'"O records from the Glacial/ Post-Glacial G.A., Grootes, P.M., Mayewski, P.A., White, transition. Annals ofGlaciology 5, 149-152. J.W.C. & Barlow, L.K. 1993: The "flickering Siegenthaler, U. & Oeschger, H. 1980: Correlation of switch" of late Pleistocene climate change. Nature "O in precipitation with temperature and altitude. 361, 432-435. Nature 285, 314-317. Vogel, J.C. & Ehhalt, D. 1963: The use of the car- van der Straaten, CM. & Mook, W.G. 1983: Stable bon isotopes in groundwater studies. In Radioiso- isotopic composition and climatic variability. In topes in Hydrology, 385-395. International Atomic Palaeoclimates and palaeowaters: A collection of Energy Agency, Vienna. environmental isotope studies, 53-64. International Walker, MJ.C. & Lowe, J.J. 1990: Reconstructing Atomic Energy Agency, Vienna. the environmental history of the last glacial/ Stuiver, M. 1968: Oxygen-18 content of atmospheric interglacial transition: evidence from the Isle of precipitation during the last 11,000 years in the Skye, Inner Hebrides, Scotland. Quaternary Great Lakes region. Science 162, 994-997. Science Reviews 9, 15-49. Stuiver, M. 1975: Climate versus changes in I3C Wohlfarth, B., Gaillard, M.J., Häberli, W. and Kelts, content of the organic component of lake sedi- K. 1994: Environment and climate in southwes- ments during the Late Quaternary. Quaternary tern Switzerland during the last termination. A Research 5, 251-262. tentative correlation of bio- and lithostrati- Stuiver, M. & Braziunas, T.F. 1987: Tree cellulose graphical records, oxygen isotopes, geomorpholo- I3C/>2C isotope ratios and climatic change. Nature gical and palaeoglaciological data. Quaternary 328, 58-60. Science Reviews 13 (in press).

Svensson, H. 1988: Ice-wedge casts and relict poly- Yurtsever, Y.( 1975: Worldwide survey of stable gonal patterns in Scandinavia. Journal of Qua- isotopes in precipitation. In Isotope Hydrology, ternary Science 3, 57-67. International Atomic Energy Agency, Vienna.

30 LUNDQUA Thesis

Stratigraphy At the Department of Quaternary Geology, Lund University, three series are QUATERNARY- published, named "Thesis", "Report" and "Uppdrag". The "Thesis" scries, contains doctor dissertations; the "Report" series primary material, often of a • Environment 'monographic character; which cannot be published in extenso in ordinary scientific jourmds; the "Uppdrag" series contains selected examples of expert reports, generally in Swedish, which may be of some general interest.-. ' . The "Thesis" and the "Rcpon" r.cncs cover different aspects of Quaternary stratii;r;iphy and environment - methods, lithostraiigraphy, biostratigraphy,' chronology, as well as iheir applicaiions within technical geology, resource .ecology, nature conservancy, etc. Complete lists of publications may he ordered from the Dcpta; iment of Quaternary Geology, Tornavagen 13,' S-223 63 Lund, Sweden. • ' ' Editorial committee: Björn E. Berglund, Svante Björck, Gunnar Digcrfcldt, Christian Hjort, Erik Lagerlund and Per Möller. Techniail editor: Per Möller • •• ,"'••"

- 2. . njclm, L. 1976: Dcglaaationcn av Småländska höglandet, . "; . 16. Adriclsson, L. 1984: Wcichsclian lithostratigraphy and glacial - : '',-'- . , •-. speciellt med avseende på dcglaciationsdynamik, ismäktighet environments in the Vcn-Glumslöv area, southern Sweden. 35 "-;•', ):'•

.' " -och tidsstiillning. (Dcglaciation of the Smaland Highland, with SEK." ••*.&,; ,,• special reference to dcglaciation dynamics, ice thickness and 17. Waldemarson, D. 1986: Weichselian lithostratigraphy, deposi- J", '',.'•• '.

•'-'• -,'chronology). 15SEK.' -•-,'£' ;. '; •• :-V ';.' '•".'•> ';"• . • ' '•-. '"• tional processes and degladation pattern in the southern Vät- - ."'• >. ••, :,3.' ,Göransson, H. 1977: The Plandrian Vcgctational History of - •'.'. tern basin, south Sweden. 40 SEK. ;,;?";--" "•' •."';!•''••"V; f'1^'?}} '..;' .'Southern Östergötland. 15 SEK: >' ,'.•" '"I,,,""'... •'-: -•-' '..'' 18. Hallsdottir, M. 1987: Pollen analytical studies of human influ- -',- •."" ;•-;'•- '4. •••Miller. U. 1977: Pleistocene deposits of the Alnarp Valley, "v>; ' •• ence on vegetation in relation to the Landnam tcphra layer in ;'-'". •''••.; &;' ..:'-- southern Sweden - Micrcfossils and their stratigraphic npplica- ).- • . '^southwest Iceland. 35 SEK. •' «^'is'. \Jr?^ " ?',"-? " j -"• p-.'V-f1^",;-!.'. ,;' tion.15SEK.V-. -"-\ ; ••-,.,„: '-'.".. '• i,. C •*-1 . •. •'•? .19. Ingolfsson, O. 1987: Investigation of the Late Wcichsclian :$~ •'.../:

;5.' Lagerlund, E. 1977: Förutsättningar för moränstratigrafiska -"'. .- gladal history of the lower Borgarfjördur region, western Ice-'--' <*"<"'• . "undersökningar p.1 Kullen i Nordvästskåne - teoriutveckling : ; s v ; J. '];iand.35SEK.v,.>V J.-,s? " v- : ' :',*•'• / '">|-.'- . ."^"••-•;;,;"V'I'r; v' ' ' och ncotcktonikfTill-stratigraphical studies in the Kullen area,' ', -. 20. Möller, P. 1987: Moraine morphology, till genesis, and degla- ^'A.v/.; . / <' NW Skänc, South Sweden: basic theory and ncotcctonics. 15 ./-:-•• ciation pattern in the Asnen aiea'i south-central Småland,Swe- V .*',.«j'- , ."•.. SEK. :,; ?v\ .' .-'. •' . "-'. / •, ', •* dcn.50SEK.'-- • .-. -1.1,v;->1' ; ./'•>;-.:"• •. ..^-v > •<<•; ^- ',.'.-• "'/"'V Vc.-V 6. '• Hilldén, A. 1979: Deglaciationsförloppcl i trakten av Berg- •A'.21- Harrison, S. 1988: Reconstructing climate from lake level '•^•;-s;-'>'- '. • hemsmoränen, öster om Göteborg (English summary: The changes.35SEK.1. "-'!" '-"?' r'';' ,-': '-' '•: >•-•''• '•"• -:" '"•"-•-•••-:":"1_'f-;-' :j ; . '.' dcglaciation in the vicinity of the Bcrghcm moraine, cast of Lcmdahl, G. 1988: Paleodimatic and paleoccological studies " ',-,'"•>-,•; ; : : ;V- ..Göteborg). 20SEK! , ::-;- '-'.<-'.;• '.'.,; " .-• ',.••""'''- based on subfossil insects from Late Wcichsclian sediments in''. •'• ; ':,';..'... .7. . Björck.S. 1979:LatcWcichselianstratigraphyofBIckinge.SE southern Sweden. 40 SEK."-'"'!i":' i:1 "'• :'-'»-i'?yi.'" -'-^ : ;''"-i'.•/'>'"f/t-'W- •• •'• '" Sweden, and water level changes in the Baltic Ice Lake. 20 ;• 23. Malmberg Persson, K. 1988: Lithostratigraphic and scdimen- ..-> "j •'.-;, .-•'•• .f Sl-K. -".._... ':•: -. .IV..:. •. "T.-','.-, ' • '.'-' - ••..:, tological investigations around the eastern boundary of the/ ' S.--'Andersson, O.H. 19S1: Borrning och dokumentation. Borr-' Baltic deposits in central Scania." 40 SEK.,: >'\?s '.''"^.SJ.- -..! l - • : ningstcknik jämte metodik för geologisk datainsamling under "'. ,24-. Licdbcrg Jönsson, B. 1988: The Late Wcichsclian macrofossil ;-'. -r>;.-': .'•• \ '. borrningsgång (Drilling and documentation). 30 SEK. '• ' v .-' flora in western Skåne, southern Sweden.' 40 SEK. ','.=... /'}^i-J.'..'.*£,'. 1 9." Hjort,Ch. 198l:StudiesoflhcQuatcrnaryinNorlhcastGrcen- r ,;.'25. Svensson, N-O. 1989. Late Weichsclian and early Holocene^y'-^•'• %'•. 1 vl'•."''" V'.land. 20SEK! ;-\-. . :> ,i .V/V'.-• •••'. •''v .'.v ';''" ' " -, •; ", shore displacement in the central Baltic, based on stratigraphi-''••: ,';';•. •• '.: 10.;Hjelmroos-Ericson, M. 1981: Iloloccnc development of Lake •J < •", , ' cal and morphological records from eastern Småland and Got- % ' ,'-"•.'"'- •-•- -•; ','=•• WiclkicGacno area, northwestern Poland. 30SEK... '"••"•' • ,'S: vy> land, Sweden. 60SEK. V .'/:v -CJUX-^h' '-"':'?'^ (V^iii^i' .': , 11.'.Liljegren, R. 1982: Paleoekologi och strandförskjutning i en 26. Thelaus, M. 1989:'.Late Quaternary, vegetation history,and :'",.;--^,'.-i''' ,':-. , . Lit'.orinavik vid Spjälkö i mellersta Blekinge (English summa- palacohydrology of the Sandsjön-Årshull area, southwestern .,'•.'.'v:>- 1 ; 1 •',')' ry: Palacoecology and shore displacement in a Littorina bay at Sweden. 40SEK. ',- j/^] •' ;^-' .t-.h --••; -A.'.,/?-:^ V?'" -'-^!."'.^; r .:'' Spjälkö, Blekinge). 30SEK. ..,- ••• 'pX.f.;'".r ^\} " .-.•'/• \: -. --. 27. Regnéll, J. 1989: Vegetation and land use during 6000 years."r-;?. V;^; '. 12.' Norddahl, H. 1983:' Late Quaternary stratigraphy on Fnjoska- Palacoecology of the cultural landscape at two lake sites in ^^iCi-'"-

; ,• dalur central North Iceland, a study of sediments, ice-lake southern Skåne, Sweden. 50SEK. •'• •' '-.'•'.: ,'o.,'V^'. "J X.^.",;-,V;/.I: ; h"', " ': strandlincs, glacial isostasy and ice-free areas. 30 SEK.' •, , - ;.;.-' 28. Olsson, S. 1991: Geochemistry, mineralogy and pore water; >'-'t 'i'-', . : 13.' Robison, J.M. 1983: Glaciofluvial sedimentation: A key to the' composition in uplifted, Late Wcichselian - Early Holoccnc:.,-/"rS;':';4.)i (

,t 't~; dcglaciation of the Laholm area, southern Sweden. 35 SEK..' clays from southern Sweden.50SEK.' •: '.y.pvi^.-' ."•" ^ .;HC/*'i/V - 14. Sandgren,' P." 1983: The dcglaciation of the Klippan area,' ••v,^ 29.' Ekström, J. 1993: TheLateQuaternary History of theUrus(Boi;. jV'l^vi' : 1 .-. V: southern Sweden, a study of glaciofluvial and glaciomarinc primigcnlus Bojanus 1827) in Sweden'. 50 SEK:'TÄi "ft- >". •..'.-', "•*-i- S^'i ," A-, sediments. 35 SEK." •":•"-!, ,\S< ?".-. ;".1..r., •} ''-:7.'-'- .* .• • ';3Ö. ' lS.-Amark', M. 1984: The dcglaciation of the castcin part of Skäne, ; ' / " southern Sweden. A study of till and stratifie d drift. 35 SEK;:. ~ 'X'" "'V b "/ ::-:-. •;:<.• j]:

CODEN: SÉ-LUNBDS/NBQK-94/31+30P ISSN: 0281-3033 :