The Pleistocene/Holocene boundary: a proposed boundary-stratotype in Gothenburg, Sweden NILS-AXEL MORNEK (Editor)

B O W

Morner, N.-A. (Editor) 1976 12 01: The Pleistocene/Holocene boundary: a proposed boundary-stratotype in Gothenburg, Sweden. Boreas, Vol. 5, pp. 193-275. Oslo. ISSN 0300-9483. The boundary between the last two geological epochs, the Pleistocene and the Holocene, is placed at ‘the date 10,000 B.P., measured in radiocarbon years’. In the European chronostratigraphy, this corresponds t o the Younger DryadPreboreal boundary, the pollen zone III/IV boundary and the Late Glacial/Postglacial boundary. The stratal sequence in the Botanical Garden of Gothenburg is proposed as a suitable boundarystratotype of the Pleistocene/Holocene that fulfils the stratigraphical rules of marine environment and accessibility. A core, labelled B 873, has been analyzed for multiple parameters by various authors. Th e suggested Pleistocene/Holocene boundary in Core B 873 is indicated by a lithologic boundary, a palynological change tentatively correlated with the pollen zone III/IV boundary, and a distinct palaeomagnetic intensity maximum, the ‘Gilon Magnetic Intensity Maximum’, identified in numerous other cores a t the Younger Dryas/Preboreal boundary and at the drainage of the Baltic Ice Lake in varved clay sequences (with the peak dated at the drainage k4 varves). This boundary is closely radiocarbon dated at 10,000 B.P. (10,000-9950 B.P.) in terrestrial-lacustrine sequences within the proposed type area in Gothenburg and in Southern Sweden, the established type region for the PleistoceneiHolocene boundary. The corresponding varve date is 9965 varves B.P. (De Geer’s varve - 1073). The various parameters directly and indirectly connected with the study of Core B 873 make global correlations possible. Because every region has its own local characteristics, however, it will be necessary to establish regional type sections, hypostratotypes. N.-A. Momer, Geological Institute, Stockholm University, Box 6801, S-I 13 86 Stockholm, Sweden, 26th April, 1976.

This study is based on an offset-printed report which was prepared for the INQUA I X Congress in New Zealand (1973). Throughout this issue, the original INQUA report is referred to as ‘the original report’ or ‘O.R. 1973’. This issue contains essentially the same material as the original report. However, some chapters have been shortened, several tables and some figures have been omitted, Chapter 12 has been rewritten, and Chapters 19 and 20 have been added. At the INQUA VIII Congress in Paris in 1969, the ‘Commission on the Study of the Holocene’ proposed (later adopted) that the Pleistocene/Holocene boundary should be placed at ‘the date 10,000 B.P., measured in radiocarbon years’. This meant that the proposed boundary was placed at the Younger DryasiPreboreal (YDIPB) boundary in the European system (the pollen zone boundary III/IV). I t also meant that from then on it 14 - Boreas 4/76

was time to start establishing local, regional, and global type sections of the Pleistocene/ Holocene boundary at the defined chronostratigraphic level. According to the stratigraphical rules, a world type section (a boundary-stratotype) has to be established in a marine environment (in a broad sense) and has to be accessible (cf. Chapter 1). Accessible sections of marine sediments of 10,008 years ago are only found in a couple of uplifted areas. Southern Sweden has widespread marine deposits of the age in question, which are correlatable with climatic changes, ice marginal Eluctuations, varve chronology, sea level changes, pollen zonation, and radiocarbon chronology in terrestrial-lacustrine sequences. Knowing the potential of the South Swedish sediments for a boundary-stratotype and the necessity of rapid establishment of a world

194 Nils-Axel Morner

type section of the Pleistocene/Holocene boundary, Fairbridge persuaded me in 1969 to start working with the establishment of a boundary-stratotype in southern Sweden. The

BOREAS 5 (1976)

field and laboraton- nork started in 1970. This issue gives a full and final report on the investigation (cf. O.R. 1973).

1. The search for a boundary-stratotype of the Holocene RHODES W. FAIRBRIDGE

Historical background About 140 years ago Lye11 (1833:52) proposed the name ‘Recent’ for the youngest epoch in geological chronology. His own hierarchy of names ‘Eocene’, ‘Miocene’, etc., was adopted and expanded by subsequent workers and for the sake of consistency a French geologist, P. Gervais in 1867, proposed ‘Holocene’ for the youngest epoch. It was not until 1855, however, at the Third International Geological Congress in Berlin, that the Portuguese delegation formally proposed that ‘holocene’ shouId be accepted into the international nomenclature of stratigraphy. Delayed already half a century, the term was only very slowly adopted. It was not until 1967 that the American Stratigraphic Commission gave their formal recognition. I n the last century, geologists in general showed extraordinarily little interest in the youngest epoch of their chronology, although they vigorously defended the concept that we should study the present as the key to the past. One may suspect that the present-day processes were simply taken for granted, so that the history of the last few thousands of years was assumed to be marked by a uniform continuum of environmental controls. Today we know that nothing could be farther from the truth. The ignorance of geologists of this varied and complex Holocene history has indeed been an important factor in delaying scientific progress in many branches of geology, palaeontology, ecology, and climatology. It is significant that the necessity for a systematic nomenclature €or the youngest geological epoch has emerged primarily in two

distinct regions, where the respective needs are most apparent. First, and most importantly, there was the Scandinavian glacio-isostatic uplift region, where large parts of the landscape were coated by sedimentary formations dating from the last few thousand years. Secondly, and somewhat less urgently, was the need felt by the prehistorians - the archaeologists - who were mapping the distribution of ancient man and the stratigraphic accumulations associated with his living sites; these deposits were best known in the caves, alluvial fills, and coastal shore-terraces of Western Europe and North Africa. For these two areas of study, specialized stratigraphic nomenclatures emerged. I n Scandinavia the term ‘Postglacial’ proposed by Forbes (1846) became formalized by Munthe (1592) to become a stratigraphic standard essentially equivalent to what is now defined as ‘Holocene’. There are, however, difficulties in extending a banal descriptive term like postglacial to world-wide validity (Morner 1973a). It has often been noted that the dramatic environmental change that accompanied the retreat of the continental ice was markedly diachronous, beginning around 15,000 yr. B.P. in the south and becoming progressively younger in the higher latitudes. Evidently, in Antarctica and Greenland no genuine postglacial phase has yet begun. For the archeologists the beginning of the Holocene corresponded approximately to the Neolithic cultural phase, followed essentially by the historical eras. But the timing of the Neolithic is also diachronous; in North Africa it is distinctly earlier than in Sweden. Thus the Quaternary specialists of all sorts

BOREAS 5 (1976)

PleistoceneJHolocene boundary

195

are urgently in need of some formal, internationally approved chronologic benchmarks with which to define the youngest epoch of geologic time.

globe. It is desirable to seek the sharpest, clearest boundary that corresponds to a moment in geologic time when a world-wide environmental change is in progress.

Defining geological units

Problems in finding a Pleistocene/ Holocene boundary-stratotype

I n the selection and definition of any unit in rock sequence (lithostratigraphy) or in geologic systematics (chronostratigraphy), the Quaternary specialists accept a geological code of nomenclature in just the same way as they recognize the conventional identification procedures adopted by the zoological and botanical unions. In this connection it is appropriate to note that the IUGS (International Union of Geological Sciences) is gradually refining the code of nomenclature (recently summarized by Hedberg 1972). Unfortunately many palynologists and others working on Holocene problems are unaware of these rules and we, therefore, repeat the key points. It is agreed that no unit in geologic chronology will be defined without first establishing a lithostratigraphic type section or stratotype, which means simply a distinctive sequence of strata which outcrop in a designated and precisely described locality. Since geological formations rarely outcrop in complete continuous sections, it is usually necessary to put together a complex assemblage or integration of sections. Above all it is desirable to specify and describe the boundary-stratotypes for every named sequence. For the Holocene the lower stratotype, the PleistoceneiHolocene boundary, is what needs to be designated. This must be a physical horizon or plane within a sequence of formations which is associated with adequate aids for correlation. Such aids include fossils, faunal or floral, material for geochemical ‘absolute’ dating, and lithic indicators of various sorts. This horizon is a benchmark established in the field. The IUGS Subcommission on Stratigraphic Classification urges that it should be selected at a site that is (1) wellsurveyed, (2) easily accessible, and (3) in a marine facies. The selected boundary-stratotype, when approved, becomes a world standard. For this reason, stress is placed upon the need for marine facies which hopefully contain fossil organisms such as pelagic foraminifera which can be correlated (eventually) over the entire

The boundary-stratotype that will serve as a chronostratigraphic timeplane for the whole world must be sought with care. It is obvious that if such events as the retreat of the glaciers, the transgression of the ocean, or the invasion by a population (flora or fauna) are diachronous, i.e. they advance step by step, then we must not assume that any of those field indications correspond to a horizontal t h e plane. They are clearly sloping. For example, the events drawing the Last Glacial stage to a close in North Africa reached in critical point about 13,000 yr. B.P. Similar events in northern Canada marking the end of the Laurentide ice sheet occurred about 6000 yr. B.P. The duty of a stratigraphic committee is therefore to choose an intermediate site that is most convenient for world-wide use. At the INQUA meeting in Warsaw of 1961, De Jong (1965) proposed a number of alternatives. Many valuable suggestions have been documented (see, for example, Mercer 1972; Morrison 1969; Neustadt 1971; Nilsson 1965). Basically there are four main favorites, beginning around 14,000, 12,000, 10,000, and 7500 yr. B.P. Most of these proposals suffered from the fact that they were based upon continental facies which would inevitably be highly timetransgressive and furthermore very difficult to correlate on a transoceanic scale. On the basis of radiocarbon dating of some deep-sea cores from tropical Atlantic, Broecker et al. (1960) indicated that the sharp change from late glacial indicators (foraminifera) to post-glacial associations occurred at about 11,000 yr. B.P. Subsequent researches from low latitude sites in all the world’s oceans and by many independent workers have shown that there is a spread of dates through about 11,000-9000 yr. B.P. for this change, which was accompanied also by a wide variety of other biological markers and by various lithologic indicators (e.g. arkosic sands and clays, see Damuth & Fairbridge 1969). There are good reasons why this change

196 Nils-Axel Morner

Fig. l t l . Idealized temperature curve for the last (Weichselian) glaciation and the beginning of the Holocene, as evidenced by various low and mid-latitude indicators. BI, 11, III=three reef-building stages on Barbados (Broecker and others); T1, T2=Termination 1 and 2 (Broecker & Van Donk 1970). W, X, Y , Z=Ericson’s pelagic zones; note that the warm ocean persists half way through the Weichselian cold stage. From Fairbridge 1972.

should occur within a narrow time-span in the low latitude oceans. I n a glacial cycle (Fairbridge 1972) that goes through anaglacial, pleniglacial, and kataglacial phases, the shape of the curves reflecting water or air temperatures, sea levels, population shifts and so on, are all skewed to the right, so that a sequence of glacials and interglacials presents a sawtooth effect (Fig. 1:l). The midpoint on the kataglacial/interglacial upsurging of any cycle has been designated the ‘termination’ by Rroecker & Van Donk (1970), although in the higher latitudes and depending upon the indicator used there are various and diachronous ‘terminations’ (Ruddiman & McIntyre 1973). Comparison of the last glacial cycle with the Milankovitch radiation curve (Fairbridge 1961: 563) shows that the last warming trend began about 17,500 yr. B.P., reaching a peak around 10,000 yr. B.P., since when it has been descending (O.R. 1973: Fig. 2; Fairbridge 1972). As recognized by Croll, nearly a century ago, a ‘retardation’ factor is involved, which must delay the visible effects of a solar radiative warming; this is due to the latent heat delay in the melting, the sluggishness of the oceanic circulation, the albedo effects of retreating snow covers and advancing vegetation, and

BOREAS 5 (1976)

various other processes. The air temperature showed an approximately 6000 yr. B.P. average (global) delay, whereas the eustatic sealevel rise showed an even greater, 10,000 yr. retardation. I n the tropics the atmospheric warm-up was most rapid and likewise that of the ocean surface, so that heavy rains began to occur in the low latitudes around 13,500 yr. B.P. Widespread sea ice and the high albedo delayed this effect in the higher latitudes until about 6000 yr. B.P. Secondary cool cycles with ice readvances contributed t o the retardation across the North Atlantic. Ruddiman & McIntyre (1973) have indicated a faunally defined deglacial warming that ranges from 13,500 yr. B.P. in the east to 6500 yr. B.P. in the west. It is hoped, therefore, that readers will appreciate that the search for a boundary stratotype, like Jason and his ‘Golden Fleece’. is bound to be beset with difficulties. Theoretically it would be possible to select one of the well-studied deep-sea cores and designate it as the standard. However, this would contravene one of the Stratigraphic Subcommission’s basic recommendations, viz. that the site selected should if possible be accessible. An accessible site of marine origin, now on land, in the 11,000-9000 yr. B.P. range is difficult to find. In the course of eight years as President of the International Shorelines Commission on INQUA, the writer made a worldwide search for an appropriate spot. The wellknown classical Quaternary sites around the Alps and in Central Europe are only poorly correlated in the marine sequences. Those of the North American craton are even more separated. The deltaic sequences of the Rhine and Mississippi are largely buried and known only in core. The classical marine Flandrian sections found in northernmost France, Belgium, and the Netherlands unfortunately are transgressive over the Late Glacial loess and so on, and only cover the time-span after about 8500 yr. B.P. The isostatic uplift area of the St. Lawrence Valley in Quebec does, however, offer real potentials for the Holocene boundary question (Terasmae 1972), although today it is still in need of more detailed studies. The Pacific Northwest of U.S. and British Columbia also offers possibilities, as do parts of Japan, New Zealand, and Scotland. At the end of this world-wide search it became apparent that by far the best surveyed, best dated, and palaeontologically best studied sec-

BOREAS 5 (1976)

tions were exposed in the glacio-isostatically uplifted parts of southwestern Sweden. Dr. N.-A. Morner had recently completed a large monograph on the region in question (1969), and over 100 radiocarbon dates were available straddling the boundary area. In 1969, we discussed potential sites and concluded that a shallow boring or excavation in the Botanical Garden of Gothenburg would disclose the critical boundary in marine environment in an accessible location that could be preserved in safety for posterity. Morner carried out the necessary boring and obtained a 14.5 ni long continuous core (labelled B 873), starting from a ground elevation of 17.4 n~ above MSL. It is essentially a marine clay, corresponding to a timespan of 12,500 to 8500 yr. B.P., the very high accumulation rate (3.8 mm/yr) providing an excellent ‘expanded scale’. In August, 1971, a joint field conference was organized in southern Sweden by the INQUA Holocene Commission and the INQUA Sub-

Plei,stocene/Holocene boundary

197

commission on Shorelines of Northwestern Europe. The regional setting was reviewed in detail and the preliminary results of the core analyses were presented. I t was clear that at a depth of 3.35 m below the ground surface there was a lithologic boundary that was closel y fixed in age at about 10,000 B.P. In anticipation of the final report of Core B 873, it was agreed that southern Sweden should be chosen as the world type region for the Pleistocene/ Holocene boundary. Summing up, (1) our world-wide search for a site exposing marine beds of an age of 10,000 yr. B.P. (they would normally be 30-40 m below MSL: Fairbridge 1961; Morner 1969) resulted in the selection of the glacio-isostatic uplift area of Gothenburg in southwest Sweden as the ideal area for establishment of the boundary-stratotype, and (2) Core B 873 from the Botanical Garden of Gothenburg disclosed an excellent profile that appears to fulfil all requirements of a world standard stratotype.

2. The Pleistocene/Holocene boundary in southern Sweden NILS-AXEL MORNER

‘The date 10,000 B.P., measured in radiocarbon years’ corresponds to the European pollen zone boundary III/IV, the Younger DryasiPreboreal (YDIPB) boundary, the Late Glacial/Postglacial boundary, the drainage of the Baltic Ice Lake (and the beginning of the Yoldia Sea in the Baltic), and the Gotiglacial/ Finiglacial boundary in De Geer’s varve chronology. I n southern Sweden, there are extensive deposits of the date in question. They are of three types: (1) Marine sediments in the southwestern part (of glacial as well as postglacial facies). ( 2 ) Glacial-lacustrine Baltic sediments in the eastern part. (3) Terrestrial-lacustrine sediments in lakes and bogs in the emerged areas of southern Sweden. Fig. 2: 1 gives the palaeogeographic landisea distribution and the position of the ice margin at 10,000 B.P. I t also shows investigated locali-

ties (up to 1973) where the Younger Dryas/ Preboreal boundary is directly or indirectly established (those in SW Sweden are listed in O.R. 1973:27-28). Out of a total of 154 localities, 3.5 occur in marine deposits (and 5 at the isolation level), 34 in Baltic deposits (and 3 at the isolation level), and 77 in lacustrine-terrestrial deposits. The YD/PB boundary is radiccarbon dated in 17, and varve dated in 30 of those localities (in addition, it has recently been palaeomagnetically identified in 17 localities; Chapter 20). Southern Sweden is thus an ideal type region for the Pleistocene/Holocene boundary, if this boundary is to correspond to the Younger Dryas/Preboreal (YD/PB) boundary. At the South Sweden Excursion in 1971, it was accordingly decided that southern Sweden should be chosen as the world type region, awaiting the final establishment of a type section (and a type area).

198 Nils-Axel Morner

F i g . 21.

BOREAS 5 (1976)

Palaeogeography of southern Scandinavia a t the Pleistocene/Holocene boundary 10,000 B.P. Dark hatching represents continental ice cover. Light hatching gives approximate extension of the sea right after the drainage of the Baltic Ice Lake. All known localities within the type region, southern Sweden, where the Younger Dryas/Preboreal boundary is directly or indirectly established are marked according to the following scheme: squares=marinz, circles= terrestrial-lacustrine, and triangles=Baltic sediments. The Younger Dryas/Preboreal boundary is radiocarbon-dated in 17 localities (filled squares and circles) and varve-dated in 30 localities (filled triangles).

The Younger Dryas Stadial The Younger Dryas Stadial, dated at 10,95010,000 radiocarbon years B.P. (Morner 1969: 177), corresponds to a period of significant and world-wide cooling (e.g. Morner 1970b, 1973b). In Scandinavia, the rapid ice recession during

the Allerod Interstadial drastically changed; along the Norwegian Atlantic coast the ice started to readvance reaching its maximum at the end of the Younger Dryas Stadial, whilst the ice recession in the Baltic area was retarded, building up three moraine lines (Figs. 212 and 2:5).

BOREAS 5 (1976)

PleistocenelHolocene boundary

199

without gyttja and the Preboreal clay being a darker, somewhat coarser clay containing some gyttja. According to Nilsson (1961a:99) ‘The boundary between clay deposition and gyttja deposition is so striking and sharp, stratigraphically as well as floristically, that it must correspond to a decisive change in the climatic evolution’. He concludes ‘Nowadays, most geologists place the (climatic) boundary between the Late Glacial and Postglacial periods at the above-mentioned level’.

1

I

Fig. 2:2. Extension of the Scandinavian ice cap during

the Younger Dryas Stadial. In Norway, the ice readvanced, reaching its maximum at the end of the YD Stadial, while the ice recession in the Baltic region was retarded, building up three moraine lines.

It has been estimated (lversen 1954) that the Younger Dryas corresponded to a temperature decrease of 3 4 ° C and that the temperature was about 6°C colder than today. The climatic deterioration during the Younger Dryas Stadial had drastic biological and sedimentological effects. The vegetational change was sudden and drastic in the whole of Europe, the Allerod forests or ‘park tundras’ being replaced by vast tundra regions. After the upper layer containing Dryas octopetala and other arctic plants in southern Sweden and Denmark (Nathorst 1870, 1872), this final tundra period was named the Younger Dryas period (Nordmann 1910). The Younger Dryas Stadial also had a drastic sedimentological effect as discussed in previous papers (Morner 1970a, 1971f). In lakes and bogs of southern Scandinavia, the Younger Dryas Stadial is often represented by an alluvial clay, the ‘Dryas Clay’. The change from gyttja to clay or clayey sediments is believed to be caused partly by death of the vegetation due t o the severe climate and partly by heavy snow melting in the springs washing out alluvial clay in lakes, bogs, and rivers. This washing out of alluvial clay also affected the coastal marine sedimentation. Marine sediments of the Kattegatt Sea seem generally to register the change from Younger Dryas to Preboreal, the Younger Dryas clay being a light grey, very fine clay

Pollen chronology and radiocarbon dating The Younger DryasiPreboreal boundary corresponds to the final general change in Europe from tundra towards forest vegetation. It corresponds to the pollen zones III/IV in Jessen’s system (1935) and zones DR-31PB in T. Nilsson’s system (1961b). In Morner’s (1971d) climatic zone system it was referred to as zones YD/PB. Berglund (1966) introduced a transitional zone (DR-3-PB) for southeastern Sweden. In Nilsson’s radiocarbon dated pollen zone system (Nilsson 1964), the base of the Preboreal was dated at 9920&150 B.P. (10,220 according to Nilsson who corrected the original dates according to the ‘new’ and unofficial half-life of the 14C isotope). Considering all available radiocarbon dates from northwestern Europe related to the Younger DryasIPreboreal boundary and corrected to the NBS standard, Morner (1969:177) arrived at the conclusion that the Younger DryaslPreboreal boundary should be placed at 10,000-9950 radiocarbon years B.P. Tauber (1970), on the other hand, dated the same boundary at 10,200 B.P. The differences in dating seem to emerge from two facts usually overlooked. First, the Younger Dryas Stadial includes three major cold peaks (Morner 1970a) aiid the end of the second peak is so distinct that it sometimes could be confused with the end of the whole Younger Dryas Stadial. Secondly, there may have been atmospheric 14C variations at about the YDIPB boundary, so that almost the same level may give dates varying by about 300 years. There are several indications that there may be a 200-300 year gap in apparent radiocarbon dates at the YD/PB boundary (10,000 and 10,250 B.P. for

200 Nils-Axel Morner almost the same level), which would be in full agreement with the magnetic intensity peak discussed in Chapter 20.

Varve chronology and the drainage of the Baltic Ice Lake The Gotiglacial/Finiglacial boundary in De Geer’s varve chronology (e.g. 1940) was originally placed at the drainage of the Baltic Ice Lake at Billingen in Sweden. De Geer dated this drainage at varve -1073 or 7912 B.C. (De Geer 1940), which with updating of the varve chronology would correspond to 9965 B.P. It is true that De Geer’s dating of the drainage of the Baltic Ice Lake was poorly based. However, the versions to follow were no better (Caldenius 1941; Nilsson 1953, 1968; Lundqvist, 1961). E. Nilsson first dated this event at 8315 B.C. (Nilsson 1953) and later changed it to 8213 B.C. (Nilsson 1968). Comparing the 8315 B.C. varve date with the 8250 B.C. corrected radiocarbon date, T. Nilsson (1964) found a perfect agreement between the varve and radiocarbon chronologies. Unfortunately, this led to the uncritical acceptance of the 8300 B.C. date as the age of the end of the Younger Dryas Stadial (e.g. Hansen 1965, West 1968), though the uncorrected radiocarbon date is 250-300 years younger, the varve date is 300 years younger and the Younger Dryas readvance in Norway (the Bergen area) reached its maximum at about 10,100-10,000 B.P. (after the dates 10,050i250 and 10,150i 300 B.P.). The drainage of the Baltic Ice Lake, the end of the Younger Dryas Stadial, and the correlations between Sweden and Finland have been discussed in several papers (see Morner 1970a: Fig. 6). (1) Hyyppa (1963) correlated the end of the Younger Dryas Stadial with the drainage of the Baltic Ice Lake at 8213 varves B.C. and the recession from the Salpausselka I11 Moraine. ( 2 ) E. Nilsson (1968) dated the drainage of the Baltic Ice Lake at 8213 varves B.C. and placed the end of the Younger Dryas Stadial at about 8100 B.C. (3) Morner (1969) found that the entire Younger Dryas Stadial covered 950 years in the varve as well as in the radiocarbon Chronology, correlated the Finnish and Swedish moraine zones (or rather retardation zones), and dated

BOREAS 5 (1976)

the end of the Younger Dryas Stadial at the drainage of the Baltic Ice Lake, the radiocarbon age being 8050 B.C. (4) Donner (1969) found morphological evidence of a regression right after the Salpausselka I1 Moraine, correlated this event with the drainage of the Baltic Ice Lake at 8213 varves B.C., and placed the end of the Younger Dryas Stadial at the same level. ( 5 ) Tauber (1970) put the end of the Younger Dryas Stadial at 8200 radiocarbon years B.C. and correlated this with E. Nilsson’s 8100 varves B.C. (6) Morner (1970a) demonstrated that there was a drastic change in the rate of recession (from 80-90 to 300 m/yr) in connection with the drainage of the Baltic Ice Lake at Billingen, and argued that this change must correspond to the YD/PB boundary. Furthermore, he correlated this event with the Finnish +292 varve. (7) Niemela (1971), revising the Finnish varve chronology, put the end of the Younger Dryas Stadial at the drainage of the Baltic Ice Lake (previously dated by E. Nilsson at 8213 varves B.C.) and correlated this event with the Finnish zero varve, which was found to correspond to a n ice marginal position ‘a few kilometers proximally t o the Second Salpausselka belt’. (8) Stromberg (1971) discussed the correlations between the Swedish and Finnish varve chronologies. He arrived at the conclusion that the two scales could not be correlated and that there must be an error in one of the scales. Realizing that all previous correlations between Sweden and Finland were in fact poorly based and that the reliable Swedish varve chronology ended at De Geer’s varve -1073 (the ingression of the salt water at Stockholm), it becanie obvious that the only way of really solving the problem was to extend the reliable time scale south of Stockholm via shortdistance correlations. During 1970171, Morner, Larsson & Liden (Morner 1977a) measured all available cuts of varved clay on Sodertorn south of Stockholm and took several cores with the Swedish foil piston corer. A main chronology was established, to which were added all measurenients (published and unpublished) made by De Geer (1940) and all published measurements made by E. Nilsson (1968) in this area. On this material, a very

PIeistocenelHolocene boundary

BOREAS 5 (1976) S

in

I

Salpausselko I

201

Finland

II I

200 m/yr

-

-

- 100 m/yr

- 10G m/yr

-

-

0 0

0

0

N I

9,900

Fig. 213.

10.000

hi:

10,100

0 0

10,200

0

Sweden

@ 5

1

10.300

10,LOO 8.P

Rate of ice recession over Sodertorn (Nynashamn-Stockholm) in &ear. At about varve -1372, the ice recession drastically increased. This must correspond to the end of the Salpausselka I1 Phase in Finland. A t about varve -1255, the rate of recession decreased, the varves became thicker, and the Dalaro Moraine was built up. This must correspond to the Salpausselka I11 Phase in Finland. At about varve -1073, there is a final increase in the rate of ice recession. Varve -1073 corresponds to the first appearance of saltwater in the Stockholm region. Furthermore, new studies have revealed that it is usually represented by a distinct drainage varve in eastern Sodertorn. Varve -1073 (9965 B.P.) must correspond t o the drainage of the Baltic Ice at Billingen. Note the close agreement between the Swedish graph and the Finnish chronology.

detailed chronology was established from Stockholm (9965 varves B.P.) to Nynishamn (about 10,500 varves B.P.). Fig. 2 3 illustrates the main results, summarized in the following way: (1) At about v a n e -1372 (10,261 B.P.) the rate of ice recession drastically changed from 10 m/yr to 120 miyr. (2) At about v a n e -1255 (10,147 B.P.) the various varves became thicker, the rate of ice recession dropped down to about 20-30 m/yr, and the Dalaro Moraine was built up.

(3) At about varve -1073 (9965 B.P.), the rate of ice recession rapidly changed from about 50 m/yr to 200-250 m/yr. I n several cores and sections, varve -1073 is a distinct drainage varve (thick varve containing sand, silt, and rounded clay balls). Varve -1073 also corresponds to the ingression of salt water to Stockholm (change from diatact to symmict varves).

Varve -1073 must correspond to the drainage of the Baltic Ice Lake at Billingen. This means that the drainage of the Baltic Ice Lake is re-dated at De Geer’s varve - 1073 (in today’s chronology corresponding to 9965 B.P.). As demonstrated by Morner (1970a), there was a similar change in the rate of ice recession at BiUingen (from 80-90 m/yr to 300 mlyr). Consequently, the Stockholm and Billingen areas can be reliably correlated via the drainage event (Fig. 2:4). Our Sodertorn chronology (Morner 1977a) also seems to provide perfect correlations between Sweden and Finland (Fig. 2:3): (1) The long retardation zone prior to varve - 1372 with only 10 m/yr recession must correspond to the Salpausselka I1 Moraine in Finland (dated at varves 0 to -183 according to Sauramo 1923). (2) The drastic change in the rate of ice re-

BOREAS 5 (1976)

202 Nils-Axel Morner Southern Sweden ice marginal positions

‘a,.:

,/’

?? ,a:

’

after 12,000 B P

cession at v a n e -1372 occurred 299 varves prior to the ingression of salt water at Stockholm. The end of the Salpausselkd I1 stage in Finland occurred 292 varves prior to the ingression of salt water in Finland. (3) The retardation zone (and the Dalaro Moraine) at about varve -1255 occurred 117 varves after varve -1372. The Salpausselka 111 Moraine in Finland was built up at some time between varves + 100 and +200 in Sauramo’s chronology (1923). (4) The ingression of salt water at Stockholm at varve - 1073 corresponds to a drainage varve correlated with the drainage of the Baltic Ice Lake. I n Finland, the same horizon (ingression of salt water) is represented by varve + 292. The nature of the drainage at Billingen (sudden or by steps; 26 ni, less than 26 m or just ingression) remains to be solved, however. At any rate, it is suggested that the ingression at Billingen corresponds t o the arrival of salt water at Stockholm and Finland. Summing up, the drainage of the Baltic Ice Lake and the end of the Younger Dryas Stadial

Fig. 2r4. Ice marginal positions in southern Sweden after 12,000 B.P.

Full lines=moraine lines, and clashed lines=varve equicesses. The Younger Dryas Stadial is represented by three cold phases (I-111) and ended with the drainage at Billingen at varve - 1073, i.e. varve 9965 B.P.

is now dated at varve 9965 B.P., i.e. at about 10,000 varve years B.P.

Ice marginal positions during the Younger Dryas Stadial The correlations between the Late Weichselian moraines and retardation zones in the southern regions of the Scandinavian Ice Cap have recently been discussed in a separate paper (Morner et al. 1975: Fig. 7). There are important geographical differences in the behaviour of the ice from the Norwegian west coast across Sweden to Finland. This is illustrated in Fig. 215. I n Sweden and Finland, the Younger Dryas Stadial is characterized by a distinct retardation in the rate of ice recession, whereas on the Norwegian west coast it is characterized by a major glacial readvance reaching its maximum just prior to 10,000 radiocarbon dated years B.P. (Fig. 2:5), as conclusively demonstrated by Mangerud (1970). The maximum in Norway corresponds in Sweden and Finland to the period just prior to the drainage of the Baltic Ice Lake. Consequently, the end of the Younger Dryas

BOREAS 5 (1976)

PleistoceneJHoloceneboundary

203

10.000

10.950

Fig. 2;5.

Ice-marginal changes in five areas during the Younger Dryas Stadial measured in km from the position at the beginning of the stadial: (1) the Bergen area, Norway, (2) southeastern Finland, (3) south-central Finland, (4) south-central Sweden, and (5) Gotland-Stockholm. The Norwegian curve is dated by radiocarbon and the Swedish and Finnish curves by the varve chronology (the two chronologies are correlated via the drainage at Billingen). The Younger D i p s Stadia1 iiicludes three phases and two intervals. Generally, the Younger Dryas Stadial is characterized by a distinct readvance in Norway and zones of retardation in Sweden and Finland.

Stadial must be put at the Swedish varve - 1073 (Finnish varve + 292) and not at varve - 1372 (Finnish zero varve).

Coiiclusions and correlations The climatic deterioration of the Younger Dryas Stadial had drastic biological and sedimentological effects. The end of the Younger Dryas Stadial corresponds to the pollen zone boundary IIIIIV, which is radiocarbon dated at 10,000 B.P. This agrees well with radiocarbon dates of the R a Moraine in Norway. In the Swedish varve chronology, the end of the Younger Dryas Stadial is represented by a distinct change in the rate of ice recession and by the drainage of the Baltic Ice Lake at about 10,000 varves B.P. The drainage of the Baltic Ice Lake at Billingen provides a basis for direct correlations (Fig. 2:6): The drainage at Billingen is registered in the varves at Stockholm and dated at varve - 1073 (corresponding to varve + 292 in Finland) which means 9965 varve years B.P. The drainage at Billingen is registered as a layer of sand and gravel in at least three lake sequences in southeast Sweden (Berg-

lund 1966: P1. X; Berglund et al. 1971:26), pollen analysed at the zone boundary IIIIIV and radiocarbon dated at about 10,000 B.P. Consequently, the end of the Younger Dryas Stadial and the pollen zone boundary III/IV are both varve and radiocarbon dated at about 10,000 B.P. Hence, the establishing of the pollen zone boundary III/IV in a marine sequence would include close indirect dating by means of varves and radiocarbon. The direct correlations are illustrated in Fig. 2:7. Summing up, the 10,000 B.P. boundary corresponds in the varve chronology to: (1) a drastic increase in the rate of ice recession in Sweden (2) the drainage of the Baltic Ice Lake at Billingen in t h e radiocarbon chronology to: (1) the drainage of the Baltic Ice Lake at Billingen (2) the pollen zone IIIIIV boundary (Younger DryasIPreboreal) (3) the sedimentological clay/gyttja boundary (4) the change from readvance to rapid ice recession in Norway. I n addition, recent studies (Chapter 20) indicate that the 10,000 B.P. boundary also corresponds to a distinct paleomagnetic intensity peak identified in varve dated as well as in radiocarbon dated sections.

204 Nils-Axel Morner

BOREAS 5 (1976) Fig. 2t6. The extension of the ice at 10,000 B.P. and six main areas for the synthesis of ice marginal changes, varve chronology, radiocarbon dating and pollen zonation. The drainage at Billingen (2) is registered in the varves in the Stockholm region (1) and in the lacustrine sequences in Eastern BIekinge (3), which are poIIen analysed and radiocarbon dated and hence correlatable with Nilsson’s standard pollen diagram from Agerods Mosse (4) and the radiocarbon dates of the ice marginal position in Norway (5 and 6).

-

0

100

200 k m

; varves

Sr, Eb.

k‘

%3,

Fig. 2t7. Relations and correlations (arrows) in southern Scandinavia between ice marginal fluctuations, varve chronology, the drainage of the Baltic Ice Lake, the pollen zone IIIiIV boundary and radiocarbon dates as established in the six main areas (cf. Fig. 2:6). The drainage of the Baltic Ice Lake at Billingen (area 2) provides a basis for direct correlations. In the Stockholm region (area l), this event is varve dated at varve 9965 B.P. In eastern Blekinge (area 3), this event is fixed a t the pollen zone III/IV boundary and radiocarbon dated at about 10,000-10,200 B.P. Th e pollen zone III/IV boundary is radiocarbon dated in numerous localities in the type region (e.g. area 4; cf. Table 1S:l) as well as in the whole of Europe at about 10,000 B.P. The Younger Dryas glacial maximum in Norway (areas 5-6) is radiocarbon dated at about 10,150-10,000 B.P. Consequently, the establishment of the pollen zone boundary III/IV in a marine sequence like Core B 873 would include close indirect dating by means of varves and radiocarbon (arrows). The same applies (to an even stronger degree) to the ‘G8ion Magnetic Intensity IMaximum’ discussed in Chapter 20.

BOREAS 5 (1976)

PleistocenelHolocene boundary

205

3. Core B 873 in the Botanical Garden of Gothenburg NILS-AXEL MORNER

O n 14 October 1970, a core, labelled as Core B 873, was taken in the Botanical Garden of Gothenburg (Figs. 3:1-2) with the Swedish foil piston corer as a potential world standard section for the PleistoceneiHolocene boundary. The exact position is: 57O41.1” Lat., 11”57.3’E Long. This place was chosen (among numerous others; cf. Fig. 2:1 and O.R. 1973:27) for three important reasons: (1) The poUen zone boundary III/IV was in marine clay. This was known from previous work in the Cothenburg area, closely discussed by Morner (1969). (2) At the time of the zone boundary IIIilV (YDIPB), the area chosen was a shallow Fig. 3:l. The Gothenburg type area for the Pleistocene/Holocene boundary and the boundarystratoype, Core B 873 (double circle). Black dots represent other boring points within the type area: H.T.= Hisinge Tunnel, T.P.=Trapiren, N.D.=Nackrosdammen, L.M.=Landala Mosse, and B.M.=Blinkens Mosse.

sea floor in an archipelago. This would offer opportunities to establish a good pollen zonation, despite the sediments being deposited in the sea (where a reliable pollen zonation is usually hard to establish). Also, there was no erosional hiatus at about the YD/PB boundary, which is common in areas further SW. (3) The zone boundary III/IV was expected to be found 2-3 m below the surface (judging from B 441; Morner 1969: Fig. S4). This means that the boundary is easily accessible. At the same time the depth is enough to prevent destruction by weathering. Furthermore, lying in the Botanical

206 Nils-Axel Morner

(

BOREAS 5 (1976)

BOTANICAL

None of the localities listed in Table 1 in O.R. 1973:27-28 and shown in Fig. 2:l were regarded as equally suitable.

GARDEN

The geology of the area

Fig. 3:2. The Botanical Garden in Gothenburg and

the location of Core B 873.

Garden of Gothenburg, the spot chosen would be protected and could easily be visited.

-LB

BO

AL

PB

YD

BO

The area was deglaciated during the Ag%rd Interstadial. The Marine Limit (ML) is at about + 93 m. The subsequent shorelevel displacement brought the sea level down to about + I 5 m, the 'Regression Maximum' or shorelevel ALV-1 being about 9700-9300 B.P. (Miirner 1969), which was followed by a series of transgressions, the highest one of which reached up to +25.5 m (PL) at 7000 B.P. This is illustrated in Figs. 3:3 and 3:4. Fig. 3:3 also shows the relation between the shorelevel displacement and the sediment surface in Core B 873. I t indicates that the water depth was 12.5 m at the time of the YD/PB boundary. In the Gothenburn area, there are several sections showing the- land 'period during ALV-I

AT

SB

-

90

- 80

ND

- 70

\

- 60

-

- LO

I

I 12

I

11

I 10

- 30

PL

I I *-. I I3

50

I

'-I

-

I

I

I

1

I

9

8

7

6

5

10

Fig. 3t3. Shorelevel displacement curve for the Gothenburg area, sediment accumulation (at the base) and distance t o the receding ice margin (hatched zone plus figures in km). The climatic zone system of Morner (1971d) is given at the top. ML=Marine Limit, RL= regression limit, and PL=Postglacial Limit. ND, LM and BM refer t o the isolation level of cored bogs w-ithin the type area. At the Pleistocene/Holocene boundary (zone YD/PB, layer 8b/Sa) the depth of the sea was 12.5 m at Core B 873 (the diatom flora indicates marine, near-shore conditions with a high influx of fresh water).

PIeistocenelHolocene boundary 207

BOREAS 5 (1976) F i g . 3:4. Land/sea distribution: (1) black area=land a t ML at about 12,400 B.P., (2) white areas (plus black)=land at P L at about 7000 B.P., which closely corresponds t o the land/sea distribution at 10,000 B.P. (the Pleistocene/Holocene boundary), and (3) grey area=sea a t the Regression maximum (ALV-1 stage) at about 9500 B.P. (RL).

(Regression Maximum) and the subsequent transgression (Morner 1969: Fig. 83). One of these sections, ‘Angggrden’ (Munthe et al. 1924: Fig. 67), lies 40 m from B 873, and another section ‘B441, Wlvaleken’ (Morner 1969:

Fig. 3.5. Stratigraphic section of the Gothenburg Botanical Garden based on four borings. Eight radiocarbon dates have been obtained. The suggested Pleistocene/Holocene boundary is marked in Core B 873.

18

-

Fig. 84), lies 180 m from B 873. These localities together with B 873 and a couple of additional borings made it possible to draw a profile of the layers in the low-lying part of the Botanical Garden (Fig. 3:s).

120 m

B LL1

Lorn

60m

B 873

The Botanical Garden in Gothenburg lL

-

sw

NE

- 1L -

BOREAS 5 (1976)

208 Nils-Axel Morner

used. The first core (B 872) failed because a plug of hard clay got stuck in the head of the corer (the preboring did not fully penetrate the ground crust). The core obtained represented the depths 1.00-1.30 m (the plug of ‘Littorina’ clay) and 7.93-11.20 in (glacial varved clay; Chapter 4, Fig. 4:2). The B 873 was taken on 14 October 1970. The core was taken in two sections, one from about 1.5 to 9.5 m and one from about 9.5 to 14.5 m depth. The corer was N-S oriented. During the boring, the corer was carefully held in this orientation with a pair of tongs. The core and the foils were cut in the scarfs between the 2.5 m long corer pipes. The pipes were sealed with muffs. The sediments in the head of the corer and in the scarf pieces between the pipes were rolled in plastic foils. The N-side and the top and the bottom were carefully marked on each core and pipe.

Organization of analyses The intention was to study the core by means

of as many significant parameters as possible,

Fig. 3t6. Taking the Core B 873 on October the 14th 1970. The corer used is the Swedish foil piston corer. In the background the Marine Botanical Institute.

Boring technique and handling of the core With the Hiller auger and a sond, three test holes (B 868-870) were made in the Botanical Garden of Gothenburg to find the most suitable place for the final core. B 868, made in the lawn outside the Marine Botanical Institute, was found most suitable, showing the ALV-1 gyttja at about 2.5 m depth and the bedrock 14.6 m below surface. The final core was taken with the Swedish foil piston corer (Kjellman et al. 19-50), which is capable of taking continuous cores of up to 12.5 m length (Fig. 3:6). In order to be able to take samples for all the various analyses at the same level, a 68 mm diameter corer was

and to have the various analyses done by specialists. When the Core B 873 was taken, the following program was set up: 1. Stratigraphy, sedimentology, molluscs, etc.: Morner (Stockholm, Sweden). 2. Pollen analysis: Berglund (Lund, Sweden). 3. Diatoms: Bjorn-Rasmussen (Gothenburg, Sweden). 4. Forams: Feyling-Hanssen and Knudsen (Aarhus, Denmark). 5. Ostracods: Du Saar (Haarlem, The Netherlands). 6. Palaeomagnetism: Hospers, Lanser and Vollers (Amsterdam, The Netherlands). 7. Radiocarban dating: Hakansson (Lund, Sweden). To this list was later added: 8. Clay minerals: Georgala and Jacobsson (Stockholm, Sweden). 9. Sedimentology: Eckhardt, Mattiat and Raschka (Hannover-Buchholz, Germany). 3b. Diatoms (quantitatively): Du Saar (Haarlem, The Netherlands). Dr. Hageman (Geol. Survey, Haarlem, The Netherlands), chairman of the INQUA ‘Commission on the Study of the Holocene, took part in the organisation of the work.

BOREAS 5 (1976)

The sampling It was calculated that samples for all the various analyses could be taken from the same level, if this was made 4 cm thick. Plastic containers of suitable size and shape were bought. It was decided that samples were to be taken at every 4 cm in the upper part and at every 8-10 cm in the lower part (especially layer 10) where the sedimentation rate had been high. I n the laboratory, the cores were removed from the pipes, preliminarily investigated, cut up in half-pipe lengths, and rolled up in plastic foils and covering paper - always with the Nside and the top and bottom carefully marked. The cores were photographed by X-ray methods (see Chapter 5). After that, the cores were opened, carefully examined (close sediment classification, occurrence of shells and other macroscopic content, measurements of varves, etc.), photographed, and cut up in pieces. I n layers 1-9, all pieces were 4 cm thick; one was used for the various analyses and the next saved as a spare sample (or used for radiocarbon dating, etc.). In layers 1&13, the analysed pieces were 4 cm thick, whereas the spare samples were S-10 cm thick (due to the high sedimentation rate). The palaeomagnetic samples needed special care. A thin furrow was cut along the N-side, into which a thin straw was inserted and the sediment simply sucked into the plastic container: a cylinder of 3.6 cm height and 3.0 cm diameter. Totally, 117 samples were taken and distributed among the members of the team. By 8 January 1971 all samples were distributed. Spare samples for further palaeomagnetic and sedimentological analyses were later added from layers 12-13 and 1-9, respectively.

The South Sweden Excursion in 1971 I n September 1971, an excursion was arranged in southern Sweden for the members of the

15 - Boreas 4/16

PleistocenelHolocene boundary

209

INQUA ‘Commission on the Study of the Holocene’ and the ‘Subcommission on Northwest European Shorelines’. The excursion started with a symposium at the Botanical Garden of Gothenburg, where the preliminary results were presented and discussed, and the stratigraphy examined (Morner 1971f). It was agreed that South Sweden was the most suitable region for a world type section for the Pleistocene/Holocene boundary. Therefore, South Sweden was proposed as the world type region for the Pleistocene/Holocene boundary. It was also suggested that Core B 873 - if chosen as the boundary stratotype - could be completed by radiocarbon dating in lacustrine sequences in the near surroundings, and that the Anggarden area (or the Botanical Garden as a whole) in Gothenburg should be proposed as the world t y p e area for the Pleistocene/Holocene boundary.

The analytical results Most investigations were completed during 1971 and 1972. By September 1972, reports on all the various investigations, except for the one on the pollen analysis, were completed. The results were first presented in the original INQUA report (O.R. 1973).

The future preservation The location in the Botanical Garden of Gothenburg guarantees future preservation. If Core B 873 is chosen as the boundary-stratotype of the Pleistocene/Holocene (at the INQUA X Congress in 1977), a small ‘house’ will be built with direct accessibility to the boundary (in connection with this construction, vertical columns of the sediments that include the boundary will be taken and distributed among interested museums and institutes).

210 Nils-Axel Morner

BOREAS 5 (1976)

4. Stratigraphy and sedinientology NILS-AXEL MORNER

Inter-marine beds of terrestrial-lacustrine deposits, as well as littoral sediments, are fre. quently found in the Gothenburg area (Morner 1969: Fig. 83). The inter-marine beds correspond to the shorelevel stage ALV-1 closely dated at 9700-9300 B.P. (Morner 1969). The stratigraphy in the Botanical Garden of Gothenburg is well documented in the profile (Fig. 3:5).The change from bluish silty clay to gyttja clay corresponds to the onset of the ALV-1 stage. The mean sea level during ALV-1 was at about + 15.5 m (Fig. 4:l): it must have been above the sediment surfaces in B 873 ( + 15.1 m) and B 869 ( + 15.2m), but below the sediment surface at + 15.7 m in B 441 (pitchy soil indicates subaerial environment, at least during low water periods) and + 16.0 m in B 870 (beach sand with drifted branches etc.). I n the hills surrounding the low area where Core 873 was taken, there are numerous lakes and bogs which were isolated from the sea before the Younger DryasIPreboreal boundary

B L11

(Figs. 3:1, 3:3, 3:4). Lake Nackrosdammen at about + 70 m was palynologically investigated and radiocarbon dated (Chapter 9). Landala Mosse at about +50 m (Sandegren & Johansson 1931: Fig. 42) and Blinkens Mosse at +27.5 m (Thomasson 1934: Fig. 61) are other pollen analysed terrestrial-lacustrine sequences within the type area (Figs. 3: 1 and 3:3). I n the Gotaalv River Valley (with thick Quaternary clay beds) numerous cores have been taken for geotechnical investigations. The cores from Trapiren (Mohren 1945), Hisinge Tunnel (Brotzen 1961, Miller in prep.), and Ingeback (Miller 1964) are of special interest as they have been palaeontologically investigated. The core from Trapiren (Bp 97) was analysed for pollen, diatoms. and forams. The core from Hisinge Tunnel was analysed for pollen and diatoms (as yet unpublished, however). The diatoms in the core from Ingeback were investigated in detail. In all these cores, the depth at the time of the Pleistocene/Holocene boundary was 50 m or more. Compared

8869

8873

B 870

MSL during ALV-1 in the Botanical Garden +16m-

- I

Gothenburg

-

Fig. 4:I.

200 m

150

100

0

-

Extension of the ALV-1 layers of gyttja, clay gyttja and gyttja clay in the low-lying area of the Botanical Garden, and the corresponding mean sea level (hatched line).

PleistocenelHolocene boundary 21 1

BOREAS 5 (1976) Table 4 : l . Stratigraphy of Core B 873.

Layer 1

Depth below surface

Description

to 1.83 m (+15.60)

banded clay gyttja (a normal ‘Tapes clay” in this area).

2

to 1.92 m (+15.51)

3

to 2.24 m (+15.19)

4

to 2.33 m (t15.10)

5

to 2.40 m (+15.03)

6

to 2.56 m (+14.87)

7

to 2.88 m (+14.55)

8a

to 3.35 m (+14.08)

8b

to 5.46 m (+11.97)

9

to 6.50 m (+10.93)

10

to 12.37 m (f5.05)

11

to 13.40 m (+4.03)

X-ray: transparent, (2), banded (1/2). W: 80-100q0, I: 3.2-9.4%, H: 42-57, C: less than 0 .5 6 , and GS: 60.25:6 (c1ay:silt:sand). g y t t p , clayey, with bands of coarse detrital gyttja. X:ray: very transparent (1). W: about 1 1 0 6 , I: 1 4 6 , H: 33, and C: 0.9%. clayey gyttja.

X-ray: transparent (2), clearly denser than layers 2 and 4. W: 105-125%, I: 10.5-13.56, C: 0.6%, and GS: 62:35:3. gyttja, clayey, with coarse detrital gyttja (a piece of wood in sample 10). Distinct boundary to layer 5. X-ray: very transparent (1). W: l15qo, I: 12%, H: 42, and C: less than 0.5%. clay gyttja, with vertical rootlets and ‘hairs’ of R u p p i a maritzma. X-ray: transparent (2), though clearly denser than layer 4. W: 75%, I: 7.5q0, H: 15, C : less than 0.570, and GS: 43:51:6. silty g y t t p clay, a piece of flint (maybe a n artifact) in sample 13. X-ray: quite dense (4). W: 60%, I: 1.8%, H: not measured, and C: less than 0.5%. blue silty clay, (a normal silty ’blue-clay’), penetrated by vertical rootlets surrounded by hard brown oxides forming pipe-like concretions (a piece of wood in sample 16). Very sharp boundary to layer 8a. X:ray: medium dense (3), interveined pattern of rootlets and concretions. W: 50-756, I: 0.54.5%, H: 23-33, C: less than 0.5% and GS: about 50:45:5. very fine-grained blue clay, (a normal ‘blue-clay’ of the Kattegatt Sea), penetrated by vertical rootlets and their pipeconcretions (a piece of wood in sample 20). Th e concretions end abruptly between samples 24 and 25 (the 8a/8b boundary). X-ray: very transparent (I), distinct interveined pattern of rootlets-concretions. W: 77-98%, I: about 4y0, H: 31-37, C: less than 0.56, and GS: 80:19:1 t o 87:12:1. very fine-grained blue clay, (a normal ‘blue-clay’), no rootlets or concretions, increasing content of FeS downwards. Peat material and sand between samples 37 and 38 indicate alluvial washing-in. X-ray: varying from transparent t o medium dense ( 1 4 ) . W: about 95%, I. about 4.04.57G, H: 28:42, C: 0.8-1.66, and GS: 88:12:0 to 90:10:0, (about 70% of fine clay). fine-grained grey clay, rich in FeS, often as distinct bands. Macroscopic and microscopic analyses suggest a somewhat coarser and more organic material than in layer 8b, which is not supported by the grain size and ignition loss analyses, however. According t o du Saar (Chapter 13) this layer is rich in clay pellets. X-ray: medium dense (3) to transparent (2) with denser bands of FeS. W: about 90-95%, I: 4.04.5%, H: about 30, C: 0.6-5.5%, and GS: about 90: 1O:O. varved clay. Each varve consists of a black FeS band, the winter unit (reducing conditions), and a grey part of clay which often contains silt and molluscs, the summer unit (oxidizing conditions). Thin silt layers occur frequently (especially in the lower portion): always in the summer units. Mollusc shells occur quite frequently in the lowermost part and in a few samples in the upper portion. They are always found in the summer units. The varves indicate annual changes between reducing and oxidizing environment. 74 distinct varves were counted (Fig. 4:2). Th e lower boundary is distinct. X-ray: transparent to medium dense (2-31, varved. W: about 1 0 0 ~ GI:, about 4%, H: 2753, and C: 0.8-6.670. microvarved to unvarved grey clay, thin layers and lumps of FeS and with fragments of Mytilus eduhs. In the upper part, 13 normal varves change over to microvarves of about 1.5 mm thickness and finally unrecognizable varves or unvarved clay. Distinct lower boundary. X-ray: quite to medium dense (4-31, numerous black patches and thin layers of FeS. W: 95-606, I: 4.3-3.5%, H: 27-36, and C: 0.6-1.9.

BOREAS 5 (1976)

212 Nils-Axel Morner Layer

Depth below surface

Description

12

to 13.60 m ($3.83)

13

to 14.49 m (+2.94)

14

to 14.60 m (f2.83)

gravelly clayey sand, rich in sheIIs (Chapter 6). Pebbles of amphibolite, gneiss, granite and chert (two pieces) were identified. A carbonate concretion was found above sample 112. X-ray: non-transoarent (5) . , with 57 kw, but medium transparent with 65 kw (Chapter 5). W: about 30% I: about 5% H: about 9, and C: about 12.5%. varved siit and clay, no shells. X-raj: non-transparent (5) with 57 kw, but medium dense with 65 kw and clcarly varved. W: 27-50%, I: 2.7-6.1%, H: 6.16, and C: 5.5-18%,. till on bedrock surface (known from probing).

to the Botanical Garden, they therefore represent deep water conditions (in the middle of the main 'sea-fiord'). Consequently, the Gothenburg area provides sediments covering the Pleistocene/Holocene boundary that represent terrestrial-lacustrine, shallow sea, and deep sea-fiord environment (Fig. 16:3). It is therefore a very suitable world type area for the Pleistocene/Holocene boundary. From the pollen analysis and radiocarbon dates in core B 441 (Morner 1969: Fig. 84), it was known that the Younger DryasIPreboreal boundary in the Angggrden area was to be found shortly below the change from clay to gyttja clay. Furthermore, it was known from work in the Viskan Valley (Morner 1969, 1971f) that there is often a lithological boundary shortly below the beginning of the ALVstage. Because the Younger Dryas Stadia1 is often in bogs and lakes characterized by washing-in of alluvial Dryas Clay (Chapter 2) and this alluvial activity is known also to have affected the fluvial (and coastal) sedimentation (Morner 1971f), it was t o be expected to find a lithologic boundary in Core B 873 representing the Younger DryasIPreboreal boundary. The boundary between layers 7 and S is distinct and pre-dates the ALV-1 stage. It was therefore a priori assumed to represent the Younger Dryas/Prebored boundary and many analyses were concentrated on this level. The results of the various analyses, however, clearly indicated that the real Younger DryasiPreboreal boundary does not correspond to the layers Sa/7 boundary but to the layers 8bi8a boundary. The stratigraphy and chronology of Core B 873 have previously been described and briefly discussed in two preliminary reports, one in connection with the South Sweden Excursion

(Morner 1971f) and one in connection with the report on the findings of the 'Gothenburg Reversed Event' (Morner et al. 1971).

Stratigraphy Core B 873 was taken with the Swedish foil piston corer (of 68 mm diameter) in two sections. The bedrock surface lies 14.5 ni below the ground surface. The uppermost part consists of weathered clay gyttja and artificial filling material. Therefore, the sampling started at 1.5 m depth. Fourteen separate layers were identified. The layers are described one by one from the top downwards. The following abbreviations are used: X-ray= X-ray photography (Chapter 5), W = water content (in 9'0 per dry weight), I = ignition loss, H = hygroscopicity, C = CaCO, (according to Passon's apparatus), and GS grain size (c1ay:silt:sand). Via the varves, the two sections of Core B 873 were correlated with Core B 872 (Fig. 4 2 ) . This correlation demonstrates that 4 varves or 25 cm are missing between the two sections of Core B 873.

Sedimentological analyses One series of samples was used for analyses of water content and ignition loss. The fresh sample was weighed, dried in 105"C, weighed, ignited at 800'C and weighed. Water content and ignition loss were calculated in percentidry weight (Fig. 168). Unfortunately, there were some hydrostatic problems with the balance which seems to have led to too high water content and too low igniticn loss in layer 7 (Fig. 16:s). I n separate series of samples, the hygroscopicity and carbonate content were

PleistoceneiHolocene boundary 213

BOREAS 5 (1976) Fig. 4 2 . Correlated varve diagrams from B 872 and B 873. Thick curves give total thickness and fine curves the thickness of winter units. The correlation indicates that 4 varves or 25 cm are missing between the two sections of Core B 873. The varve in the uppermost part of layer 11 are less distinct (and hence not combined with those of layer 10) and soon change over into microvarves.

i

" V

measured. The carbonate content was measured with the Passon apparatus, which only provides approximate values. In sample 76 from layer 10, the suinmer and winter units of one varve were measured independently:

colour grain size water content ignition loss hygroscopicity carbonate gyttja (microscop. anal.)

snmmer unit

winter

grey

black clay 103.95% 4.550/, 33.7372 2.08% more

(silty) clay 102.19q0 4.25%, 27.99qo 0.5470 less

unit

~

Nineteen selected samples were analyzed inicroscopically as to particle size and gyttja content (omitted list of analysis results are to be found in the original report on pp. 49-50). Most of the stratigraphic boundaries were easily and distinctly identified by this visual macroscopic analysis. The layers 8alSb boundary (samples 24/25) is of special interest as it corresponds to the proposed Pleistocene/Holocene boundary. This boundary is expressed in the microscope analysis as a coarsening in the grain size, an increase in the gyttja content and the onset of admixture of coarse detrital gyttja. The microscopic identification of the gyttja content of the clay layers is more precise than the ignition loss deterniinations (Fig. 16:s): the general increase starts with sample 24 and a slight increase with sample 26. The layers Sb/9 boundary is quite distinct in the microscope analysis, though this is not the case in the X-ray photography (Chapter 5) and the

sedimentological analyses described in Chapter 7.

Zonation and sedimentation rate The chronology and environmental changes of the stratigraphic units are discussed in Chapter 15. Correlations were established with the climatic zone system of the Late Weichselian of southern Scandinavia (Morner 1971d) and the subdivision of the sediments in the Kattegatt Sea (Morner 1971~).Table 4:2 gives the zonation of core I3 873 used throughout this issue (cf. Fig. 15.4). The relation to the receding ice margin, the shorelevel displacement, the bathymetric changes, and the rise of the sediment surface are given in Fig. 3:3. The mean. rate of sedimentation is 3.25 mm/year with higher figures for layers 13 and 10 (being as high as 39.2 mm/yr in layer 10) and a general mean for the other layers of 1.7 rnm/year. A detajled graph of the Sedimentation rate in the upper part of the core is given in Fig. 15:l. This figure demonstrates the precision of the direct and indirect dating of the core. Table 4:2. Zonation of Core B 873 (cf. Figs. 3:3 and 15:4).

Layer

Sample

1-6 7-8a 8b 9 10 11 12 13

1-12 13-24 25-5 1 52-64 65-101 102-109 110-11 1 112-117

Zone (Morner 1971d)

Bo

HOLOCENE

OD

PLEISTOCENE

PB YD AL

BO Fj

AG

214 Nils-Axel Morner

BOREAS 5 (1976)

5. X-ray photography NILS-AXEL MORNER

To decipher sedimentologicaf changes (especially distinct boundaries) and internal structures, the cores were photographed by X-ray. First, however, a suitable method had to be established. Dr. J.-Ch. Linde at the Royal Caroline Institute in Stockholm kindly helped with the establishment of a suitable method and the photography. It was found that ordinary hospital X-ray photography was most suitable. Various voltages, currents, times, and films were tested. The best result was obtained with 57 kw, 8 sec., 80 mA, 640 mAs, and normal film. This method was used for the photography of all core lengths. The coarser sediments of layers 12 and 13 were also photographed with 65 kw, which - in this case - gavz a better penetration. Numerous important details were revealed by the X-ray photography (Fig. 5: 1). The layers 718 and Sa/Sb boundaries show up very well.

Fig. 5:f.

Table 5.1. Transparency of the X-ray photographs (cf. Fig. 163).

Layer 1 2 3 4 5 6 7 8a 8b 9 10 11 12 13

1

transparency scale 2 3 4

5

boundary

L

sharp sharp sharp sharp sharp sharp sharp sharp

1 2 1

2 4

3 1 I

1

2 2 2

3 3 3

4

5 5

sharp sharp sharp sharp

Some X-ray photographies of Core B 873. Figures refer to the different layers. Horizontal lines and arrows mark stratigraphic boundaries.

BOREAS 5 (1976)

The same applies for the varves of layers 10 and 13. Grits, pebbles, and shells are seen. The interveined pattern in layers 7 and 8a corresponds to the formation of pyrite (X-ray det. of a ‘pipe-concretion’) along rootlets of the shore vegetation during the regression maximum of ALV-1.

PleisioceneJHolocene boundary

21 5

I n Table 5:l the transparency of the X-ray photos has been evaluated for each layer according to a tentative scale, where 1 is very transparent, 2 transparent, 3 medium dense, 4 quite dense, and 5 non-transparent (cf. Fig. 16:8).

6. Molluscs and barnacles NILS-AXEL MORNER

The shell fragments were not analysed in detail. However, in connection with the division of Core B 873 into series of samples for different analyses, the presence of molluscs and barnacles was noted. Some shells may therefore have been overlooked. Shells were found in 14 different samples and spare samples (for list of shells per sample, see O.R. 1973:57). Besides Mytilus edulis which occur in most of the 14 samples, Balanus hammeri, Macoma calcarea, Mya truncata, Saxicuva arci‘ica, and Astarte sp. were identified, all occurring in layer 12. Layer 13 does not contain any shells. Layer 12 contains a shell bearing fauna, rich in individuals and species, which rapidly decreases with the onset of layer l l . In the upper part of layer 11 and especially the lowermost part of layer 10, there is a quite distinct increase in numbers of individuals (Mytilus). In the middle and upper part of layer 10, shells were only occasionally found. I n layers 1-9, no shells were found. The fauna assemblage in layer 12 is boreoarctic, which indicates that it is younger than the faunal change in the Kattegatt Sea at 12,700 B.P. (zone LBJAG) described by Morner (1969, 1971e). The frequency changes seem to be directly related to ecoiogical changes. Layer 12 was deposited during the FjarBs Stadial, when the ice stood 5 km east of the Botanical Garden. The heavy melt water outflow from the ice caused a correspondingly heavy reaction cur-

rent bringing in water of high salinity and creating an environment which made such microorganisms as forams and ostracods flourishing, indirectly causing a flourishing of molluscs and barnacles. Rapid chemical and physical weathering of the ‘terminal grades’ of glacial comminution (silt size) into clay meant good supply of minerals, necessary for faunal and floral flourishing. With the onset of the rapid ice recession during the Bolling Interstadia1 (layer 1 l), the reaction currents probably became concentrated to the deep channel of Gota Alv. I n combination with the decrease of the mineral supply (less weathering of ‘terminal grades’), this led to a sudden decrease of forams and ostracods and therefore indirectly also of molluscs and barnacles. At the transition from the Bolling Interstadial (layer 11) to the Older Dryas Interstadial (layer 19), there is a second increase in shells. This was probably caused by increased melt water outflow, leading to increased reaction currents and supply of minerals via weathering of ‘terminal grades’, and therefore indirectly increased microfauna. During the deposition of layers 10 to 8, the environment had become stagnant with great influence of fresh water outflow, a water which was probably unsaturated as regards carbonate. One may therefore suspect that absence of shells in these layers may be apparent and caused by carbonate leaching (cf. Chapter 15).

216 Nils-Axel Morner

m

0

-N

6

BOREAS 5 (1976)

BOREAS 5 (1976)

PIeistocenelHolocene boundary

217

7. Sedimentary petrography F. J. ECKHARDT, B. MATTIAT AND H. RASCHKA

Grain-size analysis

X-ray fluorescence analysis

Thirty-three samples were dried at a temperature of 60°C and were then crushed by means of a jaw crusher with a gap width of 2 mm. The representative share of the sample being 20 g, we removed the organic matter by treatment with 150/0 -H,O, and the CaCO, constituents by treatment with diluted formic acid. After this preparatory treatment the material was washed on a filter and dried again at a temperature of 60-C. 10 g of this material was weighed for the grain-size analysis. In order to reach an optimum dispersion, the samples were twice treated in a 0.01 n ammonia solution for 2 minutes with a 20 kHz ultrasonics coupling oscillator. The grain-size distribution was determined by combined screen and pipette analyses. The boundary between both methods was around 20 ,u @I (grain diameter). Wet vibration screening by means of micro-screen with a screen aperture of 20 ,p was performed. A 0.01 ammonia solution was likewise used as dispersing agent for the pipette analysis. I n the lower section of the profile, approximately to the boundary YDIPB, extremely fine-grained sediments were found. The percentage of the fraction < 2 ,u @ (clay fraction) is in this case between 85 and 90 weight-percent. Only in the sample no. 37/38 does the clay percentage, apparently because of a finesand intercalation, fall back to approximately 74 weight-percent (Fig. 7: I). Starting with sample 24/25, which was taken from the boundary range between the pollen zones YD/PB, an increasing coarsening of the grains is to be observed in this range, though this is not yet macroscopically recognizable (in the drilling core). Below this boundary, extremely calm sedimentation conditions must have prevailed. In the upper part of the PB zone the increase of the coarse silt fraction is particularly conspicuous. In the uppermost part of the profile the clay percentage increases again and shows values ranging between 65 and 700/0,though the extremely fine-grained quality of the lower section of the profile is not reached here.

The element contents were determined with a semi-automatic Philips equipment, type PW 1220. The samples were diluted by means of lithium-ietraborate at a 1:4 ratio and, under addition of Sr as heavy absorber and of Te, Ta and Sr as an internal standard were melted to glass. The relative standard deviation for Na,O is up to 5% rel., for P28, up to 2‘70 rel., and for all the other elements less than 1% rel. The results of the RFA analyses are presented in Table I in the original report (O.R. 1973:62-63). According to the results of the grain-size analyses, also here change of the chemical composition is recognizable slightly above the pollen zone boundary YD/PB. The proportion of organic matter, determined together with CO, and H,Q as ‘ignition loss’. increases considerably towards the top. The CaO and the S O , contents show a slight increase as well. The Fe,03, AI,O, (increases slightly in the uppermost part of the profile), K,O, and MgO contents, however, are distinctly reduced. These alterations are due to the fact that the quantity of the mineral constituents changes according to the increasing frequency of the medium grain size. The percentage of clay minerals decreases at the same time the quartz and the felspar contents increase. The slight increase of the CaO contents may be due to a biogenic lime production

Clay minerals The fraction < 2 ,u @I was separated according to the Atterberg method. Specimens for texture analysis (according to Engelhardt) were prepared and analyzed by means of a Philips X-ray diffractometer and by the application of CuKu-radiation. The clay mineral spectrum remains coinpletely homogeneous over the whole profile; a boundary between YD/PB is not recognizable (Q.R. 1973: Table 2). The main clay mineral is illite, occasionally also illite and muscovite.

218 Nils-Axel Morner Besides these, kaolinite ( + chlorite) occurs. Quartz and felspar occur in minor quantities. The felspar percentage ( - 5010) is comparatively high for the fraction < 2 p @. This indicates that over the total profile the influence of chemical weathering was insignificant. I n addition, polarization-microscopic investigations of the sand fraction of some random samples did not reveal any essential differences between the mineral constituents of the upper and the lower profile section.

Conclusions

BOREAS 5 (1976)

ments starts approximately on the level of the pollen boundary YD/PB towards the overlaying strata. The medium grain size of the clays which were extremely fine-grained so far becomes coarser. Approximately 50 cm above this boundary there is a distinct coarsening of the grain size (corresponding to the layers 8a/7 boundary). I n this upper part of the profile a distinct increase of the water content, of the percentage of organic matter as well as a slight increase of the CaO and the SiO, contents (Fig. 7: l), can additionally be noted.

According to the results hitherto available, it can be said that a gradual change of the sedi-

8. Clay mineral relations DANA1 GEORGALA AND ARVID JACOBSSON

X-ray investigations were performed to obtain information about the mineralogical composition of the clays for the purpose of examining the possibility of using clay mineral variations for determining stratigraphical boundaries. Eight samples were investigated by X-ray diffraction; samples 1, 15, and 22 were of Holocene age and the others of Pleistocene age.

Methods Preliminary X-ray investigations of the untreated clay material showed that better results could be obtained by examining the material < 2 p, which is essentially a concentrate of the clay fraction. This is in accordance with previous practice (e.g. Griffin 1962). The samples were centrifuged 3-4 times to remove soluble salts. The material was rich in iron. This was indicated by the high background noise in the primary diffractograms. However, no attempt was made to remove the iron compounds or the organic material.

The Atterberg method was used in order to separate the < 2 clay fraction. Deionized water was used as the dispersing agent. The < 2 !i fraction was divided into two parts. One fraction was allowed to settle on a glass slide and the other on a quartz plate in order to produce oriented sampies. It was shown that the quartz plates gave rise to undesired high background noise in the primary diffractogram. Quartz plates were used because it was initially intended to heat the samples to over 500°C. The samples were heated to 180°C for one hour to evaporate the primary water, making it easier for the uptake of ethylene glycol. This treatment was used to investigate the swelling characteristics. The x-ray investigations were carried out with a Philips wide-angle diffractometer PW 1050, furnished with a scintillation detector. The radiation used was CoKa at 40 KV and 36 mA. The scanning speed was 0.5"/min. A monochromator with a LiF crystal was used to filter reflected radiation. Time constant 1 and a range of 2.102 C.P.S.with a divergence slit of 1' and a receiving slit 0.1 was used. O

BOREAS 5 (1976)

Results As only a qualitative analysis was required, peak intensity was approximately estimated by measuring the peak heights in mm above a base line. Fig. 3 in the original report (O.R. 1973:72) shows which minerals were found in each sample together with their peak intensities. It shows a hypothetical distribution of every mineral in each profile. The following minerals have been identified: chlorite, mica (illite and/or biotide and/or muscovite), hornblende, kaolinite, felspar, quartz and calcite. Chlorite. - Four chlorite reflections are present (O.R. 1973: Fig. 1). They are the (001) reflection at the 14 A spacing, the (092) reflection at the 7 A spacing, which is close to the (001) kaolinite, the (003) reflection at the 4.72 A spacing, and the (004) reflection at the 3.54 A spacing. The 14 A reflection can be confused with vermiculite. But heat treatment at 180°C displaces the 14.4 A reflection of Mg-vermiculite to 11.6 A and does not effect the chlorite lattice spacings (Brown 1961). The presence of chlorite was confirmed by this method (O.R. 1973: Fig. 2). Mica (illite and/or biotite and/or muscovite). The 10.1 A reflection and 4.98 A may be due to illite and/or muscovite. But since the 10.1 A reflection is very strong and 4.98 8, is weak, it is probably biotite (Brown 1961). A mica reflection occur at 3.34 A, but this is coincident with quartz. Hornblende. - Normally hornblende can be identified by the presence of a reflection at 8.43 A. But montmorillonite when swollen by treatment with ethylene glycol may also produce a 8.43-8.49 A reflection. The presence of hornblende was therefore confirmed by comparing treated and untreated samples (O.R. 1973: Figs. 1-2). Kaolinite. - Two kaolinite reflections are present (O.R. 1973: Fig. 1). These are (001) reflection at the 7.1 A spacing and (002) reflection at the 3.56 8, spacing. The 7.1 A spacing can originate froin either chlorite or kaolinite. The two minerals were distinguished by treating all samples with warm dilute HC1 which

PleistocenelHolocene boundary

219

dissolves chlorite nearly completely when they are fine-grained (Brown 1961). Quartz. - Two quartz reflections are present (O.R. 1973: Fig. I), one (100) reflection at 4.26 A spacing and one (101) at the 3.34 .A spacing. The 3.34 A spacing is coincident with a reflection from mica minerals. Calcite. - Sample 113 contains calcite. This implies that the environment may have been weakly alkaline. Felspar. - Many felspar reflections have been noted in these diffractograms, but no attempt to identify them has been made.

Clay mineral relations Clay mineral relations from the Gathenburg Botanical Gardens Quaternary core are conveniently displayed by the method of Griffin & Parrot (1964) (Figs. 8:l-3). Peak height ratios are compared for given mineral pairs and plotted relative to a base line, in this case the mean value of the peak height ratios for the given mineral pair throughout the core. Fig. 8:l shows the clay mineral provenance in terms of the 148, (chlorite) and 7A (kaolinite) peak height ratios. Since kaolinite and chlorite both have a reflection at 7A it was necessary to separate them either with dilute HCI or by heat treatment. Chlorite dissolves in dilute HC1 whereas kaolinite is insoluble. On the other hand kaolinite is heat-sensitive whereas chlorite is not. The 78, peak height used in the calculation is the kaolinite 7A reflection intensity after removal of chlorite by dissolution in dil. HCI. Fig. 8: 1 demonstrates: (1) a relatively kaolinitic zone with 14A/78, ratios considerably less than 1.7 at depth greater than 3.60 meters. (2) a relatively chloritic zone with 14A/7A ratios greater than 1.7 from 2.10-3.50 meters. (3) a surface zone, relatively more kaolinitic, going down to 2.10 meters. Fig. 5:2 shows the clay mineral distribution in terms of the 14A (chlorite) and 108, (mica) peak height ratios. The calculations distinguish: (1) a mica-rich zone with 14A/lOA ratios considerably less than 0.4, below 3.50 meters.

220 Nils-Axel Morner 1 L A / 7A

peak h e i g h t r a t i o

BOREAS 5 (1976) 1 4 h / 1 0 6 peak height

ratio

10

A/

7 h peak height r a t i o

0.1 9 0.3 0.4 0.5 0.6 07 mica - r i c h chlorite-rich 1-1

........ .......

2-

-'IS 3--22 -28

L-

/ 1. ... .. ..

d . .

5I

150 microns). It shows that calcareous fossils are absent or very poorly represented in the bulk of the samples. Ostracods only occur in samples 106 and 109 of layer 11 (Bolling), sample 111B of layer 12 (Fjaris), and samples 112, 115, and 117 of layer 13 (Agird). The faunal list and species distribution are given in O.R. 1973:118. The faunal list comprises at least 25 species (some larval stages and one internal pyrite mould could not be identified). The fauna distinctly indicates a marine environment. True brackish or fresh-water species are absent. Ostracods are rare in layer 11 (samples 106 and 109). I n the lower part of layer 13 ostracods are also rare (samples 115 and 117) or absent (samples 113, 114, and 116). I n the above-mentioned samples ostracods are too scarce t o be of any value for ecological interpretations; moreover, these ostracods may not

be autochthonous. On the other hand, the rich fauna of sample 112 (the uppermost part of layer 13; the Agfird Interstadial) is very significant for two reasons: (1) it is definitely autochthonous (most valves are preserved very well and adult forms and successive larval stages of a large number of species occur together), and (2) it is a quite typical fauna indicating fully marine (or perhaps very salty brachyhaline), sublittoral, cold environmental conditions. The fauna is certainly not indicative of arctic (frigid) but subfrigid to cold temperature conditions. At present, the southernmost boundary of these faunas is in the Skagerack and near the Shetlands Islands (the southern border of the Norwegian biogeographic province). The shell-rich sand of the F j a r b Stadia1 (layer 12, sample 111B) contains a quite rich marine ostracod fauna, also of the subfrigid to cold temperate type.

14. Palaeomagnetic investigation of Core B 873 and a possible Late Pleistocene reversal of the main geomagnetic field J. HOSPERS, J. P. LANSER, Y . VOLLERS AND N.-A. MORNER

It is customary to divide the geomagnetic reversal time-scale into normal or reversed ‘epochs’ having a duration measured in hundreds of thousands of years, and normal or reversed ‘events’ having a duration measured in tens of thousands of years. This division is essentially arbitrary, but proves to be convenient in several instances, for example, when the last 4 M.y. of the time-scale are discussed. The last, normal, epoch recognized is the

Brunhes normal epoch, which began 0.7 M y . ago and continues to the present. However, the possibility of reversed events being present in the Bruhnes nonnal epoch has been recognized for several years. Evidence of two reversed events in the Brunhes normal epoch has been reported. The older of these two is the Blake event which is about 110,000 years old and has been found by Smith & Foster (1969) in deep-sea sediment

BOREAS 5 (1976)

cores. The younger of these two is the Laschamp event, found in volcanic rocks in France by Bonhommet et al. (1964, 1969). The top of this event is placed somewhere in the age range of 20,000 to 8000 years before present (B.P.). The present paper is concerned with evidence for a relatively young reversed event in the Bruhnes normal epoch, found in sediments in §weden. The authors are well aware that the evidence adduced is still scanty and that further corroboration is required. However, given the potential usefulness of this reversed event in dating and correlating Late Pleistocene sediments, the authors considered it necessary to present the results in the INQUA report (O.R. 1973:120-133).

The Swedish sediments used for palaeomagnetic studies A first short account of this work has been published by Morner et al. (1971). The results are also discussed by Morner & Lanser (1974) and Morner (1976). The sedimentary samples discussed in the present paper are samples of known orientation. The core (labelled B 873) covers a stratigraphic range from Ag%rd Interstadial at the bottom to the Boreal zone at the top and hence represents Late Glacial and Postglacial sediments. On the basis of *4C age-determinations, reviewed by Morner (1971e), the time covered by this core ranges from about 12,500 to about 8600 years B.P. From this core, 117 samples were taken and placed in sealed cylindrical containers with a diameter of 3.1 cm and a height of 3.3 cm (six spare samples from layers 12 and 13 were later added; Fig. 16:4). These samples were obtained by means of the Swedish foil piston corer. All samples are clays.

Palaeomagnetic measurements Some of the samples were removed from their plastic containers and the containers used to check whether or not they were non-magnetic. It was found that the plastic cylindrical containers were non-magnetic, so that the great

PleistoceneJHolocene boundary

235

majority of the samples were measured in their original plastic containers. All measurements were made on an astatic magnetometer, of which the sensitivity was checked daily. The average sensitivity is approximately 5 x 10 - 7 gauss/cm3 (corresponding to approximately 4 x 10-8 e.m.u) total magnetic moment per mm scale deflection. All samples were measured in their natural state and after partial demagnetization in a field-free space and on three mutually perpendicular axes. The partial demagnetization (A.C. cleaning) was carried out at 50 C.P.S. in a field of 200 Oersted peak strength.

Discussion of results Layers 1-6 (incl.) correspond to subdivisions of the Varbergian Formation (Morner 1971~). Layers 7-13 (incl.) correspond to the Gothenburgian and Falkenbergian Formations, the Late Glacial/Postglacial boundary (10,000 years B.P.) being placed in layer 8 between samples 24 and 25. Layer 13 is presumed to represent the Aggrd Interstadial at the bottom of the Falkenbergian Formation. Layer 13 consists of varved marine clay and silt. Sample 1 is dated at about 8600 years B.P., sample 112 at about 12,400 years B.P., and sample 117 at about 12,500 years B.P. The palaeomagnetic results are shown in Fig. 14:l in terms of declination and inclination as a function of equi-spaced sample numbers. As the average inclination is high (approximately + 70" from Fig. 14:1) the inclination is a more reliable indicator of possible reversals than the declination. Fig. 14:l demonstrates that the greater part of all samples has a normal magnetization. However, there is definite evidence of samples 114-1 17 being reversely magnetized, as is apparent both from the inclinations and, to 3 lesser extent, from the declinations. We originally (Morner et al. 1971) placed the proposed reversal between samples 113 and 114, though admittedly it might be situated somewhat higher, between samples 112 and 109. on the basis of the declinations. We interpret the evidence of Fig. 14:l as possibly showing a reversal from reversed to normal, which occurred about 12,400 years B.P. As such, it may well represent the last time that the main geomagnetic field reversed polarity and ac-

BOREAS 5 (1976)

236 Nils-Axel Morner

-

* -m-a -aI

a

03

m

o

1

rn

0

m

B

O

o

e

o

0.

m

a? om

a

0

om

o

0 0

0 4

8 0

m

0

mo .3 5

o

ooom m

om m

i

Fig. 14:l. Palaeomagnetic directions are presented in terms of declinations and inclinations as a function of equi-spaced sample numbers. Both the direction of the natural remanent magnetization (N.R.M.; open circles) and that of the remaining magnetization after A.C. cleaning in fields with a peak strength of 200 Oersted (dots) are shown. Sample 117 was, by mistake, cleaned in an A.C. field of 400 Oersted peak strength. The declinations are shown with respect to a scale which differs by about 8 O from a declination scale based on true geographic North (indicated in the figure). The present-day magnetic declination at the site is 2.5" East of geographic North. Positive inclinations refer t o directions pointing below the horizontal, negative ones to directions pointing above the horizontal. Breakage of the plastic containers have caused some samples to dry and crumble; their numbers have been marked with a cross. Sample 112B was taken just below sample ll2A as an extra sample to replace, if necessary, the sandy sample 112A (other extra samples are not included; see Figs. 16:4 and 16:8).

c.

m o

08

I

L1

m

m

03

0 0

I

W

0

.O

o m m 0

co 8

o m

mo

m .o L,

om

o m

quired its present normal direction. A plot of the virtual palaeomagnetic poles on a stereographic projection corresponding to the palaeomagnetic directions of samples 117-109 shows them to be so much scattered that nothing definite can be said about the mode of reversal. The lower boundary of the reversed event is,

as yet, unknown. After having measured the spare samples of layers 12 and 13, we place the top of the reversal at the Iayer 11/12 boundary i.e. the BoiFj boundary at 12,350 B.P. (cf. Figs. 16:4 and 16:s). It is to be noted that these conclusions are in no way affected by the cleaning process;

BOREAS 5 (1976) INTENSITY

PleistocenelHolocene boundary OF M A G N E T I Z A T I O N 7

IN

237

GAUSS

%

0 0

I

I 1 I I I

D O

m

0

0

0

I0

m

I




goo

'100

I 9930-

2300 6.C

I_

Fig. 17.4. Pollen diagram from Muscotah Marsh, northeastern Kansas, USA (from Griiger 1973: Fig. 3) supplied with a Pleistocene/ Holocene subdivision according to the present author. The pollen zone boundary 3/4a is very distinct and radiocarbon dated at 99301300 yrs B.P. T h e Muscotah Marsh is here proposed as a most suitable hypostratotype for the Pleistocene/ Holocene boundary in terrestrial environment in the Midwest.

BOREAS 5 (1976) Fig. 17.3. The vegetational shift in eastern Minnesota during the last 16,000 years (from Wright 1971: Fig. 9), supplied with the Pleistocene/Holocene subdivision according to the present author. The graph demonstrates that the pollen zones are time transgressive. At 10,000 B.P. all zones are cut by a horizontal line, however. According to the present author this provides confirmation of the validity of the 10,000 B.P. climatic break in North America.

PleistocenelHoIocene boundary B

NORTH

Mixed

2000

conifer

0

w

LOBE

20

Slognonl x e

one whale vertebra and one whale rib. Shells of Hemithyris psittacea have been radiocarbon dated at 10,000* 150 B.P. and shells of Mytilus edulis at 9960+ 150 B.P. The fauna represents the very beginning of the Holocene. Gadd et al. (1972) concluded ‘the abundance and variety of shells suggest warmer, more saline water than that typical of the Champlain Sea environment, which was frigid to cold, and brackish’. The St. Nicolas section, extended into the Pleistocene, may provide a very good marine hypostratotype of the PleistoceneiHolocene boundary. In terrestrial-lacustrine sequences, there are numerous records of a drastic change at 10,000 B.P. Two examples are given here. Rampton’s (1970) diagram from Antifreeze Pond in Yukon Territory (Fig. 17:3) and Griiger’s (1973) diagram from Muscotah Marsh in north-eastern

18” - Boreas 4/76

40

t

SOUTH

forest

Sl C r o i r phase RAINY

259

I 60

--------- ........................ SUPERIOR

80

100

LOBE 120

Rodiocorban date. with one Ilondard d e V m i D n

443

160

0

im

I 20C’rn#k*

Rclectrd rodioCarbon dale

Kansas (Fig. 17:4). Both may well be used as hypostratoypes. Wright (1971: Fig. 9) demonstrated that pollen zones are time-transgressive from south to north across Minnesota. At 10,000 B.P. all pollen zones are cut by a horizontal line, however (Fig. 17:5). This is a perfect confirmation of the validity of the 10,000 B.P. climatic break (a conclusion not drawn by Wright, however). In other land and sea regions of the world, similar hypostratotypes should be established. In several regions, however, there is at present no or too poor information. This does not change the validity of the boundary-stratotype. It only means that future work is needed. The five global parameters in Core B 873 make the Pleistocene/Holocene boundary as defined in Core B 873 globally identifiable and hence give the boundary-stratotype global validity.

BOREAS 5 (1976)

260 Nils-Axel Morner

18. Proposal Core B 873 froin the Botanical Garden in Gothenburg, Sweden, fulfils all stratigraphical rules of a boundary-stratotype and exhibits a boundary that is globally identifiable via various parameters. At the INQUA I X Congress in New Zealand in 1973, we therefore proposed (O.R. 1973:163): (1) that Core B 573 should be chosen as boundary-stratotype for the Pleistocene1 Holocene, and (2) that the Botanical Garden-Angg2rdenAnnedal area in Coihenburg should be chosen as world type area for the PleistoceneiHolocene.

Because there was no radiocarbon date right at the layers Sa/8b boundary, it was, however, decided that the final decision should be postponed until the next INQUA Congress in England in 1977 so that other marine sections in Sweden, where the boundary, possibly, might be directly radiocarbon dated, could be investigated and compared with the Core B 873 records. If Core B 873 will be chosen as boundarystratotype, a small ‘house’ will be built for protection and direct accessibility to the boundary proper.

19. Report from the Holocene Commission From Bulletine 9 of the INQUA Commission for the Study of the Holocene (Hageman 1974) we quote the following regarding the decision in New Zealand t o postpone the final decision of a boundary-stratotype until the next INQUA Congress in England in 1977. ‘In this respect two decisive arguments were brought in. Firstly the fact that C14 dating at the crucial spot of the core was not possible. C14 datings from overlying layers, however, are available.’ ‘The reason why this aspect got such a heavy weight was the following. C14 is to be considered as the tool with the most general use all over the world to establish the age of layers in this specific timespan. A use that can be compared with that of a thermometer. This means that also an exact calibration - such as the freezing point of water is the reproducible reference point of a thermometer - is necessary. Consequently C14 calibration must be able on that specific Pleistocene-Holocene stratotype in order to be sure that the 10,000 C14 years (Libby halftime) - the thermometer “we

use” all over the world - is doubtless related to the standard section. From another point of view, one can say that according to the stratigraphic rules, we had to translate this date into a type section. This section - once established - then takes over the leading position as reference and becomes the highest reference in rank. It might come out later that the translation from the 10,000 C14 years into a stratotype was not correct. Then we have to drop the original leading consideration about the 10,000 C14 years and refer to the date which is represented in the stratotype which then might be an inconvenient one. Therefore the obligatory translation - commanded by the stratigraphic rules - has to be done with utmost carefulness in order to be sure that our really leading principle - the 10,000 C14 years - can be kept intact. I n other words, the most exact translation by means of reliable C14 dating on the spot is the most preferable one. Secondly the type of sedimentation in an estuarine environment makes it plausible that

BOREAS 5 (1976)

also the clayey sediments in the core are slightly reworked. We therefore may expect that the pollen diagram, although the main trend seems to be well represented, is not absolutely reliable in respect to the details. So, fine correlations on local variations are doubtful. Just these local correlations with C14 dated borings in peat layers could be very helpful to weave the proposed core B 873 in a consistent network which perhaps could overcome the objections mentioned before.

PleistocenelHolocene boundary

261

By Professor Fairbridge the following proposal was made: The commission should express its great appreciation to Dr. Nils-Axel Morner for his efforts in preparing material for the proposed Pleistocene-Holocene Boundary Stratotype in SW Sweden. The various Scandinavian authorities should be encouraged to seek comparable sections in the region, which should be considered at the next INQUA meeting.’

20. Additional information obtained in Sweden after the 3973 report NILS-AXEL MORNER

Since the INQUA Congress in 1973, new data have been obtained in Sweden that refer to the Pleistocene/Holocene boundary and the selection of a boundary-stratotype.

Radiocarbon dates from the Billingen area From the quarry at Ranstad on Mt. Billingen located 25 km outside the ice marginal position at the ‘drainage’ of the Baltic Ice Lake Thune (1974) found a bed of laminated silt and sand that contains abundant plant m a c r o f o d s of species that indicate an arctic environment; full tundra vegetation. He obtained two radiocarbon dates of these plant remains: 9980+ 175 B.P. (St-4592) 9935 180 B.P. (St-4593)

+

Consequently, a tundra vegetation - consistent with the conditions during the Younger Dryas Stadia1 - existed on Billingen at about 10,0009950 B.P. This agrees perfectly well with the dates given by Morner in Chapter 15 and elsewhere (Morner 1969, 1970a, 1973b) in opposition to the dates of about 10,300-10,200 B.P. given by Berglund (1971) and Tauber (1970) for the end of the Younger Dryas period.

Isotope records from Cotland A common Holocene sediment on the Island of Gotland (in the central Baltic) is Chara-lime which forms due to the precipitation of CaCO, that is caused by the assimilation of CO, by Chara. A core taken with the Swedish foil piston corer in Lake Tingstade Trask was analyzed as to its content of 0 1 8 and C13 and its paleonlagnetism (MBrner & Wallin 1977). The sediments of the lake had earlier been analyzed as to its content of pollen and diatoms (Lundqvist 1924). The Pleistocene/Holocene boundary is well exhibited as a drastic change from silty clay to gyttja (with 2 crn clayey sand at the contact). The paleomagnetic record (Fig. 20:6) shows a clear intensity maximum right at this boundary; an obvious record of the below-mentioned ‘Galon Magnetic Intensity Maximum’. Fig. 20:l gives the 1 8 0 and 13C variations in core B 911 (closely discussed in a separate paper by Morner & Wailin 1977). The 1 8 0 fluctuations in the organic sequence record the Holocene temperature fluctuations. The 1 8 0 content of the clay (the lowermost four analyses), on the other hand, represents the isotopic composition of the clay particles themselves, i.e. the Silurian ‘mother’ rock. The fluctuations in the clay and gyttja reflect a drastic

262 Nils-Axel Morner

BOREAS 5 (1976) Fig. 20:l. Isotope records from Lake Tingstade Trask on Gotland (Momer & Wallin 1977). Horizontal lines and dots give values of the analyses, through which the 1*O and lSC curves are drawn. The Younger Dryas/Preboreal boundary corresponds to the clay/gyttja contact. The isotope curves indicate that there was a drastic temperature rise right at the clay/gyttja contact. Th e magnetic record and the temperature change in the lower part of the core are given in Fig. 20:6.

environmental (climatic) change at the Younger Dryas/Preboreal boundary. The 13C fluctuations in the Chara-lime have a more complex background. Nevertheless, the 1 8 0 and 13C curves (Fig. 20: 1) record a drastic climatic change at the Pleistocene/Holocene boundary. Morner & Wallin (1977) calculated the corresponding temperature changes. Their temperature curves (Morner & Wallin 1977: Figs. 1516) indicate a rise of about 5°C in mean 'above +3"C' temperature and about 7" in mean summer temperature (of the lake water) during the century 10,000-9900 B.P. (Fig. 20:6) corresponding to a rise in mean air temperature of about 6°C. This drastic temperature rise is directly linked to the clayigyttja change and the magnetic intensity maximum, indicative of the Younger DryasiPreboreal boundary

Palaeomagnetic records of the 'Cilon Magnetic Intensity Maximum' In Chapter 16, it was stressed that the layers 8ai8b boundary in core B 873 corresponds to a drastic palaeomagnetic intensity increase (beginning 15 cm below the boundary which roughly corresponds to 33 years) and that a similar increase was found in core B 892 right at the pollen zone III/IV boundary. It was concluded that this intensity increase (quite independent of the sedimentological variations)

'will certainly be most useful for the global extension of the boundary. Furthermore, it is a perfect complement to the radiocarbon dating which is limited to organic material'. Further palaeomagnetic studies have recorded the same intensity maximum in 15 additional cores, which means that the intensity maximum at the Pleistocene/Holocene boundary is now being found in totally 17 cores (6 marine, 8 Baltic and 3 limnic) or all cores analyzed up to now that cover the time span in question (Table 20:l). Fig. 20:2 shows the geographical location of the palaeomagnetically analyzed cores. The palaeomagnetic intensity in sediments is a combined function of (1) the strength of the Earth's magnetic field, and (2) the number of particles per volume unit that are susceptible of magnetic orientation. Therefore, it is only the relative intensity fluctuations that have a chronological potential for correlations. The intensity maximum at the Pleistocene/Holocene boundary is recorded in all cores studied and is qualitatively independent of sedimentological changes. It is, therefore, considered to be a new and important chrono-stratigraphic marker level. The declination record usually possesses a simultaneous eastward swing reaching its peak close to the intensity peak. Table 20:1 lists the 17 cores in which the intensity maximum at the Pleistocene/Holocene boundary has been recorded, and gives the intensity values of the peak. All cores,

Plei,stocenelHolocene boundary 263

BOREAS 5 (1976)

Table 2U:l. The GBIon Magnetic Intensity Maximum in 17 south Scandinavian cores. Pre-peak and post-peak values refer to the intensity just below and just above the peak proper.

Core

B 873 C A

B B 902 B 926 B 907 B 923 B 925 B 941 B 942 B 908 B 906 MLL-20 B 911 B 903 B 892

Environment

Pleisto-Holocene (YD/PB) boundary

marine marine marine marine marine marine Baltic Baltic Baltic Baltic Baltic Baltic Baltic Baltic lake lake lake

clay clay/some gyttja clay/some gyttja clay/some gyttja clay/some gyttja seismic reflector (varved) clay clay/clay clay/clay clay/gyttja clay/clay varved clay varved clay varved clay clay/gyttja clay/gyttja fine/coarse gyttja

f

898.899

S \

3

Fig. 20.2. Location of palaeomagnetically analyzed cores (cf. Table 20:l).

Magnetic intensityx pre-peak peak

5. 0.5 1.0 0.5 0.7 4 -

3.5 131 3 30 40 (45) 50-30 0.3

0.05

400 30.5 27.0 16.5 11.2 47 161 400 755 32 115 254 875 420 3.7 8.1 0.5

emu post-peak 5. 1.0 1.8 2.8 (1.0) 10.5 5.5-1.5 1.0 21.5

5 1.5 5 10-20 25 0.3-0.1 1.2

-

except five off-shore piston cores ( B 926, 923, 925, 941, 942), were taken with the Swedish foil piston corer, giving undisturbed and continuous cores of 11 m length. The intensity fluctuations in core B 873 are discussed in Chapter 16. The declination values show a distinct eastward swing peaking in sample 27. The intensity rises rapidly from sample 27 to sample 26 where the first peak occurs (with a second peak in sample 22). The inclination curve shows a very low value in sample 23, the chrono-stratigraphical significance of which is not yet known because the other cores have not yet been analyzed as to their inclination. Cores C ( = B.P. 629), A-B ( = B.P. 300) and B 902 (=B.P. 1194) refer to marine cores taken in the Viskan Valley in SW Sweden, where the chronology and stratigraphy are known in detail (von Post 1968; Morner 1969, 1971f). All four cores possess clear intensity peaks at the Younger Dryas/Preboreal boundary. The intensity peaks are linked to distinct eastward declination peaks. The magnetic results are separately discussed by Morner & Backman (1975). Fig. 20:3 shows the intensity record of the lower part of core C. Core B 926 refers to a piston core taken in the Skagerack, NW of Stromstad. The stra-

BQREAS 5 (1976)

264 Nils-Axel Mornei

Core C

Declination (relative)

Intensity

90" I

9

180'

270"

I

I

al C al

-00

I

10

-

Fig. 20:3. Magnetic records from the lower part of Core C. OSF=the Older Sea Fiord and ALV=the Ancient Lake Veselbgen (Momer 1969, 1971f). Declination values are relative. The intensity is plottcd in two scales. The PIeistoceneiHolocene boundary corresponds t o a distinct intensity peak and a simultaneous eastward declination swing and peak.

W

C al 0

c Ln ._

11

al -

a

B 907

Intensity I

I

53

100

Fig. 20:4. Magnetic intensity record from the lower part of Core B 907 (cf. Morner 197Sa: Fig. 6 ) . The stratigraphic column includes the division in Baltic stages.

tigraphic analysis is not yet finished. The magnetic record shows a distinct intensity increase and peak at a depth of about 2.2 m, which is linked to a simultaneous eastward declination shift: a typical record obtained at the Younger Dryas/Preboreal boundary. The magnetic peak (and hence the YD/PB boundary) corresponds to a distinct acusto-seismic reflection surface in the diagram profiles and must correspond to

B 925 5,=]\

'

'-

Intensity 200 I

?m I

'L ,I

&

IWI

a major lithological change caused by the rapid change in glacial influence and climate at 10,000 B.P. (cf. Figs. 2: 1 and 20:2). Core B 907 is a long Baltic core from the Stockholm area (Morner 1975a: Fig. 6). The lowermost part consists of varved clay. The magnetic record shows a distinct intensity peak in the lower part of the varved clay, which corresponds to the boundary between the Baltic Ice Lake and the Yoldia Sea ( = D e Geer's varve -1073, or v a n e 9965 B.P.). The intensity record of the lower part of the core is shown in Fig. 20:4. Cores B 923 and 925 refer to Baltic piston cores taken in the harbour of Karlskrona (SE Sweden). The cores possess perfect magnetic records that allow close inter-core correlations (also core B 924, which stopped in a 9500 57rs old shore deposit) with a very distinct intensity maximum (Fig. 205) at the boundary corresponding to the transition from the Baltic Ice Lake to the Yoldia Sea (i.e. the Younger Dryad Preboreal boundary). In both cores, there is a typical smaller intensity peak shortly below the main peak. This smaller (older) peak is seen in many other cores (B 902, MLL-20, 911 and 903).

Ancylus

Fig. 20:5. Magnetic intensity record from the lower part of Core B 925 ranging from approximately 8500 to 10,300 B.P. The intensity peak occurs right at the sedimentological boundary that corresponds t o the Baltic Ice Lake/Yoldia Sea boundary. Layer B seems t o have been deposited within some years or (at the most) decades.

PIeistocenelHolocene boundary

BOREAS 5 (1976)

B 911

Tern peratu r e

Fig. 20.6.

265

Intensity

Magnetic intensity and temperature records from the lower part of Core B 911. Vertical scale in lo3 years B.P. At the clayigyttja boundary, there is a contemporary magnetic intensity peak; the 'Gilon Magnetic Intensity Maximum'. Th e l 8 0 and 13C records (Fig. 20:1) were recalculated to temperature in OC (Morner & Wallin 1977). Curve A gives the mean summer temperature in the lake (arrow marks the present value). Curve B gives 'mean above + 3 T temperature in the lake (arrow record (B2) respectively. The marks the present value) according to the l 8 0 record ( B l ) and the curves indicate a temperature rise of 5 - 7 T within one century (10,000-9900 B.P.).

ward declination shift (corresponding to a miCores B 941 and 942 refer to Baltic piston cores taken in the Bay of Hanobukten at gration of the magnetic pole along approxidepths of around -80-95 m. Both cores ex- mately latitude 60"N from 148"W Long. to hibit a distinct intensity peak at the boundary 30"E Long.). Cores B 908, 906, and MLL-20 refer to between the sediments of the Baltic Ice Lake and the Yoldia Sea (i.e. the Younger Dryas/ varved clay cores where the varves are carefully counted and correlated to the Swedish Preboreal boundary). Time Scale and hence absolute varve dates can Core B 911 refers to the above-mentioned core (Fig. 20:l) from Lake Tingstade Trask be obtained. The annual and seasonal magnetic (Gotland) which records a drastic temperature variations have been discussed in a separate increase at the clay/gyttja contact of the paper (Morner 1975b). Core MLL-20 from the island of Orno (Fig. Younger DryasiPreboreal boundary. The magnetic record (Fig. 20:6) shows a clear intensity 20:8) includes a complete v a n e sequence rangpeak at the clay/gyttja contact (the layer of ing from varve -1394 to -1030 in the De 2 cm clayey sand records some erosion). There Geer chronology ( = varves 10,286-9922 B.P.). is also one (or two) older peak consistent with At around the varve of the drainage of the the recorded intensity variations from the older part of the Younger Dryas period in other cores (B 902, MLL-20, 903, 923, 925). Core B 903 refers to a limnic core from Lake Soborg S o in Denmark (Morner 1975a: 8Fig. 5). The clay/gyttja contact corresponds to the Younger DryasiPreboreal boundary. The magnetic intensity (Fig. 20:7) records an intensity peak right at the clayigyttja contact (with a n older peak close below). Core B 892 from Bjorerods Mosse (Skkne) 9 corresponds Lo a stratal sequence that is closely pollen analyzed (also radiocarbon dated) by Berglund (1971). The pollen zone IIIIIV boundary corresponds to a stratigraphic change from fine to coarse detrital gyttja. The magnetic record (Morner 1976: Fig. 6) shows a tenfold 10intensity increase right - a t the YD/PB boundary Fig. 20t7. Magnetic intensity record from the lower which is linked to a simultaneous drastic east- part of Core B 903 (cf. Morner 1975a: Fig. 5).

B 903

BOREAS 5 (1976)

266 Nils-Axel Morner

MLL-20, ORNO 1970 Declination ( r e l a t i v e )

Intensity 100XO3WLw I

I

I

10 20 30

I

I

I

LO M 60

I

m

I

I

I

I

180

270

0

90

180

I

I

I

I

I

I

80 90 wo it0 120

c

1GM.1.M.

Fig. 20:8.

Varve and magnetic records from Core MLL-20. Some 300 varves were counted, ranging from varve -1394 to -1030 in the De Geer chronology (=varves 10,286-9922 B.P.). The varves a t around the varve of the drainage of the Baltic Ice Lake are contorted and cannot be properly measured. Within this contorted portion (varves 10,017-9952 B.P.), there is a distinct intensity peak; the ‘GBlon Magnetic Intensity Maximum’. In varves 160-180 (150L-10 varves prior to the drainage), there is a distinct ‘reversed’ declination swing, the ‘Orno Declination Departure’, which is also found in 8 other cores (e.g. Core B 592; Morner 1976: Fig. 6). In varves 140-100 (200+20 varves prior to the drainage), there is a second intensity maximum, found in several of the other cores, too. These exactly varve dated magnetic levels were identified in Core B 873 and used to establish a sharp date of the Pleistocene/Holocene boundary (layers Sa/Sb) in Core B 873 (Fig. 15:l).

Baltic Ice Lake (i.e. varve - 1073 or 9965 B.P.) the varves are contorted. I n this contorted varve sequence, there is a drastic magnetic intensity peak. Because of the contortion, this peak cannot be closer dated than: younger than varve - 1125 ( = 10,017 B.P.) and older than varve - 1059 ( = 9951 B.P.) i.e. within a period of 66 years. I n the lower 140 varves, there is a second intensity maximum, which obviously corresponds to the older peaks seen in cores B 902, 923, 925, 911, 903 and sample 36 of core B 873. Eastward declination peaks occur just below varve - 1059 and in varves 210-215 ( = - 1184-1179 or 10,076-10,071 B.P.) or 106-

111 years before the drainage. In varves 160180 (=varve - 1234-1214 or 10,126-10,106 B.P.), or 141-161 varves prior to the drainage and the YDIPB boundary, there is a very sharp and distinct (‘reversed’) westward declination peak (Fig. 20:8). This peak, called the ‘Orno Declination Departure’ (150f 10 yrs older than the YD/PB boundary), corresponds to similar peaks in the other cores (in sample 33 of core B 873, in cores C, B, 902, 926, 925, 911, and as a distinct departure from the main declination record shortly below the pollen zone IIIiIV boundary in core B 892). If sample 33 in core B 873 is about 1.50 years

PleistocenelHolocene boundary 267

BOREAS 5 (1976) r 7 1 1 5 /

L

1

3

1

2

I

1

1

-

- 900

B 906 -

Magnetic intensity

-800

varves 1 - 7

- 700 - 600

- so0

-LOO

- 300 - 200

- 100

-0

Fig. 20:9. Magnetic intensity fluctuations in the lowermost 7 varves of Core B 906 (Morner 1975b), taken close to Core MLL-20. The varves are closely correlated to the Swedish Time Scale (the entire sequence ranging from -1071 to -988; hence partly overlapping the contorted portion in Core MLL-20). The high values in varves 1-3 (=varve - 1071-1069) represent the top of the ‘Gblon Magnetic Intensity Maximum’.

Table 20:2. Varve dates of the ‘Gllon Magnetic In-

tensity Maximum’ obtained from core B 908 at SkalBker on GBlon (11.5 krn SSE of Stockholm). The intensity maximum occurs right at the drainage varve (+4 varves) of the Baltic Ice LakeiYoldia Sea transition, indicative of the Younger Dryas/Preboreal (Late Glacial/Postglacial) boundary, the equivalent of the Pleistocene/Holocene boundary.

varves in B 908 De Geer’s varves varves in B.P. drainage of B.I.L. duration (varves)

Main peak

High-peak

16 to 38 -1087 to -1065 9979 to 9957 -14 to t 9 23

26 t o 33 -1077 to -1070 9969 to 9962 -4 to +4 8

and sample 36 about 200 years older than the drainage of the Baltic Ice Lake, as the varves and magnetic record of core MLL-20 suggest, a straight sedimentation rate line (of 4.2mm/ yr) through these points would date the layers 8a/8b boundary in B 873 a t 10,000 B.P.; hence backing up and sharpening the dating in Chapter 15 (Fig. 15:l). Core B 906 from the island of Orno (close to MLL-20) includes so thick and coarsegrained vanes of the drainage of the Baltic Ice Lake that the corer did not penetrate them. The core includes 85 varves ranging from varve -1071 to -987 (=9963-9879 B.P.). I n varves 1-3 ( - 1071-1069), there is a very distinct magnetic intensity peak (Fig. 20:9), reaching as much as 875 intensity units. Core B 908 from Skaliker on GAlon (Sodertorn) includes a continuous sequence of 140 varves ranging from varve - 1102 (= 9994 B.P.) t o -962 (=9854 B.P.). I n a core (MLL-19) taken in 1970 for varve chronology, the basal varve sequence at this locality was measured as ranging from varve -1307 to -1116. The magnetic record (Fig. 20:10) of B 908 shows a distinct intensity peak that can be exactly dated by varves. The main peak falls within varves 16-38 (=varve - 1087-1065 or 99799957 B.P.) with the real high-peak within varves 26-33 (=varve - 1077-1070 or 99699962 B.P.). The figures are summarized in Table 20:2. Consequently, the real high-peak corresponds to the drainage of the Baltic Ice Lake i 4 years. Eastward declination peaks occur in varves 31 and 36 ( = - 1072 and - 1068). Because the magnetic intensity peak recorded in the above-mentioned 17 cores is fully recorded and exactly dated in core B 908 from Gilon, this peak is hereby named the ‘Gilon Magnetic Intensity Maximum’ (G.M.I.M.) or simply the ‘Gilon Maximum’ (G.M.). The Gilon Magnetic Intensity Maximum introduces a new tool for detailed regional correlations. The precision of the dating is extremely high (and even far better than any radiocarbon date). The intensity maximum is recorded all over southern Scandinavia (Tables 20:l and 20:3, Fig. 20:2); (1) in the proposed boundary-stratotype (B 873), (2) in marine cores related to well established chrono-stratigraphy and shorelevel displacement (cores A-C and B 902) and a distinct acusto-seismic reflector in Skagerack (B 926), (3) in varve dated

268

B 908 ~~~~

BOREAS 5 (1976)

Nils-Axel Morner Intensity , , ,1y, , ,

, , ,”,

F5;,

Declination ,2po*

, ,

360’

,2y

9r

0

L

t

i

Fig. 20: IQ. Varve and magnetic records from core B 908. The varves range from - 1102 to -962 in De Geer’s Swedish Time Scale, corresponding to varves 9994-9854 B.P. Varve 30 represents the drainage of the Baltic Ice Lake at varve -1073 or 9965 BP. The intensity peak recorded in other cores at the

Younger Dryas/Preboreal boundary, the clay/gyttja contact, the pollen zone III/IV boundary, etc. (Table 20:3), is here precisely varve dated; the main peak falling within varves 16-38 and the high. peak within varves 26-33 or a t the drainage of the Baltic Ice Lake +4 varves (cf. Table 20:2). After the location of this core, the intensity peak at the PleistoceneiHolocene boundary is named the ‘Gilon Magnetic Intensity Maximum’.

Table 20:3. The relation of the Gilon Magnetic Intensity Maximum as identified in 17 South Scandinavian cores to the Younger Dryas/Preboreal (Late Glacial/Postglacial) boundary (i.e. the Pleistocene/ Holocene boundary) as identified by: (1) the clay/gyttja change, (2) the pollen zone III/IV boundary, (3) the drainage of the Baltic Ice Lake (the Baltic Ice LakeiYoldia Sea boundary), (4) a drastic temperature rise (of about 5-7” within a century), (5) a marine seismic reflector, (6) the radiocarbon date of 10,000-9950 B.P., (7) the varve 9965 B.P. (i.e. the drainage varve - 1073), and (8) the sedimentation rate date of about 10,000 B.P. ~~

clay/gyttja pollen zone III/IV drainage B.I.L./Y.S. temperature rise seismic reflector radiocarbon date varve date sedimentation rate

873 C

A

B

* *

* *

* *

* *

902 926 907 923 925 941 942 908 906 MLL20 911 903 892

*

*

*

(*I

*

*

*

*

*

*

*

*

*

*

*

*

*

*

* *

*

*

*

*

*

*

*

*

*

*

*

*

* *

:s

*

BOREAS 5 (1976)

cores related to the drainage of the Baltic Ice Lake (B 907, 908, 906 and MLL-20), (4) in south Baltic cores right at the sediment boundary of the BaItic Ice Lake and the Yoldia Sea (B 923, 925, 941 and 942), (5) in lacustrine cores (B 911, 903, and 892) exhibiting a distinct sedimentological boundary corresponding to the Younger DryasiPreboreal boundary (usually the clay/gyttja contact), (6) in core B 892, which is closely pollen analyzed (and radiocarbon dated), at the zone III/IV (YDIPB) boundary, and (7) in core B 911 where the lSO and 13C records indicate a drastic temperature rise (right a the YD/PB boundary). Consequently, the Gilon Magnetic Intensity Maximum enables very precise correlations (Table 20:3) between the varve chronology (with a precision of a few years), the drainage of the Baltic Ice Lake, the pollen zone IIIiIV (YD/PB) boundary, the sedirnentological clay/ gyttja change (or sudden organic increase) at the Younger DryasiPreboreal boundary, radiocarbon dates of the zone YD/PB boundary (of about 10,000-9950 B.P.), and the new isotope records from Gotland showing a drastic ternperature increase at the YD/PB boundary of about 5-7” C. Chrono-stratigraphically, this magnetic intensity peak is sharper than the pollen zone III/IV boundary (cf. Fig. 2:7) and more precise than a radiocarbon date as it is exactly varve dated (Table 20:2). Its establishment in Core B 873, therefore, means (1) exact dating of the layers 8a/8b boundary, and (2) indirect correlation with all the other levels connected with this magnetic peak (listed in Table 20:3). Furthermore, the sedimentation rate dating (Fig. 15:1), the pollen zonation, and the lithological change (Fig. 16:7) recorded in Core B 873 provides a double check of the dating and correlations via the Gilon Magnetic Intensity. For the final choice of a boundary-stratotype of the Pleistocene/Holocene, the establishment and dating of the Galon Magnetic Intensity Maximum (together with the simultaneous eastward declination peak, the 150 years older Orno Declination Departure, and the 200 years older second intensity peak) means: (1) that core B 873 is even better backed up (than in the original report), that additional parameters can easily be linked to it via the magnetic records, and that it therefore is indeed a suitable boundary-siratotype; or 19 - Boreas 4/76

PIeLtoceneiHolocene boundary

269

(2) that core B 908 (Gglon) - as a new alternative - could be chosen as the boundarystratotype and the varve - 1073 (=9965 B.P.) could be chosen as the boundary proper (representing the drainage of the Baltic Ice Lake and the YD/PB boundary), because the boundary in this core can now be directly transferred via the Gilon Magnetic Intensity Maximum (with a precision of a few years) all over southern Scandinavia, from where other parameters may take over local, regional, and global correlations.

The Gothenburg Magnetic Excursion and Flip Much new information has been obtained about the Gothenburg Magnetic Excursion since the original report (Morner & Lanser 1974; Morner 1976). It is now considered to consist of (1) a period of irregular magnetism from 13,700 to 12,400 B.P., and (2) a fully reversed ‘Flip’ a t 12,400-12,350 B.P. (i.e. during the Fjaris Stadial). The present records of the Gothenburg Magnetic Flip are scattered all over the world (Morner & Lanser 1974). In Sweden, it is now found in some 10 sections, the most important ones of which are the 5 closely dated and correlated cores in Fig. 20: 11 from the Swedish west coast (Morner 1976: Fig. 7). Core B 901 reproduces the original results in a core taken 0.6 m from B 873. Core B 897 from Torsgirden represents a stratal sequence with a detailed malacological zonation (and one radiocarbon date) that is related to the climatic zonation (Morner 1969:167). Core B 896 from Agird represents the type locality of the Agird Interstadial. The stratal sequence includes a detailed malacological zonation (and four radiocarbon dates) with a clear registration of the Fjargs Stadial (Morner 1969:168). Core B 892 from Bjorkerods Mosse represent a stratal sequence which has been closely pollen analyzed by Berglund (1971). I n core B 892, the portion that corresponds to the Oldest Dryas pollen zone of Berglund could be stratigraphically subdivided in the climatic zones Fj, AG, LB, and Vi of Morner (1971d, 1971e). I n all five cores (dated and correlated on other grounds than the magnetic results), there is a reversed ‘Flip’ exactly in the Fj-zone (and

Nils-Axel Morner

BOREAS 5 (1976)

B a73 -

25 krn

70 krn

60 crn

72 krn

8 901

B 097

8 896

B 892

-90

-9 0

-90

-90

0

0-

pe

-

i $'

f

2,

zone

zone

-

5~

zone

-10

10-

'"

-15

Magnetic inclination

* E

in 5 Swedish cores

.-

zone

L

a v .

.

15-

Fig. 20:ll. Inclination records from five Swedish cores (cf. Morner 1976); all dated and correlated on other grounds than the magnetic results. Reversed inclination occurs in the Fj-zone (12,400-12,350 BP) of all cores and represents the Gothenburg Magnetic Flip (with arrows marking the inter-core correlation). Hatched areas represent parts of the cores that are not analyzed. Totally, the analyzed portion range from about 13,500 to 8500 B.P. Core B 892 represents lacustrine and the others marine environment. Reversed inclination has recently been recorded also in core 8 938 from the Baltic (cf. Fig. 20:2).

nowhere else). Fig. 20:ll gives the inclination records and the zonation of the five cores. Consequently, the original reversal in core B 873 has been fully confirmed (and sharpened). This means that the Gothenburg Magnetic Flip is an excellent tool not only for local and regional correlations but also for global correlations. The Gothenburg Magnetic Flip corresponds to the Fjar5s Stadia1 (with correlatives around the Weichselian Ice Cap) and the end of the north European pollen zone Ia or Oldest Dryas, and pre-dates the Pleistocene/Holocene boundary by 2350 years.

of the core consisting predominantly of reworked Tertitary species, in the middle part of the core being rare or absent, and in the upper part of the core from the middle of layer 8b consisting of a Late Quaternary flora. This indicates: (1) that the environment was truly marine at the layers 8a/Sb boundary, and (2) that this part of the core contains very little reworked material (which is t o be expected in this protected, shallow-water environment and even is indicated by the fine clay size particles of layers 8b and Sa, but was brought up as a 'second argument' against Core B S73; see Chapter 19).

Coccolith record from Core B 873

Conclusions

Backrnan (Chapter 20b) has identified coccoliths throughout Core B 873; in the lower part

Important additional data are presented in this chapter, which (1) sharpen the dating control

BOREAS 5 (1976)

PleistocenelHolocene boundary

considerably, (2) make detailed regional correlations on palaeomagnetic grounds both easy and very precise, (3) present new isotope data recording a drastic climatic change at the YD/PB boundary, (4) back up core B 873 as

271

a suitable boundary-stratotype, and ( 5 ) introduce core B 908 as an alternative boundarystratotype (the same might be said about core B 911).

20b. Coccoliths in Core B 873 JAN BACKMAN

Calcareous nannofossils are represented in Core B 873. Both light microscopy (LM), at 1OOOX magnification, and scanning electron microscopy (SEM) techniques have been used to delineate the abundance and composition of the nannoflora. Corrosion of the calcitic coccoliths is pronounced throughout Core B 873, especially above sample 113. Thirteen genera and twelve species were identified but due to strong corro-

sion between 6-9 species remain unidentified. According to Berger (1973), one of the most dissolution resistant forms is Cococcolithus pelugicus. This species is present in fourteen samples. Differences in dissolution resistance may well have ruled the nannofloral composition in most of Core B 873. Only two identified species, C. pelugicus and Cyclococcolithus leptoporus, are living in modern seas. McIntyre & BC (1967) have

Table 20:4. Coccoliths in Core B 873. Figures in columns 4-7 refer to number of species. ~~~

Layer

1 3 4 6 7 8a 8b

9

10

11 12 13

Sample

2 3 7 9 12 14 20 23 27 35 50 54 73 82 95 106 111 113 115

Individuals per 100 views 1 1

~~~

Quaternary Quaternary PreVaria (not t o PreQuaternary identified) Quaternary

1

2 1

3-4

rich rich rich rich rich very rich some some some soms some some some some some some some

5

som:

6

some

1 1

-

1 1 1 1 1 1 1 1

1

2

2

1 1

1 1

I

1

i 1

3 1

1

1 1-10 1-10 10 50 50

1

2

2

1

1 1

2 1 2 1 3 3

1 1

(34) 9 (14) 8 (14)

Small calcite particles

Coccolith floras

IV

111

I1

I

BOREAS 5 (1976)

272 Nils-Axel Momer

measured the present-day maximum temperature ranges of these species. C. leptoporus is distributed between 8 "C-26" C, and C. pelagicus is found in temperate waters between 7°C14°C and has its optimum living conditions within 8°C-10°C. Some Quaternary nannoplankton species are, because of their minute size (diameters around 2 microns), difficult to identify without the use of SEM. One such small coccolith species was found using the LM in seven samples, but unfortunately it was not found using the SEM. Everywhere present it was corroded, but it shows a close affinity t o Emiliunia huxleyi which occurs exclusively in Late Pleistocene and Holocene times. The lowermost two samples, 115 and 113, contain a richly varied fossil flora represented mainly by reworked Palaeogene elements. In both samples, there are four Quaternary - PreQuaternary genera present. Samples 111 to 95 show a decreasing number of individualsiview and only 2 identified and 3 4 unidentified species are present. From sample 82 the number of individuals/view never exceeds 1/100. Between sample 82 and 50 the flora becomes successively poorer in composition and abundance, with one identified and three unidentified species. I n the uppermost ten samples, C. pelagicus and Emiliunia aff. E. huxleyi are the dominant elements within the existing assemblages. Three spenolith species, S. heteromorphus, S. moriformis, and S. belemnos, were represented by one specimen each. Further, four unidentified species were found in these ten samples. One of these unidentified species shows affinity to E. huxleyi, another to Reticulofenestra hesslandii and a third to Cyclococcolitlius neogammation. This indicates that the reworked material in this part of Core B 873 is of Neogene age (and hence indicating another provenance of the reworked material than that of the lower part of the core).

Conclusions Coccoliths are present in Late Quaternary deposits of the Kattegatt Sea. Coccoliths have partly survived the strong leaching processes of calcium carbonate shells in Core B 873. The composition of the existing coccolith assemblages in most of Core B 873 is re-

stricted to more dissolution resistant species. (4) The dominant elements in the upper part of the core consist of two identified species which live in modern seas. ( 5 ) These dominant elements indicate a marine environment at the Pleistocene/ Holocene boundary in Core B 873.

REFERENCES Berger, W. H . 1973: Deep-sea carbonates; evidence for a coccolith lysocline. Deep-sea Research 20, 917-921. Berglund, B. E. 1966: Late Quaternary vegetation in eastern Blekinge, southeastern Sweden. Opera Botanica I2:l. 180 pp. Berglund, B. E. 1971: Late-Glacial stratigraphy and chronology in South Sweden in the light of hiostratigraphic studies on Mt. Kullen, Scania. Geoi. Foren. Stockh. Forh. 93, 11-15. Berglund, B. E. 8: Digerfeldt, G. 1970: A paleoecological study of the Late-Glacial lake at Torreherga, SW Scania. Oikos 21, 98-128. Berglund, B. E., Lagerlund, E., Mohren, E. & Ringberg, B. 1971: Skine and Blekinge. Guide to the South Sweden Excursion, 1971. INQUA Com. Study Holocene and Subcom, Shorelines NW Europe, 1-28. Bonhommet, N. & Bahkine, J. 1964: Sur la presence d'aimantations inversees dans la chaine des Puys. Comptes Rend. 264, 93. Bonhommet, N. & Zahringer, J. 1969: Paleomagnetism and potassium argon age determinations of the Laschamp geomagnetic polarity event. Earth Planet. Sci. Letters 13, 181-190. Broecker, W. S. et al. 1960: Evidence for an abrupt change in climate close t o 11,000 years ago. Am. I . Sci. 258, 420448. Broecker, W. S. & Van Donk, J. 1970: Insolation changes, ice volumes and the OI8 record in deepsea cores. Rev. Geophys. Space Phys. 8, 168-198. Brotzen, F. 1951: Bidrag till de svenska marina kvartaravlagringarnas stratigrafi. Geot. Foren. Stockh. Forh. 73, 57-68, Brotzen, F. 1961: An interstadial (radiocarbon dated) and substages of the Last Glaciation in Sw-eden. Geol. Faren. Stockh. Forh. 83, 144-150. Brown, G. (ed.) 1961: The x-ray identification and crystal structures of clay minerals. hlin. Soc., London. 544 pp. Caldenius, C . 1941: D s svenska lervarven och irtalen f o r istidens slutskede. Y m r r 61, 81-106. Clark, H. C. 1972: Paleomagnetism and late Pleistocene-Holocene sediments, Gulf of Mexico. A m . Ass. Petrol. Geol. Bull. 52, 852. Clark, H. C. & Kennett, J. P. 1973: Palaeomagnetic excursion recorded in latest Pleistocene deep-sea sediments, Gulf of Mexico. Earth Planetary Sci. Letters 19, 267-274. Damuth, U. E. & Fairbridge, R. W. 1970: Equatorial Atlantic deep-sea arkosic sands and ice-age aridity in tropical South Africa. Geol. SOC.Am. Bull. 81, 189206. De Geer, G. 1940. Geochronologia Suecia. Principles. Kungl. Sv. V e t . A k a d . Handl., Ser. 3, Bd 18, N o . 6.

350 pp.

BOREAS 5 (1976) De Jong, J. D. 1965: Opening notes and final resolutions of the subcommittee for the study of the holocene. I N Q U A , Report V l t h Congr., Warsaw, 1961, I , 393-396. Denham, C. R. & Cox, A. 1971: Evidence that the Laschamp polarity event did not occur 13,300-30,400 years ago. Earth Planet. Sci. Letters 13, 181-190. Donner, J. J. 1969: Land/sea level changes in southern Finland during the formation of the Salpausselka endmoraines. Bull. Geol. Soc. Finl. 41, 135-150. Fairbridge, R. W. 1961: Eustatic changes in sea level. Physics and Chemistry o f the Earth 4 , 99-185. Fairbridge, R. W. 1972: Climatology of a glacial cycle. Quat. Res. 2, 283-302. Feyling-Hanssen, R. W. 1964: Foraminifera in Late Quaternary deposits from the Oslofjord area. Norges geol. Unders. 225, 1-383. Feyling-Hanssen, R. W., Jergensen, J. A., Knudsen, K. L. & Andersen, A.-L. L. 1971: Quaternary Foraminifera from Vendsyssel, Denmark and Sandnes, Norway. Bull. geol. SOC. Denmark 21 (2-3/, 67-317. Francis, E. A. 1972: I N Q U A . Report f o r period September, 1969 to December 1971. Mimeographed report, 3 PP. Gadd, N. R., LaSalle, P., Dionne, J.-C., Shilts, W. W. & McDonald, B. C. 1972: Quaternary geology and geomorphology Southern Quebec (Guidebook Excursion A46C44). - 24th I.G.C., Montreal 1972. 70 PP. Griffin, G. M. 1962: Regional clay mineral faciesproducts of weathering intensity and current distribution in the northeastern Gulf of Mexico. Bull. Geol. SOC.A m . 73, 737-768. Griffin, G. M. & Parrot, B. S . 1964: Development of clay mineral zones deltaic migration. Bull. A m . Ass. Petrol. Geol. 48, 57-69. Griiger, J. 1973: Studies on the Late Quaternary vegetation history of northeastern Kansas. Geol. SOC. America Bull. 84, 239-250. Hageman, B. P. 1972: Bulletin 6 from the INQUA Com. Study Holocene, Sept. 1972, 6 pp. Hageman, B. 1974: Bulletin 9 from the INQUA Cam. Study Holocene, April 1974, 103 pp. Hansen, S. 1965: The Quaternary of Denmark. In K. Rankama (ed.): T h e Geological Systems. T h e Quaternary, I , 1-90. Hedberg, H. D. (ed.) 1972: Introduction (and summary) t o an international guide to stratigraphic classification, terminology, and usage. Lethaia 5, 283-295. Hessland, L. 1943: Marine Schalenablagerungen NordBohuslans. Bull. Geol. Inst. Upsala 31. 348 pp. Hyyppa, E., 1963: On the late-quaternary history of the Baltic Sea. Fennia 89, 37-SO. Irving, E. 1964: Paleomagnetism and Its Application to Geological and Geophysical Problems, 357-362. Wiley & Sons. Iversen, J. 1954: T he late-glacial flora of Denmark and its relation to climate and soil. Dan. Geol. Unders., Ser. 2, No. 80, 87-119. Jessen, K. 1935: Archeological dating of the history of North Jutland's vegetation. Acta Archaeol. 51, 185214. Johnson, E. A., Murphy, T. & Torreson, 0. W. 1948: Pre-history of the earth's magnetic field. Terr. Magn. Atmos. Elec. 53, 349-373. J@rgensen, J. A. 1971: Quaternary of Vendsyssel. I n Feyling-Hanssen, R. W., Jergensen, J. A,, Knudsen, K. L. & Andersen, A.-L. L. 1971.

PleistocenelHolocene boundary

273

Knudsen, K. L. 1971: Late Quaternary foraminifera from the Lgkken area. In Feyling-Hanssen, R. W., J@rgensen,J. A. Knudsen, K. L. & Andersen, A.-L. L. 1971. Icnudsen, K. L. 1973: Foraminifera from Postglacial deposits of the Lundegird area in Vendsyssel, Denmark. Bull. geol. SOC.Den. 22 (4), 255-282. Kjellman, W., Kallstenius, T. & Wager, 0. 1950: Soil sampler with metal foils; device for taking undisturbed samples of very great length. Roy. Swed. Geotechn. Inst., Proc. 1, 76 pp. Kukla, J. 1969: Th e cause of the Holocene climatic changes. Geol. en Mijnbouw 48, 307-334. Lundqvist, G. 1924: Utvecklingshistoriska insjostudier i Sydsverige. Sver. Geol. Unders. C-333, 1-129. Lundqvist, G. 1961: Karta over landisens avsmaltning i Sverige. Sver. Geol. Unders. Ba-18. Lyell, C . 1830-1833: Principles o f Geology. London: J. Murray, 3 vol. Mangerud, J. 1970: Late Weichselian vegetation and ice-front oscillations in the Bergen district, Western Norway. Nor. Geogr. Tidsskr. 24, 121-148. McIntyre, A,, & Be, A. W. H. 1967: Modern Coccolithophoridae of the Atlantic Ocean - I. Placoliths and Cyroliths. Deep-sea Research 14, 561-597. Mercer, J. H. 1972: The lower boundary of the Holocene. Quat. Res. 2, 15-24. Miller, U. 1964: Diatom Floras in the Quaternary of the Gota River Valley. Sver. Geol. Unders. Ca-44. 67 PP. Mohren, E. 1945: Nigot om de hydrologiska forhbllandena i Goteborgstrakten vid overgangen mellan senoch postglacial tid. Geol. Furen. Stockholm Forhandl. 67, 249-265. Morner, N.-A. 1969: The Late Quaternary history of the Kattegatt Sea and the Swedish West Coast; deglaciation, shorelevel displacement, chronology, isostasy and eustasy. Sver. Geol. Unders. C-640. 487 pp. Morner, N.-A. 1970a: The Younger Dryas Stadial, Geol. Foren. Stockh. Forh. 92, 5-20. Morner, N.-A. 1970b: Comparison between Late Weichselian and Late Wisconsin ice marginal changes. Eiszeitalter Ggw 21, 173-176. Morner, N.-A. 1971a: Eustatic changes during the last 20,000 years and a method of separating the isostatic and eustatic factors in an uplifted area. Palaeogeogr. Palaeoclim. Palaeoecol. 9, 153-181. Morner, N.-A. 1971b: Eustatic and climate oscillations. Arctic Alpine Res. 3, 167-171. Morner, N.-A. 1971c: Subdivision of the Late Glacial and Postglacial sediments of the Kattegatt Sea. Marine Geology 10, M17-Ml9. Morner, N.-A. 1971d: A Late Weichselian climatic zone system for Southern Scandinavia and related areas. Geol. Foren. Stockh. Forh. 93, 236-238. Momer, N.-A. 1971e: The age of the Agird Interstadia1 and the Fjaris Stadial. Geol. Foren. Stockh. Forh. 93, 786-713. Morner, N.-A. 1971f: The Swedish West Coast. Guide to the South Swedish Excursion, 1971. - INQUA Cam. Study Holocene and Subcom. Shorelines NW Europe, Meeting Sept. 1971. 68 pp. Morner, N.-A. 1973a: 'Postglacial' - A term with three meanings. J . Glaciol. 12, 139-140. Morner, N.-A. 1973b: Climatic changes during the last 35,000 years as indicated bv land. sea and air data. Boreas 2, 33-53.

274 Nils-Axe2 Morner Morner, N.-A. 1974: Sea level variations and climatic fluctuations. Colloque Intern. C N R S N o 219, 135141. Morner, N.-A. 1975a: Climatic cycles. J . interdisc. Cycle Res. 6 , 15-24. Morner, N.-A. 1975b: Annual and inter-annual magnetic variations in varved clay. J . interdisc. Cycle Res. 9, in press. Morner, N.-A. 1976. The Gothenburg Magnetic Excursion. Quat. Res. 7:1, in press. Morner, N.-A. 1977a: Varve chronology on Sodertorn: registration and dating of the ‘drainage’ of the Baltic Ice Lake and correlation with the Finnish varve chronology. Geol. Fdren. Stockh. Forh. 99, in press. Morner, N.-A. 1977b: Paleoclimatic records from South Scandinavia, global correlations, origin and cyclicity. Paleolimnology of Lake Biwa and the Japanese Pleistocene 4, in press. Morner, N.-A., Lanser, J. P. & Hospers, J. 1971: Late Weichselinn paleomagnetic reversal. Nature Phys. Sci. 234, 173-174. Morner, N.-A. & Rickard, D. 1974: Quantitative analyses of eustatic changes and climatic cycles. Colloque Intern. C N R S No. 219, 142-153. Morner, N.-A. & Lanser, J. P. 1974: Gothenburg Magnetic ‘Flip’. Nature Phys. Sci. 251, 408409. Morner, N.-A,, FlodBn,T., Beskow, B., Haxner, H. & Elhammer, A. 1975: Late Weichselian deglaciation of the Baltic. Baltica 6 , 33-51. Morner, N.-A. and Backrnan, J. 1975: Paleomagnetism in old and unoriented core segments. Submitted. Morner, N.-A. & Wallin, B. 1977: A 10,000 year temperature record from Gotland, Sweden. Palaeogeogr. Palaeoclim. Palaeoecol. 21, 113-138. Munthe, H. 1892: Studier ofver Baltiska hafvets quartara historia. Kungl. Svenska Vet. Akad. Handl. 18 (J), 1-20. Munthe, H. 1894: Preliminary report on the physical geography of the Litorina-Sea. Bull. Geol. Inst. Upsala 3, 1-38. Munthe, H., Johansson, H. E. & Sandegren, R. 1924: Goteborgstraktens geologi. In Goteborgstraktens Geologi - Goteborgs Jubileumspublikationer. Goteborg 1924. 180 pp. Nakajima, T., Yaskawa, K., Nutsuhara, N., Kawai, N. & Horie, S. 1973: Very short period geomagnetic excursion 18,000 yr BP. Nature Phys. Sci. 244, 8-10. Nathorst, A. G. 1870: Om nigra arktiska vaxtlamningar i en sotvattenslera vid Alnarp i Skine. Lunds Univ. Arsskr. 7. Nathorst, A. G. 1872: Om arktiska vaxtlamningar i Sk5nes sotvattensbildningar. Ofversigt Kongl. Vet. Akad. Forhandl. 29, 123-142. Neustadt, M. I. 1971: Uber die Lage der Untergrenze des Holozans nach den Vorstellungen der sowjetischen Wissenschaftler. Geol. Foren. Stockh. Forh. 93, 103-116. Niemela, J. 1971: Die Quartare Stratigraphie von Tonablagen und der Riickzug des Inlandseises zwischen Helsinki und Hameenlinna in Siidfinnland. Bull. Geol. Survey Finl. 253. 79 pp. Nilsson, E. 1953: Om sodra Sveriges senkvartara historia. Geol. Fdren. Stockh. Forh. 86, 155-246. Nilsson, E. 1968: Sodra Sveriges senkvartara historia; geokronologi, issjoar och landhojning. Kungl. Vet. Akad. Handl. Ser. 4, Bd 12, No. 1. 111 pp. Nilsson, T. 1935: Die pollenanalytische Zonengliederung der spat- und postglazialen Bildungen Schonens.

BOREAS 5 (1976) Geol. Foren. Stockh. Forh. 57, 385-562. Nilsson, T. 1961a: Grundkurs i kvartargeoiogi (3rd edition). Lunds Univ. 108 pp. Nilsson, T. 1961b: Ein neues Standardpollendiagramm aus Bjarsjoholm in Schonen. Lunds Univ. Arsskr., Avd. 2, Bd 56, No. 18. 34 pp. Nilsson, T. 1964: Standardpollendiagramme und CI4datierungen aus dem Agerods Mosse im mittleren Schonen. Lunds Univ. Arsskr., N.F., Avd. 2, Bd 59, No. 7. 52 pp. Nilsson, T. 1965: The Pleistocene-Holocene boundary and the subdivision of the late Quaternary in southern Sweden. I N Q U A , Rep. V l t h Congr., Warsaw, 1961, 1, 479494. Nordmann, V. 1910: Post-Glacial climatic changes in Denmark. In Die Verunderungen des Klimas seit dem Maximum der Letzten Eiszeit. 11th Intern. Geol. Congress, Stockholm, 313-328. Rampton, V. 1970: Late Quaternary vegetational and climatic history of the Snag-Klutlan area, southwestern Yukon Territory, Canada. Geol. SOC. America Bull. 82, 959-978. Ruddiman, W. F. & McIntyre, A. 1973: Time-transgressive deglacial retreat of polar waters from the North Atlantic. Quat. Res. 3, 117-130. Sandegren, R. 8: Johansson, H. E. 1931: Beskrivning till kartbladet Goteborg. Sveriges Geol. Underrdkning Aa-173. 141 pp. Sauramo, M. 1923: Studies on the Quaternary varve sediments in southern Finland. Bull. Comm. Geol. Finl. 60. 161 pp. Smith, J. D. & Foster, J. H. 1969: Geomagnetic reversal in Brunhes normal polarity epoch. Science 163, 565-567. Stromberg, B. 1971: Isrecessionen i omridet kring Alands hav. Forskningsrapport 10, Geogr. Inst., Univ., Stockholm. 156 pp. Tauber, H. 1966: Danske Kulstof-14 dateringsresultater 11. Medd. dun. geol. Foren. 16, 153-176. Tauber, H. 1970: The Scandinavian varve chronology and C-14 dating. I n Olsson, J. (ed.): Radiocarbon variations and absolute chronology. Proc. Nobel Symp. 12. Almqvist & Wicksell and Wiley & Sons, 173-196. Terasmae, J. 1972: The Pleistocene-Holocene boundary in Canadian context. 24th Inter. Geol. Congr., Sec. J2, 120-125. Thomasson, H. 1934: Boplatsens tidsstallning. In Alin, J., Niklasson N. and Thomasson, H.: StenBldersboplatsen p i Sandarne vid Goteborg. Goteborgs Kungl. Vet. Vitt. Sam. Handl., Femte foljden, Ser A , Bd 4, No. 2, 151-224. Thune, T. 1974: Kvartar stratigrafi och Dryas flora i Ranstad, Vastergotland. Thesis, Dept. Quatern. Res., Univ. Stockholm, 93 pp. Troels-Smith, J. 1955: Characterization of unconsolidated sediments. Dan. Geol. Unders. 4:3, 1. 73 pp. Wagner, F. 1970: Faunas of the Pleistocene Champlain Sea. Geol. Survey Canada, Bull. 181. Walton, W. R. 1964: Recent foraminifera1 ecology and paleoecology. In Imbrie, J. & Newell, N. D. (eds.): Approaches to Paleoecology, 151-237. New York: Wiley and Sons. West, R. G. 1968: Pleistocene Geology and Biology. Langmans, Green & Ltd, 377 pp. Wollin, G., Ericson, D. B. & Ewing, M. 1971: Late Pleistocene climates recorded in Atlantic and Pacific deep-sea sediments. In Turekian, K. K. (ed.): The

BOREAS 5 (1976) Late Cenozoic Glacial Ages, 199-214. Yale Univ. Press. Wollin, G., Ericson, D. B., Ryan, W. B. F. & Foster, J. H. 1971: Magnetism of the earth and climatic changes. Earth Planer. Sci. Letters, 12, 175-183. Wright, Jr., H. E. 1971: Late Quaternary vegetational history of North Amcrica. I n Turekian, K. K. (ed.): The Late Cenozoic Glacial Ages, 425464. Yale Univ. Press.

CONTRIBUTORS The following persons and institutes have contributed to the investigation. Backman, J. (Chapter 20b): Geol. Inst., Univ. Stockholm, Box 6801, 5-113 86 Stockholm, Sweden. Berglund, B. E. (Chapter 9): Geol. Inst., Lab. Quaternary Biology, Univ. Lund, Tornavagen 13, S-223 63 Lund, Sweden. Bjdrn-Rasmussen, S. (Chapter 10): Marinbotaniska Inst., Goteborgs Univ., Carl Skottsbergs Gata 22, S-413 19 Goteborg, Sweden. Eckhardt, F. J . (Chapter 7): Bundesanstalt Bodenforschung, Stilleweg 2, 3 Hannover-Buchholz, West Germany. Fairbridge, R. W . (Chapter 1): Dept. Geology, Columbia Univ., Schmerhorn Hall 604, New York, N.Y. 10027, USA. Feyling-Hanssen, R. F. (Chapter 7): Dept. Micropalaeontology, Aarhus Univ., DK-8000 Aarhus C, Denmark. Georgala, D. (Chapter 8): Geol. Inst., Univ. Stockholm, Box 6801, s-113 86 Stockholm, Sweden. Geyh, M . (Chapter 15b): Niedersachsisches Landesamt Bodenforschung, Stilleweg 2, 3 Hannover-Buchholz, West Germany.

PleistocenelHolocene boundary 275 Hageman, B. P. (administration): Rijks Geol. Dienst, Spaarne 17, Haarlem, The Netherlands. Chairman, INQUA Com. Study Holocene. Hospers, J. (Chapter 14): Geol. Inst., Division Geophysics, Univ. Amsterdam, Nieuwe Prinsengracht 130, Amsterdam, The Netherlands. Hdkansson, S. (radiocarbon dating): Radiocarbon Dating Lab., Univ. Lund, Tunavagen 29, S-223 63 Lund, Sweden. Jacobsson, A . (Chapter 8): Univ. Luleb, Division Soil Mechanics, S-95 1 87, LuleH, Sweden. Knudsen, K. L. (Chapter 7): Dept. Micropalaeontology, Aarhus Univ., DK-8000 Aarhus C, Denmark. Lanser, J . P. (Chapter 14): Geol. Inst., Division Geophysics, Univ. Amsterdam, Nieuwe Prinsengracht 130, Amsterdam, The Netherlands. Maftiat, B. (Chapter 7): Niedersachsisches Landesamt Bodenforschung, Stilleweg 2, 3 Hannover-Buchholz, West Germany. van Montfrans, H . M . (Chapter 14): Geol. Inst., Division Geophysics, Univ. Amsterdam, Nieuwe Prinsengracht 130, Amsterdam, The Netherlands. Mdm e r , N.-A. (Leader, Chapters 2-6 and 1620): Geol. Inst. Univ. Stockholm, Box 6801, S-113 86 Stockholm, Sweden. Raschka, H . (Chapter 7 ) : Bundesanstalt Bodenforschung, Stilleweg 2, 3 Hannover-Buchholz, West Germany. Tooley, M . J . (Chapter 15b): Dept. Geography, Univ. Durham, South Road, Durham, D h l 3LE, England. Vinken, R. (administration): Niedersachsisches Landesamt Bodenforschung, Stilleweg 2, 3 Hannover-Buchholz, West Germany. Secretary, INQUA Com. Study Holocene. Vollers, Y . (Chapter 14): Geol. Inst. Division Geophysics, Univ. Amsterdam, Nieuwe Prinsengracht 130, Amsterdam, The Netherlands.