The Eemian Interglaciation in Northwestern Germany

Quaternary Research 58, 49–52 (2002) doi:10.1006/qres.2002.2341

The Eemian Interglaciation in Northwestern Germany Gerfried Caspers,1 Josef Merkt, and Helmut M¨uller Nieders¨achsisches Landesamt f¨ur Bodenforschung, Stilleweg 2, D-30655 Hannover, Germany

and Holger Freund Institut f¨ur Geobotanik, Universit¨at Hannover, Nienburger Strasse 17, D-30167 Hannover, Germany Received September 20, 2001

yet been investigated (Hahne et al., 1994). Pollen records at Brokenlande and Rathenow suggest weak climate oscillations (Menke and Ross, 1967; Freund, 2000).

Hundreds of small lakes became filled with Eemian deposits following the retreat of Saalian ice in northwestern Germany. Weakly expressed climate oscillations have been inferred from some local pollen records of Late Saalian sediments. Sea level stood 7 m below the current northwest German sea-level datum (NN) during the interglacial climate optimum and decreased afterward. A uniform vegetational succession of the Eemian mixed forests, correlated with marine isotope stage (MIS) 5e, terminated with the demise of boreal woodlands marked by a steep increase in nonarboreal pollen. This is the onset of the early Weichselian Herning Stade, correlated with MIS 5d. No sharp climate oscillations during the Eemian have been documented. In the ending phase of the interglaciation, the climate deteriorated gradually. Dominant heath and grass tundras point to a substantial decline of summer temperature during the Herning Stade. Sea level then stood at −40 m NN. Sedimentological considerations, supported by counts of partly varved deposits, indicate that the Eemian lasted about 10,000–11,000 yr and the Herning for several millennia.  2002 University of Washington. Key Words: Eemian; Herning; interglacial–glacial transition; northwestern Germany.

EEMIAN INTERGLACIAL

The Eemian vegetational cycle, according to Menke and Tynni (1984), begins with the spread of birch in pollen zone (PZ) I followed by pine–birch woodlands (PZ II), which were quickly displaced by thermophilous deciduous woodlands consisting of pine and mixed oak forest (PZ III). These were succeeded by a mixed oak/hazel (PZ IVa) and hazel, yew, and lime (PZ IVb) forests. A substantial expansion of hornbeam and increasing presence of spruce characterizes PZ V. The climatic optimum was achieved in PZ IVb and PZ V. As the Eemian ended, thermophilous trees retreated and boreal-type woodland of pine, spruce, and fir expanded. Pine is dominant in PZ VI and PZ VII (note that the labeling of pollen zones used by other authors differs, cf. Selle in Fig. 2). At several locations in northwest Germany, lake deposits of the terminal phase of the last interglaciation are overlain by peat. The record of the transition to glacial conditions is frequently distorted or destroyed. Pollen profiles where the final phases are well recorded, e.g., Rederstall (Menke and Tynni, 1984), clearly show that the end of the cycle was characterized by a transitional period, probably some centuries long, during which the climate steadily deteriorated. This view is supported by the δ 18 O curve from Gr¨obern, which shows gradually decreasing values as the interglacial interval ended (Litt et al., 1996). These and other studies (Zagwijn, 1996; Aalbersberg and Litt, 1998; de Gans et al., 2000; Turner, 2000) show no major instability at any time during the Eemian interglaciation. The duration of the Eemian, as palynologically defined, has been determined on the basis of varve counts from Bispingen (M¨uller, 1974), Quakenbr¨uck (Hahne et al., 1994), and Gross Todtshorn (Caspers, 1997). Varves are either absent or poorly defined in the final phases of these lake deposits, so that estimates of the duration had to be complemented by sedimentological and palynological considerations (Fig. 2). Available

C

INTRODUCTION

Research into the last interglaciation in northwestern Europe can be traced back to the late 1800s when the Eem Zone was described by Harting (1874). The Eemian is defined on the basis of marine sediments with warm-water molluscs and terrestrial deposits of comparable age. Uniform vegetation development in this period makes correlation possible over considerable distances. At most sites (Fig. 1), the Eemian begins with calcareous deposits that lie directly on a substrate of Saalian age. The climate of the transition from glacial to interglacial conditions is poorly, or not at all, recorded. The climatic information contained in the 4000, possibly annual, late-glacial laminae underlying the Eemian lake deposit in Quakenbr¨uck (Fig. 1) has not 1

To whom correspondence should be addressed. E-mail: [email protected]. 49

0033-5894/02 $35.00 C 2002 by the University of Washington. Copyright  All rights of reproduction in any form reserved.

50

CASPERS ET AL.

level fell in the Herning to −40 m NN (Zagwijn, 1983; Streif, 1990), leading to increased continentality (Fig. 2b). The Herning was a major lengthy cold-climate episode. In the absence of varved sediments, the duration of the Herning is estimated from available palaeoecological and geological data as several millennia. The Herning correlates with marine isotope stage (MIS) 5d and was followed by the 5800 to 10,000-yr-long Brørup Interstadial (Gr¨uger, 1991), which correlates with MIS 5c.

FIG. 1. Map showing location of Eemian sites in northwestern Germany. Dashed line: Eemian shoreline at maximum sea level. 1: Bispingen; 2: Quaken¨ br¨uck; 3: Gross Todtshorn. From: Quart¨ar-geologische Ubersichtskarte Niedersachsen und Bremen 1:500 000; Hannover 1995.

evidence suggests that the Eemian in central and northwestern Germany had a duration of 10,000–11,000 yr. The margin of error is considered to be no greater than about 1000 yr (Fig. 2a). Evidence of shifts from freshwater to brackish and marine conditions in coastal areas permits the reconstruction of Eemian sea level. At the beginning of the interglacial interval, sea level was 40 m below the current German sea-level datun (NN) and then rose quickly to −7 m NN. This is a much faster rise than in a comparable interval of the Holocene. Saalian ice reached considerably farther south than during the last glaciation. At the end of the interglaciation, sea level had fallen to −20 m NN (Fig. 2b). HERNING STADIAL

A severe climatic deterioration, represented by the demise of boreal woodlands, is known as the Herning Stadial (Menke and Tynni, 1984; Litt, 1994; Freund and Caspers, 1997; Caspers and Freund, 2001). Sediments from the beginning and end of the Herning are typically rich in humus and peat; sands with low organic content are characteristic of the middle Herning. The vegetation cover remained complete, hindering major soil erosion. Heather-rich tundra dominated in the western part of the region and grass in the east. The boundary of these two vegetation types advanced westward as the stadial progressed. This vegetational change implies a substantial decline in temperature, especially in summer. Evidence for a decrease in average July temperature to less than 10◦ C is supported by pollen, macrofossil, and beetle investigations (Freund and Caspers, 1997). Sea

FIG. 2. (a) Chronology and subdivisions of the Eemian and early Weichselian. Solid bars based on varve counts (accurate); broken bars indicate some uncertainty (indicated by bar length); dots indicate estimates based on sedimentological and palynological considerations. (b) Eemian sea-level changes. Rectangles are dated sites: Solid = marine or brackish, open = peat or lacustrine (after Zagwijn, 1983, Streif, 1990, and unpublished data).

EEMIAN INTERGLACIATION, NORTHWESTERN GERMANY

BRØRUP INTERSTADIAL

The Brørup Interstadial in northern central Europe is characterized by development of woodland. Brørup deposits consist of peat and/or mud that can be organic-rich to varying degrees and, occasionally, raised bog or wooded raised bog is recorded. Although the temperature rose, average values for the warmest months were in the range of 14–16◦ C and sea levels did not rise above −40 m NN (Zagwijn, 1983). Within the Brørup Interstadial, a birch phase and subsequent a pine phase are recognized, but thermophilous deciduous trees did not play a role in either of these woodland types. In many pollen profiles, and also in stable isotope curves of deep-sea cores, a climatic deterioration is signaled, which resulted in changed woodland composition but not in woodland demise. As in the case of the Eemian, the transition from Brørup to the following cold phase appears to occur rather steadily in most records. In a high-resolution diagram from Barendorf near L¨uneburg (Lower Saxony; Freund et al., 1997), a phase of slow cooling and oscillating tree pollen representation is clearly recorded, as well as an increase in NAP that signals the beginning of the Rederstall Stadial. CONCLUSIONS

The transition from the Saalian late-glacial interval to the Eemian interglaciation occurred quickly, on the basis of pollen analytical investigations, and with at best weak climatic oscillations. These developments are reflected in central and northwestern Europe by shrub-rich vegetation. The climatic oscillations were short and of insufficient intensity to permit the establishment of woodlands, as occurred in the course of the Weichselian late-glacial interval. On the basis of sedimentological records, and also the δ 18 O curves from marine and ice cores, there is no evidence for a late-glacial phase with severe climatic oscillations. Eemian sequences from central and north western Europe show a fully developed interglacial cycle that lasted ca. 10,000— 11,000 yr, with initially pioneer woodland, followed by deciduous woodland with well-represented thermophilous elements, and finally a boreal coniferous woodland. Because varved sediments relating to the final phases of the interglaciation have not been recorded, the duration of the boreal woodlands is interpolated from the pollen and the sedimentological evidence. However, the maximum error in the estimates for the duration of the Eemian could be ca. 1000 yr, but not as much as several millennia. There is no indication of abrupt or severe climate changes in the available pollen diagrams, nor is there sedimentological evidence for such in the lake and peat deposits. Terrestrial Eemian records can be readily correlated with Eemian profiles from the German and Dutch North Sea coastal areas. Also, sealevel changes closely parallel the course of the interglacial cycle. At the beginning of the interglaciation, sea level was 40 m below

51

present levels and then rose quickly to −7 m NN in the German Bight. This represents a much faster rise than in a comparable period during the Holocene. As the interglaciation ended, sea level dropped to −20 m NN. It is difficult to explain why sea level was already at −40 m NN at the transition from the Saalian glaciation to the Eemian, even though during MIS 6 sea levels were lower than −100 m NN and the Saalian late-glacial interval was short and without pronounced climatic oscillations. The δ 18 O curves from ocean sediments indicate particularly strong ice-melt conditions, i.e., warming, in MIS 5e. The curves show a steep rise, remain high for a time, and then decline but at a slower rate than the initial increase. There are no major oscillations and, hence, there is excellent correspondence between MIS 5e, sea-level changes and the Eemian terrestrial records from northern central Europe. During the Herning Stadial, a decline in July average temperature to <10◦ C is indicated by pollen, macrofossil, and beetle evidence (Caspers and Freund, 2001). This resulted in the demise of coniferous woodlands that characterized the final phases of the Eemian. However, vegetation cover, in the form of heathdominated and later grass-dominated tundra, remained complete and so soil erosion was hindered. For this and other reasons, the usually rather thin deposits of the Herning Stadial do not necessarily indicate that this interval was of short duration. Sea level fell to −40 m NN during the Herning and did not exceed this level again until the Holocene. This fall in sea level (at least −33 m NN was achieved) from the high levels that prevailed for most of the Eemian indicates a large buildup of ice. This required considerable time and serves to emphasize the importance of the Herning as a major climatic oscillation. In the absence of varved sediments, the approximate duration of the Herning is estimated from the available palaeoecological and geological data to have lasted several millennia. Therefore, the Herning Stadial is correlated with MIS 5d, which suggests a substantial climatic oscillation. On the basis of the distinct lower sea-level and the overall palaeoecological evidence, it is not appropriate to include the Herning as a part of MIS 5e. The Brørup Interstadial, which lasted of 5800–10,500 yr, is characterized in northern central Europe by boreal, coniferoustype woodland. Even though average July temperatures were ca. 14–16◦ C, sea level at the North Sea coast failed to rise above −40 m NN. MIS 5c is characterized by a substantial rise in δ 18 O values, which are considerably higher than in MIS 5d but not as high as in MIS 5e. The Brørup Interstadial is correlated with MIS 5c. In summary, the Eemian in northwestern Germany is clearly defined on the basis of vegetation development and low sea level at the beginning and end of the cycle. This interglaciation is estimated to have had a duration of 10,000–11,000 yr. ACKNOWLEDGMENTS Michael O’Connell (Galway, Ireland) translated the final draft of an earlier more-extended manuscript, which was later condensed by George Kukla.

52

CASPERS ET AL.

Comments of two anonymous referees also contributed to improving the text. We gratefully acknowledge all these helpful contributions to this paper.

REFERENCES Aalbersberg, G., and Litt, T. (1998). Multiproxy climatic reconstructions for the Eemian and Early Weichselian. Journal of Quaternary Science 13, 367– 390. Caspers, G. (1997). Die eem- und weichselzeitliche Hohlform von Groß Todtshorn (Kr. Harburg; Niedersachsen)—Geologische und palynologische Untersuchungen zu Vegetation und Klimaverlauf der letzten Kaltzeit. Schriftenreihe Deutsche Geologische Gesellschaft 4, 7–59. [In German] Caspers, G., and Freund, H. (2001). Vegetation and climate in the Early and Pleni-Weichselian in northern central Europe. Journal of Quaternary Science 16, 31–48. de Gans, W., Beets, D. J., and Centineo, M. C. (2000). Late Saalian and Eemian deposits in the Amsterdam glacial basin. Geologie and Mijnbouw 79, 147– 160. Freund, H. (2000). Die Bohrung Rathenow (Landkreis Havelland; Brandenburg)—eine Abfolge limnischer Sedimente vom Saale-Sp¨atglazial, Eem-Interglazial bis in das Weichsel-Fr¨uhglazial. Brandenburgische Geowissenschaftliche Beitr¨age 7, 151–159. [In German] Freund, H., and Caspers, G. (1997). Vegetation und Pal¨aoklima der WeichselKaltzeit im n¨ordlichen Mitteleuropa—Ergebnisse pal¨aobotanischer,—faunistischer und geologischer Untersuchungen. Schriftenreihe Deutsche Geologische Gesellschaft 4, 7–249. [In German] Freund, H., Meyer, K.-J., Caspers, G., and H¨ofle, H.-C. (1997). Pollenstratigraphische Untersuchungen zum Eem-Interglazial und Br¨orup-Interstadial von Barendorf. Schriftenreihe Deutsche Geologische Gesellschaft 4, 157– 174. [In German] Gr¨uger, E. (1991). Late Quaternary biostratigraphy in northern Germany 150,000–15,000 years B.P. Striae 34, 7–14. Hahne, J., Kemle, S., Merkt, J., and Meyer, K.-D. (1994). Eem-, weichsel- und

saalezeitliche Ablagerungen der Bohrung ”Quakenbr¨uck GE 2.” Geologisches Jahrbuch A 134, 9–69. [In German] Harting, P. (1874). De bodem van het Eemdal. Verslogen Koninklijke Akademie Wetenschaffen 2nd Reeks, VIII, 282–290. [In Dutch] Litt, T. (1994). Pal¨ao¨okologie Pal¨aobotanik und Stratigraphie des Jungquart¨ars im nordmitteleurop¨aischen Tiefland—unter besonderer Ber¨ucksichtigung des Elbe-Saale-Gebietes. Dissertationes Botanicae 227, 1–185. [In German] Litt, T., Junge, F. W., and B¨ottcher, T. (1996). Climate during the Eemian in north central Europe—A critical review of the palaeobotanical and stable isotope data from central Germany. Vegetation History and Archaeobotany 5, 247–256. Menke, B., and Ross, P. H. (1967). Der erste Fund von Kieselgur in SchleswigHolstein bei Brokenlande s¨udlich von Neum¨unster. Mit einem Beitrag zur Gliederung des Saale-Sp¨atglazial. Eiszeitalter und Gegenwart 18, 113–126. [In German] Menke, B., and Tynni, R. (1984). Das Eeminterglazial und das Weichselfr¨uhglazial von Rederstall/Dithmarschen und ihre Bedeutung f¨ur die mitteleurop¨aische Jungpleistoz¨an-Gliederung. Geologisches Jahrbuch A 76, 3– 120. [In German] M¨uller, H. (1974). Pollenanalytische Untersuchungen und Jahresschichtenz¨ahlung an der eemzeitlichen Kieselgur von Bispingen/Luhe. Geologisches Jahrbuch A 21, 149–169. [In German] Selle, W. (1962). Geologische und vegetationskundliche Untersuchungen an einigen wichtigen Vorkommen des letzten Interglazials in Nordwestdeutschland. Geologisches Jahrbuch 79, 295–352. [In German] Streif, H. (1990). Quaternary sea-level changes in the North Sea, an analysis of amplitudes and velocities. In “Earth’s Rotation from Eons to Days” (Brosche and Sunderman, Eds.), pp. 201–214. Turner, C. (2000). The Eemian interglacial in the North European plain and adjacent areas. Geologie and Mijnbouw 79, 217–231. Zagwijn, W. H. (1983). Sea-level changes in the Netherlands during the Eemian. Geologie en Mijnbouw 62, 437–450. Zagwijn, W. H. (1996). An analysis of Eemian climate in western and central Europe. Quaternary Sciences Reviews 15, 451–469.