Scandinavia and the northern Atlantic

2.4.3.a SCANDINAVIA

After the retreat of the glaciers of the last glacial period, which covered Scandinavia, a Baltic ice lake was formed c. 13 kaBP. It existed with some interruptions until c. 10.3 ka BP. It was replaced by the Yoldia Sea, which existed until 9.5kaBP. At c. 9.9kaBP, a saline ingression increased the salinity of the water of this sea. Then, from 9.5 ka to 8 ka BP, the area was covered by a vast freshwater lake (Ancylus Lake), which was created by the retreat of the sea. Finally, from 8kaBP to the present, the brackish inland sea conditions characteristic of the contemporary Baltic Sea have existed (Bjork, 1995).

The first stratigraphical section for the post-glacial periods was established in Scandinavia by the Swedish and Norwegian botanists Rutger Sernander and Axel Blytt (Blytt, 1876; Sernander 1894) on the basis of the pioneering work by the Danish botanists

Heinrich Dau, Japetus Steestrap and Christian Vaupell, who investigated the flora of the bogs in their country. (For a fuller description of the history of the paleo-botanical research in Denmark as well as the detailed description of Denmark's environmental history, on which the following section is based, see Iversen (1973) and Bradley (1999, p. 14).) The Blytt-Sernander stratigraphical column distinguishes four periods evidenced by changes of the flora of the bogs and caused by severe climate changes. The periods are the Boreal (dry), the Atlantic (humid), the Sub-Boreal (dry and warm) and the Sub-Atlantic (humid and cool) (Fig. 2.7). As

Subbor Subatlantique
Fig. 2.7. Holocene geology of Scandinavia.

palynological research and 14C dating methods were developed, further divisions of the stratigraphical column were correlated with the Blytt-Sernander division (Bradley, 1999). A preliminary correlation between this division and that of the Levant (Fig. 2.7) shows clearly that only the long and prominent climate regimes were marked. This is true for the Boreal, which is equivalent to the warmpart of the Neolithic; the Atlantic, which is equivalent to the cold Chalcolithic and possibly the EB; the warm Sub-Boreal, which is equivalent to the MB and Late Bronze Age; and the Sub-Atlantic, which covers the Roman-Byzantine, Crusader and Little Ice Age cold periods. Consequently, although the Blytt-Sernander division is important from a scientific historical point of view, today it is outdated and it is regrettable that it is still being referred to by contemporary researchers.

The boundary between the last glacial period and last interglacial period is determined in Scandinavia according to the retreat of the ice sheet from its southernmost point in mid-Jutland. However, the boundary in pollen diagrams is set with the first appearance of Artemisia, as this is regarded as the first clear indication of a warmer climate. This was found in a section fromLake Bolling to have appeared c. 12.5 BP. Before that, the fauna was that of tundra, characterized by Dryas (D. aven) and dwarf willow; consequently this period is known as the Oldest Dryas. The first tangible warm phase started c. 12.5 ka BP and continued to 12 ka BP. It was characterized by the appearance of Betula (birch). This period is called the B0lling after the place where it was first identified. It seems that the vegetation in Denmark at that time was that of "park tundra", with birch woods in the warm places and tundra in the cold north-facing or damp areas. Moreover, vast stretches of the country were still covered by remnants of the ice cap or "dead ice". From about 12 ka to 11.7 ka BP, a cold short phase followed, the Older Dryas. During this phase, the trees disappeared and tundra conditions became dominant. About 11.7 ka BP, birch and other forest trees reappeared, which indicates a warming up of the climate. This is called the Aller0d period, based on the locality in north Zealand where it was first recognized. In this place, a layer of black organic deposit (gyttja) was observed, containing remnants of forest flora and fauna. The forest was continuous in the southeastern part of Denmark, where July temperatures reached an average of 13-14° C.

About 11 ka BP, a sudden decrease in temperature occurred, an event called the Younger Dryas, which continued until 10.8 ka BP. Again park tundra replaced the forest in the southwest of Denmark, while full tundra conditions prevailed in the northwest. The fall in the average temperature from the preceding period was approximately 3-4 °C. Precipitation during the cold periods was lower than that during the warm periods but soil moisture was higher, because evaporation rates were low. Snowfall during the warmer periods seems to have been moderate, based on the abundance of pollen of flora that cannot tolerate prolonged cover by snow.

Deep-sea sediment cores from the Atlantic show that the polar front shifted as far north as Iceland during the interstadial of 13 ka to 11 kaBP (Roberts, 1989, p. 51-52; Ruddiman and McIntyre, 1981).

The Holocene started c. 10.3 kaBP with a steep rise in temperatures, changing the open landscape of the Younger Dryas into a continuous forest. It began with the rapid spread of juniper scrub, followed by birch, aspen and pine. Evaporation as a consequence of the warm climate and transpiration by the forests caused lowering of the groundwater table, and many shallow lakes became overgrown. At c. 10 ka BP, there occurred a short recession in the warm climatic conditions: the Friesland oscillation. After this short cold period, the climate warmed up again until c. 9 ka BP. This warm period is called the Pre-Boreal in Blytt-Sernander's division and was renamed the Birch-Pine period by Iversen (1973).

At c. 9kaBP, the climate became even warmer and was characterized by forest vegetation dominated by hazel (Corylus) and pine. This is the Boreal period of Blytt-Sernander and the Hazel-Pine period of Iversen. This type of vegetation continued up to c. 8 ka BP, when the forest vegetation became more varied. Among the trees to appear were the lime (Tilia), the oak (Quercus) and the alder (Alnus). This is the Atlantic period of Blytt-Sernander and the Older Lime period of Iversen. It lasted until 5 ka BP. Thereafter, the pollen assemblage started to show the introduction of domesticated plants, as farming communities settled in the region. According to Iversen (1973), the later half of the Older Lime period is believed to be the warmest of the post-glacial periods, with temperatures at least 20 °C higher than today.

The period from 5ka to 2.5kaBP - termed by Iversen the Younger Lime period and correlated with the Sub-Boreal of Blytt-Sernander was still dominated by the lime forest, but a decline in elm and ivy occurred. The climate during this period was assumed by Iversen to have been warm, but the first signs of a decline in temperature are in evidence; however, he does not maintain that the pronounced reduction of the elm was a function of climatic changes - although this was the reason that Blytt and Sernander set the border between the Atlantic and Sub-Boreal divisions at this time. According to Iversen (1973), this disappearance is still a mystery and he attributed it either to Dutch elm disease or, more probably, to the influence of the primitive peasant culture with its domesticated animals.

According to Sernander (1894), the passage from the Sub-Boreal to the Sub-Atlantic was characterized by deterioration in the climate. This happened c. 2.5 kaBP. Iversen (1973) refers to this period as the Beech period. He maintains that this was also the period in which human interference in the ecological balance of the forest became very pronounced, especially in forest clearances.

In addition to stratigraphy based on pollen sections, Iversen also mentioned the raised bogs as climatic indicators. These formed huge sponge-like carpets, composed mainly of the peat sphagnum moss. As this moss is highly dependent on rainwater, which it absorbs through pores in its leaves and stores there, it flourishes during cool and humid periods. Consequently, profiles that show changes from dark peat, rich in heather vegetation, to a lighter one, rich in sphagnum, are indications of climate changes from drier to more humid conditions. Such a boundary sets the dividing line between the Sub-Boreal and the Sub-Atlantic and is also the dividing line between the north European Bronze and Iron Ages.

According to Iversen (1973), the Swedish scientist Granlund observed five such boundary layers, Gh I to GH V, occurring approximately at 0.8 ka, 1.6ka, 2.6ka, 3.2ka and 4.3kaBP, respectively. Of these Gh III was the most widespread. According to Iversen (1973), the Sub-Atlantic deterioration in the climate is also seen in the advance of the glaciers in Alaska, which corresponds with Gh I, II and III.

Aaby (1976) investigated a number of large open peat sections in five raised bogs in Denmark. Although he accepts that past climatic changes are reflected in raised bogs as variations in the degree of decomposition or humification of the peat, he maintains that interpretations are not straightforward. For example, the change into lighter-colored bog, which shows less humification, may be a result either of colder climate or more precipitation, or both. Nevertheless, his general conclusion is that light-colored peat was formed when wet periods were more frequent than dry ones, and vice versa. In addition to the degree of humification, he also investigated the relative distribution of two rhizopod genera, which are also indicators of the water regime of the bogs. He concluded that the bogs indicate long-term cyclic climatic variations, with a periodicity of about 260 years. He recommend that these results should be used to model future climates.

As was already discussed, the Blytt-Sernander climatic divisions determine the stratigraphy suggested by Iversen (1973), although a more critical approach could have shown the limitations of the former to describe the more detailed picture. For example, an examination of the pollen diagrams from Lake B0lling, as well as those from eastern Denmark, presented by Iversen in his book of 1973 clearly shows additional variations on top of those based on the Blytt-Sernander division.

These variations can be clearly seen in the curve of glacier fluctuations in Scandinavia during the last 9000 years (Karlen, 1991, p. 409; Fig. 2.7). The data for the curve were derived from studies of sediments from lakes downstream of small glaciers and radiocarbon dating of organic material found in the sediments. Karlen (1991) found that the mass balance of the glaciers depends on the summer temperatures in Sweden while along the Norwegian coast, it depends on winter precipitation. A correlation was also found between the mass balance of the glaciers and the tree ring records from Sweden. Narrow tree rings correlated with a positive balance and vice versa. Karlen and Kuylenstierna (1996) correlated climate change in Scandinavia, as evidenced by the advance of glaciers and fluctuations in the tree line, with the changes in solar irradiation as evidenced by 514C anomalies. They found that 17 of 19 events of low solar activity coincide with periods of cold climates.

In central southern Norway, a period of glacier advance took place between 9ka and 8kaBP. Another glacial advance was found to have occurred c. 7.5 ka BP (14C date) and another around 1 kaBP (Nesje and Dahl, 1991). Dahl and Nesje (1996) developed a new way to calculate winter precipitation during the Holocene by combining glacier equilibrium-line altitudes and the occurrence of pine trees. They found that the wettest phase was c. 8500-8300 BP. During this period, summer temperature was approximately 1.35 °C warmer than the present. These conditions changed abruptly, within 30-50 years, to a regime dominated by dry winters and summers that were a little warmer than the present. The transition can be correlated with an abrupt change into lighter oxygen isotopes recorded in Greenland ice cores.

Lamb (1984a, p. 234) cites Holmsen, who described the area covered by farming in central Norway. This was more or less unchanged from the Early Iron Age but retreated in places after 400 AD and spread abruptly between 800 and 1000 AD.

Morner and Wallin (1977) analyzed oxygen and carbon isotopes in the carbonate sediments of a lake on the island of Gotland in the Baltic Sea. They converted the isotopes ratios into temperature based on the relation between measured lake temperatures and corresponding isotope composition of the water. Their temperature versus age curve showed minimum temperatures between 10ka and c. 9.3kaBP. Later, until c. 8.5 kaBP, came a period of average temperatures a little above the present. This was followed by a warmer period extending for a few centuries and returning to average at c. 8.2 ka BP. Then came a warmer period to c. 5.8 ka BP. From 5.8 ka to 2.5 ka BP, there was a period of temperatures fluctuating a little above and around the present average. At c. 2.5 ka BP, it became colder until c. 1.2 ka BP, when there was a slight change. At the end of the section, at c. 1 ka BP, a colder trend can be seen.

Morner (1978-79), summarizing data on sea level changes along the northwestern European coasts, concluded that eustatic fluctuations could be discerned in addition to the isostatic rise of Fennoscandia (830 m since the Late Weichselian glaciation). He maintained that the eustatic fluctuations correlate with paleo-magnetism and paleo-temperatures, which suggests, in his opinion, a mutual origin. He identified and dated some 40 shorelines in the Kategat region. Because of tilting of the Fennoscandian block, the effect of the eustatic transgressions decreases inland, whereas the effect of the regressions increases inland. His curves are given in sidereal years, corrected against 14C dates, on the basis of dendrochronology and Swedish varve chronology. He observes a "regression maximum" from9.75 ka to 9.3 ka BP, a distinct marine transgression from9.3 ka to 8.3 ka BP, another distinct regression from 8.3 ka to 8 ka BP and after 8 ka BP, a transgression which reached its maximum at 7.7 ka sidereal years BP. The later oscillations were relatively small.

Ambrosiani (1984), summarizing his investigations on sea levels in Sweden, showed that the almost linear isostatic upheaval during the Post Glacial Period, causing a rise of land height of about 0.5 m per century, slowed down to 0.35 and even 0.25 m during certain periods. Consequently, the retreating shoreline curve assumes a step-like, rather than a smooth, curve. He attributed these changes to climate changes, that is, eustatic changes.

Harrison et al. (1993) used data of changing levels in seven lakes in southern Sweden, and data from Estonia, France and Greece, in a water balance model to quantify the effects of evaporation, as a function of insolation, temperatures and cloudiness, on runoff. The data from southern Sweden showed high water levels at 10 ka BP; a fall from c. 10kato 9.5 kaBP, with a minimum reached c. 9kaBP; and higher lake levels returning between 8.5 and 6 ka BP. Most lake levels were low c. 4 ka BP. Later, there was a general rise, with a relatively low level between 1.5 ka and 1 kaBP.

By analyzing bog formation in southern Sweden, Svensson (1988) also linked cold climate with more precipitation, which caused higher water tables. He observed a change to bog vegetation c. 7 ka BP, coinciding with a rise in the level of many lakes. This continued for about 1000 years and was followed by a period of drying up and humification of the peat layers and at c. 6 ka BP, a dry period evidenced by low lake levels was observed. Humid conditions started again at c. 5 kaBP. During the "Middle Sub-Boreal chronozone", presumably c. 4kaBP, there was a stage of low water levels. About 2.4kaBP, there was another stage of increased humidity. A stage of humification and low lake levels, as a function of a drier climate, was observed c. 1.7kaBP. Another period of low lake levels occurred c. 1.3 kaBP. About 1 ka to 1.2 ka BP, the climate became more humid.

Digerfeldt (1988) presented a similar, though more detailed, section of lake level changes in southern Sweden. His section was based primarily on sediments and their pollen assemblages in Lake Bysjo as well as in other lakes. He observed a distinct low level at c. 9.5ka to 9.2kaBP. This was followed by a humid period c. 7 ka BP, which corresponded to that observed by Svensson (1988). A dry period followed from c. 6.8ka to 6.5kaBP, and a colder and humid period at c. 5 ka BP corresponded with that observed by Svensson. According to Digerfeldt, a drier climate ensued, reaching a maximum from 4.9 ka to 4.6 ka BP. This event should be correlated with the dry event observed by Svensson at c. 4kaBP. This major period of dryness continued until c. 2.9ka to 2.6kaBP. A small peak of wet conditions occurred c. 3.5 ka BP. The other markedly wet period was c. 2.5 BP. The following change, which is also the last one observed by Digerfeldt and which peaked c. 1.5 BP, was dry.

Correlating the paleo-climates of the Levant with the advance and retreat curve of the glaciers of Scandinavia (Karlen, 1991; Fig. 2.8), one can see that during the main cold periods, as for example those of the Chalcolithic period (c. 6.5 ka to c. 5 kaBP), EB (c. 5 ka to c. 4 ka BP), Roman period (c. 2.3 to c. 1.7 ka BP) and the Crusader period and Little Ice Age (c. 1 ka to 0.4kaBP), the Scandinavian glaciers advanced, while during the warm periods like those of the MB (4ka to 3.5 kaBP) and Moslem-Ottoman period (0.4ka to 0.1 kaBP), the glaciers retreated. The same correlation can be drawn for temperatures, precipitation rates and levels of the lakes in Scandinavia, namely that cold humid periods in the Levant corresponded in general with low temperatures, high precipitation and high lake levels in Scandinavia.

Changes in microbiological environments in a lake situated in western Finnish Lapland were linked with changes in humidity and suggested a low phase from 8ka to 4kaBP (Hyvarinen and Albonen, 1994). As this contrasts with data from southern Finland and Sweden the authors suggest that the climate regimes in northern and southern Fennoscandia were different during the lower part of the Holocene.

2.4.3.b THE NORTHERN ATLANTIC

Kellogg (1984) investigated the percentage of sub-polar planktonic foraminifers in cores from the Denmark Strait, between Greenland and Iceland. He found high-frequency fluctuations (periods of 615-784 years) in two principal species of Globigerina. He attributed these changes to variable dissolution rates and to changes in the boundary between the Imringer and East Greenland currents, which are a function of climate changes.

Lamb (1984b) charted the changes in the penetration of polar water southwards from East Greenland into the Atlantic. This influences the surface temperature in the North Atlantic. The sea around the Faeroe Islands, which at present is under the influence of the warm saline Gulf Stream (the average temperature of which is 7.7 °C in this region), seems to have been about 5 °C colder than the average of the twentieth century, during the climax of the Little Ice Age between 1674 and 1704. Lamb (1982) noted that Western Europe encountered the warmest temperatures of the post-glacial phase between 3.1ka and 2.8kaBP (1100-800 BC).

2.4.3.c GREENLAND

Although isotopic analyses of Greenland's ice cores have revolutionized paleo-climatic research, when it comes to the Holocene, one should take into consideration a few important constraints. In the first place, the record from Greenland is that of an area well inside the Arctic belt and, therefore, the effect of minor changes may be blurred. Indeed the isotopic records show that the Holocene was a period of relative stability, with small fluctuations of 518O on the order of ±1-2%c. Moreover, there is little correlation between sites, probably because of local differences in

Fig. 2.8. Paleo-hydrology of the Vistula and Caspian Sea.

accumulation and wind drifting of snow (Bradley, 1999, p. 159) One would also expect that colder periods would be more pronounced in the Arctic region than in the more southern latitudes, the climate of which may be moderated by the tropical belt. A time lag between the start and end of climatic changes in both regions should also be expected. Indeed, a comparison between the Levant curves and Camp Century curve (Dansgaard et al., 1971) shows such discrepancies. Changes at the beginning of the EB, the impact of which were felt c. 5kaBP in the Levant, were observed in Greenland c. 5.3kaBP, if not a little earlier. The same can be said for the impact of MB changes, which started in the Levant c. 4 ka BP but in Greenland were observed c. 3.5kaBP. Also, the Roman cold period, which starts in the Levant c. 2.2kaBP, its peak being c. 2kaBP, reaches its peak in Greenland c. 1.5kaBP. There is, however, a rather good correlation between the two regions for the climate changes after the Roman period.

Porter (1981, 1986) demonstrated a close relation between the pattern of northern hemisphere glacier variations during the last millennium and volcanic aerosol production, as found in Greenland ice cores. He, therefore, maintained that sulfur-rich aerosols, generated by volcanic activity, are a primary factor in forcing climate change on the decade level. Glaciation lags behind the increase in acidity by about 10 to 15 years.

Jennings and Weiner (1996) have analyzed the lithofacies and the benthos foraminifer assemblages from two cores taken from Nansen Fjord, in eastern Greenland. They found evidence for the Medieval warm period between c. 730 and 1100 AD, an early cold interval c. 1370 AD and the severe cold period of the Little Ice Age from c. 1630 to 1900 AD. It is interesting that their record is similar to the 1000 year-long sea ice index of Iceland and, to a lesser extent, to the Crete ice core from Central Greenland.

0 0

Post a comment