The Ice Age World
Chapter I - Historial Review 2
{page 23}
"Complete" Quaternary climate records on the continents
As already mentioned, the records observed in the Quaternary sediments in various parts of the continents are generally very incomplete as compared with the marine sedimentary record. This is particularly true for the formerly glaciated regions where the glaciers frequently eroded and removed many older sedimentary beds, leaving only fragments of the older record for interpretation. However, there are areas beyond the extent of the ice where fairly continuous records of most of the last 2.5 million years have been recovered. For example, in the Rhine Delta region of Holland, which has experienced fairly continuous subsidence throughout this period, the sedimentary layers have been stacked in a thick sequence. The record presented in Fig. 2-2 was obtained from this region through analyses of cores.

Chinese loess stratigraphy
A most complete record of glaciations has been observed recently from the Chinese loess plateau. Some 40 to 50 eolian loess beds have been recorded, and all late Pliocene and Pleis- tocene glacials and interglacials are probably represented in the stratigraphic sections. Areas of high atmospheric pressure developed over the Tibetan Plateau during the glacial phases, and the resultant increased circulation (including strong katabatic winds) eroded, transported, and deposited the fine-grained, silt-sized rock material, known as loess, in central China. The intervening interglacial phases are recognized as preserved soil horizons formed by weathering in combination with vegetation. Figure 1-20A shows a magnetic susceptibility record observed in loess sections at Xifeng. Note the marked difference in magnetic susceptibility between glacials and interglacials, and how well this record corresponds with the deep-sea oxygen-isotope record. Similar, but not as complete, stratigraphic loess sections have been recorded in central Europe also. Fig. 1-21. Climate fluctuations during the last 150 000 years (three independent records).

A: Vegetation record from northern France (pollen percentages of herbs and shrubs versis trees). (Modified from G.M. Woillard, 1978.)

B: Deep-sea oxygen-isotope record. (Modified from Martinson et al., 1987.)

C: Ice-core oyygen-isotope record (8"0%,, from Vostok Station on the South Pole Plateau (see Fig. 1-24). (Modified from J. Jouzel et al., 1987.)

Red: warm, or relatively warm, phases. Blue: cold, or relatively cold, phases. There are many other "climate" graphs based on vegetation changes and on oxygen-isotope fluxuations in cores from the deep sea, ice sheets, and calcite deposits. These show much the same features, in general, as the ones presented here.

Stratigraphy of calcite deposits
Calcite deposited by calcite-saturated ground water in, for instance, limestone caves and fissures may store surprisingly good and complete records of climate fluctuations through time. The oxygen trapped in the CaCo3, is used to determine the variations in 18O content and thus the temperature fluctuations. Since calcite is well suited for Uranium-series dating, the deposits can usually be dated rather accurately. Figure 1-20B shows a record of oxygen-isotope and temperature fluctua- tions during most of the past 500 000 years.

{page 24}

Glacier ice cores
Very important information about former climate fluctuations is also stored within the beds of snow and ice in the large ice sheets. Long glacier ice cores, some more than 3000 m long, have been recovered from ice sheets on Greenland, on Baffin Island, and in Antarctica. The youngest parts of many cores display the annually accumulated snow/ice layers rather clearly, and it has been possible to count layers back to more than 30 000 years B.P. Carbon trapped as C03 has been used to date core sections younger than 40 000 years, and older sections were generally dated by means of extrapolation methods. All cores have been analyzed for various chemical components which were trapped when the snow was deposited. On the basis of analyses of the oxygen-isotope content, the climatic changes were recorded, and all climate graphs obtained from the cores are strikingly similar. However, the cores analyzed so far cover only about the last 150 000 years. The climate records obtained from the ice cores correspond very well with both the deep-sea records and the other terrestrial records (see Fig. 1-21). Among features which have become strikingly clear through the study of the ice cores are the extremely rapid changes in climate which took place during certain periods.
Glaciations older than the Cenozoic (see Fig. 1-22)
Lithifed glacial tills, and various lithified glaciofluvial and periglacial deposits, collectively referred to as "tillites", have been recorded from all of the earth's major continents. They are of different ages. Best known are the Gondwana-series Talchir tillites of Permo-Carboniferous age. About 280 million years ago India and all the southern hemisphere continents, Australia, Africa, South America, and Antarctica, were joined in the Gondwana supercontinent, which was variously covered by large ice sheets through time. Other well- authenticated tillites of earlier ages lie, for example, in the exposed rocks of the Sahara Desert of Africa. These consist of two separate units of tillite, one about 700 million years old (Precambrian) and the other about 450 million years old (Ordovician). Other 700 million year old tillites have been recorded, for instance, in Canada and Scandinavia. Such tillites are con- sidered to represent large-scale glaciations resulting from global cooling. The cold climate which resulted in large expansions of the glaciers during the last 2.5 million years represents the most recent of such major cold periods.
The Cenozoic Era cooling of the climate, 50 million - 2.5 million years ago; glacier fluctuations in the polar regions
The Cenozoic climate record was originally derived from observations on the continents. For instance, records from central Europe demonstrated that a gradual cooling occurred from about 40 million to about 2 million years ago, when a time of more dramatic climate fluctuations started (see Fig. 1-19). However, these records were incomplete and revealed little detailed information about these climate fluctuations. With the addition of information derived from the deep-sea cores during the last few decades, new and much more detailed interpretations have been made possible. Many cores with a continuous or nearly continuous sediment record of the entire Cenozoic Era have been recovered from all of the world's major oceans, and have allowed the documentation of a very interesting story. Specialists in several disciplines have analyzed the cores. Stratigraphic and climate zonations and graphs have been made based on analyses of, for example, fossil foraminifers, diatoms, radiolaria, coccoliths, oxygen isotopes, organic carbon, biosilicates and ice-rafted rock material. Mostly planktonic species are represented, but some benthic organisms have also been studied. The combined results reveal a rather consistent picture where climate interpretations based on, for instance, foram studies agree with those based on studies of other organisms, oxygen isotopes and organic carbon.

The general climatic trend observed on land has been verified in the deep-sea records, which show a general drop in temperature from the middle Eocene, 45-50 million years ago, to the late Cenozoic Ice Age period, which "started" about 2.5 million years ago (see Fig. 1-19). The early and middle Eocene climate was very warm, and no glaciers existed even in the high-latitude polar regions. The latitudinal temperature gradient, the drop in temperature from the poles towards the equator, was low, as was the vertical temperature gradient of the oceans. The cooling of the climate following the middle Eocene time was accompanied by an increased oceanographic circulation. Both the latitudinal and the vertical temperature gradients gradually increased, and they seem to have increased faster (in steps) during cer- tain intervals. A first prominent step occurred near the Eocene-Oligocene transition, and it corresponds with the first known formation of glaciers in Antarctica. Other steps seem to have occurred about 15, 10, 5, 2.5 and 0.9 million years ago. Causes for the steps have been very actively debated. However, an obviously important factor was the drastic reorganization of the ocean current system during some of these periods. Oceanic gateways were opened and closed as a result of the plate-tectonic- induced migration of the continents, and ma- rine sills/thresholds were lowered and raised. For instance, the lowering of the sill across the Atlantic Ocean between Greenland and Scotland was of immense importance for the flow of the cold deep/intermediate water current between the North and the South Atlantic. Glaciers formed much later in the Arctic than in Antarctica. The approximately 6 million year old glacial deposits in Alaska represent an old Arctic glaciation, but it has been suggested that local Alpine glaciers formed as early as 9 million years ago in some Arctic mountains. This corresponds well with observations in cores from the Arctic Ocean and the North Atlantic, where ice-rafted clasts have been found in marine sediments as old as about 10 million years.

Former forests in Antarctica and in the Arctic
Although the general climatic trend during the last 50 million years was towards a gradually colder climate, there were also times of relative warmth. Such relatively warm periods were recorded in Antarctica during middle Miocene time, about 17 to 15 million years ago, in late Miocene time, and in Pliocene time, 5 to 4.5 million years ago. The Pliocene (or Miocene? Sirius Formation tillites in the Transantarctic Mountains contain lake beds with twigs and leaves of Nothofagus trees found in an area about 5 degrees from the South Pole. This shows that the valleys in the Transantarctic Mountains at that time were more or less forested close to the South Pole, and that valley glaciers pushed into lakes on the valley floors (see Fig. 1-22).

A relatively warm climate existed in the Arctic during Paleocene and Eocene times (67-37) million years ago), and deciduous and coniferous forests covered much of the Arctic coast However, the marked global climatic cooling near the Eocene-Oligocene transition probably affected the vegetation in the Arctic also, and from then on the forest composition gradually changed to become a Taiga and forested tundra in Pliocene time. The open tundra which covers the Arctic coasts today was established as late as near the end of Pliocene time.

Figure 1-23 illustrates the warm conditions in central Europe in Miocene time, about 20 million years ago.

Climatic changes after 2.5 million years ago; formation of mid-latitude ice sheets; the "true" late Cenozoic Ice Age
Both the deep-sea oxygen-isotope record analyses of ice-rafted clasts in the glaciomarine deposits in the Arctic and Antarctica show that a cooling and a considerable expansion of glaciers took place about 2.5 million years ago. Terrestrial glacial deposits of about this age have also been recorded, for example, from Iceland, the Midwest region of USA, South America, showing that mid-latitude sheets had formed. From then on, throughout the late Pliocene and the Pleistocene, the mid- latitude glaciers existed and fluctuated. The amplitudes of the fluctuations increased 0.9 million years ago, when the largest mid- latitude ice sheets formed, supposedly in response to the 100 000 year Milankovitch cycles (see Fig. 1-26). *** During the coldest phases, the true ice ages or glacials, the ice sheets expanded over large areas of North America and northern Europe, and much of the North Atlantic Ocean was covered by sea ice (pack ice) and ice shelves. During the intermediate warmer phases, the interglacials, the climate and glacier conditions were about the same as they are today. These glacier fluctuations are recorded both in the oxygen-isotope graphs and in the terrestrial stratigraphy, which will be discussed later. VMat was the cause o the long-term and the f short-term climate and glacier fluctuations? The ultimate cause for the long-term climate and glacier fluctuations still remains much of a mystery. Over time several hypotheses have been proposed, such as: 1. long-term changes in the interstellar position of the earth; 2. varia- tions in solar activity; 3. variations in atmo- spheric carbon dioxide; 4. changes in ocean cir- culation, caused by the drift of the continents and closing or opening of ocean gateways; and 5. the changing altitude of the major mountain chains, resulting in major changes in the atmo- spheric circulation. The cause for shorter-term climate and gla- cier fluctuations during the late Pliocene and the Pleistocene has also been actively debated. The known fluctuations are shown in Fig. 1-19. Some of the coldest periods, which represent the glacial phases, had mean annual tempera- tures in the order of 10-20'C lower than today in areas beyond the ice sheets in parts of Europe and USA. In parts of the subtropics and continental tropics the temperatures were about 4-7'C lower, and along the marine equator, they were only slightly lower, or no lower, than today. The warmest phases be- tween the cold glacial phases, the interglacials, had temperatures analogous to the present, al- though in the warmest parts of some inter- glacials the mean summer temperatures were in the order of 2'C warmer than today in much of Europe and North America. Ever since it was discovered that the climate had fluctuated significantly, causing alternat- ing glacials and interglacials during the last 2.5 million years, scientists have speculated about the cause for the fluctuations. Numerous theories have been proposed, of which some of the best known suggest the cause to be changes related to: 1. astronomic factors; 2. sun- spot activity; 3. ocean currents; 4. atmospheric composition; 5. volcanic dust or dust from disintegrated meteorites in the atmosphere; 6. surges of the Antarctic Ice Sheet; 7. rising and falling of parts of the earth's crust; 8. solar- terrestrial magnetic coupling; 9. fluctuations of the upper atmospheric jet-streams; 10. asteroid impacts; and 11. interaction between ocean currents and atmospheric circulation. In addi- tion, the recently formulated "snowblitz" theo- ry s