ES 331/767 Lecture 11


James S. Aber

Table of Contents
Introduction Oxygen isotopes
Palynology Paleofauna
Related sites References


No method for directly measuring ancient climate exists; climate is not fossilized in rocks or sediments. Fortunately many kinds of climatic indicators are preserved for us to interpret. Various physical or biological processes are dependent on such aspects of climate as temperature, moisture, and seasonality. All attempts to reconstruct Quaternary climate are subject to some uncertainty and must be accompanied by age dating in order to be of any use. Some of the most valuable methods are stable isotopes, palynology and paleofauna.

World Data Center for Paleoclimatology.

Oxygen-isotope ratios

Oxygen occurs in two common, stable isotopes, 16O and 18O, of which 16O is the most abundant. The ratio of these two isotopes in water is temperature dependent and follows predictable geographic trends in the oceans, atmosphere, and glaciers--see Fig. 11-1. The ratio is altered whenever water undergoes a phase change. When sea water evaporates, for example, the heavier isotope 18O is preferentially left behind in remaining sea water, while the resulting water vapor is depleted in 18O. The oxygen-isotope composition of a water sample is expressed in delta (d) units per mil (1‰ = 0.1%) of relative concentrations with respect to the ratio of standard mean ocean water (SMOW).

         18O/16O sample - SMOW
     d = --------------------- x 1000

By definition, d is zero for standard mean ocean water. A value of d = -10 thus means the sample has an 18O/16O ratio 10‰ (or 1%) less than SMOW. The amount of water or ice required for analysis is small (5-10 g), and stable isotopes can be measured quite accurately using mass spectrographic techniques.

Under present conditions, the volume of land ice is relatively small, and this ice has d values around -30. During glacial periods, however, much isotopically light water was removed from oceans and stored in glaciers on land. This caused slight enrichment of 18O in sea water to about d = +1.5, while glacier ice had even lower d values of around -40--see Fig. 11-2. The oxygen-isotope values during past glaciations are preserved in glacier ice and in fossils buried on the sea floor. These isotopic records are primarily a measure of changing volume of glacier ice.

Oxygen-isotope ratios are also affected by temperature, for example, the water temperature in which corals grow or the air temperature in which snow crystallizes. The temperature effect is beautifully illustrated by annual oxygen-isotope layering in the Greenland Ice Sheet--see Fig. 11-3.

Coral reef, Palau Archipelago (Micronesia). Colonies of various branching coral species (Acropora) are visible just below the water surface. The oxygen-isotopes in coral skeletons are a record of water temperature. © World Data Center-A for Paleoclimatology, Educational Slide Project.

The long-term record of oxygen isotopes in the Greenland Ice Sheet extends back more than 250,000 years--see Fig. 11-4. The ice cores come from the Summit on the ice sheet's highest point; the cores reached the base of the ice sheet at > 3000 m depth. These ice cores represent the longest, most detailed record of northern hemisphere climate during the latter portion of the Quaternary Epoch. Surprising variability is evident for both glacial and interglacial episodes, in which significant short-term excursions took place frequently.

GReenland Icecore Project--see GRIP.
Greenland Ice Sheet Project II--see GISP2.

The last glaciation shows an asymmetric development with slow ice buildup following the Eemian. Maximum glacial conditions existed between 30,000 and 20,000 years ago. The change from full glacial to modern climate took place with two major jumps about 14,500 (Bølling) and 11,500 (Younger Dyras) years ago. The termination of the Younger Dryas was especially rapid. During an interval of no more than 50 years, the climate of Greenland abruptly warmed by an average 7°C (Dansgaard 1994).

This record reflects slow buildup of global glaciers and gradually colder air temperature followed by sudden collapse of ice sheets, return of glacier melt water to the sea, and warming of the atmosphere. Evidence from the base of ice cores in Greenland and Arctic Canada suggests that the Greenland Ice Sheet underwent extensive (or possibly complete) melting during the last interglaciation (Eemian), around 120,000 to 130,000 years ago (Koerner 1989).

Map of the Greenland Ice Sheet showing the Summit. GRIP and GISP2 ice cores have been drilled there, an area of thickest ice with minimal ice movement. All images © World Data Center-A for Paleoclimatology, Educational Slide Project.
The GISP2 drilling dome on the ice surface. The dome is about 105 feet (32.5 m) in diameter and encloses the lower part of the drilling tower. The dome is connected to nearby surface and buried workshops and living quarters.
Technician cleans the cutting head of the drill. Just above the cutting head, a drill barrel will hold the ice core. The equipment shown here is for cores 10 cm in diameter, the standard for most ice drilling. At GISP2, 13.2 cm cores were drilled.
Photomicrograph in cross-polarized light of ice sample from 333 m depth in the GISP2 core. Mid-sized crystals show homogeneous ice that has not been sheared by ice flow. Small bubbles appear as round inclusion, for example in the orange crystal near center. Such bubbles preserve the ancient atmosphere--fossil air.
Record of electrical conductivity from the GISP2 ice core, Summit, Greenland. The record shows annual variations for a portion of the mid-1600s. Electrical conductivity is especially sensitive to ice acidity. The major peaks represent volcanic eruptions known from the historical record. Such eruptions emit large quantities of sulfuric acid (H2SO4) into the atmosphere.

An even longer ice-core record has been retrieved from Russia's Vostok station in Antarctica (Chamot 1998). Since 1972, Russian scientists have been drilling into the ice sheet. They were joined by French and American scientists in collecting 3.7 km of core samples representing 420,000 years of glacier history. Drilling was completed in January, 1998, and the task of studying the ice core continues. Isotopes and chemical constituents are being analyzed for 2H, 1H, 18O, 16O, CO2, CH4, CH3SCH3 (dimethyl sulfide), wind-blown dust, and the remains of frozen micro-organisms. The Vostok ice record covers four glacial-interglacial cycles and represents perhaps the most detailed and complete evidence for climatic change for this entire period anywhere.

The location of Vostok Station and other research stations in Antarctica. Vostok was chosen by the Soviet Union in 1980 for deep ice drilling. The site was subsequently cooperatively operated by Russian, American, and French scientists. The bottom ice at Vostok is estimated to be half a million years old. All images © World Data Center-A for Paleoclimatology, Educational Slide Project.
Drilling site at Vostok Station, Antarctica. Ice-core drilled at this site in the 1980s gave the first long record of past atmospheric composition over the last 160,000 years.
Vostok ice core record of atmospheric CO2, temperature change, and methane (CH4). This record shows a strong correlation between greenhouse gases and temperature during the last 160,000 years. However, it is not clear what mechanisms are responsible for this relationship.
Atmospheric CO2 record from Vostok ice core and other sources. The record matches closely with paleoclimatic indicators. Recent CO2 levels are higher than any previous values in the ice-core record.

A much longer record of past glaciations is found in sediments of the deep ocean basins. Oxygen-isotope ratios of foraminifera microfossils are the basis for recognizing oxygen-isotope stages, which are dated paleomagnetically--see Fig. 11-5. Twenty-three stages are numbered for the last 900,000 years. Even numbered stages (high d values) indicate glacial conditions; odd stages (low d values) represent interglaciations. The only exception to this pattern is the last major glaciation (Wisconsin/Vistulian), which includes stages 2-4.

Marine oxygen-isotope stages have become the global standard for interpreting Quaternary paleoclimate. Nearly all other types of paleoclimatic and stratigraphic records are tied to oxygen-isotope stages via various dating techniques. The oxygen-isotope record extends back to cover the entire Quaternary. Stages 1-23 exhibit large differences between glacial and interglacial conditions with pronounced glacial stages, whereas the range is smaller for earlier stages. This presumably reflects larger glacial/interglacial oscillations during the late Quaternary compared with the early Quaternary.

Various lines of evidence confirm that a major shift in Earth's climatic regime took place between 800,000 and one million years ago. This was an interval in which many extinctions took place among deep-sea benthic foraminifera (Hayward 2001) and tropical corals (Getty et al. 2001). Glacial-interglacial cycles and sea-level fluctuations were amplified thereafter. The major classical glaciations (Wisconsin, Illinoian, Kansan, Nebraskan) all took place during the last 900,000 years. It is also apparent that many more than four glaciations occurred during this time period.


Palynology is the study of pollen. Pollen is widely preserved as a record of past land vegetation and hence is useful for reconstructing past environmental and climatic conditions. Pollen is the dust-sized (0.01-0.1 mm) male reproductive apparatus of plants. Its outer shell is composed of chemically resistant exine. Pollen grains can usually be identified on the basis of size, shape and ornamentation to the generic and sometimes specific level of classification. Most common trees and grasses are wind pollinated. These plants produce prodigious amounts of pollen grains that settle on the surrounding landscape and are preserved in lake and bog sediments.

The science of palynology began in Scandinavia, where a classic late glacial/Holocene pollen sequence has been established--see Figs. 11-6 and 11-7; and Table 11-1. The major periods correspond to phases of glaciation and represent climatic conditions that controlled vegetation--see Fig. 11-8. However, climate was not the only factor influencing vegetation. Gradual weathering, soil development, depletion of nutrients, rate of plant migration, competition between plants through time, and early man's impact on vegetation should also be taken into account when interpreting the pollen record. Plant pathogens, such as the Dutch elm disease, may also be important factors. As a result of these various influences throughout the Holocene, it must be stressed that no identical plant communities exist anywhere today, even under the same climatic conditions (Iversen 1973).

Table 11-1. Late-glacial and Holocene pollen stages for southern Scandinavia. Age ranges in calendar years BP (calibrated radiocarbon dates).
Stage Age range Vegetation Climate Human culture
Sub-Atlantic Now-2500 Beech forest Cool-wet Iron Age
Sub-Boreal 5000-2500 Mixed-oak forest Cool-moist Neolithic & Bronze
Atlantic 8000-5000 Climax linden forest Warm-moist Mesolithic
Boreal 9000-8000 Hazel-pine forest Warm-moist Mesolithic
Pre-Boreal 10,000-9000 Birch-pine forest Cool-moist Mesolithic
Late Glacial 12,000-10,000 Tundra-shrub Cold-wet Mesolithic

Based on Iversen (1973) and Rud (1979).

Palynology and related vegetation investigations have also been carried out across North America at many sites--see Fig. 11-9. However, interpretation of pollen records in North America is complicated by two factors: (1) greater geographic region covered by fewer palynologists and (2) more complex vegetation with many species per genus. Nonetheless, considerable progress has been made in reconstructing paleoclimatic conditions on the basis of pollen. A typical pollen profile showing rate of change is given for Gould Pond, Maine--see Fig. 11-10.

During full glacial conditions, plants were restricted to various refugia located: in Beringia, south of the ice sheet, in nunataks, and on exposed continental shelves. During deglaciation, plants migrated into and colonized newly uncovered terrain that consisted of various glacial sediments and bedrock, all lacking soil. The rate of species migration depended on ameliorating climate, soil development, and dispersal speed. These conditions are illustrated by tree species in the eastern United States--see Fig. 11-11. Tree species with winged-seed dispersal that could grow in open conditions and on poor soils generally migrated fastest.

A major change in vegetation took place in eastern North America during a brief period around 10,000 years ago. From northeastern Kansas to Nova Scotia, spruce forest was replaced by mixed conifer/deciduous forest or prairie grassland. This rapid transition represents the beginning of Holocene climatic conditions. The eastern mixed forest was dominated by pine, fir, birch, and oak species, which have persisted to the present. The western grassland has likewise changed only slightly during the past 10,000 years.

The effects of agriculture are unmistakable in the Holocene pollen record--see Fig. 11-12. Forest burning and clearing, planting of crops, and grazing of animals all take a heavy toll on natural vegetation. In Denmark, agriculture first appeared during the Sub-Boreal period and became pervasive during the Sub-Atlantic. With each major archeologic period, agriculture expanded to cover greater land areas and to support larger human populations. As a result of rapid growth in human population during the late Holocene, agriculture now dominates the world's modern land vegetation.

Ancient plow marks of a Middle Age agricultural field in northwestern Denmark. These plow marks were preserved under a cover of wind-blown sand that accumulated during the Little Ice Age of the 17th and 18th centuries. Photos © by J.S. Aber.
Typical agricultural scene in southern Poland, near Krakow. Virtually all arable land is cultivated in the world today, and most forests are managed for human purposes. Vegetation, soil, water, and many other environmental factors are heavily influenced by agriculture.


Most land and aquatic animals, including
insects, have restrictions in their geographic ranges based on distribution of suitable habitat. Major habitat factors include climate, vegetation, water, and terrain conditions. When habitats change quickly, populations must migrate to more favorable situations or become extinct. The story of Pleistocene fauna is one of repeated migrations due to rapid habitat changes with each glacial/climatic cycle. Owing to the relatively short time interval, evolution was modest or minimal in most groups of land and aquatic animals. In spite of drastic and frequent environmental changes, few extinctions actually took place during the Pleistocene; however, a major extinction occurred at the end of the Pleistocene.

Irish elk (Megaceros giganteus), a giant deer that lived in Ireland during late Pleistocene time. Antler spread is about 2.5 m (8 feet). Geological Museum, Leiden, the Netherlands.
Cave bear (Ursus spelaeus) from the Pleistocene of the Czech Republic. Geological Museum, Leiden, the Netherlands. Photos © by J.S. Aber.

Land animal distribution is tied most closely to vegetation, which is the basis for the animal food chain. Vegetation also controls visibility and mobility for animals. In general, large heribores and their dependent large carnivores inhabit open grassland, savanna or tundra environments. Grazing in large herds and pack hunting are typical behaviors for such animals. Forest environments, in contrast, often support smaller herivores and carnivores, and the herd/pack behavior is not so strong.

The general vegetation was a major influence on the Wisconsin faunal provinces, particularly the presence or absence of prairie openings and deciduous or conifer trees. The situation is demonstrated for the last glaciation in North America, when the distribution of land animals was much different from today. Northeastern United States was part of the Symbos-Cervalces faunal province, whereas the central and western parts of the U.S. were in the Camelops faunal province--see Fig. 11-13. The Symbos-Cervalces province was associated with spruce forest and included: mastodon (Mammut), woodland muskox (Symbos), stagmoose (Cervalces) and giant beaver (Castoroides).

The Camelops faunal province was characterized by American camel (Camelops), mammoth (Mammuthus), short-faced bear (Arctodus simus), American lion (Panthera atrox) and buffalo (Bison sp.). This province had grassland aspects with open, mixed conifer/deciduous forest conditions. The driftless area of southwestern Wisconsin and southeastern Minnesota was probably a largely treeless tundra environment with a mixed megafauna including saber-toothed cat (Homotherium serum), stagmoose, buffalo, and mammoth (Widga et al. 2012).

This floral and faunal geography came to an abrupt end between 12,000 and 9000 years ago, earlier in the south and later toward the north, at the close of the last glaciation. Conifer forests disappeared from the eastern U.S. Meanwhile, mixed deciduous forest became established in the northeast, and prairie grassland spread over the Great Plains. Similar, rapid turnover in vegetation and climatic conditions took place throughout the mid-continent region--see Fig. 11-14.

Related sites

Remember: send your comments and questions via e-mail to the instructor.

Glossary or references.

Return to icehome or schedule.
ES 331/767 © J.S. Aber (2015).