ice-age theories

ice-age theories The form of recent climatic change is becoming increasingly clear from a wide range of geological evidence. Three times in the last 800 million years the Earth has been subject to glaciation; during the Late Precambrian, the Permian and Carboniferous, and, most recently, during the Quaternary. Our understanding of the Quaternary is the most detailed because the evidence is more abundant, the range of geological techniques available is larger, and the potential for resolution of events in time and space is greater. Analysis of deep-sea cores has shown that within the last 2 million years or so of the Quaternary ice age there were over 60 isotopic stages reflecting major changes in global climate. Any attempt at explaining ice ages must therefore address two problems: first, why do ice ages start and, secondly, why were there such major alterations of climate during the Quaternary?

Taking the second question first, it seems to be accepted that Milankovich cycles are the main forcing mechanism behind the alterations of climate during the Quaternary on timescales of tens of thousands of years. In addition, recent work on ice cores from Greenland has suggested that other mechanisms, such as the turning on and off of the oceanic thermohaline circulation, may also have a profound effect on climatic change, especially on short timescales. But Milankovich cycles have operated throughout Earth history, much of which has been ice-free, so they cannot be invoked as a mechanism to initiate ice ages.

In order to answer the question ‘How do ice ages start?’ it is first necessary to review the history of the transition from an ice-free to an ice-bound world. There is evidence from oxygen isotopes at Deep Sea Drilling Program sites that ice began to build up in Antarctica about 36 million years (Ma) ago during the Eocene, following some 20 Ma of cooling. Further marked decreases in temperature during the Middle Miocene at 15 Ma probably reflect increased ice growth in Antarctica, while the decrease in the late Pliocene is probably due to ice build-up on the continents of the northern hemisphere.

Dealing with the Late Pliocene cooling in more detail, recent oxygen isotope analysis of deep-sea cores suggests that there was a period of enhanced ice volume between 4.6 and 4.3 Ma, a long-term warming trend from 4.1 to 3.7 Ma, followed by a distinct cooling trend that was initiated at 3.5Ma and which progressed through to the initiation of the large-scale northern hemisphere glaciation, known as the Quaternary ice age, after 2.85 Ma.

Many theories have been put forward to explain the initiation of the Quaternary ice ages. These include: plate motions (continental drift), uplift of the Earth's crust, changes in atmospheric circulation, variations in the gaseous composition of the atmosphere, volcanic activity, and changes in oceanic circulation.

The theory that displacements of continents over the poles might explain the initiation of glaciation is an accepted one, especially for glaciation during the Permian and Carboniferous. However, over the past 100 Ma there has been negligible continental displacement towards the poles, and this cannot explain the observed cooling since the Early Eocene, some 55 Ma ago, which eventually led to the Quaternary ice ages.

A second theory suggests that land uplift, notably of the Tibetan plateau, may have been sufficient to initiate glaciation. The Tibetan plateau is one of the most imposing large-scale topographic features on Earth. It has an average elevation of over 5 km and an area half the size of the United States. It was formed by the collision of the Indo-Australian plate with the Asian plate which began during the Middle Eocene (52–44 Ma ago) and which continues today.

The Tibetan uplift has had two main effects on atmospheric circulation. First, the modern Tibetan plateau is so high and broad that it affects atmospheric circulation on a global scale. High mountain ranges, especially those with a large north–south frontage, act as barriers to east–west circulation and force the atmospheric circulation into poleward (meridional) diversions. The large ranges of the Himalayas and also the Rockies of North America have the effect of diverting upper atmospheric circulation in a meridional sense and are probably the cause of large-scale standing Rossby waves in the upper atmosphere. Rossby waves are responsible, amongst other things, for bringing moist air from the warm south-west Atlantic towards the colder continents adjacent to the north-east Atlantic, an important source of moisture for the build-up of ice sheets.

Secondly, the Tibetan plateau is a major factor driving the Indian monsoon circulation. Because of the height of the plateau, the Sun's rays have less atmosphere to pass through. In summer the plateau heats up quickly, air rises and causes intense low pressure at the surface which draws in the monsoon winds from the south and east.

However, work by climatic modellers suggests that the development of a Tibetan plateau does not have any appreciable effect on high-latitude summer temperatures, which exert an effective control on ice sheets by melting the previous winter's snow. This suggests, therefore, that the creation of high mountain ranges in Tibet and North America is not enough by itself to initiate the growth of large terrestrial ice sheets. In 1899, T. C. Chamberlin put forward a theory linking land uplift to changes in the gaseous composition of the atmosphere that would be sufficient to initiate global cooling.

As we know today to our cost, carbon dioxide (CO₂) levels in the atmosphere influence global temperature. On short timescales the main sources of CO₂ are from the burning of organic matter, from volcanic eruptions, and from changes in the rate at which CO₂ is absorbed and stored in the oceans as carbonate rock. On timescales longer than a million years, however, the level of CO₂ in the atmosphere is controlled by the balance between the rate of input of CO₂ (which largely comes from the Earth's interior and is released through volcanic activity) and the rate of loss of CO₂ (which is largely achieved by the chemical weathering of silicate rocks which absorb CO₂ from the atmosphere as part of the chemical reaction).

Maureen Raymo of the Massachusetts Institute of Technology and Bill Ruddiman of the University of Virginia suggest that Tibetan uplift both raised and exposed silicate rocks to chemical weathering, thereby abstracting CO₂ from the atmosphere. At the same time the enhanced monsoon climates produced by the Tibetan plateau would flush away the weathering products down the steep slopes into the rivers, and continually expose a fresh supply of silicate minerals for renewed chemical weathering.

One requirement of this theory, however, is for there to be a mechanism for adding to the levels of CO₂—for if the process were to run unchecked, the atmosphere would lose all its CO₂ within a few million years. Researchers suggest that this can be achieved by, first, a decrease in the amount of carbon being buried, resulting in more CO₂ being available, and secondly, by a decrease in the rate of carbonate precipitation on the ocean floor (which would abstract CO₂) relative to the rate of silicate deposition (which does not abstract CO₂). This hypothesis has become increasingly popular in recent years and is seen by many as providing a convincing mechanism for inducing the downward climatic spiral which began in the Miocene, continued through to ice build-up on Antarctica, and ended with the Quaternary ice age.

Volcanic activity has also been promoted as a possible cause of glaciation. The geological record suggests that Quaternary explosive volcanism was about four times higher on average than that during the Miocene and Pliocene. Explosive volcanism can eject large amounts of dust and debris into the stratosphere which can have a cooling effect. However, the main climatic effects are probably caused by sulphur, ejected into the stratosphere, which rapidly converts to sul-phuric acid, an aerosol which cools the troposphere by back-scattering incoming solar radiation.

Recent studies suggest that dust does not remain in the atmosphere for long after a volcanic event and that volcanic aerosols do not have a long-lasting effect on climate. It is thought, for instance, that the 1815 eruption of Tambora led to an average fall of 0.7 °C in the northern hemisphere the following year but that temperature soon recovered. Similar short-term effects have been noted from the 1980 eruption of Mount St Helens. However, the events noted above are comparatively small scale. By contrast, the eruption of Toba in Sumatra some 73 500 years ago was the largest known explosive event in the Quaternary. The eruption seems to have coincided with a period of rapid global ice accumulation at the start of the last major cold period. It is sug-gested that the eruption could have led to a cooling of about 12 °C in average summer temperatures for 2–3 years in the areas of growth of the Laurentide ice sheet which built up over northern North America. It is not impossible, therefore, that a very large volcanic eruption in conjunction with other climatic factors might be capable of initiating an ice age.

Changes in ocean circulation have also been put forward as a possible mechanism for initiating an ice age. The closure of the gap between the Atlantic and Pacific oceans in the Central American region which took place gradually between 13 and 1.8 Ma is seen as a significant event, mostly for its effects on oceanic circulation in the North Atlantic.

One major determinant of climatic change during the Quaternary both on long and short timescales has been the strength of the thermohaline circulation. In the North Atlantic cold, dense saline water is formed which flows southwards at depth, eventually reaching all the world's oceans. A return flow of warmer water is induced which greatly influences temperatures in the North Atlantic and adjoining continents. One major requirement of the thermohaline circulation is the formation of dense saline water. When the Panama gap was open and the Pacific and Atlantic were joined, it is suggested that more water flowed from the Pacific to the Atlantic than vice versa, limiting the build-up of salinity in the North Atlantic. Once the gap closed, the Gulf Stream flow increased, producing a region of high evaporation off eastern North America and an increase of salinity in the North Atlantic.

There is also evidence that during the Miocene a sill surrounding the Arctic Ocean, known as the Greenland–Scotland Ridge, subsided, allowing more cold polar water to escape into the North Atlantic. As the salinity of the North Atlantic grew and as outflow of cold polar water increased, so the thermohaline circulation increased in vigour, providing the mild winter temperatures and large amounts of moisture to the North Atlantic, which are prerequisites to the build-up of the large continental ice caps on the adjacent cold continents.

It therefore seems likely that the cause of the Quaternary ice age is a very complex question to answer. The climatic cooling induced by Tibetan uplift and consequent decrease in atmospheric CO₂ may have started the climatic downturn in the Miocene, but the closing of the Central American isthmus seems to have been the trigger for the build-up of ice in the northern hemisphere. This then provides the opportunity for Milankovich cycles to become the pacemaker of the ice ages, causing alternating glacials and interglacials.

B. A. Haggart

Bibliography

Andersen, B. G. and and Borns, H. W. (1994) The ice age world.Scandinavian University Press.
Imbrie, J. and and Imbrie, K. P. (1979) Ice ages: solving the mystery. Erslow Publishers, Short Hills, New Jersey.