Jack Haven

"The Beginning of an Ice Age"

A Comprehensive Analysis of the Origins of Glacial Climates during the Pleistocene Era

The last great ice epoch began around 2.4 million years ago (Ma), when major ice sheets began to appear throughout North America and Europe. From the moment we started to think about such things, the question of how this occurred has been bombarded by hypotheses that include everything from changes in the atmospheric composition of CO2 and other trace gases (Plass) to the increased volcanic activity that took place in the late Cainozoic (Kennett and Thurnell). Of all the hypotheses proposed to explain why Earth began to go through cyclical periods of glaciation and ablation at the beginning of the Pleistocene, four stand out. Their relation to each other is difficult to ignore; each begins with patterns that the planet has been going through for millions upon millions of years. These patterns were forced to change around three Ma as Earth's many continents continued to slide along the surface of the planet and crash into one another. The surface of the planet changed, and this affected how the patterns had settled climate during the Pliocene and before.

The most important of these hypotheses was proposed by a Serbian mathematician after whom these astronomic cycles were named. The Milankovitch cycles are paramount to the theory of glaciation at the beginning of the Pleistocene. They serve to push and pull back the effects of the other hypotheses, thus serving to regulate the cycles of glaciation and ablation that we have seen over the past 2.4 Ma. That the Milankovitch cycles cannot be solely responsible for glaciation is a fact most scientists have not failed to observe. The Milankovitch cycles, like the movement of the jet stream and the currents of the oceans (both of which will be discussed in this report), have been going on since Earth first coalesced from a sea of burning space debris around four billion years ago. The changes that took place on the surface of the planet to cause these ever-repeating cycles to hurtle us into the beginning of a an ice age, the rest period within of which we are nearing the end, is at the core of the matter that we are about to discuss.

Milankovitch, Obliquity, and Eccentricity

The Milankovitch cycles are, or at least were in 1966, the only one of the many hypotheses proposed to explain the beginning of the Pleistocene that can rightfully be called a theory, since it is the only one that can be convincingly tested (Broecker 299). The study of deep-sea sediments, performed by the Columbia of University during the 1950s, provided a wealth of information that empirically verified the hypothesis of orbital variation proposed by James Croll in 1875 and mathematically supported by Milutin Milankovitch in 1941. Sediment cores extracted from deep in the ocean, where the sedimentation rate is uniform and small, contain the forams of long-dead benthic and planktonic foraminifera. The benthic species of foraminifera has a shell that is formed from calcium carbonate (CaCO3) taken from the carbonate that has been dissolved into the oceans. As glaciers form and recede, the ratio of oxygen isotopes (16O and 18O in particular) changes, which in turn affects the isotopic concentration of CaCO3 in the benthic species' forams. By measuring the abundance of oxygen isotope ratios, or δ18O , at different distances from the top of a sediment core, one can form a picture of global ice volume over time (Muller and MacDonald 215). From this, it can be inferred that high negative values of δ18O occurred every time the relatively warm climate we are experiencing today took place, approximately in cycles of 100,000 years (Johnson 11-15).

The Milankovitch cycles are three: They correspond to obliquity, eccentricity, and precession. Obliquity refers to the tilt of Earth's axis, which cycles every 41,000 years. Eccentricity refers to variations in the shape of Earth's orbit around the sun, cycling every 100,000 years. Precession refers to the wobble of the Earth's axis, changing its direction but not its tilt. This last one cycles every 19,000 to 23,000 years (Muller and MacDonald 215; Willis, et al. 568). Milankovitch came to believe that the amount of solar energy entering the atmosphere at any given latitude would vary depending on the obliquity, eccentricity, and precession of Earth at a given moment in time. This is called insolation, and it came to be discovered that global solar insolation corresponded rather well with oceanic δ18O .

For the Milankovitch theory of ice ages, the level of summer insolation in the Northern hemisphere is the key to glaciation. During the winter, in high Northern latitudes, moisture in the air precipitates as snow. During the summer, if the temperature is high enough, the snow will melt and this cycle will continue next year. If the Earth's orbit around the sun is such that summer insolation in the Northern hemisphere is low, then the summers will not be hot enough to melt the snow. Snow fields will become ice sheets, and these become glaciers. The presence of glaciers cool the surrounding environment, which facilitates their continued growth. If insolation remains low, the glaciers continue to spread every year, and an ice age occurs (Wilson, et al. 61-64).

For the first million years of the Pleistocene, a period known as the Matuyama chron, the dominant Milankovitch cycle affecting the cycles of glaciation on Earth was the obliquity cycle. At about 0.735 Ma, glaciations intensified, and the dominant Milankovitch cycle became the eccentricity cycle; this began the period called the Brunhes chron. This discovery was made because of the rhythmic variance in oceanic δ18O that had existed until about 0.9 Ma, which showed that the 41,000-year cycle was still dominant at the time. After 0.9 Ma, however, the variations in oceanic δ18O became more complex, and they could no longer be correlated with the 41,000-year cycle. By 0.45 Ma, the eccentricity cycle had become unquestionably dominant, and δ18O varied accordingly (Ruddiman and Raymo 411-417).

Oceanic δ18O only explained that the change happened, but it failed to account for why. There are several hypotheses given to explain this. Muller and MacDonald proposed that an increase in the amount of interplanetary dust falling to Earth might have caused the shift. Before the dust increase, they argued, insolation based on obliquity would have been the primary factor affecting glaciation. Afterwards, dust accretion would have been the driving force. This was tested by studying the presence of 3He in sediment, which was found to follow the 100,000-year cycle of eccentricity. The dust in question would have come from a narrow ring around the sun formed by asteroid collisions (217-218). Another hypothesis is that the beds across which ice sheets flowed became rough due to the erosion being caused by glaciers as they advanced. Since glaciers flow more easily along a soft surface than a hard one, they began to slow, resulting in thicker ice sheets. These thicker ice sheets might have been able to survive the increase in summer insolation during the obliquity and precession cycles; they would have had to wait for the eccentricity cycle to provide enough warmth to melt the ice. (Wilson, et al. 152). More likely is the possibility that CO2 and trace gases in the atmosphere declined past a certain point around this time, leading to the formation of thicker ice sheets that could not have formed with the earlier atmospheric concentrations of greenhouse gases. These thicker ice sheets appear to have formed during periods of interaction between eccentricity and precession, and to a much lesser extent obliquity, leading to glacial and interglacial cycles of between 80,000 and 120,000 years. This seems to match up with the rhythmic variations of oceanic δ18O (ibid. 152).

Milankovitch is able to explain the regularity of the cycles of glaciation and interglaciation, but he is not able to explain why ice ages began at all. The Pliocene was a period of relatively warmer weather. What caused the beginning of the ice ages? Several supporting hypotheses are attached to the Milankovitch theory of ice ages, without which it would be difficult to argue how predictable variations in Earth's orbit that have been occurring since it first formed can create epochs during which the world is entirely tropical along with epochs during which the world is an ice ball. The key to these hypotheses is moisture. For the Milankovitch cycles to create ice ages, moisture must get to where it is very cold, and it must fall as snow. If there is not enough snowfall, then the little snow there does fall will melt quickly and no ice sheets will form. The second criterion regards summer insolation. Summers must be cold enough that they allow for ice sheets to form. If the summer is too warm, then ablation will occur: The snow will melt, and ice sheets will not form. If snowfall is high and ablation is low, then the stage is set for an ice age. We now consider the hypotheses that pick up where Milankovitch left off.

The Rise of the Central American Isthmus and the Gulf Stream

The Gulf Stream existed five Ma, but much of the Atlantic North Equatorial Current passed between North and South America into the Pacific Ocean. At the same time, the Central American isthmus was beginning to rise out of the water. As the strait between North and South America continued to shoal, the equatorial current could no longer send as much water into the Pacific. By 3.8 Ma, the rising of Costa Rica and Panama had closed off the Caribbean Sea, forcing the equatorial current to redirect its waters toward the northwest, and intensifying the Gulf Stream (Keigwin 352). It is not difficult to show when the Central American isthmus rose from the sea. All one has to do is perform radioisotope dating on marine fossils that had been common in both the Caribbean and the Pacific. The intensification of the Gulf Stream can be observed by looking at the winnowing and erosion that took place to the sediments in the Yucatan channel (ibid., Johnson 29-30).

The rise of the Central American isthmus had three primary effects. Warm water being propelled northeast along the Atlantic showered northern Europe with warm, humid wind. This warm wind continues to give western Europe much more comfortable weather than Canada has, even though they are at the same latitude. Second, the humidity that is sent northeast by the Gulf Stream offers the northern latitudes the moisture needed to create abundant quantities of snow: more than enough to form ice sheets. Moisture is paramount for the creation of an ice sheet thick enough to survive the ablation that tends to take place during the summer. This is made clear by the presence of an ice sheet over Greenland that, for some reason, does not seem to cover its northern coast. This is because the precipitation that is brought up north by the Gulf Stream does not reach the northernmost shoreline of Greenland (Johnson 31).

Third, the Gulf Stream began to send very saline water up north. The Atlantic is much more saline than the Pacific. This is because of its relative isolation from the other oceans. It is separated from the Pacific by the American continent, from the Indian Ocean by Europe and Africa, and from the Arctic Ocean by shallow underwater sills. Its salt content becomes especially concentrated in the Caribbean Sea, where it loses a great deal of water to evaporation. This water then travels west over Central America into the Pacific Ocean. The Gulf Stream pushes the warm, especially salty water of the Caribbean into the arctic zone. As this saline water travels north, it cools. When the temperature of the north-moving water becomes comparatively cool, it begins to fall. Salt water is denser than freshwater, and water that is as saline as that which arrives from the Caribbean will easily slip under the ocean around Iceland, only to be replaced by still more warm salt water following behind. The cooler water is continuously falling and becoming displaced by warmer water from the Caribbean, and an immense cycle of temperature transfer is formed. This is called the oceanic conveyor belt, and it is one of our world's most effective temperature control mechanisms (Johnson 73-74).

The Gulf Stream creates an ice age via a complex, negative feedback mechanism that conspires to shut down the oceanic conveyor belt. When glacial ice is turned to water, it decreases the relative quantity of salt in the Atlantic. The Atlantic is less saline as a result, and salt does not collect in the Caribbean Sea as it does now. When the Gulf Stream pushes warm water north, it will not fall under the cooler, more-or-less equally saline water of the Arctic. Since the Atlantic is no longer more dense than the Arctic, the water will take longer to cool to the point that it becomes dense enough to fall. The conveyor belt, thus, will fail, and cold water will not return south to cool the tropics. Since the Gulf Stream is only sending moisture up north, this moisture will fall as snow. The snow will collect, year after year, creating ice sheets. These ice sheets will form glaciers, they will grow, and an ice age will come about. The Gulf Stream works in the inverse. During a glacial period, water will leave the oceans to form thick glaciers on land. As these glaciers expand, the sea level will fall, and salt in the oceans will grow more concentrated. The Gulf Stream, continuing to push water up north, will end up collecting and pushing up more and more saline water each year. Eventually the water will be saline enough to fall below the cooler, less saline water up north, and the conveyor belt will start up again. Once the oceanic conveyor belt is functioning, the Milankovitch cycles regulate insolation, leading to net ablation in the arctic. Eventually the glaciers begin to recede, and the world enters an interglacial period.

The Himalayas and the Jet Stream

As rising land goes, Central America is not the only part of the world that is blamed for causing the ice ages. The Himalayas have been rising steadily for some 35 Ma, ever since India first collided with Eurasia. By studying pollen assemblages scientists have proven that the Himalayas were much lower than they are today as early as the Miocene. It is believed that the Himalayas only began to serve as a barrier for animal migration since the late Pliocene, very recently in geological time (Ruddiman and Raymo 418-419).

The jet stream air currents travel west to east along latitudinal lines. They have, however, been known to meander, or veer off course. One reason why they do this is related to temperature. The different ways in which land and sea absorb heat affect the distribution of mass and pressure in the tropopause, where the jet stream air currents whisk over the planet's surface. The other reason the jet stream air currents meander is related to the location and height of mountain ranges. The Himalayas, which are the highest mountains in the world, can force the jet stream air currents to meander. Unlike temperature, orography is consistent. Temperature changes depending on a location's absorption of solar energy during a given day, but the mountains are in the same place, doing the same thing, for millions of years. This suggests that orography tends to exert a greater influence on jet stream air current meandering than temperature (ibid. 422-423).

The polar jet stream air currents carry very cold air along the surface of the planet. When the Himalayas rose high enough to cause the jet stream to meander, it redirected the chilled tropospheric winds southward. This led to chillier climates farther south than had ever existed before. Colder climate does one thing: It decreases ablation. If the summer climate of a given place is not warm enough to melt the snow, the snow forms into an ice sheet and creates glaciers. The redirection of the polar jet stream allowed global climate to cool enough so that the Milankovitch cycles could step in and create an ice age where, normally, one might not have occurred.

Albedo and North America

One interesting hypothesis that was not much discussed was the possibility that the movement of North America toward the arctic region facilitated the beginning of the ice ages. The hypothesis has to do with albedo. It was faintly mentioned by several authors (among them Ruddiman and Raymo, Ewing and Donn, and Pauly); it was only described in passing. Maybe there is nothing to this hypothesis, or maybe it has not yet struck the interest of someone who has felt it had merit enough to consider. In any case, it seems prudent to dedicate at least a paragraph to it.

During the past few million years, North America, along with the other continents, has been moving along the surface of the planet. Pauly suggested that, at other moments in history, the world has experienced intense ice ages that, for some reason, did not cause very much glaciation at all. He holds that this is because there have been times, such as during the ice ages that took place during the Mesozoic, when the pole has been right in the middle of the ocean. Glaciation requires landmass; if there is no landmass, then glaciation cannot occur. When North America moved toward the north pole, it offered a willing landmass to be glaciated. As all the other factors fit into place (i.e. the Central American isthmus came up, the Himalayas caused the jet stream to meander, and Milankovitch kicked in), snow began to accumulate in the northernmost latitudes of North America.

North America is a huge and contiguous landmass. It offers an immense expanse upon which a glacier could expand. The more a glacier grows, of course, the more it can. Snow and ice, apart from being cold, are white. As the ice grows wider and thicker, the solar radiation that is meant to cause ablamation and stop glaciers from forming cannot do its work. White snow and ice have a high albedo and reflect solar radiation. Therefore, the larger the North American glaciers grew, the colder it became, and the easier it was for the glaciers to continue to grow. This is a positive feedback mechanism that continues to make climates colder and colder until the inevitable outcome, an ice age, comes to pass.

 

There are many other hypotheses besides these. Just after the Summer 2005 class, for which this paper was a requirement, was ending, Gies and Helsel published an article called "Ice age epochs and the sun's path through the Galaxy" in Astrophysical Journal. They argued that orbital variations are not the only astronomical fact to affect our cycles of glaciation and ablamation. In fact, the sun's travels along the outer edge of the Milky Way has a key role in bringing about these major ice ages, not only the one now in the Pleistocene, but the ones in the Mesozoic and the Paleozoic as well. When the sun travels through the spiral arms of the Milky Way, this exposure to a higher than normal cosmic ray flux may cause increased cloud cover on Earth. This cloud cover could be responsible for bringing about our long ice ages. Certainly the data for when the sun has traveled through these galactic spiral arms appears to correspond with data outlining the periods during which Earth has experienced long-term cycles of glaciation and ablamation. It will be exciting to see how this new information is received and whether it will affect the Milankovitch theory of ice ages.

It seems unlikely, however, that any new discovery could toss out the Milankovitch cycles as paramount to the understanding of all the factors that cause ice ages, not only now but in the distant past. It is apparent that Milankovitch alone cannot account for everything, but it is an integral part of any attempt to understand climate change of this dramatic sort. Along with Milankovitch, in order to finally come to grasp the complex mechanism involved in turning the northern latitudes of our planet into a very cold, white, permanently-winter wonderland, one must take into account the rise of the Central American isthmus; the rise of the Himalayas and their redirection of the jet stream air currents; even interstellar dust, albedo, continental drift, and—yes—the transgalactic voyages of our entire solar system.

Works Cited and Consulted

Broecker, Wallace S. "Absolute Dating and the Astronomical Theory of Glaciation." Science 151 (1966): 299-304.

Erwing, Maurice, and William L. Donn. "A Theory of Ice Ages." Science 123 (1956): 1061-1066.

Gies, D. R., and J. W. Helsel. "Ice age epochs and the sun's path through the Galaxy." Astrophysical Journal 626.2 (2005): 844-848.

Johnson, Robert G. Secrets of the Ice Ages. Minnetonka, Minnesota: Glenjay Publishing, 2002.

Keigwin, Lloyd. "Isotopic Paleoceanography of the Caribbean and East Pacific: Role of Panama Uplift in Late Neogene Time." Science 217 (1982): 350-353.

Kennett, J. P., and R. C. Thunell. "Global increase in Quaternary explosive vulcanism." Science 187 (1975): 497-503.

Muller, Richard A., and Gordon J. MacDonald. "Glacial Cycles and Astronomical Forcing." Science 277 (1997): 215-218.

Pauly, Karl A. "The Cause of the Great Ice Ages." The Scientific Monthly 75.2 (1952): 89-98.

Plass, G. N. "The carbon dioxide theory of climatic change." Tellus 8 (1956): 140-156.

Ruddiman, W. F., and M. E. Raymo. "Northern Hemisphere Climate Regimes During the Past 3 Ma: Possible Tectonic Connections." Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 318.1191 (1988): 411-429.

Shackleton, N. J. "The Climate System in the Recent Geological Past." Philosophical Transactions: Biological Science 341.1297 (1993): 209-213.

Willis, K. J., et al. "The Role of Sub-Milankovitch Climatic Forcing in the Initiation of the Northern Hemisphere Glaciation." Science 285 (1999): 568-571.

Wilson, R. C. L., S. A. Drury, and J. L. Chapman. The Great Ice Age. New York: Routledge, 2000.