Frozen Earth: The Once and Future Story of Ice Ages

by Doug Macdougall

  We do not know if frequent shifts in climate also characterized earlier ice ages, or if the long, warm intervals between the Earth’s major ice ages were monotonously stable, because we don’t have any high-resolution records from those times. However, we do have a very high quality record of climate through the last few millennia of the Pleistocene Ice Age, and we also have written and archeological evidence from the same time period. It is very instructive to compare the two, for it provides insight into the intersection between the continuously varying ice age climate and developing human civilization.

  There are now several good examples of past civilizations collapsing as a result of abrupt climate change. In most cases, the archeological evidence for collapse has long been known, but the causes have not. Many different factors—competition, war, technology development, disease, and others—have been proposed as being important, but none have been entirely convincing. The evidence of a climate change connection has come from improved techniques for measuring or inferring local environmental conditions such as precipitation and temperature, and from new information about the timing of changes in these conditions. And although temperature is often the parameter that attracts the most attention in climate records, in many cases, rapid swings in the amount of precipitation have been most important. A classic example comes from the Akkadian civilization of the Middle East, which was thriving before it experienced sudden collapse about 4,200 years ago. More than a century of archeological research had documented this decline in great detail, but there was no agreement about its cause. However, on the basis of regional climate data that are now available, it appears that the collapse can be linked to an abrupt transition into a time of severe drought. The Akkadians had adapted to life in an arid region, but they still depended on predictable seasonal rainfall for irrigation. When the regional climate changed and the rain suddenly failed, they were in trouble. Whole cities were abandoned and a prosperous empire crumbled. Hordes of refugees moved southward, in what is present-day Iraq, from northern Mesopotamia to regions along the Euphrates River that had a more dependable water supply. But the drought was widespread, and the “barbarians” from the north—as they are termed in the writings found on clay tablets from the region—so burdened already stressed resources that these cities too were brought to the verge of collapse. A very similar fate appears to have befallen the Mayan civilization of Central America several millennia later. Like the Akkadians, their society collapsed rapidly for reasons that are hard to discern from archeological evidence alone. But they too depended on seasonal rainfall for irrigation and survival. Sediment cores from the nearby Caribbean show that between 1,100 and 1,200 years ago, the region experienced an extended period of drought, punctuated by short but repeated spells of even more severe aridity. That is precisely the time when Mayan civilization collapsed. Like the Akkadians, they were unable to sustain their formerly prosperous cities when the rains failed.

  The Akkadians and the Mayans present interesting but isolated examples of the influence of climate on mankind. The periods of drought that affected them may have been local, but they were most probably related to more global shifts in average climate as the Earth emerged from the most recent glacial period. However, what has really caught the interest of a small but growing group of anthropologists, archeologists, and historians who are examining history from the perspective of climate is the past 1,000 years. This period has the advantage that there is an abundance of proxy records from ice cores, tree rings, and sediments, for the most part providing very high time resolution. In many cases, even year-to-year changes can be tracked with confidence. Furthermore, climate information can be compared directly with the written historical record. Temperatures deduced from central Greenland ice cores can be checked against statistics for wheat harvests in Europe or the northern limits of vineyards. Perhaps not surprisingly, many intriguing coincidences of timing have been discovered between climate change and historical events.

  In the climate record of the past millennium, two periods stand out—a warm one that has been called the “Medieval Warm Period,” and a colder interval dubbed the “Little Ice Age.” The latter is a misnomer in technical terms, because it was far from being an ice age. In reality, it was nothing more than a slightly cool interval within the current interglacial period, but it was characterized by expansion of glaciers in many parts of the world—this was especially obvious in the relatively heavily settled Alps—which led to the name. Compared with the longer-term fluctuations of the 100,000-year ice age cycles, or even shorter-term shifts such as those of the Younger Dryas period, these two slightly abnormal intervals experienced only small changes in average temperature, little more than one or two degrees Celsius. However, the historical record indicates that even fluctuations of this magnitude had direct, and sometimes severe, consequences for human society. Again, it was not just the temperature change that was important, but the overall weather patterns that accompanied the change—especially the amount and distribution of precipitation, and the frequency of storms.

  Eight or nine hundred years ago, Europe was in the midst of the Medieval Warm Period. Although there were the inevitable spells of “bad weather,” summers were generally warm and crops plentiful enough that there were few serious famines. Temperature data from Greenland ice cores corroborate the inference made from the written record that this was a time of relatively warm, benign climate in the region. Vineyards flourished in England to the point that the French became concerned about the impact on their own wine industry. In the North Atlantic, ice packs retreated northward, especially in summer, and the incidence of severe storms decreased. The Vikings, who were already expert sailors and had raided coastal Ireland, Britain, and Europe centuries before, sailed north and west under the more favorable conditions and established settlements in Iceland and coastal Greenland. They conducted regular supply and trade voyages between these colonies and Scandinavia, and from Greenland they also explored the eastern shore of North America. Although they never colonized the New World, they did found small settlements about 1,000 years ago, such as the one that tourists now visit at L’Anse aux Meadows, in northern Newfoundland. In Europe, the warm and equable climate meant that kings and landowners prospered, and laborers and peasants rarely went hungry. The amount of land devoted to agriculture increased, and, especially in northern countries, fields crept up hillsides to elevations that had not been farmed before. The range of warmth-loving crops moved steadily northward. The general prosperity meant that there were funds to fight the Crusades and to employ generations of stonemasons who traveled throughout Europe building beautiful, massive, and very expensive cathedrals. In harsher times, such feats might not have been possible.

  Europeans living at this time could not have anticipated the rigors that the Little Ice Age was about to bring. As the thirteenth century drew to a close, the Medieval Warm Period was also nearing its end. The Greenland ice cores, together with other climate records, show that temperatures decreased beginning around 1300 and stayed low—although with some intervening warm intervals—well into the nineteenth century (figure 25). Where, exactly, to place the beginning and end of the Little Ice Age is somewhat subjective. Some scientists restrict the term to the period between 1600 and 1800, when Europe experienced some especially cold weather and Alpine glaciers grew substantially (in the early 1600s, one observer reported from Switzerland that a glacier was advancing daily by as far as one could shoot a musket). Other climatologists take the 1600–1800 period simply to be the climax of the Little Ice Age, and use the term for the entire time from 1300 until about 1850, when average temperatures began a slow climb toward today’s values. But however one defines the Little Ice Age, throughout even this geologically short timespan, there were multiple temperature excursions bringing both warmer and colder weather. It was only in an average sense that the interval was cold.

  Figure 25.Temperature variations through the last millennium as recorded in a Greenland ice core. The scale is
relative, with today’s average fixed at zero. The Little Ice Age lasted from approximately 1300 to 1850, and its coldest period was near its end.

  What caused the Little Ice Age, or even what caused the short-term cold-to-warm-and-back-again flip-flops within the Little Ice Age, is unknown. Climate science is notoriously difficult, because there are so many interconnected variables at work that cause and effect are often impossible to discern with confidence. The oceans, however, are thought to play a central role, because they store a vast amount of heat energy, which they disperse worldwide through the mechanism of ocean currents. The atmosphere is important too, and there are close connections between what happens in the oceans and the atmosphere. In recent years, continuous global monitoring of winds, ocean temperatures, clouds, and many other parameters that affect weather have provided a good empirical view of how some aspects of the climate system work. However, a clear understanding of the causes is still elusive. A good example is the El Niño phenomenon—over the past decade, a very detailed picture has emerged of how ocean temperatures and atmospheric pressure change in the equatorial Pacific Ocean during El Niño cycles, yet exactly why this happens, or why the cycles occur every three to seven years, or how events in the equatorial Pacific affect the weather in the United States or Australia or India is not well understood. In the North Atlantic region, where most of the historical records of the Little Ice Age are found, there are also periodic shifts in the overall weather patterns. They seem to occur somewhat less regularly than the El Niño events of the Pacific (at least over the relatively short span of time they have been observed), and they last longer. These climate changes are related to a phenomenon known as the North Atlantic Oscillation—NAO—and they bring Europe, and eastern parts of North America, alternating periods of warmer and colder average temperatures. They also have a significant effect on the distribution of rainfall. The “oscillation” part of the NAO refers to variations in the strength and position of atmospheric high and low pressure zones in the North Atlantic, which in turn affect the location of the jet stream and the path that winter storms follow across eastern North America, the North Atlantic, and Europe. Often one mode of the oscillation dominates for a decade or two, followed by an abrupt shift to the other. Although the NAO is defined in terms of an atmospheric phenomenon, many researchers contend that it is closely linked to changes in the circulation pattern of cold and warm waters in the North Atlantic Ocean, again emphasizing the close coupling between the ocean and atmosphere. Most likely prolonged times of warm, stable climate such as the Medieval Warm Period, or generally cool and wet periods such as the Little Ice Age, are manifestations of something like the different climate modes that characterize the present-day North Atlantic Oscillation.

  Even if the prevailing weather of the Little Ice Age can be linked to an observed phenomenon such as the North Atlantic Oscillation, however, that still doesn’t solve the problem of cause. For that it’s useful to recall James Croll’s maxim, which was that in order to understand nature, one has to understand the underlying physical laws. Croll realized that ice ages might occur when lower than normal amounts of solar energy are received on the Earth because of orbital changes. Heat energy from the sun is also what drives atmospheric winds and ocean currents and is ultimately responsible for phenomena such as the North Atlantic Oscillation. Recognizing this, scientists have been searching for clues about how the input of solar energy might have varied over the timescales of the Medieval Warm Period and the Little Ice Age. Orbital parameters of the kinds investigated by Croll and Milankovitch occur much too slowly to explain these shorter-term climate fluctuations, so evidence for other types of variation has been examined.

  Climatologists are nothing if not ingenious. They have developed a bevy of proxies to investigate past climate, and several of them provide good measures of the sun’s energy output. Although it doesn’t fluctuate very much, the sun’s “activity,” a measure of how much heat energy it releases into space, does vary. A well-known example is the eleven-year solar cycle, the peak of which is marked by intense solar storms, great eruptions of matter in flares from the sun’s surface that hurl ions out into space—a symptom of a more energetic sun. The magnetic fields that accompany the solar flares play havoc with telecommunications on Earth, and those who live at high latitudes get to see spectacular displays of the aurora borealis. Times of high solar activity are also marked by sunspots—phenomena that have been observed and recorded for centuries, initially because they were thought to be portents of the future, and later out of scientific interest. Two thousand years of combined and (almost) continuous Chinese and European observations paint a vivid picture of the sun’s activity. They show that the number of sunspots recorded during the Little Ice Age was low, by implication the sun ever so slightly cooler. The change is small enough that it cannot, on its own, account for the climate variations. But even variations of this magnitude may be part of the explanation, through what climatologists refer to as forcing. Just as orbital changes do on a longer timescale, they seem somehow to pace the temperature changes associated with events such as the Medieval Warm Period or the Little Ice Age.

  Other proxies confirm the sunspot observations. When the sun’s activity is high, the strength of its magnetic field is also high, and cosmic rays (high-energy particles that originate outside the solar system and permeate interstellar space) are deflected away from the Earth in greater numbers than usual. This happens because cosmic ray particles carry an electric charge, and when electrically charged particles enter a magnetic field, their direction of travel is changed—they curve away from their original path. Normally, a combination of the sun’s and the Earth’s magnetic fields deflect away most cosmic rays, but some especially energetic ones are only minimally diverted and penetrate through to the atmosphere. When this happens, they smash into nitrogen and oxygen atoms and make radioactive isotopes such as carbon-14. The amount of carbon-14 formed is thus a fairly sensitive measure of solar activity—the stronger than normal magnetic field that characterizes times of high solar activity deflects more cosmic ray particles away from the Earth, resulting in lower carbon-14 production. Tracing how carbon-14 has changed in the atmosphere, which can be done by measuring the isotope in ice cores or tree rings, is therefore another way to gauge the sun’s activity. The carbon-14 data corroborate the sunspot record and indicate that solar activity was generally high during the Medieval Warm Period, and low during the Little Ice Age.

  As solar activity decreased near the beginning of the Little Ice Age, the signs of a deteriorating climate in the North Atlantic region are abundantly clear both from scattered historical records and archeological evidence. In the islands of the Canadian arctic, the Inuit people began to migrate south. By the 1340s, sailors traveling between Iceland and Greenland had to follow longer, more southerly routes in order to avoid ice and treacherous weather. The Norse settlements in Greenland went into decline and were eventually abandoned as an already unforgiving environment became even harsher. Skeletons from graves at some of the Greenland settlements show that by 1500, near the end of the settlement period, even the average size of the few people still living there had decreased significantly. In the worsening weather, farming became increasingly difficult, and their diets correspondingly less varied and nutritious. In the low-lying coastal areas of Germany, Denmark, and Holland, fierce storms obliterated large tracts of agricultural land in the thirteenth and fourteenth centuries, and by most estimates claimed more than a hundred thousand lives. And in the Alps, glaciers began to encroach on farms and pastures that had been established during earlier, warmer times.

  In spite of these undeniable facts, there is still much debate about the degree to which historical events such as economic decline, famine, the “Black Death” (plague), and changes in political fortune were due to the deteriorating climate as opposed to other factors. Perhaps climate is just a “forcing factor,” in the same way that the sun’s activity or the Earth’s orbital parameters are
forcing factors for climate itself, not wholly responsible, but important enough to tip the balance in one direction or another. However, some historical events of the Little Ice Age are so widespread throughout Europe that an external factor such as climate seems likely to be the predominant cause.

  An example is the disaster that struck across Europe during the early years of the fourteenth century. The relative prosperity and benign weather of earlier times had resulted in an expanding population. Much marginal land had been occupied and farmed. Then a series of cold, wet years struck. Grain rotted in the fields, or sometimes couldn’t be planted at all. The cost of food rose rapidly. Livestock and people began to succumb to famine and disease, and even the rich had difficulty getting enough to eat. Social unrest was rampant; gangs of desperately hungry peasants roamed the countryside searching for food, and in some places there were reports of cannibalism. By the 1320s, land and even whole villages had been abandoned throughout Europe, and population fell precipitously, in places by more than half. This was indeed abrupt change; it occurred within a human lifetime, at a time when life expectancy was half what it is today. A few decades later, the first of a series of waves of bubonic plague, the infamous Black Death, hit an already reeling Europe. Huge segments of the population, especially in crowded urban areas, succumbed to the disease. The first half of the fourteenth century is also the period that historians identify with the dismantling of the feudal system that had prevailed throughout Europe. Can all of this be ascribed to climate? Certainly, the poor harvests of the early 1300s, and the fungus and other diseases that affected crops, can be attributed directly to a series of very wet and generally cold years. People and livestock, weakened by shortages of food, were especially susceptible to disease. The Little Ice Age was in its early stages, but already, it seems, it was having a profound effect on society.