Some 4,300 years ago, circa 2300 BC, an era that archaeologists call the Bronze Age, any rich and thriving civilization must have looked to all the Old World like a fixture of permanence and security. The transformation from subsistence farming to organized irrigation agriculture had given rise to culture and organized economies across a large swath of Earth from the Levant to Asia. Cities were large. Trading networks were prosperous. Classes of professionals were engaged in writing and mathematics, and nobles were articulating public policy.
In northern Africa, along the Egyptian Nile, the pharaohs of the Old Kingdom had built the pyramids at Giza. In Mesopotamia, between the Tigris and the Euphrates, humanity's first empire, that of Sargon the Akkadian, sprawled over 800 miles from the Persian Gulf to the northern Mediterranean. In southern Asia, across an area larger than Europe, the Harappan civilization of the Indus Valley had invented writing and employed artisans and traders in large and prosperous cities.
And then they were gone. The Harappan of the Indus Valley abandoned their large cities of Mohenjo-Daro and Harappa while northern India grew in population. In the "posturban" farming culture that survived, writing lay fallow for 1,400 years. So steep was the decline of the Old Kingdom that according to an inscription on the tomb of Ankhtifi, a regional governor in southern Egypt, "all of Upper Egypt was dying of hunger to such a degree that everyone had come to eating their children." The Akkadian empire was no more.
What in the world happened? Pondering this question for more than a century, generations of archaeologists and Near East scholars have been embroidering on orthodox historical and anthropological explanations about economic, political, and military upheavals of one kind or another. Societies are complex systems, after all, subject to any number of internal and external forces and to their own nonlinear behavior. For one reason or another, the idea that climate changes played an important role in this history has not been a very big part of the mix of answers. Since the 1990s, however, advances in paleoclimatology and other earth sciences have been forcing themselves on this question. And the evidence continues to accumulate: Abrupt climate change took hold of humanity's first great civilizations and shook them until they collapsed.
Only recently have researchers come to realize that the last 10,000 years—the epoch known as the Holocene—which saw the rise of humanity, has not been the benevolently stable climate that scholars and scientists have assumed. And they are beginning to tap the potential of Greenland ice and other archives to shed new light on events on the scale of human history. On this scale, of course, it is not the global but the regional climate changes that push humanity around. Hidden inside the variable of global temperatures is the more powerful circumstance of their differences between one place and another. Climate scientists know that changes in these temperature differences alter the circulation of the atmosphere, and this is where climate hits the pavement of human experience. This is what is most important to societies: not the temperature changes themselves, but how these changes affect precipitation patterns over time—where in the world it rains or snows and how little or how much. In the 1990s, research by ecologist Kathleen R. Laird of sediments in Moon Lake, North Dakota, developed a striking record of abrupt, large, and long-lived changes in wet and dry conditions over North America during the last 2,000 years. Societies large and small are resilient and almost—but not quite—infinitely resourceful. Any society—however large or small, however resourceful—can reach a point, a threshold at which it is exquisitely sensitive to the availability of water.
In the 1980s, archaeologist Harvey Weiss was studying a site called Tell Leilan in northern Mesopotamia when he hit on the connection between abrupt climate change and the collapse of Old World civilizations. Tell Leilan is in a part of the Near East called the Habur Plains, where humans have been cultivating barley and wheat since the Neolithic Natufians first took up farming 8,000 years ago. As Sargon the Akkadian extended his reach in 2300 BC, these grain fields of the northern Habur Plains would become the breadbasket for his empire. During the 1980s, Weiss was developing a theory around a pattern he was noticing in the excavations. The abandoned cities and settlements of northern Mesopotamia were in a region that, then as now, enjoyed a Mediterranean climate of wet winters and dry summers. Farmers depended on rainfall to water their crops. Yet, suddenly, for some reason, by the thousands, Akkadian citizens migrated downstream to southern Mesopotamia, where irrigated agriculture was employed.
In 1993, Weiss and colleagues reported finding in the dust of abandonment in Tell Leilan, site of the ancient Akkadian city of Shekhna, evidence for the sudden onset of drought that struck in 2200 BC and lasted 300 years. After years of conjecture among historians and archaeologists about the role of climate in the rise and fall of civilizations, Weiss's research had reached a turning point, a potent new collaboration. Coming just as the Greenland ice core projects were confirming the reality of abrupt climate change and hinting at its global extent, the Weiss studies pioneered a fascinating reevaluation of two very different lines of research. Archaeologists began seeing events in terms of climate, and paleo-
Drought Record from Moon Lake, North Dakota
A reconstruction from studies in the 1990s of fossil algae in sediments from this High Plains lake by Kathleen R. Laird and colleagues shows many long, intense droughts during the last 2,000 years. Notice the abrupt change in the pattern about AD 1200. The famous "Dust Bowl" drought of the 1930s appears to have been relatively minor and brief compared to other episodes in this record. Other climate reconstructions, such as tree ring records in California, show that these droughts were widespread.
climatologists began seeing episodes of Holocene change in terms of human history. At the Shekhna excavation was the French geologist and soil scientist Marie-Agnès Courty, who had pioneered new techniques that proved especially valuable at Tell Leilan. Courty provided the hard data that made the case.
Among the layers of earth representing successive eras of habitation at the site was a large section nearly two feet thick that dated from the time of the abandonment of Shekhna in 2200 BC. Unlike the rest of the layers, this sandy material contained no organic matter and showed no wormholes or other indications of moisture.
Powerful confirmation soon came in the work of paleoceanographers Heidi M. Cullen and Peter B. deMenocal, who analyzed the dust in a sediment core taken from the Gulf of Oman some 1,300 miles downwind of Tell Leilan. Isotopic analyses identified the dust as Mesopotamian. The chemistry of a volcanic ash layer in the ocean sediment matched the tephra layer recovered at Tell Leilan. And, according to Cullen and deMenocal, the calibrated dates of "the aridification and social collapse events are indistinguishable." In 2000, in an article in the journal Geology, the team concluded: "All available evidence indicates that this event records a dramatic mid-Holocene change in regional climate "
The team pointed to a variety of other paleoclimate records from the Middle East that document a sudden shift to dry, windy conditions across the region at the same time. In Lake Van in eastern Turkey, at the headquarters of the Tigris and Euphrates Rivers, sediments recorded a 30- to 60-meter fall in the lake level. To the south, the level of the Dead Sea fell by 100 meters. Lake levels fell in East Africa and North Africa as far west as Morocco.
Everything implicated abrupt climate change as "a key factor" in the collapse of the Akkadian empire. As Weiss had surmised, immigrants fleeing the famine to southern Mesopotamia swelled the cities beyond their capacity. Deprived of grain from the north, the Akkadians found their resources overwhelmed. They built a 100-mile wall against the "invaders" from the north, but soon the highly sophisticated society collapsed in violence and anarchy. From beginning to end, the imperial empire had lasted only a century. "Furthermore," wrote Cullen and deMenocal, "these responses occurred despite the fact that the Akkadians had implemented sophisticated grain-storage and water-regulation technologies to buffer themselves against historical ... variations in rainfall."
Using the same northeastern Arabian Sea sediment cores, geochemist Michael Staubwasser examined the link between the abrupt change 4,200 years ago and the collapse of the Harappan civilization in the Indus River Valley. Oxygen isotope shifts in a sediment core revealed a sharp decline in the outflow from the Indus River 4,200 years ago that transformed the Indus Valley civilization from "a highly urban phase to a rural post urban phase. In particular," wrote Staubwasser, "cultural centers, such as the large cities of Mohenjo-Daro and Harappa, were almost completely abandoned while locations in northern India grew in population." The coincidence of the Harappan migrations and the drop in Indus River discharge 4,200 years ago was hard to ignore. "A possible explanation is that a reduction of the average annual rainfall over the Indus river watershed restricted Harappan farming in the Indus valley and left large city populations unsustainable."
Along the Nile, meanwhile, crop failure and famine that struck the Old Kingdom ended the 90-year reign of Pepy II in 2152 BC. "Within the span of 20 years, fragmentary records indicate that no less than 18 kings and possibly one queen ascended the throne with nominal control over the country," the historian Fekri Hassan wrote in 2001. So severe was the famine, wrote Hassan, that people "were forced to commit unheard of atrocities such as eating their own children and violating the sacred sanctity of the royal dead."
To Hassan, there is no doubt that abrupt climate change—sudden drought—led to devastatingly low flows of the Nile, that famine and poverty caused the collapse, and that "the Nile can be considered as the force which destroyed the civilization that it had nurtured." The Nile-fed Lake Faiyum, more than 200 feet deep, dried up entirely. As the Nile was to Egypt, climate was to the Nile. The sudden onset of low flows faithfully reflected suddenly dry conditions in the Nile headwater regions of Ethiopia and equatorial Africa. According to Hassan, a 1999 study from West Africa of dust in the Kajemarum Oasis of northern Nigeria recorded a "pronounced shift in atmospheric circulation" around 2150 BC that led to "less rainfall and a reduction of water flow in a vast area extending from Tibet to Italy." According to archaeologist Lauren Ristvet, the abrupt "aridification event" 4,200 years ago "has been recorded in 41 paleoclimate proxies in the Old World, from Kilimanjaro, Tanzania to Rajasthan, India, East Asia and the Pacific."
Beyond the Old World climate crashes, powerful and abrupt climate changes during the past 10,000 years have been detected in climate archives throughout the world, but most particularly in records from the energy-rich and water-rich lower latitudes that plot changes in the global distribution of rain and snow. As the French researcher Françoise Gasse of Université Aix-Marseille argues, the concept of a "fairly stable" Holocene may have more to do with the history of the science than with the history of Earth's climate. The idea came from polar ice and North Atlantic deep-sea core records, where fluctuations during the Holocene were much lower than the transition from the ice age, such as the Younger Dryas. "On the contrary," wrote Gasse in 2000, "dramatic hydrological changes have long been apparent in Africa, and appear as large in magnitude as their glacial counterparts. Dramatic changes in water resources have enormous consequences on human populations, generating famines, migrations, civilization foundations and collapses."
The last of the abrupt changes that paleoclimatologists say resembles the pattern of events during the ice age was a sudden but relatively brief episode of sharp cooling and drying that shows up in polar ice cores and North Atlantic sediments after the Younger Dryas, about 8,200 years ago. To paleoclimate researchers, this event has the earmarks of another ice sheet meltwater flood that changed the density balance of the thermohaline circulation and curbed the northerly flow of warm water into the North Atlantic. The 2002 report on abrupt change by the National Research Council noted that, coming more than 3,000 years after the Younger Dryas, this episode "punctuated a time when temperatures were similar or even slightly above more recent levels, demonstrating that warmth is no guarantee of climate stability." Measuring argon and nitrogen isotopes in air bubbles trapped in the Greenland ice cores as well as oxygen isotopes in the ice itself, Takuro Kobashi of Scripps Institution of Oceanography in San Diego and colleagues produced a striking profile of the event. In a period of only five years, temperatures plunged and the climate remained "locked" in a cold regime for about 60 years.
From Weiss's new archaeological perspective, this 6200 BC climate crash had a portentous impact on Neolithic farmers, who for 1,000 years had been engaged in rain-fed cereal agriculture in northern Mesopotamia, when drought suddenly struck. Some of them migrated downstream along the steep banks of the Tigris and Euphrates Rivers into the southern alluvial delta floodplain near the Persian Gulf, where streamflow could be managed and small levees could be breached to water their fields. Irrigation agriculture was twice as productive as dry farming in the region, Weiss noted in 2003, but the new, more labor-intensive methods also required new social organization. The climate crash 8,200 years ago, he said, "provided the natural force for Mesopotamian irrigation agriculture and surplus production that were essential for the earliest class-formation and urban life."
For the rest of the Holocene, that is, for the last 8,000 years, abrupt changes in precipitation patterns leave the trample marks of big boots through climate records close to where human societies have flourished and fallen. None was more dramatic than the changes 4,000 years ago that transformed the Sahara from a green, wet savannah of annual grasses and low shrubs that characterized the landscape since the beginning of the
Holocene period into a vast desert, a dramatic change that has puzzled climate researchers for years. Subtle changes in the geometry of Earth's orbit are ultimately responsible—but why so sudden in the mid-Holocene, and why so severe?
As the great meteorologist Jule G. Charney had argued in 1975, the dynamic response of the atmosphere to vegetation loss begets the growth of a desert. In 1999, the German climate modeler Martin Claussen described how a powerful feedback mechanism provoked the most profound abrupt climate change of the last 6,000 years. Writing in Geophysical Research Letters, Claussen concluded that while the timing depends on global circumstances, "the abrupt desertification in North Africa during the mid-Holocene can be explained only in terms of internal, mainly regional, vegetation-atmosphere feedbacks in the climate system."
In 1991, a collaboration between anthropologist Izumi Shimada and tropical ice core specialist Lonnie G. Thompson secured one of the first solid connections between abrupt climate change and culture in the Western Hemisphere. Thompson's Quelccaya ice core from the high Andes of southern Peru allowed researchers to devise a detailed picture of the circumstances of abandonment and reformation of the Moche civilization in the sixth century. Centuries of imperial coastal Moche culture came to an end during a 30-year drought that was followed by severe El Niño flooding. The Moche capital was destroyed. Sand dunes filled the irrigation channels. There was widespread famine. Between AD 600 and AD 750, the succeeding Moche culture relocated its capital farther north and farther inland, near Andean rivers, where the supply of water was more dependable, and implemented new agricultural and architectural techniques.
In the Northern Hemisphere, nothing in history has equaled the disintegration of the great Maya civilization of the Yucatan Peninsula in what is now Mexico, Honduras, and Guatemala. For more than 3,000 years, from about 2000 BC, this society flourished through a "dark age" of European habitation and may have reached a peak population of 15 million. With their writing system and solar calendar, and their traditions of religion and nobility, the Maya engaged in astronomy and mathematics and practiced the architectural arts, building great temples and pyramids, palaces, and observatories.
Between the middle of the eighth century and the middle of the tenth century, the end came in several episodes of political consolidation, famine, disease, and disorder.
To explain the disaster, archaeologists rounded up their usual suspects, citing a weak economic base, a rigid political hierarchy, and palace intrigue by nobles vying for power. Some pinned the blame for the collapse on the effects of overpopulation and deforestation. Although problems such as these almost certainly affected the Maya at one time or another, 3,000 years of sustained existence would seem to be its own best argument against the notion that internal weaknesses caused the demise.
As archaeologists are inclined to emphasize, great episodes of history do not yield to simple explanations, to pat answers. In the case of the collapse of the Maya, in particular, arguments for the role of a changing climate have been hard-pressed to explain the complexity of this civilization's multistage, multifarious decline. The contraction began in the late eighth century with the abandonment of cities in the western rain forest; in the ninth century, in the central lowlands; and finally early in the tenth century, in the Maya heartland, where archaeologists estimate that by AD 930 95 percent of the population was gone. Early pollen studies were inconclusive, mainly because researchers could not distinguish between the effects of human deforestation and those of a changing climate. In 1996, geologist David Hodell reported results of chemical analyses of a Yucatan saline lake's sediment record that pointed to "the first unambiguous evidence for climate drying between AD 800 and 1000." This period, which saw the Maya collapse, was the driest episode of the past 7,000 years in Hodell's record.
In 2003, the sediments of the Cariaco Basin, the best climate archive in the Tropics, yielded a more detailed picture of what happened during the Maya collapse. By a fortunate coincidence, the seasonal movement of the Intertropical Convergence Zone, the thunderstorm belt in which northern and southern trade winds come together, brings the same climate of dry winters and wet summers to both the Yucatan Peninsula and this ocean basin north of Venezuela. An international team led by German geologist Gerald H. Haug tracked rises and falls in the presence of titanium, which varies with patterns of rainfall and river runoff into the Cariaco Basin. Haug's results showed that within the general 150-year dry period identified by Hodell were three distinct spikes of extreme drought at AD 810, 860, and 910 that mirrored periods of Mayan social collapse.
"We suggest that the rapid expansion of Maya civilization from AD 550 to 750 during climatically favorably (relatively wet) times resulted in a population operating at the limits of the environment's carrying capacity, leaving Maya society especially vulnerable to multiyear droughts," the team reported in Science. Already in a dry period, already near the threshold, the Maya may have been pushed over the edge by distinct pulses of abrupt climate change that brought collapse first in one region and then another and, finally, early in the tenth century, in the heartland itself. Sadly, just two years later, according to the Cariaco Basin record, wet summers returned to Yucatan.
A similar pattern seems to have run its course south of the Equator with the disintegration of a southern Andean Peruvian culture that coincided with "abrupt, profound climate changes" detected in sediment cores taken from the bottom of Lake Titicaca, site of a large urban population center of the Tiwanaku civilization of a thousand years ago. The Tiwanaku engineered an elaborate cultivation system of raised fields that so improved agricultural production it stimulated a dense population growth that could not be sustained during dry periods. The story unfolds in research by geographer Michael W. Binford, anthropologist Alan L. Kolata, and others who constructed a timeline spanning 3,500 years in the southern Andes highlands. Dry-land agriculture came to the region of Lake Titicaca with the Chiripa culture about 1500 BC during a time of plentiful water, when the lake level rose 65 feet, and continued after the emergence of Tiwanaku about 400 BC. Raised-field cultivation was developed about AD 600; by AD 1000 it was the main source of local food production. The end came about AD 1150, a time when drought caused the lake level to fall as much as 55 feet. "The lack of water simply made the physical and biological functions of the fields impossible," the team wrote in 1997.
The period from the tenth century until the early fourteenth century was a relatively benign climate epoch of warm temperatures around the North Atlantic, at least, that saw a dramatic expansion of European civilization. Vineyards in sunny England began producing good wines. Fields of barley and wheat swayed in the mild breezes of Iceland, and Norse livestock farmers secured their settlements in Greenland. Whether the Medieval Warm Period was a global phenomenon is a question of continuing debate, although it certainly was not a uniformly benevolent time. Very different and less salubrious changes in the climate of North America were under way. Again, they were punctuated not by changes in temperature but by changes in the distribution of rain and snow. Tree ring studies and other innovative methods are continuing to bring the history of climate forward in time, although the science of the last thousand years is young and the picture is far from complete.
In a region where most native peoples did not erect enduring monuments or settle in large, permanent population centers, direct archaeological clues to the climate's role in human events are few. Whether the 26-year drought in the late thirteenth century in the American Southwest was the primary cause for the abandonment of many elaborate settlements by the Anasazi is a matter of continuing debate. Recent modeling studies suggest that a more complex set of circumstances led to abandonments as populations fell below a certain threshold, although there is no doubt about the impact of drought on the overall viability of desert agriculture. Nor is there any argument about the evidence for enormously long and intense droughts in the climate record of the last millennium in North America.
In 1994, geographer Scott Stine reported remarkably clear evidence for extremely long droughts in California's Sierra Nevada. Using radiocarbon dating, Stine ascertained the age of the wood in tree stumps he found rooted in the beds of modern mountain lakes, streams, and marshes. The mountain range had endured more than two centuries of drought from AD 892 to about AD 1112, Stine reported, and another 140 years of drought that began about AD 1209 and came to an end around AD 1350.
Across the plains and prairies of interior North America, vast sand dunes lie under a thin skin of vegetation that has been peeled back by climate more than once. The dunes have drifted in the winds of droughts that far exceeded the intensity and duration of anything in the modern record of the region, including the 1930s "Dust Bowl," the United States' iconic experience with climate change. In the last quarter of the thirteenth century the first of two "mega-droughts" of the past thousand years gave flight to the sands of the region during a time that coincides with Stine's Sierra Nevada drought and the abandonment of the Anasazi settlements of the Southwest.
In the 1540s, when Europe was experiencing what climate researchers call the Little Ice Age, a second mega-drought spread from northern Mexico. By the end of the century, it extended over much of the United States. This was the worst drought to hit North America in 500 years, and it was implicated in a recent study of a hemorrhagic fever epidemic that led to a colossal loss of native Mexican lives in the second half of the sixteenth century. A 2002 study by Mexican epidemiologist Rodolfo Acuna-Soto and
American tree ring researcher David W. Stahle pointed to climate-driven spikes in rodent populations as its probable cause. Epidemics in 1545 and 1576 killed an estimated 17 million people, more than twice the number of victims of the smallpox outbreak that the Spanish delivered to Mexico in 1520. "In absolute and relative terms the 1545 epidemic was one of the worst demographic catastrophes in human history," Acuna-Soto and Stahle wrote in the journal Emerging Infectious Diseases. This was the drought that greeted Sir Walter Raleigh's hapless colonists in 1587 as they stepped onto the beach at Roanoke Island.
New high-resolution climate profiles from tree rings, deep-sea and lake sediments, and ice cores are rewriting some interesting episodes of human history, although, as Gerald Haug and colleagues noted in their Maya study, technical and other problems remain to be solved. "Unfortunately," they wrote, "the limitations of temporal resolution and chronology in paleoclimatic records still present a major obstacle to the development of a globally meaningful view of Holocene climatic changes and their role in social change." Its own origins in geology and in polar ice have given paleoclimate research an odd, upside-down appearance. Ironically, because abrupt change was recognized first in ice age archives, researchers have more highly developed ideas about events of a distant past than they do about more recent human experience with climate.
Writing in Science in 1996, University of Arizona geoscientist Jonathan T. Overpeck, a leading Holocene climate researcher, observed that both the early Greenland ice core record and the modern instrumental climate record of the past 150 years have led to some seriously mistaken but "comforting conclusions" about the character of events of the past 10,000 years. In contrast to the ice age, Overpeck noted, "the current warm interglacial climate is often characterized as relatively stable, leaving the impression that climates of the future are likely to be more or less well behaved." And because it is the only period when changes were widely recorded by instruments, many people seem to believe that the record of the last century represents the full range of natural climate variations. With advances in research techniques has come a new picture.
"It is now clear that climate variability in many regions of the world, including Greenland, was significantly greater during the last 10,000 years than during the last 150 years," Overpeck wrote. Most of these warm-era events were smaller than ice age changes, but many were much bigger than any that instruments have recorded in the last 150 years. "More importantly," he wrote, "many of these past Holocene events appear to have been large enough that, if they were to recur in the future, they would have major impact on humans." David Stahle makes the point: "Year-in and year-out, over the long haul, drought extracts the most from humanity."
Was this article helpful?