by Brian Fagan
The arid north coast of what is now Peru was one pole of an interconnected highland and lowland Andean world, where long-term droughts and El Niños were a constant threat to civilization.
TELECONNECTIONS, LINKS BETWEEN contemporary climatic events in different parts of the world, are fundamental to understanding the Medieval Warm Period. Climatologists tie together such events as droughts or major El Niños by using ice cores, tree rings, deep-sea cores, and other well-dated proxies, but even the best-documented links remain tentative. However, for the first time, we are beginning to learn just how widespread droughts were during the warm centuries. A grid of tree-ring sequences reveals prolonged dry cycles in the North American West and Southwest between the tenth and thirteenth centuries that appear to be linked to the Pacific Decadal Oscillation and to cool conditions in the eastern Pacific. The Cariaco basin core from the Caribbean and borings from Yucatán lakes document harsh droughts in the Maya lowlands of Central America between the eighth and eleventh centuries that can be linked to the Intertropical Convergence Zone lingering in a more southerly position. Can we, then, add Andean droughts to this emerging pattern of dry cycles? If so, can we reconstruct how the wealthy civilizations of the coastal Peruvian desert adapted to lengthy droughts?
Locations mentioned in this chapter. Some obscure places are omitted.
The eighteenth-century German naturalist Alexander von Humboldt was the first European scientist to explore the Andes. He marveled at the great variety of animals and plants that thrived in the harsh and varied landscape of the high peaks. Humboldt observed the constant assault on higher altitudes waged by mountain farmers, who lived among terraced fields on the sides of deep valleys. The mountains tumble through steep foothills to the low-lying Peruvian coast, a shelf at the foot of the Andes crossed by rivers that descend from the mountains to the Pacific. The bleak coastal plain is itself effectively rainless. It is carved by some forty rivers and streams fed by mountain runoff, but much of the low-lying terrain is too sharply undulating to permit irrigation. Two large areas on the north coast boasted of topography that allowed farmers to link their field systems to canals: the Chicama-Moche valleys and the Motupe-Lambayeque-Jequetepeque region.
Few human communities anywhere, let alone civilizations, have ever faced such a challenging environment—not only one of the driest on earth, but one where short- and long-term climatic events could alter the landscape drastically almost overnight. In good years, there was enough water for farmers to cultivate maize, beans, cotton, and other staples, provided they managed the runoff carefully and irrigated their fields. For agriculture, everything depended on canals, reservoirs, and expert water conservation strategies. Fortunately for the farmers, the people of the coast did not live on agricultural produce alone. The Pacific was a bountiful source of shellfish, and, above all, of anchovies, which abounded by the million close inshore thanks to the natural up-welling of nutrients stimulated by the cold, north-flowing current (now named for Von Humboldt). Before industrial exploitation depleted the fishery, the swarms of anchovies were almost unbelievable. In 1865, the American archaeologist and diplomat Ephraim Squier described rowing in a small boat through the sluggish swell, passing through “an almost solid mass of the little fishes which were apparently driven ashore by large and voracious enemies in the sea.” A 1.2-mile-long (2-kilometer) belt of fish lay so close to shore that women and children were scooping them up “with their hats, with basins, baskets, and the fronts of their petticoats.”3
Against seemingly insurmountable odds, a series of wealthy states flourished along Peru’s north coast for many centuries. A combination of agriculture, fishing, and long-distance trade provided at least a partially successful buffer against an arid, demanding environment.
IN GOOD RAINFALL years, agriculture in coastal river valleys could be highly productive, provided water was carefully managed. Even in good times, the people of each town and village conserved every drop they could, for they well knew that even a one- or two-year drought could bring hunger. But ice cores tell us that the Andeans faced far longer dry cycles than that. Lonnie Thompson of Ohio State University has long worked on ice cores from the Quelccaya ice cap in the southern Peruvian Andes. His sequences reveal four widespread regional droughts between the sixth and early fourteenth centuries A.D., culminating in a major dry spell between 1245 and 1310.4 This catastrophic thirteenth-century drought had drastic effects on highland societies, especially the pre-Inca state of Tiwanaku, centered along the southern shores of Lake Titicaca.5 Lake core data from Titicaca itself date the onset of the drought here to about 1150, somewhat earlier than elsewhere. The lake level dropped between 39 and 56 feet (12 and 17 meters) over five decades in the late twelfth century, which reflects an estimated drop of between 10 percent and 12 percent in rainfall from the modern average. As I have described elsewhere, Tiwanaku was heavily dependent on raised-field agriculture nourished by streams and high groundwater levels. In a time of persistent drought, Tiwanaku’s agriculture was no longer sustainable. The city collapsed and its population dispersed into much smaller communities and turned to the herding of native, drought-adapted llamas and alpacas and opportunistic dry farming.6
The same drought cycles affected a much wider area. Lago Argentino, in extreme southern Patagonia, lies in the southern hemisphere’s westerly wind belt, which brings rainfall to the region.7 Like the Sierra lakes, Lago Argentino rises and falls like a rainfall barometer. When the water rises, it buries the southern birch trees that have grown in the exposed basin. The birches grew for between a half century to a full century before perishing in about A.D. 1051. One hundred and twenty miles (200 kilometers) farther north, Lake Cardiel recovered from one of its lowest levels at about the same time, drowning shrubs rooted on the shore. Radiocarbon dates tell us they died between 1021 and 1228, remarkably close to the dates for the end of droughts in California.8 As we saw in chapter 7, the mid-latitude storm track of the northern hemisphere remained north of California during these centuries, either because the circumpolar vortex (the circulation of the jet stream in the high troposphere) contracted, bringing warmer conditions north and south, or on account of a persistent ridge of high pressure. Both these conditions have occurred in modern times. Vortex contraction would also account for the droughts in southern Patagonia, but for different reasons. The Patagonian lakes lie to leeward of the Andes and in the zone where westerly winds are strongest. This would have had the effect of intensifying the Andes rain shadow (the area to leeward of a mountain range where much less rain falls) and triggering drought. The aberrant atmospheric circulation of medieval times brought much greater changes in rainfall than in temperature. This is certainly true of the Andes, where ice cores provide an unusually complete proxy of the rainfall shifts that affected not only Tiwanaku but also Chimor, the Chimu state of the arid north coast. However, unlike Tiwanaku, Chimor survived the same drought cycle and even thrived, which is surprising given the other climatic villain on the stage: El Niño.
EL NIñO, THE so-called Christmas Child, has been part of the climatic dramatis personae in some of the earlier chapters, but it plays a major role in the Andean historical drama. Unlike droughts, El Niños are short-term climatic events, which bring torrential rainfall to the normally arid Peruvian coast. An El Niño can wipe out years of irrigation works in hours and reduce crop yields dramatically for months, if not years. El Niños also bring tropical sea currents that replace the cold, north-flowing Humboldt Current. Upwelling close inshore slows and the anchovies move away to cooler water. In the short space of a few months, the two foundations of coastal life crumble without warning, leaving thousands of people without sustenance.
In 1892, a Peruvian sea captain, Camilo Carrillo, published a paper in the Bulletin of the Lima Geographical Society, in which he described a tropical countercurrent known by local fisherfolk as El Niño (“the Child Jesus”), “because it has been observed to appear immediately after Christmas.”9 A flurry of scientific papers s
ubsequently drew attention to what appeared to be a local phenomenon, which temporarily decimated the anchovy fisheries and brought heavy rainfall to the mountains and coast. El Niño remained a local meteorological curiosity until 1969, when the UCLA oceanographer Jacob Bjerknes linked atmospheric circulation in the Pacific with variations in sea temperatures in the tropical ocean. He linked El Niños with the seesawlike movements of the Southern Oscillation and identified them as not merely a local phenomenon but rather a global climatic force, now known to be second only to the seasons in its influence on global climate (for more about this, see the sidebar).
El Niño, La Niña, and the Southern Oscillation
The Southern Oscillation is an irregular, seesawlike pattern of reversing surface air pressure between the eastern and western tropical Pacific. When surface pressure is high in the east, it is low in the western Pacific, and vice versa. Ocean warmings and pressure reversals are usually simultaneous. The British meteorologist Sir Gilbert Walker discovered the Southern Oscillation in the 1920s. He realized that when atmospheric pressure was high in the Pacific, it tended to be low in the Indian Ocean. The erratic swings of the Southern Oscillation changed rainfall patterns and wind directions in both areas. During the early 1960s, another meteorologist, Jacob Bjerknes, connected the Southern Oscillation to El Niño events, whence the term “El Niño/Southern Oscillation” (ENSO), which commonly appears in the scientific literature.
When warming occurs in the eastern Pacific, the sea surface temperature gradient between east and west declines. The trade-wind flow weakens, but the weaker east–west pressure gradient has to accompany the reduced trades. Such changes require pressure changes between the eastern and western tropical Pacific—the seesaw effect of the Southern Oscillation.
El Niño. Every few years, the northeast trade winds in the Pacific slacken and the inexorable forces of gravity kick in. Westerly winds increase over the Pacific east of New Guinea, generating subsurface Kelvin waves that push surface water to the east. The trade winds have piled up warm water in the west. As the winds slacken, that water flows eastward. More than 650 feet (200 meters) below the surface of the Pacific, the water temperature becomes much cooler. This thermocline is much closer to the surface in the eastern Pacific. When the Kelvin waves push east, the thermocline sinks in the east and warm water cascades above it toward the American shoreline. The Southern Oscillation changes direction and an El Niño is born. Now the cooler water is in the west. Searing droughts affect Australia and Indonesia, while rain clouds form over the arid Galapagos Islands and the Peruvian coast. The warm, moist air over South America causes the jet streams to lurch north, bringing storms to the Gulf of Mexico and heavy rain to California. The effects of a major El Niño can be severe on a global scale, as the map shows.
The global effects of a major El Niño; compiled from a variety of sources.
La Niña. Once warmer waters have spread into the central and eastern Pacific, some of the Kelvin waves rebound off the South American coast. The reflected waves eventually hit Asia and rebound again. Now the thermocline deepens in the west and shallows in the east. The easterly trades strengthen and the warm water pool in the western Pacific thickens. Upwelling renews, cooling surface waters in the east. El Niño gives way to cooler conditions, which in their extreme, and irregular, manifestations become La Niña, “the Young Girl,” the cool and dry opposite of the Christmas Child. La Niña is still very much of a mystery, but it appears to last longer and have just as serious effects on human societies as its opposing twin, especially those subject to droughts caused by a cooler eastern Pacific.
The ENSO cycle is, and was, a powerful engine of climatic change, constantly swinging, totally unpredictable, and almost as powerful in its global impact as the seasons.
Despite increasingly sophisticated computer models, generations of climatologists have failed to establish any predictable pattern for ENSO events. Historical records over the 270-year period from 1690 to 1987 chronicle eighty-seven such events in Peru, spaced between two and ten years apart, but earlier than that the climatic record is still uncertain, partly because El Niños rarely last more than a couple of years and because they leave inconspicuous traces in the geological record, usually flood deposits in archaeological sites.
By all accounts, Chimor suffered from numerous El Niños. Chimu sites in the Moche and Jequetepeque valleys document massive floods. Dozens of Jequetepeque Valley excavations have documented four major floods, followed by extensive rebuilding, between 2150 B.C. and A.D. 1770.10 A seabed core from a sheltered basin on the Pacific shelf 50 miles (80 kilometers) west of Lima shows a remarkably low concentration of El Niño events between A.D. 800 and 1250, although some did occur, evidenced in a lake in the Ecuadorian highlands.11 From today’s point of view, the significant one took place in A.D. 1230, when a great El Niño descended on the coast and caused widespread flooding and damage. These major events are but a few of the many ENSOs that affected coastal Peru over the centuries, each of them marked by thick flood deposits of rock and gravel carried by cascading water and preserved in river valley walls. There is also abundant evidence of dune activity during arid cycles, when fine sand blew over human settlements, preserving them in thick wind-blown layers in the same deposits. While droughts and El Niños occurred over wide areas, the effects were localized, with more damage in one place than in another, depending on local topography and other factors.
Droughts and El Niños were realities etched into every coastal farmer and fisher family’s minds. Both arrived without warning and at irregular intervals. So did earthquakes and other tectonic activity that uplifted the coast suddenly and caused major landslides, some of which buried valley field systems. Strong winds contributed to desertification and caused sand dune shifts that smothered fertile land. The list of potential hazards was daunting. How, then, did Chimor adapt to such potential catastrophes?
AN EVENING IN A.D. 400. The sun casts long shadows across the plazas and pyramids of Cerro Blanco, near the modern city of Trujillo in the Moche Valley. A huge crowd of artisans, farmers, and petty officials fills the great courtyard in the shadow of the Pyramid of the Sun. A small thatched temple stands high above the plaza, its dark, open doorway in shadow. Drums beat; smoke from sacred fires swirls across the slope of the artificial mountain. Suddenly, the crowd falls silent, all eyes raised to the pyramid summit. The rays of the setting sun bathe a man clothed in gold and silver, who has emerged from the temple. He stands upright, scepter in hand, looking rigidly toward the western horizon. The polished metal glows fiery orange as it catches the slanting light; the living sun god has appeared before his subjects.12
We cannot understand the Chimu without going back a considerable distance in history, for many of their institutions went back centuries to the Moche, who ruled over much of the north coast during the early first millennium A.D.13 The Moche kingdom covered the Lambayeque and several other valleys, probably as a series of fiefdoms governed by a small number of noble families with close kin ties. Political and economic power was in only a few hands, apparently those of men perceived to have supernatural powers. By the time the Chimu rose to prominence along the coast after 1100, the Moche were but a distant memory. The deeds of their mythic lords must have been richly embroidered in oral tradition.
The Moche state collapsed around A.D. 650, in part because of changing political conditions and competition from ambitious neighbors, but also because of the effects of a series of major drought cycles and massive El Niños. But their political institutions and religious beliefs seem to have survived, albeit in slightly altered forms. After A.D. 900, the center of political gravity shifted to another lordly dynasty, that of the Sicán. According to legend, the last Sicán lord, Fempellec, moved an ancient stone idol brought from afar by a royal ancestor, Naymlap, from the Lambayeque Valley to his capital at Batán Grande in the Leche Valley. Rains, floods, and disease promptly ensued. Fempellec’s angry subjects cast him into the Pacific, and Chimu armies fro
m the neighboring Moche Valley conquered the kingdom.
The Chimu had become a significant political power in the Moche Valley after A.D. 800. Their civilization had deep roots in the past, and among its sources were the by-then-almost-forgotten Moche. Perhaps the first Chimu rulers were descendants of once powerful Moche nobles, whose abandoned centers littered the coastal plain. Over the next four centuries, Chimor’s lords began to wield expanding political authority. By 1200, they dominated a broad swath of the north and north-central Peruvian coast. Many of the cultural and political institutions forged by these people and their powerful lords, such as compulsory labor for the state and an elaborate road system, became part of the fabric of the Inca civilization that dominated the Andes when the Spanish arrived.
The Moche lords and their Sicán successors had invested in elaborate ceremonial centers, dominated by artificial mountains fashioned from adobe bricks. The last of these pyramid towns was Batán Grande, which fell before a huge El Niño as the Chimu conquered Sicán. Chimor’s lords had apparently learned a lesson. They built an entirely different form of capital. Instead of investing in pyramids, they channeled resources into securing reliable food supplies. Rather than adobe mountains, they built themselves large, walled compounds where they dwelled in splendid isolation at their capital, Chan Chan, near the mouth of the Moche Valley.14
