The A.D. 1130–1180 Megadrought in the Northern Southwest

Posted June 5, 2024

Dr. Susan C. Ryan, Executive Vice President of the Research Institute

The effects of changing climatic conditions on human and non-human populations are some of the most discussed topics throughout the world today. Climate change affects how modern humans interact with their environments (and vise-versa), as has been the case for the past 100,000 years! For this reason archaeologists are in a unique position to study past human behaviors to better understand how we can navigate changing conditions in the present. A large body of work within Southwestern archaeology has focused on the role environmental downturns played in the mass-migrations of ancestral peoples from the northern Southwest to areas including the northern Rio Grande region, Zuni Pueblo, Acoma Pueblo, and the Hopi Mesas at the end of the 13th century A.D. In addition to a myriad of social, political, and economic factors, there is consensus the Great Drought of A.D. 1276–1299 played a role in the final depopulation of the northern Southwest. However, in terms of duration and reduced precipitation, the Great Drought was less severe than previous droughts whose impacts on ancestral peoples are less clear, including droughts in the early decades of the A.D. 800s and A.D. 1130–1180.

The A.D. 11301180 Drought

The single greatest megadrought recorded for North America occurred in the western half of the continent from A.D. 1140–1162. This period had 23 consecutive years of negative Palmer Drought Severity Index values. Resulting in a period of prolonged moisture deficiency in the Colorado Plateau spanning five decades, from A.D. 1130–1180, this megadrought was nothing to scoff at! During a study of drought in the northern Southwest, paleoclimate researchers reconstructed annual Colorado River flows at Lees Ferry, Arizona for the period A.D. 762–2005 using tree-ring samples and stream flow data beginning in the year 1906. Although this study was based solely on stream flow—which is affected by factors other than precipitation—it clearly indicates Colorado River flows in the mid A.D. 1100s were the lowest of the past 1,200 years!

Natural and Cultural Effects of Drought

The mid-A.D. 1100s megadrought certainly had environmental repercussions including depressed water tables, eroded floodplains, decreased climatic variability, reduced precipitation, and diminished agricultural productivity. Additionally, the mid-A.D. 1100s drought had significant cultural impacts on occupants in the Southwest due to its duration, intensity, and persistence. Yet relatively little attention has been paid to the cultural impact of the A.D. 1130–1180 drought despite that this period coincides with the decline and reorganization of the Chaco regional system. In general, construction of great houses in Chaco Canyon ceased by A.D. 1130. Is this a coincidence? To further evaluate the effects of climate change on ancestral populations in the northern Southwest, two questions require attention: Were great house communities in the Mesa Verde region depopulated during the A.D. 1130–1180 drought? Did deteriorating environmental and/or social conditions lead to migration events? We turn to dendroarchaeology for answers.

The Tree-Ring Record for A.D. 1130–1180

Trees harvested for construction materials, firewood, and other domestic needs are frequently preserved in the archaeological record of the American Southwest. The likelihood of preservation increases when wood has been charred as fungi are unable to break down burned wood as easily as unburned wood. Archaeologists commonly infer the settlement history of regions and villages from ancient wood harvesting activities. Gaps in tree-cutting activities may indicate periods of reduced harvesting and/or the depopulation of a village or region.

Archaeologists note relatively few cutting dates (dates generated when the outside, annual ring and bark are present on a wood sample) have been recovered from several periods in the Mesa Verde region, including three decades between A.D. 1150–1170 coinciding with the environmental downturn. If regional tree-ring patterns are indeed reflective of human occupation, wood harvesting and construction projects should be evident in the tree-ring record when populations were static or decreasing. Even though few sites were demonstrably constructed during the A.D. 1130–1180 period, it seems highly unlikely the region was depopulated due to the drought.

A Pause in Construction

Why haven’t we recovered more tree-ring samples dating to the A.D. 1130–1180 period in the Mesa Verde region? Does this trend indicate populations moved from the area? Answers may be threefold. First, the use-life of wood roofs in earthen pit structures was much shorter than the use-life of roofs constructed in masonry buildings (wood in direct contact with the ground deteriorates much quicker due to bacteria, fungi, and termites feasting on the posts). Examples of historic and Prehispanic tree harvesting hiatuses—despite documented occupation—come from Walpi Pueblo, Zuni Pueblo, and Pueblo Bonito. The Walpi cutting-date distribution, which includes more than 200 dates, indicates two gaps lasting longer than a decade during the 1780s and 1790s and then again in the 1940s and 1950s. The Zuni Pueblo cutting-date distribution also indicates several long gaps at A.D. 1700–1750, 1750–1770, 1770–1800, and 1810–1830. Both Walpi and Zuni Pueblos were continuously occupied during these years. Moreover, the occupation of Pueblo Bonito was continuous (based on pottery and stratigraphic evidence), yet the tree-ring record indicates multiple decade-long gaps in harvesting activities from A.D. 864–915, 924–1020, and 1110–1125. Thus, these examples suggest people are known to continually occupy masonry structures for up to three generations without needing to harvest trees for repairs.

Second, recycling behaviors in the past contribute to the reduced number of tree-ring dates from the A.D. 1130–1180 period. When structures weren’t burned during decommissioning events, construction materials were often reclaimed for other purposes. Many unburned roof beams were recycled into roofs of new structures. An example of recycling rates comes from excavations at Sand Canyon Pueblo, a canyon-head village occupied from A.D. 1260–1280. Sixty-eight percent of the 275 cutting dates represent recycled construction beams—30 percent of which were harvested prior to A.D. 1225 despite the fact the village was continuously occupied until A.D. 1280.

Finally, sampling strategies may also bias our understanding of tree-ring data. For example, a single kiva excavated in the Mesa Verde region yielded 73 cutting dates from the A.D. 930–970 period which, when added to the regional tree-ring record, essentially erased an existing gap.

Seeking Multiple Lines of Evidence

Wood harvested by ancient peoples is not always present or preserved in the archaeological record, therefore multiple lines of diagnostic evidence—including stratigraphy, pottery accumulation rates, and architecture—must be examined in combination with tree-ring data to determine occupational histories of villages and regions. Examining pottery accumulations from associated middens to determine the number of years a structure was occupied will not provide an adequate understanding of how long wooden elements within structures can exist without repair.

Occupation of the Mesa Verde Region During the A.D. 11301180 Drought

With data generated from the Crow Canyon Archaeological Center Research Database, the map provided at the top of the story illustrates numerous sites with tree-ring cutting dates between A.D. 1150–1170 from sites located in present-day southwest Colorado. Although these data do not necessarily indicate occupation of each village or residence during this period (for example, Castle Rock Pueblo was occupied in the A.D. 1200s), dates do indicate people were in the immediate vicinity harvesting wood during the drought. These data suggest people continued to harvest timbers for construction, albeit at a reduced rate, and were making informed decisions for how best to socially mediate shifting environmental conditions. In contrast, areas including Cedar Mesa, in present-day southeast Utah, appear to have been depopulated during the drought. The central and eastern portions of the Mesa Verde region may have been one of the best locations to thrive during the megadrought.

The Past Meets the Present

Gaining insights into past social dynamics during the A.D. 1130–1180 drought provides a much-needed contrast to the mass-migration events in the northern Southwest during the Great Drought at the end of the 13th century A.D. No two environmental downturns are the same and there is no one-size-fits-all strategy for how humans mediate changing conditions. Severe and extended droughts are an inevitable part of natural climactic cycles across the globe and—although we can’t control them—we can predict droughts and employ a myriad of human strategies developed over countless generations for how to respond, including relocation, the recycling of construction materials, and reducing large-scale construction projects. I encourage you to consider the ways we rely upon water in our daily lives for drinking, cooking, bathing, electricity, recreation, washing, raising plants and animals, and so much more. Have you taken water for granted? What choices will you make to adapt to a consistently changing climate?