Skip To Content
Cambridge University Science Magazine
Water is essential for life. In this article, we explore the palaeoclimatological evidence linking societal change to periods of drought, with a specific focus on the Maya civilisation. It has been the subject of recent sensationalist news articles, so we ask whether science can conclusively confirm how the great society collapsed.

“We definitely consider ourselves palaeoclimatologists,” Nick Evans tells me, when asked his opinion on the matter. A former PhD student in the Department of Earth Sciences, Nick worked extensively on the reconstruction of palaeoclimatic conditions and their relationship with the collapse of the Maya. He is keen to emphasise that his work is not a detailed archaeological study, nor is it sufficient to claim that drought caused the collapse.

Existing across South America from 2000 BCE until 1539 CE, the Maya civilisation is famed for its art, architecture and hieroglyphic scripts. From the highlands of Sierra Madre to the lowlands of Mexico’s Yucatán Peninsula, Maya cities seemingly flourished for centuries. The demise of the Lowland Classical Maya Civilisation during the Terminal Classic Period (800-1000 CE) is thus of great interest to archaeologists, with warfare and strain on resources often implicated in the collapse. In 1995, however, scientists studying the past climate of our planet weighed in on the debate, suggesting in fact that the regional climate had a role to play.

Cambridge professor David Hodell - Evans’ supervisor - carried out the initial work in this area, publishing data from Lake Chichancanab in Yucatán Peninsula. These data consist of the oxygen isotope composition of shells from lake organisms and imply a climatic change synchronous with the collapse of the Maya. These organisms make their shells from calcium carbonate which is precipitated directly from the lake water. In theory, their composition reflects to some degree that of the water in which they resided . This water is in turn influenced by the prevailing climatic state, making these shells powerful indicators of past climate.

The ratio of a heavier oxygen isotope, 18O, to the more abundant and lighter 16O, is thought to depend on the amount of precipitation occurring across a region. The heavier isotope of oxygen falls as rain more quickly than its lighter counterpart, meaning that as clouds and storms travel along their path, the rain falling contains an ever-increasing percentage of the lighter oxygen isotope. This so-called ‘amount effect’ dictates that during times of drought, the oxygen isotopic composition of the water is heavier than at times when more rain falls. This arises because more of the isotopically lighter water can remain locked in clouds and storm fronts when less rain falls, maximising the effects of the fractionation. During times of peak rainfall, however, this lighter water also falls as rain, shifting the observed oxygen isotope composition to lighter values. Prof. Hodell’s 1995 paper found a peak in 18O values during the Terminal Classic period, and thus inferred drought coincident with the collapse of the Maya.

Further work on sediments and speleothems (stalagmites and stalactites) has strengthened the weight of evidence in favour of drought conditions correlating with the collapse of the Maya. Despite this, Evans tells me,



“Until now, no one has been able to quantify the drought.”



The problem, he says, is that the double oxygen isotope method used previously responds to multiple parameters in addition to precipitation, such as humidity and atmospheric conditions. As a result, answers to key questions such as ‘How big was the drought?’ and ‘By how much did precipitation rate decrease?’ have remained elusive. In a 2018 paper, Evans and colleagues were able to combine traditional oxygen isotope measurements with a third isotope, 17O, and two isotopes of hydrogen. As each of these isotopes display different responses to the competing environmental parameters, a clearer picture of precipitation levels can be formed. Such simultaneous isotopic measurements cannot be made on calcitic material, but a convenient alternative is gypsum, a hydrated mineral of calcium sulphate. Precipitating from lake waters during times of high evaporation rates, gypsum traps water molecules in its structure, meaning scientists can release and measure the isotopic composition of this water. In theory, this water reflects the isotopic composition of the lake at the time of gypsum formation and thus records the prevalent climatic conditions at the time.

This innovative approach has allowed for the first quantification of conditions of humidity and precipitation during the presumptive drought. By formulating a complex model of lake dynamics, Evans and colleagues were able to convert the measured isotope measurements of gypsum into estimates of climatic conditions.



The model outputs suggest a reduction in average precipitation rate of 41-54% and a decrease in humidity of 2-7% during the Terminal Classic. Furthermore, it suggests the drought was multidecadal, representing a prolonged period of decreased precipitation rate.



This quantification of the Yucatán drought during the Terminal Classic represents a substantial improvement on previous data and revived tabloid sensationalism regarding the collapse of the Maya civilization. However, despite answering the question of the scale of the drought, Evans is under no illusions regarding the limits of palaeoclimatic data. It alone cannot address the question of how much impact the drought would have had on the Maya civilisation. It tells us nothing of how the drought would have impacted crop yields or of how the Maya people might have been able to adapt their infrastructure to suit the drier landscape. Rather, the data provide a potentially powerful tool for helping us answer these questions. Evans proposes that in the future, progress might be made by producing computer models of societally relevant parameters, such as crop yields, and forcing these models with his climatic data.

The outcomes of palaeoclimatic studies alone are insufficient to conclude whether drought contributed to the collapse of the Maya civilisation. Furthermore it is unclear even in the modern day exactly how climate and society interact. A major drought affected the Levant region of Africa from 2006 to 2011, with the associated displacement of thousands of Syrians from rural communities into urban areas in an attempt to access secure food, water and economic resources. Commentators regularly cite the subsequent stress on infrastructure and social relationships as a major contributor to the anti-government protests in 2011 and the outbreak of civil war. If drought did indeed cause large societal changes in Syria, then it did so through a complex web of interactions. Evans summarises nicely when he says that there is rarely a one-to-one correlation between climate and society.

Despite this complexity, the new data contributes to growing evidence that societies do not respond well to changes in patterns of rainfall. If drought can be linked to war, human suffering and large-scale societal collapse, then it represents a key vector in the discussion on climate change. These problems are intensely personal: while pictures of a collapsing ice shelf might invoke mild indignation, the very real prospect of a struggle for fresh water supplies is a much harder image to shake and is a powerful tool for climate action groups. Nonetheless, the scientific community has a responsibility to draw attention to the limits of its data and to communicate its findings responsibly: it is premature to say from palaeoclimatic evidence alone that drought caused the collapse of the Maya

James Kershaw is a 4th year Earth Scientist at Emmanuel college.

Artwork by Alexander Bates, @as_bates.