The following is a guest post by Ornob Alam, a graduate student in Michael Purugganan’s lab at New York University. Ornob’s PhD projects examine the demographic and evolutionary history of domesticated Asian rice in the context of past climate change and human migrations; he is on Twitter as @genomeinquirer.
While modern, anthropogenic climate change threatens to be unprecedented in rapidity and scale, humans have always lived at the whim of an unpredictable planet. For decades, paleoclimatologists have been using tree rings, ice cores, lakebed sediments, and other natural records to reconstruct ancient climates, revealing a dynamic tapestry of global weather patterns through time. This has allowed archaeologists, paleontologists, and others concerned with understanding the past to place historical events within their climatic context, notably implicating climate change as having helped to drive the Neolithic shift from hunter-gatherer to agricultural societies, the rise and fall of civilizations, and megafaunal extinctions.
Recently, this alliance between archaeology and paleoclimatology has welcomed genomics into the fold, with the emergence of studies that incorporate all three. As any single kind of historical record is incomplete at best and misleading at worst, such interdisciplinary collaboration strengthens our ability to make inferences about the past. Whole genome sequencing, ancient DNA extraction technologies, and population genetics methods allow us to not only study the demographic history of humans using modern and ancient human DNA, but also many aspects of their behavior and culture by exploring their biotic environments. Two studies published in 2020 used non-human genomes to show long-lasting effects of ancient climate change on human societies in different parts of eastern Asia.
Rice finds a way
The domestication of Asian rice (Oryza sativa) by about 8,000 years ago transformed societies living in the Yangtze valley. But for thousands of years, the influence of rice agriculture did not extend far beyond its local sphere in southern China (explore rice archaeobotanical sites here). Japonica rice, the original domesticate, begins to sparsely appear in the archaeological record in Southeast Asia between 5,000 and 4,000 years ago, and was likely more prestige food than a staple in the region until as recently as 2,000 years ago. Meanwhile, indica rice arose in the Indian subcontinent, also sometime between 5,000 and 4,000 years ago, following hybridization between imported japonica varieties and locally cultivated proto-domesticated rice. Over in the temperate northeast zones of Korea and Japan, rice archaeobotanical sites begin to multiply after 3,000 years ago.
While the archaeological record is inherently flawed through its dependence on chance, these trends over a large, densely sampled area suggest a major geographical expansion of rice agriculture starting between 5,000 and 4,000 years ago and intensifying afterwards. Paleoclimate data, in turn, has revealed a decline in global temperature starting from about 5,000 years ago and culminating in the steep drop 4,200 years ago described as the 4.2K event. This climatic shift was hypothesized to have resulted in the emergence of temperate-adapted rice from the ancestral tropical japonica rice in China. It is in this context that Gutaker et al. entered the conversation in 2020.
Population genetic methods attain remarkable power to infer demographic and evolutionary history when applied to high-resolution whole-genome sequence data. Gutaker et al. employed these methods to analyze genome sequences of more than a thousand traditional rice landraces across Asia. After showing that the landraces could be grouped into genetic clusters or subpopulations that largely corresponded to geography, they were able to use coalescent modeling– which is based on reconstructing genealogical paths to a common ancestor – to date the origin of the temperate japonica subpopulation to about 4,100 years ago, and the diversification of rice in Southeast Asia to about 2,500 years ago. Notably, the latter date, while later than the oldest archaeological rice remains in the region, is consistent with more widespread adoption of rice agriculture in the region about 2,000 years ago.
The dates were thus consistent with northeastern and southern expansions of temperate and tropical rice, respectively, following the global temperature decline that peaked with the 4.2K event. Providing further support for a climate-driven expansion of rice agriculture, they also showed that temperature explained a substantial portion of genetic variation across rice subpopulations. Finally, the paper employed paleoclimate reconstruction in conjunction with data on habitats favored by tropical and temperate rice to show that the climate in China and northeast Asia could no longer support tropical japonica rice after about 4,000 years ago. It appears, then, that the people who relied on rice agriculture at the time responded to climate change by migrating to more favorable environments, and through the adoption – whether or not they developed it through conscious selection – of temperate-adapted rice.
A shrinking wilderness
Over to the southwest, hunter-gatherer groups living near the northeastern margins of the Tibetan plateau subsisted on big game hunting until about 5,200 years ago. Millet agriculture was then introduced from the north, but hunting still likely played a major role until about 4,000 years ago. This gave way to large, fully farming-based societies by about 3,600 years ago, with the arrival of the cold-hardy Near Eastern crops wheat and barley, as well as domesticated animals like sheep and cattle. The adoption of wheat and barley likely helped buffer against the temperature decline at the time, and facilitated settlement of higher altitudes than allowed by millets.
Did animal husbandry replace big game-hunting because of higher productivity, greater compatibility with a sedentary lifestyle, or a decline in big game populations? It was likely a combination of all three, but in the context of the accompanying climatic transition, understanding the fate of the big game animals can help clarify relationships between climate change, human activity, and biodiversity.
Chen et al. analyzed animal remains from an archaeological site in the northeastern Tibetan plateau dated to about 5,200 years ago. The assemblage was largely composed of wild animals rather than domesticates, and remarkably for a modern day temperate region, included several tropical species such as bamboo rats, rhinoceroses, and unidentified wild bovids. They sequenced ancient DNA from the rhinoceroses and bovids at the site, and identified the species as Sumatran rhinoceros and tropical gaur, which are both only found much further south in tropical climates today. This suggested that the range of these species had extended to the northeastern Tibetan plateau before 5,000 years ago. They report paleoclimate proxy data from neighboring regions indicating warmer summer temperatures around the time, which would have allowed the species to graze in the north.
Using the newly sequenced ancient gaur genomes, along with sequences of modern gaurs, the authors reconstructed the population history of the species, and detected a rapid decline in population size about 5,000 years ago, which continued for more than a thousand years. This must have coincided with a shrinking habitat range to the north, as indicated by decreasing annual precipitation and summer temperatures in the region between 5,000 and 4,000 years ago. To strengthen their case for a climate-driven range contraction, they looked at the proportion of wild vs. domesticated animals in zooarchaeological sites across the Yellow River basin between 8,000 and 3,000 years ago, and found a steep reduction in wild mammal diversity between 5,000 and 4,000 years ago.
An innovative synthesis of population genetic, paleoclimate, and archaeological data had thus provided robust evidence that climate change contributed to the decline in available big game in the region, which likely accelerated the transition to animal husbandry. Are humans then completely absolved of any role in this loss of biodiversity? Not quite. The authors also found an association between increasing human activity, as reflected by the number of archaeological sites, and decreasing wild mammal diversity across the Yellow River basin between 8,000 and 3,000 years ago. It seems plausible that overexploitation of already shrinking populations hastened the end for many wild mammals in the region. But humans not only survived the loss of a major food source, they actually thrived, owing to the adoption of new subsistence strategies.
These studies are the culmination of many advances but at the same time, they’re just the tips of an iceberg. Understanding the history of dispersal of crops and the associated climatic drivers as in Gutaker et al. paves the way toward investigations of genetic adaptations to local environments, perhaps in the vein of another innovative study – on the chocolate tree – from last year. Identifying adaptations to specific climates may be key to developing crop improvement strategies to mitigate the effects of ongoing climate change. Ancient DNA helps benchmark population genetic analyses by providing accurate historical snapshots, analogous to how fossils are used to calibrate macroevolutionary timescales. Greater sampling of non-human ancient DNA as in Chen et al. will not only reveal aspects of our own sociocultural history, but also help in the study of processes like extinction, domestication, and adaptation in diverse organisms, especially in the face of environmental changes. Viewed in light of their present and past contexts, the study of genomes can unlock a greater appreciation of the historical contingency of the biology within and around us.
R. M. Gutaker et al. 2020. Genomic history and ecology of the geographic spread of rice. Nature Plants, 6(5). doi: 10.1038/s41477-020-0659-6
N. Chen et al. 2020 Ancient genomes reveal tropical bovid species in the Tibetan Plateau contributed to the prevalence of hunting game until the late Neolithic. Proc. Natl. Acad. Sci., 117(45):28150–28159. doi: 10.1073/pnas.2011696117