National Science Review PERSPECTIVE 10: nwad082, 2023 https://doi.org/10.1093/nsr/nwad082 Advance access publication 23 March 2023 EARTH SCIENCES Catastrophic climate change and the collapse of human societies Josep Peñuelas1,2,∗ and Sandra Nogué1,2,3 The scientific community has focused the may easily overwhelm societal adaptive cal instability, systemic financial risks, the agenda of studies of climate change on capacity. spread of infectious diseases and the risk lower-end warming and simple risk anal- These climate projections could be of spillover. For example, there is evi- yses, because more realistic complex as- even more detrimental if models would dence that the 2007−10 drought con- sessments of risk are more difficult, the not neglect, as they currently do, feed- tributed to the conflict in Syria [11]. benchmark of the international targets back in the carbon cycle and potential Anthropogenic climate change inter- is the Paris Agreement goal of limiting tipping points that could generate higher acting with these other stressors could warming to <2◦ C, and the culture of cli- GHG concentrations [4]. Examples in- thus cause a global catastrophe, in a mate science is to try to avoid alarmism clude the apparent slowing of dampen- worldwide societal collapse. Kemp et al. [1]. Current fires, prolonged droughts, ing feedbacks such as the natural carbon- [1] have reminded us that although we floods and heat waves, together with the sink capacity [5,6], the loss of carbon have reasons to suspect it, such potential consequent food insecurity, civil unrest due to increasing frequencies and inten- collapsing futures are rarely studied and and migrations, however, are opening the sities of fire at northern latitudes [7], poorly understood. The closest research eyes not only of most scientists but also droughts and fires in the Amazon [8] or is the search for evidence of tipping of most people all over the world to the the thawing of Arctic permafrost that re- dynamics and estimating thresholds, need for considering, at least, the poten- leases methane and CO2 [9]. This feed- timescales and impacts of potential tial catastrophic effects of the collapse of back is also likely not proportional to tipping points [4]. We advocate for ecosystems and society due to the current warming, as is sometimes assumed. In- considering them while using the avail- emergency of climate change. stead, abrupt and/or irreversible changes able knowledge acquired from historical The projections for the climate of may be triggered at a temperature thresh- and prehistorical examples of local and the coming decades are, as we all know, old [7]. Particularly worrying is a ‘tipping regional collapses, transformations and worrying. The worst-case scenarios in cascade’ in which multiple tipping ele- resilience of human societies also driven the 2022 Intergovernmental Panel on ments interact in such a way that tipping by climate and unsustainable use of Climate Change (IPCC) report project one threshold increases the likelihood of resources (Fig. 1). temperatures by the next century that tipping another [4,10]. Climate change together with turbu- last occurred in the Early Eocene, re- Climate change also interacts with lent political rivalries and resource mis- versing 50 million years of cooler cli- other anthropogenic stressors such as management that have hindered solv- mates within two centuries. The Pliocene changes in land use, loss of biodiver- ing underlying problems have played key and Eocene provide the best analogues sity, nutrient imbalances, pollution and roles in the collapse or transformation for near-future climates [2]. Climates an overuse of available resources that are of numerous previous human societies. like those of the Pliocene are likely to crossing the planetary safety boundary Variables associated with temperature prevail as soon as 2030 and unmiti- limits and operating as a possible catas- have been identified as the most strongly gated scenarios of emissions of green- trophic mix. This mix may exacerbate so- limiting factors to the growth of popula- house gases (GHGs) will produce cli- ciety vulnerabilities and cause multiple tions of European hunter-gatherers dur- mates like those of the Eocene for the indirect stresses such as economic dam- ing the late Pleistocene [12]. Droughts coming decades. This situation is par- age, loss of land and water, and food have also been identified as drivers of ticularly alarming because human soci- insecurity that can merge into system- stress and collapse. Thanks to analysis eties are locally adapted to a specific cli- wide synchronous failures. These cascad- of tree rings from Juniperus trees cou- matic niche with a mean annual tempera- ing effects are not only biophysical or pled with 13 C stable isotope analysis, it ture of ∼13◦ C [3]. We can thus logically biogeochemical, but they also affect hu- was possible to determine the impact expect that current and future warming man society, generating conflicts, politi- of a multi-year drought on the Ancient  C The Author(s) 2023. Published by Oxford University Press on behalf of China Science Publishing & Media Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Natl Sci Rev, 2023, Vol. 10, nwad082 Figure 1. Integration of elements indicating: (i) examples of approaches to determine ancient environmental change and societal complexity; (ii) a regional/local ancient collapse described by an example from Mayapan, the largest Postclassic Mayan city [20]; and (iii) the modern potential global collapses described by Kemp et al. [1]. The figure shows the interconnected nature of climate change, unsustainable use of resources and civil unrest. It also highlights the relevance of ancient technologies and management strategies to mitigate the consequences of the complexity of climate change (e.g. Refs [20–25]). Anatolian empire of the Hittites that The study of collapses of previous human the diet was potentially complemented collapsed >3000 years ago [13]. More societies, in many cases associated with with drought-resistant indigenous plants recently, the Central American Mayan changes in climate and resource scarcity [21]. Other examples of cultural and civilization built cities supporting pop- such as those that currently character- technological practices are the reservoirs ulations of ≤100 000 and reached its ize anthropogenic global change on our built around the temples of Angkor Wat, peak about 800–1000 CE before col- planet, sheds light on worst-case scenar- the Sri Lankan reservoirs that have pro- lapsing due to a major transformation ios for the coming years and decades of vided irrigation water for cultivation for of its political and social structures and climate change, but also on how to over- 2000 years and the terraces built to mit- the unsustainable use of their environ- come them, on possible resistance and re- igate the risk of flood and to support agri- ment, especially by deforestation and ero- silience [18,19]. For example, bioarche- culture [22,23]. sion [14]. Similar collapsing processes ological data from Mayapan (Yucatán, The data on local and regional occurred in, e.g. the Roman Empire, Mexico) indicate that Mayan political collapses thus provide very useful infor- Greenland-Iceland Norse culture [15] and economic structures in the shape mation not only on the reconstruction of and Easter Island in the Pacific Ocean of a network of small Mayan states per- past climates and past societal vulnerabil- [16,17] sisted until the Spanish colonization, in- ities, but also on the ways and the ancient These prehistorical and historical data dicating some amount of regional resis- strategies and technologies that miti- demonstrate that societal collapse rarely tance and resilience [20] (Fig. 1). In ad- gated aspects of past changes in climate, has a single cause but is often due to dition, the diet of the lowland Maya may such as impacts of extreme droughts many biophysical and socio-economic have been more varied than what was on food security and breadbaskets stressors, including climate, overexploita- previously thought. According to ethno- [24]. Analysing the mechanisms for the tion of resources and societal conflicts. graphic and ethnobotanical assessments, extreme consequences of climate change, Page 2 of 3 Natl Sci Rev, 2023, Vol. 10, nwad082 overexploitation of resources and mis- ∗ Corresponding author. 13. Manning SW, Kocik C and Lorentzen B et al. management of people, but also the E-mail: josep.penuelas@uab.cat Nature 2023; 614: 719–24. resilience of some of these previous 14. Kennett DJ, Breitenbach SFM and Aquino VV human societies, is strongly warranted. REFERENCES et al. Science 2012; 338: 788–91. This needed research must also consider 15. Dugmore AJ, McGovern TH and Vésteinsson O 1. Kemp L, Xu C and Depledge J et al. Proc Natl Acad the differences in technologies, social et al. Proc Natl Acad Sci USA 2012; 109: 3658– Sci USA 2022; 119: e2108146119. complexity and cognitive levels to those 63. 2. Burke KD, Williams JW and Chandler MA et al. cases that occurred during prehistoric 16. DiNapoli RJ, Crema ER and Lipo CP et al. Nat Proc Natl Acad Sci USA 2018; 115: 13288–93. and historical periods. Such analyses Commun 2021; 12: 3939 . 3. Xu C, Kohler TA and Lenton TM et al. Proc Natl could help to better convince and inform 17. Rull V. Quat Sci Rev 2021; 252: 106751. Acad Sci USA 2020; 117: 11350–5. managers and politicians, and society in 18. Butzer KW. Proc Natl Acad Sci USA 2012; 109: 4. Armstrong McKay DI, Staal A and Abrams JF et al. general, to take action to mitigate and 3632–9. Science 377: eabn7950. adapt to climate change, improve re- 19. Middleton GD. Urban Hist 2021; 48: 724–48. 5. Peñuelas J, Ciais P and Canadell JG et al. Nat Ecol silience and better prepare and diversify 20. Kennett DJ, Masson M and Lope CP et al. Nat Evol 2017; 1: 1438–45. possible emergency responses. Commun 2022; 13: 3911 . 6. Wang S, Zhang Y and Ju W et al. Science 2020; 21. Fedick SL and Santiago LS. Proc Natl Acad Sci 370: 1295–300. FUNDING 7. Descals A, Verger A and Yin G et al. Remote Sens USA 2022; 119: e2115657118 . This work was supported by the TED2021- 22. Brown AG, Fallu D and Walsh K et al. Geomor- Ecol Conserv 2022; 9: 76–89. 132627B-I00 grant, funded by MCIN and the phology 2021; 379: 107579 . 8. Yang Y, Saatchi SS and Xu L et al. Nat Commun European Union NextGeneration EU/PRTR, 23. Evans DH, Fletcher RJ and Pottier C et al. Proc 2018; 9: 3172. and the European Research Council grant Natl Acad Sci USA 2013; 110: 12595–600. 9. Kille N, Chiu R and Frey M et al. Geophys Res Lett ERC-CoG-2021-101045309 TIME-LINE. 24. Boivin N and Crowther A. Nat Ecol Evol 2021; 5: 2019; 46: 3990–8. 273–84. Conflict of interest statement. None declared. 10. Klose AK, Wunderling N and Winkelmann R et al. 25. Kemp L and Cline EH. In: Izdebski A, Haldon J Environ Res Lett 2021; 16: 125011. and Filipkowski P (eds.). Perspectives on Public Josep Peñuelas1,2,∗ and Sandra Nogué1,2,3 11. Kelley CP, Mohtadi S and Cane MA et al. Proc Natl 1 CREAF, Cerdanyola del Vallès, Spain; 2 CSIC, Global Policy in Societal-Environmental Crises: What the Acad Sci USA 2015; 112: 3241–6. Future Needs from History. Cham: Springer Inter- Ecology Unit CREAF-CSIC-UAB, Spain and 12. Ordonez A and Riede F. Nat Commun 2022; 13: 3 Universitat Autònoma de Barcelona, Spain national Publishing, 2022, 207–23. 5140. Page 3 of 3