A climate model running for thousands of simulated years has produced severe, decades-long droughts over the Maya lowlands without any volcanic eruptions, solar shifts, or other external triggers. The droughts that emerged from the model’s own internal rhythms match the intensity and duration of dry spells that paleoclimate records link to the political fragmentation and population decline of the Classic and Terminal Classic Maya period, roughly the 8th through 10th centuries CE.
The results, published in Quaternary Science Reviews in early 2026, challenge a longstanding assumption in Maya climate research: that droughts severe enough to destabilize complex societies must have been set in motion by some identifiable outside force.
Thousands of simulated years, no external push
The research team, led by climate scientists working with the EC-Earth3 coupled atmosphere-ocean model developed by a European consortium, ran what is known as an 8K simulation. The model held external conditions constant, meaning no volcanic aerosols were injected into the virtual atmosphere and solar output did not fluctuate. Under those fixed conditions, the simulated climate still generated clusters of dry years over the Yucatan Peninsula and northern Central America that persisted for multiple decades.
Precipitation and precipitation-minus-evaporation data from the simulation, archived in a public Zenodo repository, show that internal variability operating on multi-decadal, centennial, and multi-centennial timescales can stack dry years into prolonged episodes. The spatial footprint of the simulated drying covers much of the region that hosted dense Maya populations, including the northern Yucatan and adjacent lowlands.
For decades, researchers have discussed shifts in the Intertropical Convergence Zone and ocean-atmosphere coupling as mechanisms behind Maya-era droughts. Those discussions, however, typically assumed that something external had to initiate the process. The EC-Earth3 experiment strips that assumption away. As the study’s authors write in Quaternary Science Reviews, the coupled atmosphere and ocean, left to their own dynamics, produced drought episodes “comparable in intensity and duration” to those reconstructed from cave minerals and ocean sediments.
What cave and ocean records already told us
The simulation does not exist in a vacuum. Independent proxy records from across the Caribbean and Central American region have documented repeated droughts during the centuries when major Maya cities were abandoned.
A high-resolution speleothem study led by researchers at the University of Cambridge, published in Science Advances, reconstructed wet-season and dry-season rainfall variability in Northwest Yucatan, Mexico, between 871 and 1021 CE. The record identified multiple multi-year droughts and linked reduced construction activity and political disruption at nearby sites to specific dry intervals.
A widely cited speleothem-based reconstruction by Douglas Kennett and colleagues, drawn from a cave in Belize and published in Science in 2012, tied precisely dated dry periods to phases of political fragmentation across Maya polities. And high-resolution sediment records from the Cariaco Basin off northern Venezuela, reported by Gerald Haug and colleagues in a 2003 Science paper, provided a long-duration drought framework that many subsequent studies have used as a timeline anchor for regional hydroclimate shifts during the Classic-to-Terminal Classic transition.
These proxy lines of evidence, drawn from caves and ocean floors, converge on the same broad conclusion: severe droughts struck the Maya lowlands repeatedly during the 8th through 10th centuries. The EC-Earth3 results now offer a physical mechanism for that pattern, demonstrating that internal climate variability alone can account for the timing and severity of such droughts within a physically consistent model.
Where the records disagree
Agreement on the big picture does not mean agreement on the details. A synthesis published in Climate of the Past in 2018 evaluated regional paleoclimate records and highlighted persistent disagreements among archives. Different proxies, whether speleothems, lake sediments, or marine cores, sometimes point to different drought timings or intensities. Some suggest stepwise drying; others favor punctuated extremes separated by partial recoveries.
The Belize cave record and the Cariaco Basin sediments illustrate this tension. One samples mineral deposits in a Maya heartland cave; the other reads ocean-floor chemistry hundreds of kilometers away. Both are frequently cited, but they capture different parts of the climate system and can yield different temporal signatures. Dating uncertainties, seasonal biases in mineral growth, and local hydrology all complicate direct comparisons. Researchers still disagree about the exact onset, duration, and spatial reach of the key drought episodes.
The EC-Earth3 team acknowledges in the paper that their experiment tests internal variability in isolation. “Real-world droughts almost certainly involved some combination of internal cycles and external triggers,” the authors note, adding that disentangling those contributions in the proxy record remains difficult. Volcanic aerosols, modest changes in solar output, and feedbacks from land-surface changes may all have modulated the background variability the model reveals.
The human factor the model does not capture
The simulation demonstrates that the climate system can produce Maya-scale droughts on its own, but it does not account for how human activity may have amplified or accelerated drying at the local level. Archaeological evidence from across the Maya lowlands documents extensive forest clearance, agricultural terracing, and elaborate water-management infrastructure. Deforestation, in particular, can reduce local moisture recycling and increase surface temperatures, potentially deepening droughts that the broader atmosphere-ocean system initiates.
These land-use feedbacks are not represented in the EC-Earth3 experiment. Coupled climate-archaeological models that integrate natural variability with human land modification remain a next step rather than a completed analysis. No published responses from Maya archaeologists integrating the EC-Earth3 results with site-specific excavation data have appeared alongside the study as of May 2026.
That gap matters because the debate over the Maya “collapse” has never been purely about climate. Some researchers foreground environmental stress; others emphasize warfare, trade disruption, or internal political dynamics. The new simulation does not resolve those debates. It does, however, establish that the climate backdrop was capable of generating the kinds of droughts that feature prominently in many of those narratives, without requiring a volcanic eruption or solar anomaly to set them off.
Droughts without warning signs: what the EC-Earth3 experiment means for rainfall-dependent regions
If the climate system can generate multi-decadal droughts severe enough to strain complex societies through its own internal oscillations, the implications extend well beyond archaeology. The EC-Earth3 simulation showed that the Yucatan Peninsula and northern Central America received drought clusters lasting decades under perfectly stable external conditions. Present-day communities across those same lowlands still depend heavily on seasonal rainfall for maize agriculture and municipal water supplies, and they would face similar precipitation deficits without any identifiable external warning sign like a major eruption.
Anthropogenic warming adds another layer. Rising baseline temperatures increase evaporative demand, meaning that even modest reductions in rainfall can produce more severe agricultural drought than the same precipitation deficit would have caused centuries ago. How warming interacts with the kinds of internal oscillations highlighted by the EC-Earth3 experiment is not yet well understood. Answering that question will require models that bridge physical climate processes, land-use change, and the social systems that ultimately experience drought as either a manageable dry spell or a full-blown crisis.
The ancient Maya case offers a warning not because modern societies are destined to repeat it, but because it demonstrates that prolonged, society-stressing droughts can emerge from the climate system’s own variability. Planning for that possibility means looking beyond eruptions and solar cycles and taking seriously the droughts that need no external trigger at all.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
