The tree had been dead for millennia when scientists cut into its rings. Preserved in sediment near the French Alps, its wood held a secret: a radiocarbon anomaly so extreme it rewrote the upper limit of what the Sun can do.

Around 12,350 BC, during the waning days of the last glacial period, the Sun unleashed an extraordinary burst of energetic particles toward Earth. The evidence survives as a sharp spike in radiocarbon—the isotope carbon-14—locked inside the growth rings of subfossil trees. That spike measures roughly 40 parts per thousand, nearly double the size of any similar event discovered within the past 12,000 years. Until now, scientists lacked the tools to assess what such a spike meant under ice age conditions. A new atmospheric model has changed that.

The finding establishes this event as the strongest known extreme solar particle event in Earth's history. It also provides a rare chronological anchor for dating ancient materials from a period when precision timelines grow frustratingly vague.

When the models couldn't reach back far enough

Radiocarbon spikes tied to extreme solar storms are rare. Only eight had been identified across the Holocene—the warm epoch stretching from roughly 10,000 BC to the present. The most famous occurred in 775 AD, leaving a signature of about 20 parts per thousand in tree rings worldwide. Researchers confirmed its solar origin by cross-referencing the radiocarbon signal with beryllium-10 and chlorine-36 isotopes trapped in polar ice.

The 12,350 BC event was discovered recently in tree samples from southwestern Europe. Its radiocarbon signal stood out immediately: almost twice as large as the 775 AD benchmark. Yet translating that raw measurement into an estimate of the solar storm's strength proved impossible. Traditional carbon-cycle models rely on simplified "box" structures where large reservoirs—atmosphere, ocean surface, deep ocean, biosphere—exchange carbon at fixed rates. Those models assume relatively stable climate conditions. They work well for the Holocene. They fail for glacial climates, when atmospheric carbon dioxide levels, ocean circulation, ice coverage, and vegetation patterns differed sharply from today.

The problem wasn't just academic. Without proper modeling, scientists couldn't determine whether the 12,350 BC storm merely appeared larger due to environmental factors, or whether it genuinely surpassed every other known event.

A new model built for frozen worlds

Researchers developed a next-generation tool called SOCOL:14C-Ex, a chemistry-climate model that simulates radiocarbon production and atmospheric transport in three dimensions. Unlike box models, it accounts for winds, stratosphere-troposphere exchange, regional topography, and latitudinal variations in carbon sinks. Crucially, it can operate under glacial boundary conditions: lower carbon dioxide (240 parts per million versus 285 ppm during the Holocene), extensive ice sheets, colder sea surfaces, reduced vegetation cover.

The model divides the atmosphere into grid cells roughly 2.8 degrees on a side, spanning 39 vertical levels from ground to 80 kilometers altitude. It tracks how radiocarbon—produced high in the polar stratosphere when solar particles slam into nitrogen and oxygen—descends through large-scale circulation, mixes across latitudes, and gets absorbed by oceans and plants. Each location's carbon sink depends on surface type, albedo, latitude, and season. Ice-covered regions have no sink. Open water absorbs carbon year-round. Mid-latitude land absorbs carbon only during the local growing season.

Before applying it to the glacial event, the team validated the model against 775 AD. They simulated the known solar particle spectrum for that storm—based on the extreme solar event of January 2005, scaled upward by a factor of 455—and compared the model's atmospheric radiocarbon predictions to actual tree-ring measurements from Germany, Russia, the United States, New Zealand, Argentina, and Canada. Agreement was excellent. The model reproduced not just the peak radiocarbon values but also the shape of the rise, the timing differences between hemispheres, and the regional spread. No adjustable parameters. No post-hoc tuning.

Disentangling ice, air, and magnetism

Three factors control how a given solar storm translates into a measured radiocarbon spike: geomagnetic shielding, atmospheric carbon dioxide levels, and climate-driven circulation patterns.

Geomagnetic shielding matters because Earth's magnetic field deflects lower-energy solar particles away from mid and low latitudes. Only the polar regions remain fully exposed. When the field is weak—as it was around 12,350 BC, with a dipole moment of 6.3×10²² ampere-square-meters compared to 9.5×10²² for 775 AD—more particles penetrate deeper into the atmosphere. The model shows this boosts radiocarbon production by roughly 25 percent for the same particle flux.

Carbon dioxide levels affect the radiocarbon-to-total-carbon ratio. Since radiocarbon measurements express the abundance of carbon-14 relative to carbon-12, a lower atmospheric CO₂ concentration amplifies the apparent signal. The late glacial atmosphere held 240 ppm; by 775 AD it had risen to 285 ppm. That difference reduces the measured spike by about 16 percent for identical radiocarbon concentrations.

Climate patterns influence how quickly stratospheric radiocarbon reaches the ground and how efficiently it gets removed by sinks. The colder glacial climate slowed the large-scale Brewer-Dobson circulation, delaying the descent of radiocarbon-rich air. It also reduced oceanic and biospheric uptake rates due to colder surface waters and sparser vegetation. These effects nearly canceled each other out at the peak of the radiocarbon pulse, producing only a one-percent difference in maximum values between glacial and Holocene simulations.

Together, the combined effect of weaker geomagnetic shielding, lower CO₂, and altered climate patterns meant that the same solar storm would produce a radiocarbon signal 1.5 times larger in 12,350 BC than in 775 AD.

Twice the signal, eighteen percent more fury

Working backward from the observed spike, the researchers fitted their model to the tree-ring data. They treated two parameters as unknowns: the storm's strength relative to 775 AD, and the date it occurred.

The best fit placed the event in early March 12,350 BC, with the storm about 18 percent stronger than the 775 AD benchmark. The uncertainty spans roughly plus-or-minus 11 percent. The timing estimate allows a broader window—anywhere from early January through late April—though early March emerges as most probable. That seasonal range comes from comparing the first-year radiocarbon rise to the second-year peak. If the storm had struck in autumn, more radiocarbon would have reached the ground before the growing season ended, producing a higher first-year value. A late-winter or early-spring event matches the observed pattern.

The storm's particle fluence—the total number of particles integrated over the entire event—likely reached 1.4×10¹⁰ particles per square centimeter for energies above 200 megaelectronvolts. That exceeds the 775 AD event's fluence by the same 18 percent margin. For context, the strongest solar storm directly observed by satellites, which occurred in February 1956, delivered a fluence roughly 500 times weaker.

What this means for Earth, then and now

An event of this magnitude would have dramatically increased radiation doses in the upper atmosphere, though the effects on surface life remain speculative. The cosmic-ray-induced ionization of the atmosphere would have spiked, potentially affecting cloud formation, lightning rates, and atmospheric chemistry. However, the dense glacial ice sheets and colder climate likely minimized biological impacts compared to what a similar storm would inflict today.

Modern society would not fare as well. Solar storms inject high-energy particles into Earth's magnetosphere, generating intense radiation belts that can destroy satellite electronics, disrupt GPS navigation, and induce damaging electrical currents in long-distance power grids and pipelines. The 1989 Quebec blackout resulted from a comparatively minor geomagnetic storm. A 12,350 BC-scale event striking today could cripple satellite constellations, disable communication networks, and trigger cascading infrastructure failures across continents.

The discovery also provides a chronological gift to archaeologists and paleoclimatologists. Radiocarbon dating relies on knowing the past atmospheric concentrations of carbon-14. Sharp, well-dated spikes serve as "tie points"—fixed markers that allow researchers to synchronize floating tree-ring chronologies, refine calibration curves, and anchor sediment cores. All eight previously known extreme events occurred during the Holocene, when chronologies are already relatively secure. The 12,350 BC spike is the first—and so far only—such marker extending into the late glacial period, a time when dating uncertainties span decades or centuries. Its discovery tightens the timeline for ice-core records, pollen sequences, and archaeological sites from that era.

The shape of solar violence

Eight extreme solar particle events now populate the Holocene record. Two—775 AD and 7176 BC—rank as the strongest, comparable in size within measurement uncertainties. The others trail by factors of 1.5 to 3. The newly characterized 12,350 BC event joins that top tier, possibly edging ahead as the single most powerful.

Statistical patterns remain elusive with such small numbers, but the clustering of event sizes hints at a natural upper limit. The Sun may have a characteristic maximum for how much energy it can release in a single eruptive burst, at least over the timescales recorded in cosmogenic isotopes. Whether that ceiling reflects fundamental magnetic energy storage limits, flare physics, or particle acceleration processes remains an open question.

The model that made this analysis possible—SOCOL:14C-Ex—now stands ready for broader applications. It can simulate radiocarbon transport under any climate and geomagnetic configuration, opening the possibility of studying solar activity across the entire sweep of the last glacial cycle and beyond. Future work will require matching the radiocarbon data with measurements of beryllium-10 and chlorine-36 from ice cores to reconstruct the full energy spectrum of the 12,350 BC solar particles.

For now, one conclusion holds firm: 14,300 years ago, beneath a colder sun and weaker magnetic shield, Earth absorbed the fiercest solar onslaught yet known. The trees remember.

Credit & Disclaimer: This article is a popular science summary written to make peer-reviewed research accessible to a broad audience. All scientific facts, findings, and conclusions presented here are drawn directly and accurately from the original research paper. Readers are strongly encouraged to consult the full research article for complete data, methodologies, and scientific detail. The article can be accessed through https://doi.org/10.1016/j.epsl.2025.119383