(Credit: © athapet - stock.adobe.com)
In a Nutshell
- Sixteen major floods struck Europe between late 1341 and 1343, making 1342 the worst flood year in the past 700 years on the continent.
- Four of those floods were extreme enough to carry expert-estimated return periods of 500 to 1,000 years, yet they all happened within the same two-year window.
- Researchers warn that modern flood management systems are rarely designed to handle rapid sequences of massive floods, and that such sequences should be treated as a real possibility worth planning for.
A medieval catastrophe buried in old chronicles and monastery records is forcing modern flood experts to rethink how prepared the world really is for what rivers can do.
A new study published in Nature has uncovered something hiding in plain sight for nearly 700 years. What historians long believed was a single catastrophic flood in 1342 was actually part of a relentless, back-to-back sequence of 16 major floods that battered much of Europe over just two years. Researchers compiled more than 1,000 original historical source entries, including chronicles, legal records, merchants’ reports, and weather diaries, to piece together what happened across the continent between late 1341 and 1343. What they found was startling. Four of those 16 floods were so severe that experts estimate they would occur only roughly once every 500 to 1,000 years under normal circumstances, though the researchers stress these are reconstructed estimates based on expert judgment rather than precisely measured probabilities. All of them happened within roughly 24 months.
To convey the scale, the study’s authors compared this medieval sequence to some of the worst floods Europe has seen in the last 60 years. As an illustrative analogy rather than a literal equivalence, they describe 1342–1343 as comparable to the great European floods of 1999, 2002, 2005, 2006, and 2013 plus about a dozen other 10-to-100-year floods, all compressed into less than two years. Rapid sequences of massive floods like this are rarely factored into risk management planning, meaning current flood defenses may not be built for that kind of back-to-back disaster.
The Floods That History Forgot
For centuries, historians and scientists focused almost exclusively on one event from this period: the Magdalena Flood of late July 1342, named after the Christian feast day of Saint Maria Magdalena. In southern Germany especially, it was considered the single largest flood of the last thousand years, responsible for enormous destruction along the Rhine and the Main. Thousands of people died. Bridges were wiped out. Entire settlements along riverbanks were destroyed. Magdalena left such a scar on the landscape and collective memory that local authorities began mounting flood-level markers and memorial plaques on the walls of prominent buildings to shape awareness of the disaster’s scale.
But the Magdalena Flood, as devastating as it was, turns out to have been just one chapter in a much longer nightmare. Before it, a severe flood the researchers call the “Candlemas Flood,” named after the Christian feast celebrated on 2 February, struck Western and Central Europe in mid-February 1342, with estimated return periods of 200 to 1,000 years in places like Prague, where it swept away an iconic stone bridge that had stood since 1172. Three spring floods followed. Then came Magdalena in July, then two autumn floods. In 1343, the pattern continued with the “Jacob Flood” in late July and the even larger “Bartholomew Flood” in late August, both tearing through the Alps and beyond.
So why did most of this get forgotten? Part of the answer is surprisingly human. According to the study, closely spaced floods tend to be remembered as continuations of one disaster rather than as separate events, and institutional memory crystallizes around a single dramatic episode. Magdalena was the one that got the flood markers. Magdalena was the one that made it into the chronicles in vivid detail. Everything else faded.
How Researchers Reconstructed 700-Year-Old Floods
Piecing together the scale and timing of floods that happened before modern measurement tools existed required an enormous detective effort. About 77% of the flood data entries the team used came from sources written at the time or very shortly after the events, not from later retellings. Researchers prioritized these firsthand accounts over later interpretations, a choice they say sets this study apart from earlier work and makes the findings more reliable.
Flood severity was estimated several ways. Preserved flood markers and written descriptions of how high water rose against known landmarks, like church podiums or town walls, let researchers compare medieval conditions with modern measurements while accounting for how rivers and landscapes have changed over the centuries. In other cases, they inferred flood magnitude from how unusual contemporaries described an event. When a medieval chronicler reached for the Latin word “diluvium,” meaning a deluge of biblical proportions, that was a strong signal. When an account described a flood as surpassing anything within living memory, the researchers read that as a sign of a rare event and inferred a return period of around 100 years.
Researchers also identified modern flood events from 1960 to 2020 that most closely matched each of the 16 historical events in timing, geography, and cause. That comparison helped fill gaps in the historical record and gauge just how extraordinary the medieval sequence truly was.
What Drove 16 Catastrophic Floods in Two Years?
One of the most pressing questions the study takes on is why so many extreme floods piled up in such a short window. Researchers point to two major culprits working together, though they describe both as plausible contributors rather than confirmed explanations, since direct causal proof using current climate models is not yet possible.
First, a cluster of volcanic eruptions around 1340 to 1341, including an explosive eruption of Iceland’s Hekla volcano in May and June of 1341, may have injected material into the upper atmosphere and cooled the Northern Hemisphere. That cooling, according to the study, probably shifted major weather systems southward and may have weakened the polar vortex, the ring of cold air that normally keeps Arctic cold locked near the poles. When the polar vortex weakens, cold air spills southward into Europe, creating conditions that can produce intense, prolonged rainfall.
Second, Arctic sea ice had been retreating since the mid-1330s. The researchers propose that less sea ice left warmer North Atlantic surfaces, which pump more moisture into the air, and that warmer seas paired with cooler, dust-laden air favored more frequent and intense storms carrying heavy rain into the European interior. Alongside these two factors, the researchers add that natural climate variation likely played a role too, and that each factor’s contribution remains uncertain.
Beyond the floods themselves, the consequences reshaped European society in ways still visible today. Transportation across the continent broke down as roads turned impassable and major river crossings vanished, stalling military operations. Lost harvests triggered food shortages and famine across much of Southern Germany, Switzerland, and Austria, and food prices soared across Western and Central Europe from 1342 to 1344. Salt production in Italy collapsed, reportedly because flooding ruined the salt pans, forcing imports from as far away as Cyprus.
Water-borne illness spread through flood-affected and famine-weakened populations. A condition described in historical sources as a “bleeding illness” (presumably dysentery or another form of bloody diarrhea) swept through affected regions, and in some areas floodwaters never fully receded between events, leaving communities inundated and diseased for years. Disruption of trade routes may also have fed the financial crisis in the Italian banking system that began in 1343, the study notes, citing contemporary merchants’ reports. That crisis, tied to the collapse of major Florentine banking houses, is usually blamed on other causes, but the researchers say the floods could have been a contributing factor.
One lasting consequence was a fundamental shift in how European communities built flood defenses. After 1342, bridges were rebuilt in stone, town walls were reinforced and raised, and earthen barriers were constructed along riverbanks. In Prague, the Candlemas Flood had swept away the Judith Bridge, the iconic stone crossing that had stood since 1172. Its replacement, the Charles Bridge, absorbed the hard lessons directly, with broader, higher arches, sharply pointed ice-breaking piers, and riverbanks elevated by up to 9 meters. Whether modern society is adapting as quickly is a question the study raises head-on.
Are Modern Flood Defenses Ready for What’s Coming?
Flood management today is built largely on data from recent decades, occasionally supplemented by records of isolated historical disasters. Modern risk planning, the researchers write, rarely accounts for the possibility that multiple catastrophic floods could strike the same broad region within months of each other, before defenses can be rebuilt, budgets replenished, or communities fully recovered.
Climate projections, the paper notes, suggest extreme storm activity in Europe may increase. Historical evidence now shows that rapid-fire sequences of massive floods are not merely theoretical. They have happened before, on a continent-wide scale. What medieval Europe endured in 1342 and 1343 is a reminder of what rivers are capable of doing, and the researchers argue that quick, proactive strategies may be needed before the next cluster arrives.
Paper Notes
Limitations
Reconstructing flood events from medieval historical records carries inherent uncertainty. Only around 77% of the flood data entries came from contemporary sources; the remainder were drawn from later medieval sources built on earlier firsthand accounts. Estimating return periods for floods that occurred before instrumental measurement existed requires expert judgment and a relative, rank-based approach rather than direct numerical comparison to modern gauge data. Direct comparisons between medieval and modern flood levels are complicated by substantial changes in river shape, floodplain management, and land use over the past 700 years. Documentary sources from this period are also unevenly distributed geographically, tending to favor politically and economically central regions where records survived intact, so regions with fewer surviving archives may be underrepresented. The proposed causes involving volcanic eruptions and Arctic sea ice decline are described by the authors as plausible contributors rather than confirmed explanations, since direct causal proof using available climate models is not yet possible. Return period estimates shown in the study’s figures are described explicitly as expert-based, subjective estimates.
Funding and Disclosures
This paper was published open access under a Creative Commons Attribution 4.0 International License. No specific funding sources or competing-interest disclosures were identified in the portion of the paper reviewed for this article; full author contributions and competing-interest statements are noted as available in the paper’s online content.
Publication Details
Paper Title: “Cascading continental-scale floods across Europe in 1342–1343”
Authors: Andrea Kiss, Alberto Viglione, Mariano Barriendos, Silvia Enzi, Martin Bauch, Kris Decker, Nicolas Maughan, Dag Retsö, Astrid Ogilvie, Miriam Bertola, Tim Soens, Libor Elleder, Rudolf Brázdil, Kathleen Pribyl, Juraj Parajka, Ioannis Telelis, Inês Amorim, Josep Barriendos, Oliver Böhm, Gerardo Benito, Chiara Bertolin, Dario Camuffo, Denis Coeur, Gaston Demarée, Radoslaw Doktor, Markus Dotterweich, Rüdiger Glaser, Jürgen Komma, Neil Macdonald, Hrvoje Petrić, Christian Rohr, Petra Schmocker-Fackel, Lothar Schulte, Willem H. J. Toonen, Oliver Wetter, and Günter Blöschl
Journal: Nature
Published Online: August 12, 2026
DOI: 10.1038/s41586-026-10888-8
Received: October 24, 2024 | Accepted: July 6, 2026







