Thera Eruption Date: The Century Nobody Has Closed

One body of evidence says the volcano blew in 1609-1560 BCE. Another says the pottery cannot be that old and the eruption belongs somewhere near 1500 BCE. The gap is about a century, it has been open for forty years, and in the last five years it has not closed — it has been re-argued at higher resolution by both sides. Most popular accounts resolve it by quietly printing 1600 BC, or by splitting the difference at 1550. This dossier does neither.

Silence Index — D-191

Busy

Ranks 97 of 99 scored dossiers — quieter than 2% of the others we have scored. A low score is not a warning; it describes who else will be standing next to you.

How the total is weighted. These are the fixed weights of the four inputs, the same on every dossier — not this place’s scores. We do not publish per-input bars, because the per-input values were never recorded article by article and inventing four plausible numbers that sum to the total would be a fabrication dressed as a chart.

Footfall pressure — visitors measured against the ground that absorbs them30%

Access friction — permits, caps, seasons, boats, weather windows30%

Acoustic floor — how close the nearest engine is20%

Unresolved questions — what the record still cannot settle20%

This dossier was scored before 2 September 2026, when the fourth input became night darkness. It keeps the input it was reasoned with until it is re-scored at its next source re-check; the methodology page records the change.

The index is a desk instrument scored by this publication, not a field measurement. How it is calculated, and where it fails.

What this dossier is built from

This is desk research. Nobody from this publication has been to Santorini, has stood in the shelter over Akrotiri, or has handled a gram of Theran tephra. What follows is assembled from peer-reviewed papers in Science Advances, Scientific Reports, PLOS ONE, Radiocarbon, PNAS Nexus, Antiquity, the Journal of Greek Archaeology, Nature Geoscience and the Journal of Near Eastern Studies; the Smithsonian Institution's Global Volcanism Program; the United States Geological Survey's earthquake catalogue, queried directly; the National Observatory of Athens; the Hellenic Statistical Authority; and the GeoNames gazetteer.

It does not tell you when Thera erupted. It tells you why two independent bodies of evidence disagree by roughly a hundred years, why the disagreement has hardened rather than dissolved, and why the confident single year printed on so many labels is a decision somebody made rather than a fact anybody established.

The two witnesses

The first witness is a stack of tree rings. In 2018 Charlotte Pearson and colleagues published, in Science Advances, an annual-resolution radiocarbon time series running from 1700 to 1500 BCE, built on wood whose calendar years are known ring by ring rather than averaged in five- or ten-year blocks. Their finding was that the international calibration curve carried an offset across that interval, and that correcting for it produced a shift in the calibrated age range for Thera toward the 16th century BCE. Their own framing of the dispute is worth quoting because it is the cleanest statement of it we found: a discrepancy between radiocarbon dating, which gives the late seventeenth to early sixteenth century BCE, and archaeological dating, which gives the mid sixteenth to early fifteenth.

The second witness is a sequence of pots and a chain of Egyptian regnal dates. Sturt Manning, no friend to it, sets out its logic plainly in the Journal of Greek Archaeology in 2024. Late Minoan IB objects turn up in Egypt in contexts associated with Tuthmosis III. His reign, give or take twenty-five years of argument, sits in the first half of the fifteenth century BCE. That gives a minimum date for the end of Late Minoan IA, and therefore for the eruption, which was conveniently summarised as about 1500 BCE. Manning's verdict on that number is that it came not from any specific evidence but from the convenience of the summary.

Each camp regards the other's anchor as the soft one. To the radiocarbon side, a ceramic synchronism is a chain of inferences each of which could slip a generation. To the archaeological side, a calibrated radiocarbon range is a statistical object whose apparent precision conceals a calibration curve that has been revised repeatedly, most recently in ways that moved the answer. Neither characterisation is unfair. That is the problem.

What popular coverage gets wrong

The received version, which circulates in guidebooks, in television documentaries, in museum and site interpretation, and in the standard telling of Aegean prehistory for general readers, runs like this. Thera erupted in about 1600 BC. The eruption destroyed Minoan Crete. The memory of it reached Plato as Atlantis. Each of those three claims is in a different kind of trouble.

Take the year first. The Smithsonian's Global Volcanism Program — a reference work, not a popularisation — lists the Minoan eruption in Santorini's eruptive history as 1610 BCE, flat, with no bracket attached, while its geological summary says the youngest caldera formed about 3,600 years before present, around 1600 BCE. No paper we opened supports a single year. The most precise published figure we found, Manning's Bayesian model in PLOS ONE in 2022, is a range: 1606-1589 BCE at 68.3% probability, 1609-1560 BCE at 95.4%. Pearson and colleagues, working on a carbonised olive shrub from Therasia in 2023, report increased probability for a mid sixteenth century date instead. Those are not the same answer, and neither of them is 1600.

The destruction of Crete is worse. On the radiocarbon dating, the eruption predates the Late Minoan IB destruction horizon on Crete by something over a century. Manning's own summary is that placing the eruption in the Second Intermediate Period makes it very difficult to see any association between Thera and the LM IB destructions at all. The two events were separated by roughly a hundred years, not joined by cause and effect. Note the shape of that: the very same early dating that makes Thera sound more dramatic is what severs it from the collapse it is famous for causing.

And Akrotiri was not Pompeii. The stratigraphy at the town describes an abandonment before the eruption, with most portable valuables removed and stored foodstuffs left carefully secured behind. The interval between the last human occupation and the ash has been estimated as anything from weeks or months to several years. People left. The town is empty of them.

As for Atlantis, no source in the primary literature we opened treats the connection as evidence of anything. It rests on a reading of a Greek philosophical text written roughly eleven centuries after the eruption on any chronology on offer.

Finally, the compromise. A great many accounts, having noticed that one camp says roughly 1600 and the other roughly 1500, print 1550 BC. We will not. There is no evidence for 1550 BC. It is the arithmetic mean of two positions, and averaging two disagreeing witnesses produces a number that neither of them would testify to.

The olive branch, and what it cannot bear

The single most cited object in this argument is a branch of olive wood, buried alive by the eruption and recovered from the tephra. Radiocarbon dating of its outermost wood, calibrated on IntCal13, gave 1656-1609 BCE at two sigma and 1632-1615 BCE at one sigma. That is a seventeenth century answer, and for a decade it was treated as close to decisive.

It has been under sustained attack since 2014, when Antiquity ran a debate feature in which Paolo Cherubini and colleagues argued the olive dating was unreliable on several grounds, with a reply from the dating team and comments from other specialists. The most damaging single result came in 2018, in Scientific Reports, where Yael Ehrlich, Lior Regev and Elisabetta Boaretto took modern olive wood — a trunk cross-section and a living branch — and dated it at near-annual resolution using the radiocarbon bomb peak, where the true answer is independently known. Their finding was that the last formed wood around the circumference of an olive is not chronologically homogeneous and can differ by up to a few decades. Their conclusion is the one that matters here: the outermost wood layer does not necessarily represent the date of the last year of growth, and the Santorini branch could predate the eruption by a few decades. They also note that intra-annual density fluctuations in olive can be visually mistaken for true rings — the mechanism by which a branch acquires more years than it lived.

A second olive, a carbonised shrub from neighbouring Therasia killed by the same eruption, was published by Pearson, Sbonias, Tzachili and Heaton in 2023. It dates slightly younger than the branch, with its youngest samples calibrating to 1610-1510 BCE, or about 1602-1502 BCE once an offset is applied. Manning re-analysed the same shrub data the following year in Radiocarbon, arguing that with informative priors from stratigraphy, context and sample properties the same measurements support the late seventeenth to early sixteenth century. Two teams, one shrub, two answers. We print both.

The plateau, the ice, and the years nobody can choose between

Underneath all of this sits an awkward piece of physics. The atmospheric radiocarbon record across this exact interval is close to flat — a calibration plateau running roughly 1620 to 1540 BC, which is to say across most of the disputed century. On a plateau, probability spreads. The IntCal20 curve publication says as much: IntCal13 was built for this period on twenty-, ten- and five-ring samples and included a plateau, and the annual dataset has refined the definition of that plateau, which the authors describe as very important for establishing true possible calendar age ranges. Refining a plateau is not the same as removing one.

Two further complications are live. The first is regional offset: Manning and colleagues, in Scientific Reports in 2020, reported offsets of roughly zero to twenty-two radiocarbon years between Mediterranean-Anatolian wood and IntCal20, attributable partly to measurement technology and partly to differing growing-season lengths. Twenty-two years is small; in a century-wide argument it is not nothing. The second is the search for a sharp production spike — a Miyake event — that would pin the sequence absolutely. Pınar Erdil and colleagues at Groningen went looking, publishing about ninety annual measurements between 1660 and 1507 BCE in Radiocarbon in 2026. They found no compelling evidence of any rapid increase in radiocarbon production, only minor differences between laboratory datasets that may be species, region or pretreatment related. Their framing of the state of play is blunter than anything we would have written ourselves: in recent years, the disagreements have escalated with the introduction of IntCal20.

The ice cores were supposed to be the tiebreak, and they are not. For decades a large sulphate spike at 1628 BCE in Greenland ice was read as Thera. In 2022 Pearson and colleagues, in PNAS Nexus, used sulphur isotopes and tephra geochemistry across Greenland and Antarctic cores to demonstrate that the 1628 BCE layer belongs to Aniakchak II in Alaska. Thera was evicted from its own famous date. What the ice offers instead is a menu: constrained stratospheric sulphur injections at 1611 BCE, at 1561, 1558 and 1555 BCE, and at about 1538 BCE, all of them below 14 plus or minus 5 teragrams of sulphur, and therefore all of them implying a climate forcing well below Tambora in 1815. No Theran tephra has been matched to any of them. The ice narrows the field to four or five candidate years spanning seventy-three years. It does not choose.

Egypt's side, and its 2025 complication

The Egyptian half of the argument turns on where the New Kingdom starts, because the eruption is either before Ahmose, in the Second Intermediate Period, or after him, in the early Eighteenth Dynasty. In September 2025, Hendrik Bruins and Johannes van der Plicht published in PLOS ONE the first direct radiocarbon comparison between that dynastic transition and the eruption: a mudbrick from the Ahmose temple at Abydos and a linen burial cloth associated with Satdjehuty, both from the British Museum, and wooden stick shabtis from Thebes in the Petrie Museum.

Because the objects cannot be arranged in a stratigraphic sequence, they could not run a Bayesian model, and instead compared uncalibrated radiocarbon ages directly. Their conclusion is that the two datasets have a different time signature and that the Minoan eruption is older than the reign of Nebpehtire Ahmose. They also note, usefully for anyone trying to keep score, that Egyptological estimates for Ahmose's reign themselves range from 1580-1557 BCE to 1524-1499 BCE — a spread of more than fifty years in the fixed point everyone is measuring against.

One casualty of that paper deserves flagging, because it has been in circulation for a decade. Ritner and Moeller's 2014 argument in the Journal of Near Eastern Studies connected the storm described on Ahmose's Tempest Stela at Karnak to the eruption. On the 2025 radiocarbon comparison that link cannot hold: if the eruption predates Ahmose, the stela is not describing it.

What a hundred years actually buys

Three things hang on this century, and they are not small.

The first is causation. If the eruption falls in the late seventeenth or earliest sixteenth century, it is separated from the Late Minoan IB destructions on Crete by a hundred years or more, and the single most repeated sentence about it — that it destroyed Minoan civilisation — becomes untenable. If it falls near 1500, the connection is at least arguable. The second is the world the Aegean was trading with: an eruption in the Second Intermediate Period puts Late Minoan IA into a Canaanite-Levantine dominated system under the Hyksos, while an eruption in the early New Kingdom puts it into an entirely different political geography. The third is everything downstream. Thera is used as a marker horizon to synchronise the archaeological chronologies of the Aegean, Egypt and the Near East, and to anchor ice-core, speleothem and lake-sediment records. A marker horizon that can move a century moves everything pinned to it.

It is also worth recording that the eruption's physical effects are being revised downward at the same time as its date is being argued over. Modelling published in Scientific Reports in 2015 puts the erupted magma at about 39 km³ and finds the eruption released far more halogens than sulphur — between 50.6 and 675 teragrams of chlorine against 0.34 to 36 teragrams of sulphur — with potentially severe stratospheric ozone consequences but, on the ice-core sulphur figures, a modest climate forcing. The same paper dates the eruption to the beginning of August, from insects preserved in the tephra. That is the one part of the timing on which we found no dispute at all: nobody knows the year, and there is a published case for the month.

The town, the ash, and ground that is still moving

Akrotiri itself is a Late Bronze Age town of multi-storey houses, preserved in metres of pumice and ash, entered today under a modern shelter. Its tsunami reached Crete: a deposit identified at Malia in 2021 shows inundation up to 400 metres inland, with a run-up not exceeding eight metres above sea level, bracketed by radiocarbon dates of 1744-1544 BCE below and 1509-1430 BCE above — a bracket wide enough to accommodate every date in this dossier, which is precisely the difficulty.

The volcano is not finished. Between January 2011 and 2012, GPS and satellite radar recorded 10 to 20 million cubic metres of magma intruded beneath the caldera, described at the time as the only volumetrically significant intrusion since 1955 and equivalent to between 10 and 50 per cent of the volume of recorded dome-forming eruptions. Then, in 2025, the ground moved again. We queried the United States Geological Survey's earthquake catalogue directly for the box containing Santorini and Amorgós between 25 January and 15 March 2025: 86 earthquakes of magnitude 4.5 or greater, the largest two both magnitude 5.3, on 31 January and 10 February, at depths clustering between about eight and fourteen kilometres. A study in Seismica by two seismologists at the National Observatory of Athens traces activity emerging beneath Santorini and inside the caldera in late summer 2024, migrating deeper around Kolumbo and then shallowing north-east toward the islet of Anýdros, and attributes it to a complex interaction between tectonic and magmatic processes rather than to either alone.

That swarm is why anyone planning a visit should check current conditions rather than trust a dossier written at a desk. It is also, in its way, the point: the island that cannot agree on when it last erupted catastrophically spent six weeks of 2025 being shaken by something nobody could immediately classify.

Why the Silence Index is 34

Low, and deliberately so. Physical remoteness is moderate: an Aegean island, but one with an airport and year-round ferries. Access friction is close to nil: a ticketed site with a road to it. Visitor volume is very high. The fourth input we have scored as unresolved questions rather than night darkness, and it is the only one running near the top of the scale — a hundred-year hole in the dating of the defining event of the site, contested by two research communities, still open after five years of new data. A place can be loud, easy and comprehensively unresolved at once.

What we will not do

As of August 2026 the honest description of the state of play is this. Within the radiocarbon community the argument has narrowed: Manning, writing in 2024, describes the scope of dispute as having tightened to around fifty years, against a gap of over a century a couple of decades ago, with the live options being roughly 1611 BCE or around 1600 BCE on one hand and about 1561 BCE on the other. That is real progress, and we say so. But narrowing is not closing, the archaeological tradition's roughly 1500 BCE has not been withdrawn by its own side, the Groningen group described the disagreements in 2026 as having escalated rather than resolved, and the Egyptian radiocarbon comparison published in September 2025 arrived at the question from a completely different direction and did not land on a year either.

So we will print the ranges and name the people who produced them, and we will not average them. An admitted hole in a chronology is a better thing to hand a reader than a tidy number with nobody's name on it. If you visit Akrotiri and a board gives you a single year, the board is not lying to you; it is doing what institutions do with an unresolved argument, which is to pick the most quotable side of it and stop talking. This publication would rather keep talking, and say clearly that we do not know.

Corner Codex — D-191

Place: Archaeological site of Akrotiri, Santorini (Thíra), South Aegean, Greece
Coordinates: About 36.3578°N, 25.3975°E — the gazetteer entry is the modern village, not the dig
Volcano: Santorini, 36.404°N 25.396°E, 367 m, shield with calderas and lava domes (Smithsonian)
Eruption, radiocarbon: 1609-1560 BCE at 95.4% (Manning, PLOS ONE, 2022)
Eruption, ice-core options: 1611 BCE, 1561/1558/1555 BCE, or about 1538 BCE (Pearson et al., 2022)
Eruption, archaeological tradition: Mid-16th to early 15th century BCE; conventionally about 1500 BCE
The famous olive branch: 1656-1609 BCE at 2 sigma on IntCal13 — and contested since 2014
Calibration plateau: Roughly 1620-1540 BC, which is most of the disputed century
Magma erupted: About 39 km³, with 0.34-36 Tg sulphur and 50.6-675 Tg chlorine modelled
Season of eruption: Beginning of August, from insects preserved in the tephra
Distance from the LM IB destructions on Crete: Something over a hundred years, on the radiocarbon dating
Current seismicity: 86 earthquakes of M4.5 or greater between 25 Jan and 15 Mar 2025 (USGS)
Silence Index: 34 / 100

The Traveler's Panel · checked 25 August 2026

Changing information. Prices, permits, seasons and road access change, sometimes at short notice. Everything here is a starting point for your own confirmation — not a quote, and not a guarantee. Verify with the operator and the relevant official body before you book.

The famous date	About 1600 BC, printed as fact. No paper we opened supports a single year
Radiocarbon range	1609-1560 BCE at 95.4% (Manning, 2022); mid-16th century (Pearson et al., 2023)
Archaeological tradition	Mid-16th to early 15th century BCE; conventionally about 1500 BCE
Ice-core candidates	1611, 1561/1558/1555, or about 1538 BCE — none matched to Theran tephra
What we refuse	The 1550 BC compromise. It is a midpoint, not a measurement
Access	Open, ticketed archaeological site under a modern shelter. Price not verifiable from a state source
Live hazard	86 earthquakes of M4.5+ near Santorini and Amorgós, 25 Jan-15 Mar 2025 (USGS)

Getting there — Athens or Piraeus → Santorini (Thíra) by air or ferry → the south of the island by road

Santorini is one of the easiest places in this archive to reach and one of the hardest to write about honestly. The Greek National Tourism Organisation gives flights from Athens at about fifty minutes; ferries run from Piraeus and between the Cyclades year round. The archaeological site sits in the south of the island, inland of the modern village of Akrotíri, which the GeoNames gazetteer places at 36°21′28″N 25°23′51″E. The gazetteer carries no separate entry for the excavation itself.
We are not going to pretend to practicalities we could not source. The Greek Ministry of Culture's own pages for the site could not be reached from here; the Odysseus cultural portal, which for years carried the state's site descriptions, no longer resolves at all; and the state ticketing platforms returned a 404 and a 403. Anyone planning a visit should check the Ministry of Culture directly rather than trust a number copied from a reseller.
The site is covered by a modern shelter and is walked on raised walkways, which means it is one of the few Bronze Age excavations that can be visited in bad weather and in high summer heat. It also means the experience is a managed one: this is a roofed, ticketed, signposted archaeological attraction on a heavily visited island, not a remote ruin.
Check current seismic and volcanic conditions before travelling. Between 25 January and 15 March 2025 the USGS earthquake catalogue records 86 events of magnitude 4.5 or greater in the box between Santorini and Amorgós, the two largest both magnitude 5.3. Conditions of that kind affect ferry schedules, site closures and cliff-path access, and none of it can be forecast from a desk months in advance.
The Institute of Geodynamics at the National Observatory of Athens publishes real-time seismicity, earthquake announcements and ShakeMaps in English, and is the authority to read rather than any aggregator. The Smithsonian's Global Volcanism Program, by contrast, holds no weekly activity reports for Santorini at all, so the international weekly channel most readers would check simply has nothing on it.

Indicative costs — verify before booking

Admission	Not stated here. The Ministry of Culture's pages could not be opened and the state ticketing platforms returned errors. We will not quote a reseller
Why that gap matters	This is a state-run site with a statutory tariff. If we cannot read the tariff, printing a figure would be a guess dressed as a fact
Visitor numbers	Unknown. ELSTAT publishes admissions and receipts for archaeological sites by month in aggregate, not by named site
What the island costs	Not our subject, and heavily seasonal. Every figure we found came from businesses selling the thing being priced
The real cost of the argument	Forty years of laboratory time, and a marker horizon that moves everything pinned to it whenever it shifts

What to pack

Scepticism about signage. If a board gives you a single year for the eruption, treat it as an institutional decision rather than a finding, and note which year it picked.
Sun and heat protection for the walk in and out. The excavation is roofed; the approach and the rest of a Santorini day are not.
Footwear for raised timber and metal walkways over an uneven excavation, not for a beach.
A current ferry and site-closure check made the week you travel, not the month you booked. The 2025 sequence is recent enough that access arrangements are still worth verifying.
Something to read on the argument itself. The 2018 Science Advances paper and the 2022 PLOS ONE paper are both open access, and the disagreement is far more interesting in the primary literature than in any summary of it, including this one.

Where to stay

There is no accommodation at the archaeological site and no reason to expect any: Akrotíri is a working village in the south of a small island, and the excavation is a day visit from anywhere on Thíra. Firá, the island's administrative seat, sits at about 36.42°N 25.43°E with a resident population of 2,376 on the gazetteer's figure, and Oía at about 36.46°N 25.38°E with 1,087; Emporeío, at about 36.36°N 25.45°E, is the larger settlement at 3,704 and is the closest substantial village to the site. Those are resident populations, and they are the reason the Silence Index here is low rather than high: a caldera rim with several thousand permanent residents and a summer visitor load many multiples of that is not a quiet place, whatever the photographs suggest. Anyone whose interest is the chronology rather than the view should know that the argument is not conducted on the island at all. It is conducted in radiocarbon laboratories in Arizona, Groningen, Oxford, Zurich and Rehovot, and in the pages of four or five journals, and none of it is visible from a hotel terrace.

Safety & responsible travel

Do not repeat 1600 BC as an established date, and do not repeat 1550 BC as a compromise. Name a range and name whose range it is.
Do not repeat that Thera destroyed Minoan Crete. On the radiocarbon dating the Late Minoan IB destructions follow the eruption by something over a century, which the published argument describes as making any association very difficult to see.
Do not describe Akrotiri as a town caught unawares. The stratigraphy describes an evacuation: portable valuables removed, stored food left secured behind. Comparisons to Pompeii mislead in both directions.
Stay on the walkways and photograph without flash where asked. The wall paintings and organic impressions that make this site scientifically useful are the fragile part.
Treat the volcano as a working system rather than a viewpoint. Satellite geodesy recorded 10 to 20 million cubic metres of magma intruded beneath the caldera in 2011-2012, and the ground moved again in 2025.

Nearby, and quieter

Thirasía, the island on the far side of the caldera at 36°26′6″N 25°20′20″E with a settlement of 140 people, is where the carbonised olive shrub at the centre of the 2023 and 2024 arguments was found — a place that has done more to unsettle the eruption date in the last three years than Akrotiri itself. Nea Kameni, the young cone in the middle of the caldera at 36.40467°N 25.398067°E, has produced every historically recorded eruption here and is the surface above the magma body measured by satellite radar in 2011 and 2012. Further out, the uninhabited islet of Anýdros at 36°37′30″N 25°41′3″E sits on the line between Santorini and Amorgós toward which the 2025 seismic sequence migrated, and close to where the largest events in the USGS catalogue were located. And Crete, across roughly a hundred kilometres of open sea, holds the other half of the story: the tsunami deposit identified at Malia in 2021 records inundation up to 400 metres inland with a run-up under eight metres, bracketed by radiocarbon dates so wide — 1744-1544 BCE below, 1509-1430 BCE above — that they accommodate every rival chronology at once.

Getting there, in order

1Athens or Piraeus
↓
2Santorini (Thíra) by air or ferry
↓
3the south of the island by road

Each leg is a separate booking and a separate chance to be weathered out. Build spare days into the chain, not onto the end of it.

Where it is

Akrotiri, Santorini (Thíra), South Aegean, Greece · 36.3578°N 25.3975°E · Open in Google Maps · OpenStreetMap

Key locations

Akrotiri (village and archaeological site) — gazetteer entry is the village; no separate entry exists for the excavation
36.3578°N 25.3975°E
Santorini volcano (summit reference) — Smithsonian Global Volcanism Program, 367 m, last eruption 1950 CE
36.4040°N 25.3960°E
Nisída Néa Kamméni — intra-caldera cone; site of the 2011-2012 magma intrusion measured by InSAR and GPS
36.4047°N 25.3981°E
Firá — island administrative seat, population 2,376 on the gazetteer figure
36.4200°N 25.4300°E
Oía — northern village on the caldera rim, population 1,087
36.4600°N 25.3800°E
Nisída Thirasía — source of the carbonised olive shrub dated in 2023 and re-analysed in 2024
36.4350°N 25.3389°E
Therasia (settlement) — population 140; the only settlement on the far side of the caldera
36.4347°N 25.3447°E
Nisída Ánydros — uninhabited islet toward which the 2025 seismic sequence migrated
36.6250°N 25.6842°E

Coordinates are for orientation and are accurate to roughly the width of the feature named, not to a doorway. Several of these points are inside protected areas or need a permit, a guide or a boat — read the access notes above before you plan a route to any of them.

Sources & further reading — checked 25 August 2026

University of Arizona Campus Repository — Pearson, Brewer, Brown, Heaton, Hodgins, Jull, Lange and Salzer, 'Annual radiocarbon record indicates 16th century BCE date for the Thera eruption', Science Advances 4 (2018). The full abstract: an annual-resolution radiocarbon time series 1700-1500 BCE from calendar-dated tree rings; an offset from the international radiocarbon calibration curve; a shift in the calibrated age range for Thera toward the 16th century BCE; and the discrepancy stated as radiocarbon late 17th-early 16th century versus archaeological mid 16th-early 15th century BCE
Cambridge Core / Antiquity — Manning, Höflmayer, Moeller, Dee, Bronk Ramsey, Fleitmann, Higham, Kutschera and Wild, 'Dating the Thera (Santorini) eruption: archaeological and scientific evidence supporting a high chronology', Antiquity 88(342), 2014, pp. 1164-1179. The nine-author case for integrating typological, stratigraphic and radiometric dating and placing the eruption in the late seventeenth century BC — the reference statement of the Aegean high chronology position, and the paper against which the later 16th-century results are argued
Nature / Scientific Reports — Ehrlich, Regev and Boaretto, 'Radiocarbon analysis of modern olive wood raises doubts concerning a crucial piece of evidence in dating the Santorini eruption', Scientific Reports 8 (2018). Bomb-peak dating of a modern olive trunk cross-section and living branch showing that the last formed wood around the circumference is not chronologically homogeneous and can differ by up to a few decades; the statement that the outermost wood layer does not necessarily represent the last year of growth; and the Santorini branch calibrated on IntCal13 to 1656-1609 BCE at 2 sigma and 1632-1615 BCE at 1 sigma
PLOS ONE — Manning, 'Second Intermediate Period date for the Thera (Santorini) eruption and historical implications', PLOS ONE 17(9) e0274835, 20 September 2022. The Bayesian model giving 1606-1589 BCE at 68.3% and 1609-1560 BCE at 95.4%; the placement of the eruption in the Second Intermediate Period; the account of Akrotiri's abandonment with portable valuables removed and stored foodstuffs left; and the finding that the eruption predates the Late Minoan IB destructions on Crete by a century or more
PLOS ONE — Bruins and van der Plicht, 'The Minoan Thera eruption predates Pharaoh Ahmose: Radiocarbon dating of Egyptian 17th to early 18th Dynasty museum objects', PLOS ONE 20(9) e0330702, 10 September 2025. Radiocarbon dates on a mudbrick from the Ahmose temple at Abydos and the Satdjehuty linen burial cloth, both British Museum, and Theban stick shabtis from the Petrie Museum; the use of uncalibrated comparison in place of Bayesian modelling; the conclusion that the eruption is older than Ahmose's reign; and Egyptological ranges for that reign from 1580-1557 BCE to 1524-1499 BCE
Nature / Scientific Reports — Pearson, Sbonias, Tzachili and Heaton, 'Olive shrub buried on Therasia supports a mid-16th century BCE date for the Thera eruption', Scientific Reports 13 (28 April 2023). A carbonised olive shrub from Therasia killed by the same eruption deposits, dating slightly younger than the earlier olive branch, with the youngest samples calibrating to 1610-1510 BCE, or about 1602-1502 BCE with an offset applied, and overlapping multiple volcanic sulphate markers in ice-core records
Cambridge Core / Radiocarbon — Manning, 'Problems of Dating Spread on Radiocarbon Calibration Curve Plateaus: the 1620-1540 BC Example and the Dating of the Therasia Olive Shrub Samples and Thera Volcanic Eruption', Radiocarbon 66(2), April 2024. The calibration plateau of roughly 1620-1540 BC across which probability spreads; the argument that informative priors from stratigraphy, context and sample properties sharpen the Therasia results substantially; and the re-analysis returning a late 17th to early 16th century BC eruption date from the same measurements Pearson and colleagues published
Cambridge Core / Radiocarbon — Erdil, Kuitems, Scifo, Brown and Dee, 'Investigating potential radiocarbon anomalies around the time of the Minoan eruption of Thera: A new high-resolution dataset from Groningen', Radiocarbon 68(2), 2026, pp. 264-280. About ninety annual radiocarbon measurements between 1660 and 1507 BCE; no compelling evidence of rapid increases in radiocarbon production; minor inter-dataset differences possibly species, region or laboratory pretreatment related; and the statement that in recent years the disagreements have escalated with the introduction of IntCal20
Cambridge Core / Radiocarbon — Raj, Regev and Boaretto, 'Calibration of Multiple Tree-Ring Blocks and Its Implication on the Debate of Minoan Eruption of Santorini Around 17th-16th Century BCE', Radiocarbon 65(3), 2023. A moving-average calibration method applied to the olive branch dates, returning calibrated ages younger than previous estimates in the region of 1618-1541 BCE, favouring a mid-16th century date as more probable while stating that the range still spans both the 17th and 16th centuries BCE
Oxford Academic / PNAS Nexus — Pearson, Sigl, Burke, Davies, Kurbatov, Severi, Cole-Dai, Innes, Albert and Helmick, 'Geochemical ice-core constraints on the timing and climatic impact of Aniakchak II (1628 BCE) and Thera (Minoan) volcanic eruptions', PNAS Nexus 1(2), 2022. Sulphur isotope and tephra analysis attributing the 1628 BCE ice layer to Aniakchak II in Alaska rather than Thera; the remaining dating options for Thera as constrained stratospheric sulphur injections at 1611 BCE, 1561/1558/1555 BCE and about 1538 BCE; all below 14 ± 5 Tg sulphur; and cores GISP2, GRIP, NGRIP, EGRIP, WDC and EDML
Archaeopress / Journal of Greek Archaeology — Manning, 'Thera, the Aegean, Egypt, the Hyksos and Anatolia: rethinking the orthodox synchronisations and histories', JGA 9 (2024). The derivation of the traditional about-1500 BCE date from Late Minoan IB objects in Tuthmosis III contexts rather than from specific evidence; the statement that the scope of dispute is now narrowed to around fifty years against over a century two decades ago; and the author's preference for 1611 BCE or broadly around 1600 BCE
Nature / Scientific Reports — Manning, Wacker, Büntgen, Bronk Ramsey, Dee, Kromer, Lorentzen, Tegel and others, 'Radiocarbon offsets and old world chronology as relevant to Mesopotamia, Egypt, Anatolia and Thera (Santorini)', Scientific Reports 10 (2020). Offsets of roughly zero to twenty-two radiocarbon years between Mediterranean-Anatolian wood and IntCal20, attributed partly to a divergence between modern AMS and older counting technologies and partly to differing growing-season lengths and timings; and the statement that accurate dating is needed for any assessment of Thera's impact
Cambridge Core / Radiocarbon — Reimer and 40 others, 'The IntCal20 Northern Hemisphere Radiocarbon Age Calibration Curve (0-55 cal kBP)', Radiocarbon 62(4), August 2020, pp. 725-757. The calibration curve every date in this dossier depends on; the large influx of annual tree-ring data new to IntCal20; and the specific note that IntCal13 was based on 20-, 10- and 5-ring samples for this period and included a plateau which the annual dataset has refined
Cambridge Core / Antiquity — Cherubini and colleagues, with responses, 'Bronze Age catastrophe and modern controversy: dating the Santorini eruption', Antiquity 88(339), 2014, p. 267. The debate feature in which Cherubini and colleagues argue the Santorini olive-branch dating is unreliable on several grounds, with a reply from the authors of that dating, comments from other specialists and a closing reply — the point at which the branch stopped being treated as settled evidence
Nature / Scientific Reports — Cadoux, Scaillet, Bekki, Oppenheimer and Druitt, 'Stratospheric Ozone destruction by the Bronze-Age Minoan eruption (Santorini Volcano, Greece)', Scientific Reports 5 (2015). Petrological and atmospheric modelling on about 39 km³ of magma, giving minimum yields of 0.34 Tg sulphur, 50.6 Tg chlorine and 0.1 Tg bromine and maximum yields of 36 Tg sulphur, 675 Tg chlorine and 1.5 Tg bromine; modelled ozone column reductions of 20 to over 90% at northern high latitudes; and the eruption dated to the beginning of August from insect fossils in the tephra
Nature / Scientific Reports — Lespez, Lescure, Saulnier-Copard, Glais, Berger, Lavigne, Pearson, Virmoux, Müller Celka, Pomadère and others, 'Discovery of a tsunami deposit from the Bronze Age Santorini eruption at Malia (Crete)', Scientific Reports 11 (2021). Coastal coring at Malia showing an erosional event followed by a high-energy marine sand layer, inundation up to 400 metres inland, a run-up not exceeding 8 metres above sea level, and bracketing radiocarbon dates of 1744-1544 BCE below the deposit and 1509-1430 BCE above it
University of Chicago Press / Journal of Near Eastern Studies — Ritner and Moeller, 'The Ahmose ‘Tempest Stela’, Thera and Comparative Chronology', JNES 73(1), 2014, pp. 1-19. The 2014 argument connecting the storm described on Ahmose's Tempest Stela at Karnak to the Thera eruption — the reading that the September 2025 Egyptian radiocarbon comparison now contradicts, since an eruption predating Ahmose cannot be what his stela describes
Nature Geoscience — Parks, Biggs, England, Mather, Nomikou, Palamartchouk, Papanikolaou, Paradissis, Parsons, Pyle, Raptakis and Zacharis, 'Evolution of Santorini Volcano dominated by episodic and rapid fluxes of melt from depth', Nature Geoscience 5 (2012). InSAR and GPS measurement of 10 to 20 million cubic metres of magma intruded beneath Santorini since January 2011, equivalent to 10-50% of the volumes of recorded dome-forming eruptions, and described as the only volumetrically significant intrusion since 1955
Seismica (McGill University) — Fountoulakis and Evangelidis, National Observatory of Athens, 'The 2024-2025 seismic sequence in the Santorini-Amorgos region: insights into volcano-tectonic activity through high-resolution seismic monitoring', Seismica 4(1), 29 May 2025. A machine-learning phase-picked, relocated catalogue showing seismicity emerging beneath Santorini and within the caldera in late summer 2024, migrating deeper around Kolumbo and then shallowing toward Anydros along south-west to north-east structures, and attributed to a complex interaction between tectonic and magmatic processes
United States Geological Survey — Earthquake Catalog (FDSN event service), query for 36.2-36.9°N, 25.0-26.3°E, 25 January to 15 March 2025, M4.5 and above. 86 events of magnitude 4.5 or greater in the seven weeks queried; the two largest both M5.3, on 31 January 2025 at 05:30:54 UTC at 13.5 km depth and 10 February 2025 at 11:36:28 UTC at 10 km depth; and hypocentres clustering between roughly 8 and 14 km
Smithsonian Institution — Global Volcanism Program, Santorini (212040). Coordinates 36.404°N 25.396°E, summit elevation 367 m, described as a shield volcano with calderas and lava domes; last known eruption 1950 CE; the youngest caldera formed about 3,600 BP; and the eruptive history listing the Minoan eruption flatly as 1610 BCE with no uncertainty attached
Smithsonian Institution — Global Volcanism Program, Santorini weekly activity reports index. The Global Volcanism Program's explicit statement that it holds no Weekly Volcanic Activity Reports for Santorini — meaning the 2025 unrest is not covered by the international weekly reporting channel most readers would check
Institute of Geodynamics, National Observatory of Athens — institute pages and earthquake announcement service. The Greek national seismological institute's remit across seismology, tsunami, seismotectonics, tectonic geodesy, geophysics and engineering seismology, and its public services: real-time seismicity, earthquake announcements, the EIDA node, ShakeMaps and the 'Did you feel it?' reporting system
Hellenic Statistical Authority (ELSTAT) — Museums and Archaeological Sites Attendance survey (SCI21). The monthly national survey of admissions and receipts at museums and archaeological sites, published to March 2026 — and published in aggregate by month rather than by named site, which is why this dossier gives no visitor figure for Akrotiri
Hellenic Organization of Cultural Resources Development (ODAP). The Greek state body responsible for the commercial management of museums, monuments and archaeological sites, whose public English-language pages carried no schedule of admission fees at the time of writing
Greek National Tourism Organisation — Visit Greece, Santorini. The official state tourism description of the island: the caldera and active volcano, the component landmasses Thera, Thirasia, Aspronisi, Palea Kameni and Nea Kameni, flights of about 50 minutes from Athens — and no mention of the archaeological site of Akrotiri at all
Institute for Aegean Prehistory (INSTAP). The privately funded American non-profit, established in 1982, that supports research into Aegean civilisations from the Palaeolithic to the beginning of writing in Greek — one of the principal funders standing behind the fieldwork this argument is conducted on
The Archaeological Society at Athens. The society founded on 6 January 1837 to discover, restore and supplement the antiquities of Greece, which conducts excavations across the country and holds the excavation archives, diaries, plans and photographs behind much Greek fieldwork
GeoNames gazetteer — Akrotiri, Santorini. Akrotíri as a populated place at 36°21′28″N 25°23′51″E and the Akrotíri lighthouse at 36°21′27″N 25°21′25″E — and the absence of any separate gazetteer entry for the archaeological site itself, which is why this dossier's coordinate is the village
GeoNames gazetteer — populated places on Thíra. Firá at about 36.42°N 25.43°E, seat of a third-order administrative division, population 2,376; Emporeío at about 36.36°N 25.45°E, population 3,704; and Oía at about 36.46°N 25.38°E, population 1,087
GeoNames gazetteer — Nisída Néa Kamméni. The island of Nea Kameni at 36.40467°N, 25.398067°E — the young intra-caldera cone that has produced every historically recorded eruption at Santorini, and the surface expression of the system measured by satellite geodesy in 2011-2012
GeoNames gazetteer — Thirasía. Nisída Thirasía, island, at 36°26′6″N 25°20′20″E, and the settlement of Therasia at 36°26′5″N 25°20′41″E with a population of 140 — the island that produced the carbonised olive shrub central to the 2023 and 2024 arguments
GeoNames gazetteer — Anýdros. Nisída Ánydros in the Cyclades at 36°37′30″N 25°41′3″E — the uninhabited islet on the Santorini-Amorgós line toward which the 2025 seismic sequence migrated, and close to the epicentres of the largest events in the USGS catalogue query

Official and scientific bodies are the authority for safety, regulatory and scientific claims. Commercial operators are used only for practical detail such as tour length, meeting points and indicative pricing — never for a safety or scientific statement.

How this was made

Desk research. Compiled from official bodies, scientific and conservation sources, established journalism and licensed operators. Not based on a personal visit.

Editor’s method note — what we checked, corrected and could not verify

Desk research only. Nobody from this publication has been to Santorini or to Akrotiri, and nothing in this dossier rests on observation of the site.
Every URL cited was opened and read before citation. Sources that could not be opened are listed at the end of this note and are cited nowhere in the piece.
We do not state a date for the eruption anywhere in our own voice. Where sources disagree we print every published range and name the team that produced it: Manning's 1609-1560 BCE at 95.4% (2022), Pearson and colleagues' mid-16th century from the Therasia shrub (2023), Manning's late-17th to early-16th re-analysis of the same shrub (2024), Raj and colleagues' 1618-1541 BCE (2023), and the archaeological tradition's roughly 1500 BCE.
We explicitly refuse the 1550 BC compromise that circulates in general coverage. It is the midpoint of two positions and is supported by no source we opened. Averaging disagreeing witnesses is not a finding.
The commission for this dossier flagged the risk that a 2021-2026 paper might have closed the question. We checked. It has not closed, but it has moved, and the article is dated to the state of play in August 2026. Manning (2024) describes the dispute as narrowed to about fifty years; Erdil and colleagues (2026) describe the disagreements as having escalated with IntCal20. Both statements are quoted, and we do not choose between them.
The 86-earthquake figure for the 2025 swarm is our own direct query of the USGS FDSN event service for a defined bounding box and date window, not a figure taken from reporting. The query URL is cited in full so that it can be re-run. A different box or a lower magnitude threshold would return a different number, and Greek national catalogues, which pick smaller events, will show far more.
Coordinates come from the GeoNames gazetteer and, for the volcano, from the Smithsonian's Global Volcanism Program. GeoNames carries the village of Akrotíri but no separate entry for the excavation, so the panel coordinate is the village and is labelled as approximate.
We give no admission price and no opening hours. The Greek Ministry of Culture's site pages for Akrotiri could not be reached, the Odysseus cultural portal no longer resolves, and the state ticketing platforms returned errors or refused access. We would rather print nothing than a price copied from a commercial reseller.
Bibliographic discovery used the Crossref REST API and the National Library of Medicine's PubMed index to enumerate 2021-2026 literature; neither is cited as a source for any claim, and every paper found through them was then opened at its publisher.
Sources we could not read, and therefore did not cite: the Science version of Friedrich and colleagues' 2006 olive-branch paper (403); the PNAS paper on Theran ash, victims and tsunami debris at Çeşme-Bağlararası (403 at PNAS, robots-blocked at Europe PMC, a CAPTCHA at PMC); Johnston and colleagues' revised eruption-volume paper in the Journal of the Geological Society (403); the Earth and Planetary Science Letters and Journal of Volcanology and Geothermal Research papers behind Elsevier's link resolver, which returned no content; the Greek Ministry of Culture's Akrotiri pages and the Odysseus portal, which no longer resolves; the ODAP and Hellenic Heritage ticketing platforms (404 and 403); Fraport Greece's traffic statistics (404); Perseus Digital Library's text of Plato's Timaeus (robots-blocked); UNESCO's Greek tentative list pages, which returned only navigation; and Nature's and Europe PMC's search interfaces, both robots-blocked or rate-limited.
Things we could not verify: the year of the eruption, by any route; whether the Therasia olive shrub or the earlier olive branch preserves true annual rings, the two teams disagreeing on exactly this; which of the ice-core sulphate events at 1611, 1561, 1558, 1555 or about 1538 BCE is Thera, no Theran tephra having been matched to any of them; the interval between the abandonment of Akrotiri and the eruption, published estimates ranging from weeks to several years; the current admission price and opening hours at Akrotiri, no state source being readable; how many people visit Akrotiri in a year, ELSTAT publishing only aggregate monthly totals; what interpretation the site's own boards and signage actually give as the eruption date, which we cannot know without going; and whether the 2025 seismic sequence represents magmatic recharge, tectonic release, or both, the published position being that it is a complex interaction of the two.

AI is used to draft this text, and a human checks it. Every figure, date and quotation above was verified by Tony Hall against the source cited for it in the reference list; where sources disagree, the disagreement is shown rather than resolved, and where a claim could not be confirmed it is either marked unverified or left out. The illustrations are AI-created and labelled as such — no photograph on this site is presented as a record of a real place. Full disclosure · Editorial policy.

Access details change — recheck before you travel. Found an error? Tell us — corrections are dated and shown in place.

Two Witnesses, One Missing Century: The Date of the Thera Eruption