In the spring of 1258 the European chronicles record an unusually cold and wet summer — much of England and the Low Countries experienced a partial harvest failure; the Italian chronicles record unusual late-spring frosts. The English Benedictine chronicler Matthew Paris, working at St Albans, recorded several specific events in the Chronica Majora for 1258: a violent winter that did not seem to end, atmospheric optical phenomena including unusual sunset colours, and a substantial famine that killed an unknown but reportedly large number of London poor over the following winter (the worst documented English famine before the Great Famine of 1315-1322).
Modern climate science has identified the substantive single cause of the 1258 cold summer. It was the largest volcanic eruption of the last seven thousand years — substantially larger than Krakatoa, substantially larger than Tambora, substantially larger than Laki, substantially larger than every other documented historical eruption. It produced a sulphate-aerosol cloud that circled the globe and reduced average global temperatures by approximately 1°C for several years. The medieval chroniclers across Europe and East Asia recorded its effects in substantial detail but had no idea what was causing them.
The eruption itself was identified only in 2013.
The mystery
The eruption’s atmospheric-chemistry signature had been recognized since the 1980s. Greenland and Antarctic ice cores show a substantial sulphate-aerosol spike dated to approximately 1258-1259, of roughly twice the magnitude of the 1815 Tambora signal. The signal was symmetric — present at substantially equal magnitudes in both polar ice records — suggesting an equatorial source latitude. The size of the signal suggested an eruption substantially larger than any historical eruption on record.
The mystery was that no contemporary chronicle anywhere — in Europe, in the Islamic world, in China, in pre-Columbian America — recorded a volcanic eruption in 1257 or 1258. Volcanoes that produce eruptions of this scale (volcanic explosivity index 7 on a 0-8 scale, approximately ten times larger than Mount St Helens 1980) are normally observed at distance and recorded in some surviving documentary source. The 1257 eruption was not.
The proximate atmospheric effects — the cold summer of 1258, the famine, the unusual atmospheric optical phenomena, the global sulphate-cloud signal in the polar ice — were documented. The volcano that caused them was unidentified for 755 years.
The 2013 identification
The identification was published in October 2013 by a team led by the French geographer Franck Lavigne in the Proceedings of the National Academy of Sciences. The source was Mount Samalas, a now-vanished volcano on the Indonesian island of Lombok, approximately 50 kilometres east of Bali. The crater that the eruption produced — the Segara Anak caldera — survives as a 6-kilometre-wide lake-filled depression on the western side of the modern volcanic complex called Mount Rinjani (Lombok’s highest peak, at 3,726 metres). The pre-eruption Mount Samalas had been approximately the same height as modern Rinjani; the May 1257 eruption removed approximately 1,700 metres of the mountain’s vertical profile and produced approximately 40 cubic kilometres of ejected tephra — comparable to the largest eruption of the Tambora caldera and substantially larger than Krakatoa.
The identification used five separate converging lines of evidence:
Tephra geochemistry: the chemical composition of glass shards recovered from the Greenland and Antarctic ice cores matches the Lombok caldera’s pyroclastic deposits to high precision.
Carbon-14 dating: charcoal samples recovered from the Lombok pyroclastic deposits date to the mid-13th century with overlap with the 1258 ice-core signal date.
Indonesian textual evidence: an 18th-century Old Javanese chronicle, the Babad Lombok (the Chronicle of Lombok, compiled around 1700 from earlier oral and written sources), records a catastrophic mountain eruption in the kingdom of Pamatan that buried the royal capital and killed the king. The chronicle dates the event approximately to the period corresponding to 1257.
Local archaeology: the surviving Pamatan-period archaeological sites on Lombok show an abrupt termination layer dated approximately to the mid-13th century, with volcanic-ash deposits overlying the latest occupation surfaces.
Climate modelling: the magnitude of the 1258 ice-core sulphate signal and the recorded European-climate effects of 1258 are quantitatively consistent with an Indonesian-latitude eruption of approximately the Samalas magnitude.
What it did
Modern climate-modelling reconstructions of the Samalas eruption suggest:
The eruption injected approximately 158 million tonnes of sulphur dioxide into the stratosphere — about three times the Tambora injection of 1815.
The resulting global stratospheric sulphate aerosol cloud reduced average northern-hemisphere temperatures by approximately 1°C for approximately 2-3 years. Summer temperatures in northwest Europe in 1258 may have been 1.5-2°C below the surrounding-decade average.
The agricultural disruption across Europe and the Mediterranean was but probably less catastrophic than the comparable Tambora-1815 or Laki-1783 episodes — partly because 13th-century European agricultural systems had less concentrated grain-monoculture dependence than later periods, partly because the Samalas timing fell at a moment of overall European agricultural prosperity.
The eruption contributed substantially — by the strongest current interpretation — to the climatic transition into the Little Ice Age that would dominate northern-hemisphere climate from approximately 1300 until 1850. The pre-1257 period had been the substantially-warm Medieval Climate Anomaly; the period from 1258 onwards shows progressively cooler average temperatures with extended cold-summer episodes (1315-1317, 1452-1454, 1601, 1816, etc.) that the Samalas-initiated stratospheric aerosol cloud may have catalyzed by pushing the climate system across some threshold of stratospheric-volcanic-aerosol persistence.
The site
The Segara Anak caldera is now part of the Gunung Rinjani National Park and is a trekking destination. The walk from the south coast Lombok village of Sembalun Lawang to the caldera’s rim takes approximately two days. The lake at the bottom of the caldera — about 50 metres deep — sits approximately 2,000 metres above sea level and contains a small active cone (Mount Barujari) that has been intermittently erupting since the 19th century. The 1257 caldera’s outer rim is approximately 8 kilometres in diameter.
The site has been continuously inhabited; the modern population of Lombok is approximately 3.4 million, predominantly Sasak Muslim with Hindu Balinese and Buginese minorities. The Pamatan kingdom destroyed in 1257 was the ancestor of the medieval Lombok Sasak political tradition; the post-eruption Lombok political-cultural reorganisation produced the small competing Sasak kingdoms that the 17th-century Balinese conquest would absorb.
The volcano that produced the largest recorded eruption of the last seven thousand years sat for 755 years between its actual occurrence and its scientific identification. The interval is unusual in the documentary history of volcanology. Most of the comparable-magnitude prehistoric eruptions identified in the polar ice-core record — Vesuvius 79 AD, Hekla 1104, Taupo c. 230 AD, Toba c. 74,000 BP — have either contemporary documentary evidence (Vesuvius), well-developed regional folklore (Hekla, Taupo), or pre-historical biological-evolutionary signals (Toba). Samalas had effectively none of the three until the 2013 work. It was, between 1258 and 2013, the largest historical event that nobody knew had happened.