1257 Samalas eruption


In 1257, a catastrophic eruption occurred at Samalas volcano on the Indonesian island of Lombok. The event had a probable Volcanic Explosivity Index of 7, making it one of the largest volcanic eruptions during the current Holocene epoch. It created eruption columns reaching tens of kilometres into the atmosphere and pyroclastic flows that buried much of Lombok and crossed the sea to reach the neighbouring island of Sumbawa. The flows destroyed human habitations, including the city of Pamatan, which was the capital of a kingdom on Lombok. Ash from the eruption fell as far as away in Java; the volcano deposited more than of rocks and ash.
The eruption was witnessed by people who recorded it on the Babad Lombok, a document written on palm leaves. It left behind a large caldera that contains Lake Segara Anak. Later volcanic activity created more volcanic centres in the caldera, including the Barujari cone, which remains active. The aerosols injected into the atmosphere reduced the solar radiation reaching the Earth's surface, cooling the atmosphere for several years and leading to famines and crop failures in Europe and elsewhere, although the exact scale of the temperature anomalies and their consequences is still debated. The eruption may have helped trigger the Little Ice Age, a centuries-long cold period during the last thousand years. Before the site of the eruption was known, an examination of ice cores around the world had found a large spike in sulfate deposition around 1257, providing strong evidence of a large volcanic eruption having occurred somewhere in the world. In 2013, scientists linked the historical records about Mount Samalas to these spikes.

Geology

General geology

Samalas was part of what is now the Rinjani volcanic complex, on Lombok, in Indonesia. The remains of the volcano form the Segara Anak caldera, with Mount Rinjani at its eastern edge. Since the destruction of Samalas, two new volcanoes, Rombongan and Barujari, have formed in the caldera. Mount Rinjani has also been volcanically active, forming its own crater, Segara Muncar. Other volcanoes in the region include Agung, Batur, and Bratan, on the island of Bali to the west.
Lombok is one of the Lesser Sunda Islands in the Sunda Arc of Indonesia, a subduction zone where the Australian plate subducts beneath the Eurasian plate at a rate of. The magmas feeding Mount Samalas and Mount Rinjani are likely derived from peridotite rocks beneath Lombok, in the mantle wedge. Before the eruption, Mount Samalas may have been as tall as, based on reconstructions that extrapolate upwards from the surviving lower slopes; its current height is less than that of the neighbouring Mount Rinjani, which reaches.
The oldest geological units on Lombok are from the Oligocene-Miocene, with old volcanic units cropping out in southern parts of the island. Samalas was built up by volcanic activity before 12,000 BP. Rinjani formed between 11,940 ± 40 and 2,550 ± 50 BP, with an eruption between 5,990 ± 50 and 2,550 ± 50 BP forming the Propok Pumice with a dense rock equivalent volume of. The Rinjani Pumice, with a volume of dense rock equivalent, may have been deposited by an eruption from either Rinjani or Samalas; it is dated to 2,550 ± 50 BP, at the end of the time range during which Rinjani formed. The deposits from this eruption reached thicknesses of at distance. Additional eruptions by either Rinjani or Samalas are dated 11,980 ± 40, 11,940 ± 40, and 6,250 ± 40 BP. Eruptive activity continued until about 500 years before 1257. Most volcanic activity now occurs at the Barujari volcano with eruptions in 1884, 1904, 1906, 1909, 1915, 1966, 1994, 2004, and 2009; Rombongan was active in 1944. Volcanic activity mostly consists of explosive eruptions and ash flows.
The rocks of the Samalas volcano are mostly dacitic, with a content of 62–63 percent by weight. Volcanic rocks in the Banda arc are mostly calc-alkaline ranging from basalt over andesite to dacite. The crust beneath the volcano is about thick, and the lower extremity of the Wadati–Benioff zone is about deep.

Eruption

The events of the 1257 eruption have been reconstructed through geological analysis of the deposits it left. The eruption probably occurred during boreal summer two or three months of September that year, in light of the time it would have taken for its traces to reach the polar ice sheets and be recorded in ice cores and the pattern of tephra deposits. It began with a phreatic stage that deposited of ash over of northwest Lombok. A magmatic stage followed, and lithic-rich pumice rained down, the fallout reaching a thickness of both upwind on East Lombok and on Bali. This was followed by lapilli rock as well as ash fallout, and pyroclastic flows that were partially confined within the valleys on Samalas's western flank. Some ash deposits were eroded by the pyroclastic flows, which created furrow structures in the ash. Pyroclastic flows crossed of the Bali Sea, reaching the Gili Islands to the west of Samalas, while pumice blocks presumably covered the Alas Strait between Lombok and Sumbawa. The deposits show evidence of interaction of the lava with water, so this eruption phase was probably phreatomagmatic. It was followed by three pumice fallout episodes, with deposits over an area wider than was reached by any of the other eruption phases. These pumices fell up to to east, against the prevailing wind, in Sumbawa, where they are up to thick.
The deposition of these pumices was followed by another stage of pyroclastic flow activity, probably caused by the collapse of the eruption column that generated the flows. At this time the eruption changed from an eruption-column-generating stage to a fountain-like stage and the caldera began to form. These pyroclastic flows were deflected by the topography of Lombok, filling valleys and moving around obstacles such as older volcanoes as they expanded across the island incinerating the island's vegetation. Interaction between these flows and the air triggered the formation of additional eruption clouds and secondary pyroclastic flows. Where the flows entered the sea north and east of Lombok, steam explosions created pumice cones on the beaches and additional secondary pyroclastic flows. Coral reefs were buried by the pyroclastic flows; some flows crossed the Alas Strait between Sumbawa and Lombok and formed deposits on Sumbawa. These pyroclastic flows reached volumes of on Lombok, and thicknesses of as far as from Samalas. The phases of the eruption are also known as P1, P2, P3 and P4. The duration of the P1 and P3 phases is not known individually, but the two phases combined lasted between 12 and 15 hours. The pyroclastic flows altered the geography of eastern Lombok, burying river valleys and extending the shoreline; a new river network developed on the volcanic deposits after the eruption. The eruption column reached a height of during the first stage, and of during the third stage ; it was high enough that in it and its isotope ratio was influenced by photolysis at high altitudes.
Volcanic rocks ejected by the eruption covered Bali and Lombok and parts of Sumbawa. Tephra in the form of layers of fine ash from the eruption fell as far away as Java, forming part of the Muntilan Tephra, which was found on the slopes of other volcanoes of Java, but could not be linked to eruptions in these volcanic systems. This tephra is now considered to be a product of the 1257 eruption and is thus also known as the Samalas Tephra. It reaches thicknesses of on Mount Merapi, on Mount Bromo, at Ijen and on Bali's Agung volcano. In Lake Logung away from Samalas on Java it is thick. Most of the tephra was deposited west-southwest of Samalas. Considering the thickness of Samalas Tephra found at Mount Merapi, the total volume may have reached. The dispersal index of the eruption reached during the first stage and during the third stage, implying that these were a Plinian eruption and an Ultraplinian eruption respectively.
Pumice falls with a fine graining and creamy colour from the Samalas eruption have been used as a tephrochronological marker on Bali. Tephra from the volcano was found in ice cores as far as away from Samalas, and a tephra layer sampled at Dongdao island in the South China Sea has been tentatively linked to Samalas. Ash and aerosols might have impacted humans and corals at large distances from the eruption.
There are several estimates of the volumes expelled during the various stages of the Samalas eruption. The first stage reached a volume of. The phreatomagmatic phase has been estimated to have had a volume of. The total dense rock equivalent volume of the whole eruption was at least. The magma erupted was trachydacitic and contained amphibole, apatite, clinopyroxene, iron sulfide, orthopyroxene, plagioclase, and titanomagnetite. It formed out of basaltic magma by fractional crystallization and had a temperature of about. Its eruption may have been triggered either by the entry of new magma into the magma chamber or the effects of gas bubble buoyancy.
The eruption had a Volcanic Explosivity Index of 7, making it one of the largest eruptions of the current Holocene epoch. Eruptions of comparable intensity include the Kurile lake eruption in the 7th millennium BC, the Mount Mazama eruption in the 6th millennium BC, the Cerro Blanco eruption about 4,200 years ago, the Minoan eruption between 1627 and 1600 BC, and the Tierra Blanca Joven eruption of Lake Ilopango in the 6th century. Such large volcanic eruptions can result in catastrophic impacts on humans and widespread loss of life both close to the volcano and at greater distances.
The eruption left the wide Segara Anak caldera where the Samalas mountain was before; within its high walls, a deep crater lake formed called Lake Segara Anak. The Barujari cone rises above the water of the lake and has erupted 15 times since 1847. A crater lake possibly already existed on Samalas before the eruption and supplied its phreatomagmatic phase with of water. Alternatively, the water could have come from aquifers. A collapse structure cuts into Rinjani's slopes facing the Samalas caldera.
The eruption that formed the caldera was first recognized in 2003, and in 2004 a volume of was attributed to this eruption. Early research considered that the caldera-forming eruption occurred between 1210 and 1300. In 2013, Lavigne suggested that the eruption occurred between May and October 1257, resulting in the climate changes of 1258. Several villages on Lombok are constructed on the pyroclastic flow deposits from the 1257 event.

Research history

A major volcanic event in 1257–1258 was first discovered from data in ice cores; specifically increased sulfate concentrations were found in 1980 within the Crête ice core associated with a deposit of rhyolitic ash. The 1257–1258 layer is the third largest sulfate signal at Crête; at first a source in a volcano near Greenland had been considered but Icelandic records made no mention of eruptions around 1250 and it was found in 1988 that ice cores in Antarctica – at Byrd Station and the South Pole – also contained sulfate signals. Sulfate spikes were also found in ice cores from Ellesmere Island, Canada, and the Samalas sulfate spikes were used as stratigraphic markers for ice cores even before the volcano that caused them was known.
The ice cores indicated a large sulfate spike, accompanied by tephra deposition, around 1257–1259, the largest in 7,000 years and twice the size of the spike due to the 1815 eruption of Tambora. In 2003, a dense rock equivalent volume of was estimated for this eruption, but it was also proposed that the eruption might have been somewhat smaller and richer in sulfur. The volcano responsible was thought to be located in the Ring of Fire but could not be identified at first; Tofua volcano in Tonga was proposed at first but dismissed, as the Tofua eruption was too small to generate the 1257 sulfate spikes. A volcanic eruption in 1256 at Harrat al-Rahat near Medina was also too small to trigger these events. Other proposals included several simultaneous eruptions. Estimated diameters and positions of the calderas left by the eruption ranged from, close to the equator and probably north of it.
While at first no clear-cut climate anomaly could be correlated to the 1257 sulfate layers, in 2000 climate phenomena were identified in medieval records of the northern hemisphere that are characteristic for volcanic eruptions. Earlier, climate alterations had been reported from studies of tree rings and climate reconstructions. The deposits showed that climate disturbances reported at that time were due to a volcanic event, the global spread indicating a tropical volcano as the cause.
The suggestion that Samalas/Rinjani might be the source volcano was first raised in 2012, since the other candidate volcanoes – El Chichón and Quilotoa – did not match the chemistry of the sulfur spikes. El Chichon, Quilotoa and Okataina were also inconsistent with the timespan and size of the eruption. The conclusive link between these events and an eruption of Samalas was made in 2013 on the basis of radiocarbon dating of trees on Lombok and the Babad Lombok, a series of writings in Old Javanese on palm leaves that described a catastrophic volcanic event on Lombok which occurred before 1300. These findings induced Franck Lavigne, a geoscientist of the Pantheon-Sorbonne University who had already suspected that a volcano on that island may be responsible, to conclude that the Samalas volcano was this volcano. The role of the Samalas eruption in the global climate events was confirmed by comparing the geochemistry of glass shards found in ice cores to that of the eruption deposits on Lombok. Later, geochemical similarities between tephra found in polar ice cores and eruption products of Samalas reinforced this localization.

Climate effects

Aerosol and paleoclimate data

Ice cores in the northern and southern hemisphere display sulfate spikes associated with Samalas. The signal is the strongest in the southern hemisphere over the last 1000 years; one reconstruction even considers it the strongest of the last 2500 years. It is about eight times stronger than that of Krakatau. In the northern hemisphere it is only exceeded by the signal of the destructive 1783/1784 Laki eruption; The ice core sulfate spikes have been used as a time marker in chronostratigraphic studies. Ice cores from Illimani in Bolivia contain thallium and sulfate spikes from the eruption. For comparison, the 1991 eruption of Pinatubo ejected only about a tenth of the amount of sulfur erupted by Samalas. Sulfate deposition from the Samalas eruption has been noted at Svalbard, and the fallout of sulfuric acid from the volcano may have directly affected peatlands in northern Sweden. In addition, the sulfate aerosols may have extracted large amounts of the beryllium isotope from the stratosphere; such an extraction event and the subsequent deposition in ice cores may mimic changes in solar activity. The amount of sulfur dioxide released by the eruption has been estimated to be 158 ± 12 million tonnes. The mass release was greater than for the Tambora eruption; Samalas may have been more effective at injecting tephra into the stratosphere, and the Samalas magma may have had a higher sulfur content. After the eruption, it probably took weeks to months for the fallout to reach large distances from the volcano. When large scale volcanic eruptions inject aerosols into the atmosphere, they can form stratospheric veils. These reduce the amount of light reaching the surface and cause colder temperatures, which can lead to poor crop yields. Such sulfate aerosols in the case of the Samalas eruption may have remained at high concentrations for about three years according to findings in the Dome C ice core in Antarctica, although a smaller amount may have persisted for an additional time.
Other records of the eruption's impact include decreased tree growth in Mongolia between 1258–1262 based on tree ring data, frost rings, light tree rings in Canada and northwestern Siberia from 1258 and 1259 respectively, thin tree rings in the Sierra Nevada, California, U.S. lake sediments recording a cooling episode in northeastern China, a very wet monsoon in Vietnam, droughts in many places in the Northern Hemisphere as well as in southern Thailand cave records, and a decade-long thinning of tree rings in Norway and Sweden. Cooling may have lasted for 4–5 years based on simulations and tree ring data.
Another effect of the eruption-induced climate change may have been a brief decrease in atmospheric carbon dioxide concentrations. A decrease in the growth rate of atmospheric carbon dioxide concentrations was recorded after the 1992 Pinatubo eruption; several mechanisms for volcanically driven decreases in atmospheric concentration have been proposed, including colder oceans absorbing extra and releasing less of it, decreased respiration rates leading to carbon accumulation in the biosphere, and increased productivity of the biosphere due to increased scattered sunlight and the fertilization of oceans by volcanic ash.
The Samalas signal is only inconsistently reported from tree ring climate information, and the temperature effects were likewise limited, probably because the large sulfate output altered the average size of particles and thus their radiative forcing. Climate modelling indicated that the Samalas eruption may have reduced global temperatures by approximately, a value largely not replicated by proxy data. Better modelling with a general circulation model that includes a detailed description of the aerosol indicated that the principal temperature anomaly occurred in 1258 and continued until 1261. Climate models tend to overestimate the climate impact of a volcanic eruption; one explanation is that climate models tend to assume that aerosol optical depth increases linearly with the quantity of erupted sulfur. The possible occurrence of an El Niño before the eruption may have further reduced the cooling.
The Samalas eruption, together with 14th century cooling, is thought to have set off a growth of ice caps and sea ice, and glaciers in Norway advanced. The advances of ice after the Samalas eruption may have strengthened and prolonged the climate effects. Later volcanic activity in 1269, 1278, and 1286 and the effects of sea ice on the North Atlantic would have further contributed to ice expansion. The glacier advances triggered by the Samalas eruption are documented on Baffin Island, where the advancing ice killed and then incorporated vegetation, conserving it. Likewise, a change in Arctic Canada from a warm climate phase to a colder one coincides with the Samalas eruption.

Simulated effects

According to 2003 reconstructions, summer cooling reached in the southern hemisphere and in the northern hemisphere. More recent proxy data indicate that a temperature drop of occurred in 1258 and of in 1259, but with differences between various geographical areas. For comparison, the radiative forcing of Pinatubo's 1991 eruption was about a seventh of that of the Samalas eruption. Sea surface temperatures too decreased by, triggering changes in the ocean circulations. Ocean temperature and salinity changes may have lasted for a decade. Precipitation and evaporation both decreased, evaporation reduced more than precipitation.
Volcanic eruptions can also deliver bromine and chlorine into the stratosphere, where they contribute to the breakdown of ozone through their oxides chlorine monoxide and bromine monoxide. While most bromine and chlorine erupted would have been scavenged by the eruption column and thus would not have entered the stratosphere, the quantities that have been modelled for the Samalas halogen release would have reduced stratospheric ozone.

Climate effects

Samalas, along with the Kuwae eruption in the 1450s and Tambora in 1815, was one of the strongest cooling events in the last millennium, even more so than at the peak of the Little Ice Age. After an early warm winter 1257–1258 resulting in the early flowering of violets according to reports from France, European summers were colder after the eruption, and winters were long and cold.
The Samalas eruption came after the Medieval Climate Anomaly, a period early in the last millennium with unusually warm temperatures, and at a time when a period of climate stability was ending, with earlier eruptions in 1108, 1171, and 1230 already having upset global climate. Subsequent time periods displayed increased volcanic activity until the early 20th century. The time period 1250–1300 was heavily disturbed by volcanic activity, and is recorded by a moraine from a glacial advance on Disko Island, although the moraine may indicate a pre-Samalas cold spell. These volcanic disturbances along with positive feedback effects from increased ice may have started the Little Ice Age even without the need for changes in solar radiation, this theory is not without disagreement. The Little Ice Age was a period of several centuries during the last millennium during which global temperatures were depressed; the cooling was associated with volcanic eruptions.
Other inferred effects of the eruption are:
Other regions such as Alaska were mostly unaffected. There is little evidence that tree growth was influenced by cold in what are currently the Western United States, where the eruption may have interrupted a prolonged drought period. The climate effect in Alaska may have been moderated by the nearby ocean. In 1259, western Europe and the west coastal North America had mild weather.

Social and historical consequences

The eruption led to global disaster in 1257–1258. Very large volcanic eruptions can cause significant human hardship, including famine, away from the volcano due to their effect on climate. The social effects are often reduced by the resilience of humans.

Lombok Kingdom and Bali (Indonesia)

Western and central Indonesia at the time were divided into competing kingdoms that often built temple complexes with inscriptions documenting historical events. However, little direct historical evidence of the consequences of the Samalas eruption exists. The Babad Lombok describe how villages on Lombok were destroyed during the mid-13th century by ash, gas and lava flows, and two additional documents known as the Babad Sembalun and Babad Suwung may also reference the eruption. They are also – together with other texts – the source of the name "Samalas" while the name "Suwung" - "quiet and without life" - may, in turn, be a reference to the aftermath of the eruption.
The city of Pamatan, capital of a kingdom on Lombok, was destroyed, and both disappeared from the historical record. The royal family survived the disaster according to the Javanese text, and there is no clear cut evidence that the kingdom itself was destroyed by the eruption, as the history there is poorly known in general. Thousands of people died during the eruption although it is possible that the population of Lombok fled before the eruption. In Bali the number of inscriptions dropped off after the eruption, and Bali and Lombok may have been depopulated by it, possibly for generations, allowing King Kertanegara of Singhasari on Java to conquer Bali in 1284 with little resistance. The western coast of Sumbawa was depopulated and remains so to this day; presumably the local populace viewed the area devastated by the eruption as "forbidden" and this memory persisted until recent times.

Oceania and New Zealand

Historical events in Oceania are usually poorly dated, making it difficult to assess the timing and role of specific events, but there is evidence that between 1250 and 1300 there were crises in Oceania, for example at Easter Island, which may be linked with the beginning of the Little Ice Age and the Samalas eruption. Around 1300, settlements in many places of the Pacific relocated, perhaps because of a sea level drop that occurred after 1250, and the 1991 eruption of Pinatubo has been linked to small drops in sea level.
Climate change triggered by the Samalas eruption and the beginning of the Little Ice Age may have led to people in Polynesia migrating southwestward in the 13th century. The first settlement of New Zealand most likely occurred 1230–1280 AD and the arrival of people there and on other islands in the region may reflect such a climate-induced migration.

Europe, Near East and Middle East

Contemporary chronicles in Europe mention unusual weather conditions in 1258. Reports from 1258 in France and England indicate a dry fog, giving the impression of a persistent cloud cover to contemporary observers. Medieval chronicles say that in 1258, the summer was cold and rainy, causing floods and bad harvests, with cold from February to June. Frost occurred in the summer 1259 according to Russian chronicles. In Europe and the Middle East, changes in atmospheric colours, storms, cold, and severe weather were reported in 1258–1259, with agricultural problems extending to North Africa. In Europe, excess rain, cold and high cloudiness damaged crops and caused famines followed by epidemics, although 1258–1259 did not lead to famines as bad as some other famines such as the Great Famine of 1315–17.
In northwest Europe, the effects included crop failure, famine, and weather changes. A famine in London has been linked to this event; this food crisis was not extraordinary and there were issues with harvests already before the eruption. The famine occurred at a time of political crisis between King Henry III of England and the English magnates. Witnesses reported a death toll of 15,000 to 20,000 in London. A mass burial of famine victims was found in the 1990s in the centre of London. Matthew Paris of St. Albans described how until mid-August 1258, the weather alternated between cold and strong rain, causing high mortality.
The resulting famine was severe enough that grain was imported from Germany and Holland. The price for cereal increased in Britain, France, and Italy. Outbreaks of disease occurred during this time in the Middle East and England. During and after the winter of 1258–59, exceptional weather was reported less commonly, but the winter of 1260–61 was very severe in Iceland, Italy, and elsewhere. The disruption caused by the eruption may have influenced the onset of the Mudéjar revolt of 1264–1266 in Iberia. The Flagellant movement, which is first recorded in Italy in 1260, may have originated in the social distress caused by the effects of the eruption, though warfare and other causes probably played a more important role than natural events.

Long term consequences in Europe and the Near East

Over the long term, the cooling of the North Atlantic and sea ice expansion therein may have impacted the societies of Greenland and Iceland by restraining navigation and agriculture, perhaps allowing further climate shocks around 1425 to end the existence of the Norse settlement in Greenland. Another possible longer term consequence of the eruption was the Byzantine Empire's loss of control over western Anatolia, because of a shift in political power from Byzantine farmers to mostly Turkoman pastoralists in the area. Colder winters caused by the eruption would have impacted agriculture more severely than pastoralism.

Four Corners region, North America

The 1257 Samalas eruption took place during the Pueblo III Period in southwestern North America, during which the Mesa Verde region on the San Juan River was the site of the so-called cliff dwellings. Several sites were abandoned after the eruption, which had cooled the local climate. The Samalas eruption was one among several eruptions during this period which may have triggered climate stresses, which in turn caused strife within the society of the Ancestral Puebloans; possibly they left the northern Colorado Plateau as a consequence.

Altiplano, South America

In the Altiplano of South America, a cold and dry interval between 1200 and 1450 has been associated with the Samalas eruption and the 1280 eruption of Quilotoa volcano in Ecuador. The use of rain-fed agriculture increased in the area between the Salar de Uyuni and the Salar de Coipasa despite the climatic change, implying that the local population effectively coped with the effects of the eruption.

Northeast Asia

Problems were also recorded in China, Japan, and Korea. In Japan, the Azuma Kagami chronicle mentions that rice paddies and gardens were destroyed by the cold and wet weather, and the so-called Shôga famine may have been aggravated by bad weather in 1258 and 1259. Other effects of the eruption include a total darkening of the Moon in May 1258 during a lunar eclipse, a phenomenon also recorded from Europe; volcanic aerosols reduce the amount of sunlight scattered into Earth's shadow and thus the brightness of the eclipsed Moon. The effects of the eruption may also have hastened the decline of the Mongol Empire, although the volcanic event is unlikely to have been the sole cause.