Australia: The Land Where Time Began

A biography of the Australian continent 

Mt. Paektu volcano – Quantifying Gas Emissions from the “Millennium Eruption” of Paektu Volcano, North Korea/China

Paektu/Changbaishan volcano straddles the border between North Korea and China. The Millennium Eruption (ME) erupted 23³ km or rock equivalent at about 946 CE, with the result that copious amounts of magmatic volatiles, such as H₂O, CO₂, sulphur and halogens, were released into the atmosphere. For assessing the volcanogenic climatic impacts accurate quantification of volatile yield and composition is critical, in particular in events that occurred earlier than the satellite era. A geochemical technique was used in this study and quantifies the composition of volatiles, as well as the upper bounds of yields, for the Millennium Eruption by examination of trends in incompatible trace and volatile element concentrations in melt inclusions that were crystal-hosted. It was estimated that as much as 45 Tg of sulphur could have been emitted to the atmosphere. This is greater than the quantity of sulphur that was released by the eruption of Tambora in 1815, which contributed to the “the year without summer.” The maximum yield of gas that was estimated in this study for the Millennium Eruption places it among the strongest emitters of climate-forcing gases in the Common Era (CE). Greenland ice cores record, however, only a very weak sulphate signal that is attributed to the Millennium Eruption. Iacovino et al. suggest that other factors came into play in the minimisation of the glaciochemical signature. In this paradoxical case, in which there is high emissions of sulphur that do not result in a strong glacial sulphate signal may present a way forward in building more generalised models for the interpretation of which volcanic eruptions have produced large climate impacts.

There are profound impacts on planetary atmospheres of volcanic emissions that drive climate change over a range of temporal and spatial scales (1-3). Explosive eruptions that generate stratospheric clouds are the principal source of sulphur in the stratosphere (4) and these clouds can result in long-lasting (in essence, several years) effects on the atmosphere by the injection of SO₂ which oxidises to form sulphate aerosol and promotes global cooling. Magmas that are rhyolitic feed explosive volcanic eruptions, though they are characteristically sulphur-poor, may still contribute a substantial amount of sulphur to the atmosphere that is sourced from a pre-eruptive vapour phase that is rich in sulphur (5-13). In highly silicic magmas that are highly evolved and require protracted crystallisation during crustal storage, which results in the exsolution of vapour via second boiling (14,15).  It is possible, in some cases, to constrain it by examination of geochemical trends in erupted samples (16,17), though the presence, amount and composition of such pre-eruptive gas are not recorded directly in the rock record.

It is possible that direct measurements may be made from satellites or in situ remote sensing of sulphur yields and subsequent climate impacts from modern volcanic eruptions. In order to evaluate the impacts of volcanoes on the atmosphere and climate over geologic time, a way in which to assess sulphur yields of ancient eruptions, or eruptions that are not monitored is required. The degree of climate forcing that imposed by a volcanic eruption is dictated by many factors, which include though are not limited to, gas yield and composition. According to Iacovino et al. a large yield of sulphur may be required for, though not necessarily be indicative of, strong climate forcing. The extent to which a sulphur-rich volcanic eruption will cause a climate perturbation can be determined by other factors, such as latitude, the season in which the eruption occurred, and the eruption column height. Efficacy of the atmospheric transport of the volatiles that are emitted can be elucidated by the concentrations of sulphur in polar cores and proper interpretation thereof, are believed to be the best records of atmospheric chemistry changes over time. A more complete picture of the role of volcanism on climate can be by combining independent analyses of :

i)                   The total yield of gas from large explosive eruptions as is recorded in volcanic rocks, and

ii)                Perturbations in atmospheric chemistry.

A method for constraining the total gas budget of large silicic eruptions by the examination of geochemical trends in glasses, crystals, crystal-hosted melt inclusions (MIs), which represent various stages of the history of crystallisation of magmas is demonstrated in this study. It was revealed by petrological and thermodynamical modelling that the evolution of volatiles in pre-eruptive melt and a coexisting vapour phase in the buildup to the comenditic Millennium Eruption of the Paektu volcano, about 946 CE. The gas generated prior to, as well as during, the eruption and contributions to the volatile budget from the solid phases (e.g. sulphide), was considered by this technique. The total amount of gas that was generated during the crustal magma storage was calculated by the use of this method, and so the values that are reported in this paper represent maximum possible volatile yields. Evidence for the generation of a pre-eruptive vapour phase, that was rich in sulphur, which supplies most of the sulphur that was erupted, was found by this study. If the true gas yields were close to the maximum values in this study, then the new estimates of gas yields would place the Millennium Eruption among the largest emitters of climate-forcing gases in the Common Era. On its own, this implies the potential for volcanogenic climate effects that were not unlike those that were seen following the eruption of Tambora in 1815 (18), which was responsible for the “the year without summer” in 1816 and the deaths of more than 71,000 people. Conversely, however, sulphate deposits in Greenland ice cores are relatively low that have recently been linked to the Millennium Eruption (19) suggest there was only a modest release of sulphur to the atmosphere. Earlier work that cited low sulphur yield estimates (2 Tg) appears to be consistent with this interpretation, though this value excludes the contribution from pre-eruptive fluid or breakdown of solid phase within magmas from the Millennium Eruption (20).

Volatile budgets for ancient silicic eruptions and the excess gas problem

Modern eruptions may be monitored and direct measurement of their gas yields carried out via satellite or remote sensing in situ. In the case of ancient eruptions that were not monitored, clues in the rock record are relied on to determine the amount and composition of erupted gas. This is often achieved by comparing geochemical signatures in rocks that recorded volatile histories before and after the eruption being studied. It is common to characterise pre-eruptive magmatic volatile contents by the use of dissolved volatile concentrations in phenocryst-hosted melt inclusions: small beads of liquid that have been trapped within crystal hosts and quench the glass on eruption (21). The melt inclusion is to a large extent isolated from changes in the magmatic system, in essence, magma differentiation and degassing, and so acts as a time capsule that records snapshots of melt chemistry, which includes dissolved volatiles, throughout the pre-eruptive evolution of the magma interstitial liquid between crystals, conversely [matrix glass (MG)] is privy to changes within the system and therefore most of its dissolved volatile content will be degassed upon eruption with the ascent of the magma with the accompanying decrease of confining pressure. It is therefore possible to calculate the proportions of volatiles that is lost during eruption as the difference between volatile concentrations in matrix glass and those in the most evolved melt inclusion, which represents the magma at a time just prior to the eruption. The mass of volatiles that are released during eruption can be calculated if the volume and density of the erupted material are known, this is the “petrological method” and it is often applied to ancient eruptions or eruptions that are not monitored for which there is no direct measurement of gas yield.

The way to test the petrological method by comparing calculated to measured gas yields has been provided by advances in remote sensing (22). The discovery of what is known as the “excess S” or “excess gas” problem: sulphur yields quantified by the petrologic method are often much lower than those measured by remote sensing (7,12,13,23,24). Excess gas is most pronounced in silicic magmas in which melts are characteristically S-poor, where the degree of excess degassing, which is defined as the ratio of measured SO₂ yield to the petrologic method estimate, can range from 10 to 100 (112).

The exsolved C-O-H-S vapour present in the vapour phase that is within the magma prior to the eruption (5-13) predominantly supplies the “excess” S. The important role of a vapour phase rich in sulphur in determine the total gas budgets of large silicic eruptions was famously demonstrated by the eruption of Pinatubo in 1991 (25), and similar conclusions were drawn for the Valley of Ten Thousand Smokes (8), Redoubt (9), Mount St Helens (26), as well as others. According to Iacovino et al. these findings are consistent with experimental studies that have been carried out on the fluid/melt partitioning of sulphur in these rhyolitic melts that are characteristically poor in sulphur (27).

Methods for quantifying a pre-eruptive magmatic gas phase at unmonitored eruptions was established in earlier studies (12,16,17) and have found that pre-eruptive gas can total up to 6% by weight (wt%) [about 30% by volume (volume %) of crustal magma, which implies that a significant  proportion of gas that was erupted may have been sourced from fluid generated during the storage and evolution of magma. It was suggested by several lines of evidence, which included the detection of substantial excess gas, systems like Paektu, that were rich in silica, may retain most of the vapour until eruption. The migration of bubbles through viscous rhyolitic melts is not fast enough for volatile transfer on eruptive time scales (13) and is not efficient, particularly in melts that are crystal poor (28). It is not likely that magmatic convection, which is probably an important mechanism for pre-eruptive degassing of low viscosity basaltic magmas (11), occurs to an appreciable extent in kinetically sluggish high-velocity silicic systems (13). It is important to evaluate the presence, amount, and the composition of any pre-eruptive fluid phase that may contribute to the total gas budget. The petrologic method, which records only the degassing of dissolved volatiles, and cannot account for the existence of a pre-eruptive vapour phase, may result in a significant underestimate of yields of gas from large silicic eruptions, which leads to assessments that are not accurate of hazard and the potential for climate change.

Geologic background and the melt inclusion tephra

Paektu, located at 42.0056^oN, 128.0553^oE, is an intraplate volcano that has a summit lake of 37 km² is bisected by the border between North Korea (DPRK) and China. Relatively little is known about Paektu, a volcano that has erupted explosively multiple times (volcanic explosivity index ≤6) including the Millennial Eruption, one of the largest volcanic events on Earth in the last 2,000 years.

There were 23 ± 5 km³ of dense rock equivalent (DRE) of material deposited from the Millennial Eruption in 2 phases which were chemically distinct in the form of ash, pumice and pyroclastic flow (20) deposits: an initial nearly aphyric (≤3% by volume phenocrysts) (95% tephra by volume) comendite pumice and a late stage phenocryst-rich (10-20% by volume) trachytic pumice (5% tephra by volume). The petrology and mineralogy of 5 pumices from the Millennium Eruption (4 comendite and 1 trachyte), as well as trace element and volatile element concentrations in matrix glasses and more than 100 comenditic and trachytic melt inclusions that are hosted in anorthoclase, sanidine, clinopyroxene, olivine and quartz were characterised by electron microprobe, Fourier transform infrared (FTIR) spectroscopy, and ion microprobe (SHRIMP). In this analysis melt inclusion used are glassy and free of bubbles, with the exception of a small number of melt inclusions that were rehomogenised and analysed for their content of CO₂. Among the melt inclusions that were used none showed signs of leakage or devitrification. It was revealed by dissolve volatile concedntrations that the melt inclusion with moderate H₂O (<4 wt %), minimal sulphur (<300 ppm) and CO₂ (<23 ppm in homogenised melt inclusion) and significant halogens (<4,200 ppm Cl, <4,000 F).

The impact of climate of the Millennium Eruption

It is suggested by the significant sulphur and halogen yields, taken alone, from the Millennium Eruption that it had the potential to affect the atmosphere of the Northern Hemisphere and surface temperatures in the years following the eruption. It can be difficult to quantify the impacts of ancient eruptions and depend on many factors, which include the latitude of the volcano, the season of eruption, as well as the yield of sulphur. Following eruptions at high latitudes, which are much smaller and shorter that if the same eruption occurred in the tropics (55) e.g., perturbations in climate. This effect is compounded if the eruption occurred in winter when removal of aerosols from the stratosphere was enhanced (56). It is suggested by this, therefore, that climate forcing following the Millennial Eruption, which occurred at 42^oN latitude and likely during winter in the Northern Hemisphere (based on tree ring estimates) (57), may have been diminished.

It is sometimes possible to identify signatures of climate forcing following large eruptions as sharp increases in sulphate deposits in polar ice cores, which preserve a timeline of the chemistry of palaeoclimate extending back in time several thousand years (58,59).

Sulphate loading as determined from ice core records is currently the best proxy for determining the climatic impact of the Millennium Eruption, as historical records that detail possible climatic effects after the eruption, and records of the actual eruption are ambiguous (60,61). The Greenland ice cores shows a spike in the deposition of sulphate at around 940 CE and this has been positively linked to the Millennium Eruption by matching chemical fingerprints in simultaneously emplaced tephra to Millennium Eruption ash (19). Sun et al. (19) concluded that the relatively small sulphate load of 9 kg/km² to be indicative of minimal climate forcing from the Millennium Eruption when it is compared to the sulphate that was deposited by the 2 tropical eruptions of Tambora, about 40 kg/km², and Krakatau, about 15 kg/km², that were followed by global surface temperature declines (18,62). Other authors have shared this conclusion (56,60), all of whom cite low emissions of sulphur, based on the 2-Tg estimate of Horn and Schmincke (20), and an important factor as well as seasonality and high latitude of the Paektu eruption.

The new gas yield estimation of Iacovino et al. considers all possible gas that was emitted during the Millennium Eruption (all pre-eruptive vapour is assumed to have been retained until eruption), and though this represents a maximum value, they stress that the range of possible volatile yields must be considered in silicic magma systems, where there may be substantial exsolution and retention of a pre-eruptive vapour phase. Iacovino et al. suggest, given their new upper bound for the yield of sulphur from the Millennium Eruption, that if large scale surface cooling following the eruption was minimal, it was likely to be due to seasonal and latitude controls on the sulphate aerosol transport to the Arctic, as has been proposed (Kravitz & Robock, 2011), and does not reflect the release of sulphur from Paektu.

Sources & Further reading

1. Iacovino, K., et al. (2016). "Quantifying gas emissions from the “Millennium Eruption” of Paektu volcano, Democratic People’s Republic of Korea/China." Science Advances 2(11).