Modelling massive sulphate aerosol pollution, following the large 1783 Laki basaltic eruption

https://doi.org/10.1016/j.epsl.2005.04.046Get rights and content

Abstract

The climatic impact of volcanic eruptions depends mainly on the amount of sulphur-rich gases released into the stratosphere. These are rapidly converted to sulphate aerosols, which result in cooling of the lower troposphere and warming of the lower stratosphere. These gases are retained in the stratosphere and then concentrated by atmospheric processes, forming acid fog near the Earth's surface, with potentially dramatic environmental consequences on plants, animals and humans. We have modelled such emissions in the case of the unusually large 1783–1784 basaltic fissure eruption of Laki (Iceland), using an Atmospheric General Circulation Model. Results show good agreement with historical observations, such as typical time taken by individual pulses to reach continental Europe or geographical extent of the deadly haze that covered much of the northern hemisphere. The model could be adjusted to predict the climatic consequences of very large eruptions, up to the scale of massive flood basalts, and test the proposal that these are the main agent in most mass extinctions of life on Earth.

Introduction

The year 1783 was marked by numerous geological and geophysical events: a high degree of seismic activity in Europe, notably in Calabria, accompanying an eruption of Etna (Italy) on 17 February 1783 and of Stromboli, a few days afterwards. These were followed by other eruptions: Asama volcano in Japan from May to August, Vesuvius in August and most importantly Laki (Iceland) from June 1783 to February 1784. Additionally, in late February 1783, a new island appeared then rapidly disappeared on the southwest coast of Iceland after a submarine eruption. Laki was the second largest basaltic lava flow event in the last 2000 yr, following the 934 Eldgjá eruption [1], [2]. In a thorough reassessment of all aspects of the eruption, Thordarson and Self [2] noted that it produced widespread sulphuric acid aerosol haze. They reconstructed the eruptive sequence and quantified sulphate aerosol emissions, concluding in favour of a strong link between volcanism and climate. Grattan and co-authors [3], [4], [5], [6], [7], [8], [9] vindicate the environmental impact of volcanoes. Early observations and accounts come from the personal journals of de Montredon [10], Franklin [11] and White [12], who pointed out (as early as a few months after the onset of eruption for de Montredon and almost as fast for Franklin) the connexion between unusual climate and volcanic activity in Iceland. Descriptions of fog impact are also found in van Swinden [13] and Brugmans (1787), in [7]. These reports and hundreds of other accounts identified for instance by Thordarson and Grattan in their papers (see references), together with Kington's study of weather in the 1780s over Europe [14], allow a reconstruction of daily meteorological conditions before, during and after the Laki eruption. The purpose of the present study is to attempt to reproduce in a quantitative way the distribution and propagation of sulphate aerosols at the time of the Laki, and to see to what extent the model can account for observations reported during this “Annus Mirabilis”.

Section snippets

The scenario of Laki injections into the atmosphere

Based on eyewitness accounts, a thorough search of literature and chemical analyses, Thordarson and Self [2] have reconstructed the time sequence of lava eruption and gas emissions. Laki volcano belongs to the Grimsvötn volcanic system in the southeastern part of Iceland. Its 1783 eruption began on June 8 and ended on February 7, 1784. During 8 months, 14.7 ± 1 km3 of lava were produced. The successive phases took place along a 25 km-long fissure, but also involved explosive events. Ten eruptions

Modelling sulphate aerosol distribution induced by the eruption

In order to propose a “hind” prediction of volcano-driven climate change, we have simulated the distribution of sulphate aerosols both in space and time, in an attempt to reproduce the propagation of the unusual volcanic haze. We have forced an Atmospheric General Circulation Model (AGCM), coupled with an atmospheric chemistry module, under the scenario representing the rate of sulphate injection into the Earth's atmosphere as reconstructed by [2]. The LMDzT-INCA model (for more detailed

Results: aerosol distribution on a global scale

The output of our simulations allows us to draw maps of tephra and sulphate aerosol distribution on a global scale or in any particular region. Rather expectedly, we find that space and time dispersion of aerosols depends directly on atmospheric circulation and wind patterns at different atmospheric levels. The Icelandic depression is a dominant factor driving tropospheric circulation over Greenland and Europe. Aerosols injected in the heart of the low-pressure zone are rapidly scattered within

Environmental impact of the sulphate aerosol haze

We next compare our model output with reported haze observations [2]. We find the same sense of propagation, from Iceland alternately towards Asia and North America. A large quasi-circular area, including Polar Regions, and approximately centred on the northern tip of Scandinavia, was covered by the famous dry haze of de Montredon and Franklin [2], [10], [11]. This has been reported on our late-June map (Fig. 3a). It fits reasonably well with the aerosol concentrations deduced from our model,

Conclusions

The first results of our model of climatic impact of the Laki eruption appear to be promising. The LMDzT-INCA model reproduces with good accuracy the observed aerosol distribution. The reality of hemispheric distribution and propagation of sulphate aerosols is attested by reports of environmental effects on livestock, crops and humans. Therefore, that cooling was mostly due to stratospheric sulphate aerosols appears to be vindicated in some detail by our simulation. It should soon allow us to

Acknowledgements

We are grateful to Yves Balkanski and Didier Hauglustaine for providing the LMDz-INCA model. We thank Christiane Textor for her help with the LMDz-INCA model and Emmanuel Le Roy Ladurie for helpful discussion. We thank John Grattan, Steve Self and Mike Widdowson for friendly and useful interactions on this research program. Alan Robock provided incisive reading of an earlier version of the paper, which helped to clarify assumptions and advances made in this study. We are grateful to Steve Self

References (38)

  • J. Grattan et al.

    An amazing and portentous summer: environmental and social responses in Britain to the 1783 eruption of an Iceland volcano

    The Geogr. J.

    (1995)
  • J. Grattan et al.

    Non-climatic factors and the environmental impact of volcanic volatiles: implications of the Laki fissure eruption of AD 1783

    The Holocene

    (1994)
  • M.M. de Montredon

    Recherches sur l'origine et la nature des vapeurs qui ont régné dans l'atmosphère pendant l'été de 1783

    Mém. Acad. R. Sci., Paris

    (1784)
  • B. Franklin

    Meteorological imagination's and conjectures

    Manchester Lit. Philos. Soc., Mem. Proc.

    (1784)
  • G. White

    The Natural History of Selbourne

    (1789)
  • S.P. van Swinden

    Observations on the Cloud Which Appeared in June 1783

  • J. Kington

    The Weather of the 1780s Over Europe

    (1988)
  • T. Thordarson et al.

    The Laki (Skaftár Fires) and Grímsvötn eruptions in 1783–1785

    Bull. Volcanol.

    (1993)
  • T. Thordarson et al.

    Sulphur, chlorine, and fluorine degassing and atmospheric loading by the 1783–1784 AD Laki (Skaftár Fires) eruption in Iceland

    Bull. Volcanol.

    (1996)
  • Cited by (70)

    • Tracing volcanic emissions from the Central Atlantic Magmatic Province in the sedimentary record

      2021, Earth-Science Reviews
      Citation Excerpt :

      The supraregional distribution of volcanic gases from the Laki fissure eruptions (1783–84), and recent air traffic disruptions due to volcanic eruptions on Iceland (Eyjafjallajökull volcano in 2010, Grimsvotn volcano in 2011), can be explained by an anticyclonic circulation pattern (Grattan, 2005). A low-pressure zone known as the Icelandic Depression is a dominant driving force of the tropospheric circulation over Greenland and Europe (Chenet et al., 2005). Volcanic gases that are injected in the heart of this low-pressure zone will be rapidly distributed to the upper troposphere/lower stratosphere (Chenet et al., 2005), and later drawn down in a high pressure cell and efficiently transported and distributed to the surface (Grattan, 2005), demonstrating the importance of the atmospheric circulation patterns.

    • Tracking timescales of short-term precursors to large basaltic fissure eruptions through Fe-Mg diffusion in olivine

      2016, Earth and Planetary Science Letters
      Citation Excerpt :

      Magmatism on the EVZ over the Holocene period has been typified by flood lava eruptions (>1 km3), which include the 8.6 ka Thjorsárhraun, AD 934–938 Eldgjá and AD 1783–1784 Laki eruptions (e.g. Thordarson et al., 2003b; Thordarson and Höskuldsson, 2008). These large-volume, long-lived eruptions have historically produced significant global climatic, environmental and societal impacts (e.g. Thordarson et al., 1996; Larsen, 2000; Thordarson et al., 2001; Thordarson and Self, 2003; Chenet et al., 2005; Oman et al., 2006; Fei and Zhou, 2006; Schmidt et al., 2010), and similar effects may be expected in the event of future flood lava eruptions (Schmidt et al., 2011). In this study, we use diffusion chronometry to characterise the timescales of crystal storage and transport in the lead-up to the AD 1783 Laki eruption.

    • The drought and locust plague of 942-944 AD in the Yellow River Basin, China

      2016, Quaternary International
      Citation Excerpt :

      The eruption process of Icelandic volcanoes is prolonged and complex, thus complicating the environmental effects (Self et al., 2006; Thordarson and Larsen, 2007). For instance, the 1783–1784 AD Lakagígar (Laki) eruption lasted for eight months and was composed of at least 10 eruptive episodes (Thordarson et al., 1996; Thordarson and Self, 2003; Chenet et al., 2005). The Eldgjá eruption lasted several years, and its magnitude was significantly larger than the Lakagígar eruption.

    View all citing articles on Scopus
    View full text