EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Toba volcano super eruption destroyed the ozone layer and caused a human population bottleneck

Sergey Osipov1, Georgiy Stenchikov2, Kostas Tsigaridis3,4, Allegra LeGrande3,4, Susanne Bauer3,4, Mohamed Fnais5, and Jos Lelieveld1
Sergey Osipov et al.
  • 1Max Planck Institute for Chemistry, Mainz, Germany (
  • 2King Abdullah University of Science and Technology, Thuwal, Saudi Arabia
  • 3Center for Climate Systems Research, Columbia University, New York, USA
  • 4NASA Goddard Institute for Space Studies, New York, USA
  • 5King Saud University, College of Science, Riyadh, Saudi Arabia

Volcanic eruptions trigger a broad spectrum of climatic responses. For example, the Mount Pinatubo eruption in 1991 forced an El Niño and global cooling, and the Tambora eruption in 1815 caused the "Year Without a Summer." Especially grand eruptions such as Toba around 74,000 years ago can push the Earth's climate into a volcanic winter state, significantly lowering the surface temperature and precipitation globally. Here we present a new, previously overlooked element of the volcanic effects spectrum: the radiative mechanism of stratospheric ozone depletion. We found that the volcanic plume of Toba enhanced the UV optical depth and suppressed the primary formation of stratospheric ozone from O2 photolysis. Sulfate aerosols additionally reflect the photons needed to break the O2 bond (λ < 242 nm), otherwise controlled by ozone absorption and Rayleigh scattering alone during volcanically quiescent conditions. Our NASA GISS ModelE simulations of the Toba eruption reveal up to 50% global ozone loss due to the overall photochemistry perturbations of the sulfate aerosols. We also consider and quantify the radiative effects of SO2, which partially compensated for the ozone loss by inhibiting the photolytic O3 sink.

Our analysis shows that the magnitude of the ozone loss and UV-induced health-hazardous effects after the Toba eruption are similar to those in the aftermath of a potential nuclear conflict. These findings suggest a “Toba ozone catastrophe" as a likely contributor to the historic population decline in this period, consistent with a genetic bottleneck in human evolution.

How to cite: Osipov, S., Stenchikov, G., Tsigaridis, K., LeGrande, A., Bauer, S., Fnais, M., and Lelieveld, J.: Toba volcano super eruption destroyed the ozone layer and caused a human population bottleneck, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-4131,, 2020


Display file

Comments on the display

AC: Author Comment | CC: Community Comment | Report abuse

displays version 1 – uploaded on 04 May 2020
  • CC1: Comment on EGU2020-4131, Alexander Archibald, 06 May 2020

    Dear Dr Osipov, 

    Thank you for your very clear slides and very interesting work! 

    It would be nice to know some of the details of how you changed the photolyis code, but my question is more on the conclusion of the effects of the ozone hole. From your figures it looks like the tropical ozone hole is short lived (a few months). Maybe short is quite subjective -- so my question is, is the length of the hole long enough to contribute significantly to the historic population decline? How do we convert the change in ozone column and changes in UV to this? 


    • AC1: Reply to CC1, Sergey Osipov, 06 May 2020

      Dear Dr. Archibald,

      Thank you for your positive feedback and interest in this work! 

      The "ozone hole" conditions after Toba persisted for almost a year. I agree that it may seem "short-lived" at first, but it is not. WHO considers UV index (UVI) of 10 as "extreme," which is background state in tropics. Even 15 minutes of exposure causes sunburn and eye damage (erythema, photokeratitis). Please have a look at slide 9 for more details and references on UV-induced health-hazardous effects.

      UVI index is an excellent overall diagnostic to asses the consequences of ozone depletion. We quantified that 250->125 DU ozone depletion translates into 12-> 60 UVI increase. Yet, SO2 and sulfate aerosols compensate partially for the loss of the ozone shield, which translates into an effective UVI increase to 28. SO2 spectral absorption is similar to ozone, and during the initial stage of the eruption, volcanic plume + O3 (which is unperturbed yet) completely shuts down UV flux at the surface. During the later stage of the eruption, SO2 is replaced by less efficient SO4 aerosols, which do not compensate entirely for the O3 depletion, which explains the UVI to increase at the surface.

      Please, also note that halogens (not included in this study) will drastically increase not only the strength of the ozone hole but also prolong its lifetime. Thus, this estimate represents the lower limit of the potential impact. One should also take into account an "entire cocktail of the environmental stress factors": UV exposure, the decay of the habitat, etc. For more details on the climate impacts of Toba, please, refer to

      Please, refer to the slide 10 for some details on the Fast-J2 changes. The in-depth details and the source code will be made public once the paper is published (currently under review). Briefly, the original fast-j2 code splits the wavelength spectrum into two regions: above 291 nm (full scattering code) and below 291 nm (pseudo-absorption). The value of 291 nm is not random and was picked to resolve O3 photolysis. For the computational efficiency purpose, the "below 291 nm" range treats the Rayleigh scattering as purely absorbing (with a 0.57 correction factor for the optical depth). In the volcanically unperturbed atmosphere, this is a code approximation, but it fails to capture the scattering by sulfate aerosols and does not represent actinic flux increase above the plume and decrease below it. I have remedied this but extending the scattering calculations to 200 nm (which covers O2 photolysis). The challenge is purely numerical/technical, as the overall OD grows exponentially into shorter wavelengths. I've also coupled the online SO2 from the chemistry module to Fast-J2 and included SO2 cross-section to the photolysis module.

      Please, let me know if you have further questions.

      Best regards,
      Sergey Osipov