Sensitivity of the Northern Hemisphere Winter Stratosphere to the Strength of Volcanic Eruptions

Major tropical volcanic eruptions lead to a large amount of aerosols in the equatorial stratosphere. Besides cooling the earth surface by backscattering the incoming solar radiation, these volcanic aerosols absorb solar near-infrared and terrestrial long-wave radiation, resulting in a positive temperature anomaly in the equatorial stratosphere. The positive temperature anomaly increases the meridional temperature gradient, which, via the thermal-wind relation, has been proposed to strengthen the polar vortex in the first northern hemisphere (NH) post-eruption winter. The strengthened polar vortex shifts the North Atlantic Oscillation to a positive phase, leading to higher temperatures over Northern Europe and Siberia and lower temperatures over the Labrador Sea and Southern Europe. Reconstruction of surface temperatures after large tropical eruptions show a signal over Europe and Siberia in boreal wintertime, with temperature anomalies resembling a positive phase of the North Atlantic Oscillation. However, the current generation of coupled climate models do not, on average, produce a significantly strengthened polar vortex for the largest tropical eruptions since 1850 as suggested by observational records. Here we use the Earth system model of the Max-Planck –Institute (MPI-ESM-LR) to investigate the impact of tropical eruptions on the NH polar stratosphere. We explore for two different eruption magnitudes, if the volcanically-induced meridional temperature gradient anomaly has a direct impact on the polar vortex and to what extent a volcanic eruption might influence the ensemble variability of the northern polar stratosphere in wintertime. We combine the two largest eruptions of the historical simulations performed within CMIP5 (the 1883 Krakatau eruption and the 1991 Pinatubo eruption) and compare the dynamical impact with simulations of the even larger Tambora eruption in 1815. We analyze an ensemble of 20 simulations of large eruptions (10 of each Krakatau and Pinatubo) and a 20-member ensemble of the very large Tambora eruption focusing on the position of the maximum meridional temperature gradient. The relatively large ensemble size enables us not only to assess the mean state of the polar stratosphere in post-volcanic winters, but also to investigate changes in ensemble variability of the NH polar temperatures in the stratosphere. All model experiments show the expected positive zonal-mean temperature anomalies in the equatorial stratosphere due to the absorption of radiation by the volcanic aerosols. The temperature anomalies are four times larger after the Tambora eruption compared to the eruption of Krakatau and Pinatubo. We find that for both eruptions strengths the strongest temperature-gradient is not in the region of the polar vortex, but occurs equator-wards at approximately 30°N, at the edge of the bulk of the aerosol cloud. Consequently, we find significant zonal wind anomalies in both experiments at 30°N in 50 hPa – 10 hPa, especially in early winter. At this position we obtain significant westerly zonal wind anomalies. Under the very strong forcing of the Tambora eruption, the westerly wind anomalies are sufficiently strong to alter the propagation of planetary waves. These waves are refracted towards the equator and cannot break in the region of the northern hemisphere polar vortex. Hence, we obtain a significantly stronger polar vortex in the case of the Tambora eruption. Because the radiative perturbation of the volcanic aerosols does not directly affect the polar vortex, the response of the polar vortex to the eruptions of Krakatau and Pinatubo is less robust. The high latitude meridional temperature gradient anomaly is primarily due to dynamical heating by an enhanced residual circulation. Depending on the spatial structure of the polar downwelling branch of the residual circulation anomalies, the vortex can be either strengthened or weakened. The ensemble variability of the zonal mean temperature and zonal wind anomalies during mid- and late winter in the northern polar stratosphere is significantly reduced under the forcing of the Tambora eruption. The very large Tambora eruption constrains the NH polar vortex to a stronger-than-average state. In the case of the smaller Krakatau/Pinatubo eruptions the variability is not reduced in the following winter. Internal variability plays a dominant role in the model for the state of the polar vortex in the NH post-eruption winter and can therefore mask the impact of an eruption of the size of Pinatubo or Krakatau in the MPI-ESM-LR.