“…We identified robust responses using the statistical significance of the anomalies, as well as degree to which the ensemble agreed on the sign of the response. The main results from the CESM1(WACCM) Laki simulations are summarized as follows: - The Laki aerosols as simulated by CESM1(WACCM) have a peak effective radius ranging from about 0.2 to 0.4 μm, with an average effective radius near 0.3 μm in the NH lower stratosphere.
- SO 2 gas injected into the upper troposphere/lower stratosphere at high latitudes and the subsequent SO 4 aerosols are transported to the SH at significant mixing ratios (up to 5 ppbv SO 4 in the SH tropics and ~0.5 ppbv SO 4 at the South Pole).
- Robust wind and temperature responses are simulated in both hemispheres in the months after the onset of the NH high‐latitude eruption.
- The SAM response extends from the stratosphere to the troposphere, lasting about 3 months and peaking about 5 months after the beginning of the eruption, in SH spring.
- The NAM response is longer lasting but more confined to the troposphere; this response is strongest in the boreal late winter/spring and is related to the ENSO forcing as well as pressure changes at the surface (Zambri et al, ).
- The choice of base period for computing anomalies, that is, subtracting the mean of some control period or the mean of an equivalent ensemble without the volcanic perturbation, can significantly alter the perceived “forced response” to a volcanic eruption, in agreement with the analysis in Pausata, Grini, et al ().
…”