Including ash in UKESM1 model simulations of the Raikoke volcanic eruption reveals improved agreement with observations

Abstract

In June 2019 the Raikoke volcano, located in the Kuril Islands northeast of the Japanese archipelago, erupted explosively and emitted approximately 1.5 Tg ± 0.2 Tg of SO2 and 0.4–1.8 Tg of ash into the upper troposphere and lower stratosphere. Volcanic ash is usually neglected in modelling stratospheric climate changes since larger particles have generally been considered to be short-lived particles in terms of their stratospheric lifetime. However, recent studies have shown that the coagulation of mixed particles with ash and sulfate is necessary to model the evolution of aerosol size distribution more accurately. We perform simulations using a nudged version of the UK Earth System Model (UKESM1) that includes a detailed two-moment aerosol microphysical scheme for modelling the oxidation of sulfur dioxide (SO2) to sulfate aerosol and the detailed evolution of aerosol microphysics in the stratosphere. We compare the model with a wide range of observational data. The current observational network, including satellites, surface-based lidars, and high-altitude sun photometers means that smaller-scale eruptions such as Raikoke provide unprecedented detail of the evolution of volcanic plumes and processes, but there are significant differences in the evolution of the plume detected using the various satellite retrievals. These differences stem from fundamental differences in detection methods between, e.g. lidar and limb-sounding measurement techniques and the associated differences in detection limits and the geographical areas where robust retrievals are possible. This study highlights that, despite the problems in developing robust and consistent observational constraints, the balance of evidence suggests that including ash in the model emission scheme provides a more accurate simulation of the evolution of the volcanic plume within UKESM1.