No abstract

Abstract. For over 6-months, the 2014–2015 effusive eruption at Holuhraun, Iceland injected considerable amounts of sulphur dioxide (SO2) into the lower troposphere with a daily rate of up to one-third of the global emission rate causing extensive air pollution across Europe. The large injection of SO2, which oxidises to form sulphate aerosol (SO42−), provides a natural experiment offering an ideal opportunity to scrutinise state-of-the-art general circulation models (GCMs) representation of aerosol-cloud interactions (ACIs). Here we present Part 1 of a two-part model inter-comparison using the Holuhraun eruption as a framework to analyse ACIs. We use SO2 retrievals from the Infrared Atmospheric Sounding Interferometer (IASI) instrument and ground-based measurements of SO2 and SO42− mass concentrations across Europe in conjunction with trajectory analysis using the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model to assess the spatial and chemical evolution of the volcanic plume as simulated by 5 GCMs and a chemical transport model (CTM). IASI retrievals of plume altitude and SO2 column load reveal that the volcanic perturbation is largely contained within the lower troposphere and that the spatial evolution and vertical variability of the plume is reasonably well captured by the models, although the models underestimate the mean plume altitude. HYSPLIT trajectories are used to attribute to Holuhraun emissions 184 instances of elevated sulphurous surface mass concentrations recorded at 22 air monitoring stations across Europe. Comparisons with the simulated concentrations show that the models underestimate the elevated SO2 concentrations observed at stations closer to Holuhraun whilst overestimating those observed further away. Using a biexponential function to describe the decay of observed surface mass concentration ratios of SO2-to-SO42− with plume age, in-plume gas-phase and aqueous-phase oxidation rates are estimated as 0.031 ± 0.002 h−1 and 0.22 ± 0.16 h−1 respectively with a near-vent ratio of 31 ± 4 [μgm−3 of SO2 / ugm−3 of SO42−]. The derived gas-phase oxidation rates from the models are all lower than the observed estimate, whilst the majority of the aqueous-phase oxidation rates agree with the observed rate. This suggests that the simulated plumes capture the observed chemical behaviour in the young plume (when aqueous-phase oxidation is dominant), yet not in the mature plume (when gas-phase oxidation is dominant). Overall, despite their coarse resolution, the 6 models show reasonable skill in capturing the spatial and chemical evolution of the Holuhraun plume which is essential when exploring the eruption impact on ACIs in the second part of this study.

No abstract

No abstract

No abstract