Sensitivity of volcanic ash dispersion modelling to input grain size distribution based on hydromagmatic and magmatic deposits

Abstract

The size distribution of volcanic ash is rarely measured in real time and Volcanic Ash Advisory Centres (VAACs) often rely on a default particle size distribution (PSD) to initialise their dispersion models when forecasting the movement of ash clouds. We conducted a sensitivity study to investigate the impact of PSD on model output and consider how best to apply default PSDs in operational dispersion modelling. Compiled grain size data confirm that, when considering particles likely to be in the distal ash cloud (<125 μm diameter), magma composition and eruption size are the dominant controls on grain size distribution. Constraining the PSD is challenging but we find that the grain size of deposits from large hydromagmatic eruptions remains relatively constant with distance, suggesting that total (whole-deposit) grain size distributions (TGSDs) for these eruptions could be estimated from a few samples. We investigated the sensitivity of modelled ash mass loadings (in the air and on the ground) to input PSDs based on coarse to fine TGSDs from our dataset. We found clear di_erences between modelled mass loadings and the extent of the plume. Comparing TGSDs based on ground-only and ground-plus-satellite data for the Eyjafjallajokull 2010 eruption, we found that basing input PSDs on TGSDs from deposits alone (likely missing the finest particles) led to lower modelled peak ash concentrations and a smaller plume.