The Evolution of Icelandic Volcano Emissions, as Observed From Space in the Era of NASA's Earth Observing System (EOS)

Abstract

AbstractVolcanoes are natural phenomena that have global environmental impacts. Satellite remote sensing can help classify volcanic eruptions and track the dispersion of emissions. We assess multiple volcanic eruptions in Iceland (Eyjafjallajökull 2010, Grímsvötn 2011, and Holuhraun 2014–2015), using space‐borne observations to infer information about the geological dynamics of each volcano and the properties and evolution of plume particles. We derive qualitative constraints on plume particle size, shape, and light‐absorption characteristics from Multiangle Imaging SpectroRadiometer (MISR) space‐borne imagery. With the MISR Research Algorithm (RA), we distinguish sulfate/water‐dominated volcanic plumes from Holuhraun and ash‐dominated plumes from Eyjafjallajökull and Grímsvötn, and even identify subtler changes in ash particle size and light‐absorption within plumes. Additionally, plume heights are retrieved geometrically from MISR. These are combined with surface thermal anomalies from the MODerate resolution Imaging Spectroradiometer (MODIS) and SO2 concentrations derived from the Ozone Monitoring Instrument (OMI) to synthesize eruption remote‐sensing chronologies. Signals related to differences in particle properties are identified and linked to evolving magma composition at Eyjafjallajökull. The results illustrate the potential to distinguish qualitative differences in eruptive magma composition based on particle light absorption and plume profile from remote‐sensing. For the sulfate‐rich Holuhraun plumes, the influence of aerosol hygroscopic growth during transport is inferred from such data. Three processes appear to dominate plume evolution in Iceland: downwind aerosol formation, particle hydration, and particle deposition. This work demonstrates enhanced MISR capabilities and, more generally, remote‐sensing analysis that can be applied globally, especially where suborbital volcano observations are limited or entirely absent.