Reconstruction of flux and altitude of volcanic SO2 emissions from IASIsatellite observations: implications for volcanological and atmosphericalstudies.

International audienceVolcanic SO2 degassing is a crucial indicator of the sub-surface volcanic activity, which is widelyused today for volcano monitoring and hazard assessment purposes. Volcanic SO2 is also importantregarding atmospherical studies. More easily detectable from space, SO2 can be used as a proxy ofthe presence of ash to anticipate air traffic issues caused by explosive eruptions. Moreover, volcanicSO2 strongly impacts air quality but also climate following its conversion to radiatively-activesulphate aerosols. However, the accurate assessment of these various impacts is currently hamperedby the poor knowledge of volcanic SO2 emissions, which can substantially vary with time, in termsof flux and altitude.To fulfil this need, we propose a strategy relying on satellite observations, which consequently allowsfor monitoring the eruptive activity of any remote volcano. The method consists in assimilatingsnapshots of the SO2 load, provided every ~12 hours by IASI, in an inversion scheme that involvesthe use of a chemistry-transport model to describe the dispersion of SO2 released in the atmosphere.Applied on Eyjafjallajökull (Iceland) and Etna (Italy) eruption case-studies, this procedure allows forretrospectively reconstructing both the flux and altitude of the SO2 emissions with an hourlyresolution. We show the improvement gained in the simulations and forecasts of the location andmass load of volcanic SO2 clouds using such a detailed reconstruction of emissions.For calibration-validation purpose, we compared our satellite-derived time-series of the SO2 fluxwith ground-based observations available on Etna. This comparison indicates a good agreementduring ash-poor phases of the eruption. However, large discrepancies are observed during the ashrichparoxysmal phase as a result of enhanced plume opacity affecting ground-based ultravioletspectroscopic retrievals. Therefore, the SO2 emission rate derived from the ground is underestimatedby almost one order of magnitude. This result calls for the necessity to revisit currently availableinventories of the global budget of sulfur released by volcanoes, because they heavily rely on groundbasedobservations. It also shows that volcano observatories cannot rely solely on ground-basedspectroscopical observations for the monitoring of ash-rich explosive eruptions.Moreover, we will discuss the assimilation procedure of recently-developed IASI products whichdeliver snapshots of the volcanic SO2 cloud altitude. Such improvement renders the inversionprocedure, used for the reconstruction of the altitude of emissions, independent of the wind shearprerequisite.Eventually, building on our accurate source of precursory SO2 gas emissions, we can explore theformation and lifecycle of sulphate aerosols in volcanic plumes. Remote sensing of troposphericsulphate aerosols from moderate eruptions is not straightforward. However, we will show how acombination of chemistry-transport modelling and space-borne CALIOP lidar observations allowsfor tracking these aerosols despite their small concentration. The latest promising developments fromhigh-resolution infrared sounders will bring further constraints on sulphate aerosol load andcharacteristics in dispersed volcanic clouds