Strength of TROPOMI Satellite Observations in Retrieving Hourly Resolved Sources of Volcanic Sulfur Dioxide by Inverse Modeling

Volcanic eruptions release sulfur dioxide (SO2) gas, affecting air quality, ecosystems, aviation, and potentially climate. To comprehensively assess these atmospheric effects as well as local volcanic hazards, high-temporal-resolution information on SO2 mass flux and injection height is crucial and can be delivered on an hourly basis by inverse modeling. We show here the strength of assimilating high-spatial-resolution SO2 column measurements from Sentinel-5P/TROPOMI hyperspectral observations, compared to coarser resolution Suomi-NPP/OMPS data, through inverse modeling using the CHIMERE Eulerian chemistry-transport model. As a case study, we investigate the dynamics of the 2018 Ambrym eruption (December 13–18, Vanuatu) starting with the extrusion of an intra-caldera lava flow, coinciding with lava lake draining, followed by the lateral propagation of a voluminous intrusion of magma triggering a submarine eruption. Prior to this eruption marking the end of a decades-long sustained passive degassing associated to lava lake activity, Ambrym held the distinction of top ranking volcanic SO2 emitters in the world. The assimilation of TROPOMI data by inverse modeling proves efficient for accurately characterizing Ambrym SO2 emissions during periods of intense degassing that can be strongly underestimated with OMPS data. We show that this advantage relies on the high spatial resolution of TROPOMI observations which allows for robustly capturing near-source highly concentrated SO2 plumes emitted a few hours before satellite overpass. TROPOMI measurements, also facilitate the detection of volcanic SO2 even at the noisy swath edges, thanks to a superior signal-to-noise ratio and less pixel geometric distortion than OMPS. This distinctive attribute of TROPOMI has enabled the characterization of pre-eruptive SO2 emissions missed by OMPS. Moreover, our inverse modeling procedure effectively discerns, and traces hidden SO2 plumes beneath clouds by assimilating SO2 column data from successive satellite overpasses, hence mitigating satellite retrieval limitations. Furthermore, this approach proves to be less susceptible to interference from ash emissions, as compared to flux estimation derived from near-source observations by the geostationary Himawari-8/AHI sensor. Our study unveils that the 2018 Ambrym eruption discharged approximately 42 +/-16 kt (by OMPS) to 52+/-13 kt (by TROPOMI) of SO2 during the event. The hourly-resolved SO2 flux time series retrieved by inverse modeling sheds light on the eruption’s progression, pinpointing distinct sources of SO2 emissions from either lava flow or shallow magma intrusions. In summary, the assimilation of TROPOMI data into inverse modeling procedures holds substantial promise for advancing our comprehension of magma transport and the environmental repercussions associated with volcanic eruptions.