SO2 emissions, plume heights and magmatic processes inferred from satellite data: The 2015 Calbuco eruptions

Quantifying time-series of sulfur dioxide (SO2) emissions during explosive eruptions provides insight into volca- nic processes, assists in volcanic hazard mitigation, and permits quantification of the climatic impact of major eruptions. While volcanic SO2 is routinely detected from space during eruptions, the retrieval of plume injection height and SO2 flux time-series remains challenging. Here we present a new numerical method based on forward- and backward-trajectory analysis which enable such time-series to be determined. Using this method applied to GOME-2 satellite imagery we investigate the SO2 emissions from two sub-Plinian eruptions of Calbuco, Chile, produced in April 2015. Our results show a mean injection height of 15 km for the two eruptions, with overshooting tops reaching 20 km. We calculate a total of 0.295 ± 0.045 Tg of SO2 emitted, with 0.140 ± 0.033 Tg produced by the first eruption and 0.155 ± 0.031 Tg by the second one. Using standard models we convert plume heights to mass eruption rates (MER). Comparing gas flux and MER we discover quite different volcanic processes driving the two eruptions, with the first eruption producing an SO2 flux three times higher than the second one, while they both had similar MERs. We propose that this difference re- flects different exsolved volatile contents before the onset of the two eruptions, with the first eruption richer in pre-exsolved gas than the second one. This hypothesis is supported by melt inclusion measurements of sulfur concentrations in plagioclase phenocrysts and groundmass glass of tephra samples through electron microprobe analysis. Combining the satellite and petrological analysis, we propose that the overpressure caused by the pre- exsolved volatile phase (not only SO2, but also probably H2O and CO2) may have triggered the eruption. These results demonstrate that our new methodology produces constraints on SO2 flux and plume height time- series permitting new insights into sub-surface processes using satellite SO2 data.