Towards monitoring phreatic eruptions using seismic noise

Phreatic and hydrothermal eruptions remain among the most difficult to forecast. The frequent absence of clear precursor signals challenges volcanologists' ability to provide timely and accurate hazard advice. They remain poorly understood and have recently caused human fatalities. It is therefore paramount to better investigate such eruptions by integrating new methodologies to fully understand the preparatory processes at play and improve our ability to forecast them.

Among the different approaches to monitor volcanoes, seismology forms the basis, and most active volcanoes are nowadays equipped with at least one seismometer. Seismology is unique amongst the Earth Science disciplines involved in volcano studies, as it provides real-time information; as such, it is the backbone of every monitoring program worldwide. With data storage capabilities expanding over the last decades, new data processing tools have emerged taking advantage of continuous seismic records. Recent advances in volcano monitoring have taken advantage of seismic noise to better understand the time evolution of the subsurface. 

The well-established seismic interferometry has allowed us to detect precursory changes (dv/v or decorrelation) to phreatic eruptions at different volcanoes, thereby providing critical insights into the triggering processes. More recent approaches have provided insights into the genesis of gas-driven eruptions using seismic attenuation (DSAR: Displacement seismic amplitude ratio) and correlation with tidal stresses (LSC). Yet, puzzling observations have been made at different volcanoes requiring the use of numerical models and machine learning-based approaches, as well as complementary dataset to reach a more comprehensive understanding. This presentation will review recent insights gained into precursory processes to phreatic eruptions using seismic noise and how we could possibly forecast them. These tools are freely available to the community and have the potential to serve monitoring and aid decision-making in volcano observatories.