Future Directions in Volcano Seismology

Abstract:

Volcano seismology provides three main contributions in volcano science: 1) eruption forecasts; 2) crustal characterization; and 3) constraining volcanic processes. Better crustal characterization and understanding of volcanic process commonly feeds back into better eruption forecasting. Future directions in this field include more consistent analyses and modeling of a wider variety of seismic signals made possible partially through better constrained subsurface models.

In seeking insights into volcanic processes, we need improved understanding of the spatial and temporal relationships among different volcanic signals such as tremor, very-long-period (VLP), long-period (LP) and volcano-tectonic (VT) earthquakes. Waveform modeling has come a long way toward understanding long-period sources but additional work is warranted on more diverse events at more volcanoes. In order to make a holistic interpretation of all earthquake types in space and time, we need to utilize new methods of detection and location using non-standard techniques that allow these exotic sources inclusion in standard earthquake catalogs. The inclusion of complementary datasets, such as infrasound, is also important in understanding shallow processes.

Traditional travel-time tomography, combined with novel uses of noise, has identified large-scale magma “chambers”; but to-date, not in enough detail to be useful for monitoring purposes. High-density nodal deployments of portable seismometers have illuminated finer-scale volcanic structures, especially in the shallow crust. Combining improved velocity models with high-precision earthquake locations, especially incorporating smaller earthquakes through matched filtering, has proven critical in understanding magma storage and plumbing systems.

Many techniques discussed above provide useful insights after an eruption, and automation has the potential to advance real-time eruption forecasting. By incorporating constraints on processes and structures with constraints from other fields (e.g., petrology, geodesy) in a physics-based model, volcano science can move toward process-based forecasts, resulting in longer forecast windows for emergency responders and the public.