Nonlinear ground response due to air-to-ground coupling of volcanic explosions; a study case from Mt. Etna volcano, Sicily.

Ground response (GR) refers to the amplification and damping of seismic wavefield components under linear and nonlinear elastic conditions. While seismic waves are the primary triggers of GR, other dynamic phenomena, such as explosions and strong acoustic waves, can also induce GR once they couple into the ground. In volcanic environments, natural explosions frequently interact with unconsolidated near-surface materials, making GR a critical factor in assessing volcanic hazards.

To investigate GR in such contexts, we selected Mt. Etna as a study site due to its persistent volcanic activity, which generates a wide frequency range (0.01–100 Hz) of seismo-acoustic signals. Additionally, Mt. Etna features complex ground structures, such as faults, dykes, and unconsolidated scoria deposits, making it an ideal natural laboratory for examining GR phenomena. In 2019, a multi-parametric network was deployed near its summit crater, comprising broadband seismometers, infrasound sensors, and a buried fiber optic cable (30 cm depth) for distributed dynamic strain sensing (DDSS).

We compiled a catalog of over 8,000 volcanic explosions. Our observations reveal emergent high-frequency (10–50 Hz) acoustic waves embedded within the low-frequency signals of the explosions. These high frequencies are amplified when the explosions couple into the scoria material of the deposit, as evidenced by the DDSS and broadband seismometer data.

To characterize the local response of the near-surface material during air-to-ground coupling of the explosions, we analyzed stress-rate vs. strain-rate relationships derived from peak-to-peak (p-p) amplitudes of GR signals and classified explosion events. Explosions were classified using waveform similarity, while GR in the DDSS signals were classified using a modified approach that incorporates both temporal and spatial dimensions. These relationships reveal hyperelastic behavior of the scoria material, described by three distinct and consecutive elastic stages: linear, softening, and stiffening.

The hyperelastic curves enable the extraction of key elastic parameters, which we use to model GR at Mt. Etna with waveform propagation codes employing lattice methods. We validate this approach by estimating Vp velocities from elastic parameters and comparing them with direct Vp measurements from tap test on the fibre optic cable. Preliminary modeling results demonstrate the potential of lattice methods to capture the nonlinear dynamics of geomaterials and provide deeper insights into the elastic parameters influencing GR. These findings underscore the importance of incorporating GR into volcanic hazard assessments and enhance our understanding of near-surface material dynamics in volcanic environments.