De Siena, Luca

Seismogram envelopes recorded at Campi Flegrei caldera show diffusive characteristics as well as steep amplitude increases in the intermediate and late coda, which can be related to the presence of a non-uniformly scattering medium. In this paper, we first show the results of a simulation with a statistical model considering anisotropic scattering interactions, in order to match coda-envelope duration and shape.We consider as realistic parameters for a volcanic caldera the presence of large square root velocity fluctuations (10 per cent) and two typical correlation lengths for such an heterogeneous crust, a = 0.1 and 1 km. Then, we propose the inclusion of a diffusive boundary condition in the stochastic description of multiple scattering, in order to model intermediate and late coda intensities, and particularly the sharp intensity peaks at some stations in the caldera. Finally, we show that a reliable 2-D synthetic model of the envelopes produced by earthquakes vertically sampling a small region can be obtained including a single drastic change of the scattering properties of the volcano, that is, a caldera rim of radius 3 km, and sections varying between 2 and 3 km. These boundary conditions are diffusive, which signifies that the rim must have more scattering potential than the rest of the medium, with its diffusivity 2–3 orders of magnitude lower than the one of the background medium, so that the secondary sources on its interface(s) could enhance coda intensities. We achieve a good first-order model of high-frequency (18 Hz) envelope broadening adding to the Monte Carlo solution for the incident flux the secondary source effects produced by a closed annular boundary, designed on the caldera rim signature at 1.5 km depth. At lower frequencies (3 Hz) the annular boundary controls the intermediate and late coda envelope behaviour, in a way similar to an extended diffusive source. In our interpretation, the anomalous intensities observed at several stations and predicted by the final Monte Carlo solutions are mainly due to the diffusive transmission reflection from a scattering object of increased scattering power, and are controlled by its varying thickness.

Identifying deformation and pre-failure mechanisms preceding faulting is key for fault mechanics and for interpreting precursors to fault rupture. This study presents the results of a new and robust derivation of first motion polarity focal mechanism solutions (FMS) applied to acoustic emission (AE). FMS are solved using a least squares minimization of the fit between projected polarity measurements and the deviatoric stress field induced by dilatational (T-type), shearing (S-type), and compressional (C-type) sources. 4 × 10 cm cylindrical samples of Alzo Granite (AG, porosity <1%) and Darley Dale Sandstone (DDS, porosity ≈14%) underwent conventional triaxial tests in order to investigate the relationships between increasing confining pressure (5, 10, 20, and 40 MPa), deformation and failure mode, and role of microstructural features. Results highlight that S-type events occur in very low numbers with poor spatial correlation to fault structure. Instead, deformation is driven by a complex interplay between compactant (C-type) and dilatant (T-type) regions of deformation. C-type events are the earliest precursor related to crack nucleation and T-type events mark new cracks opening, with the onset of fracture growth characterized by periodic cycles of coalescence. For AG a single sequence is able to lead to dynamic failure, while for DDS several cycles are needed for coalescence to take place due to the competition between dilatant and compactant deforming regions induced by multiple fracture nucleation sites. The occurrence of C- and S-type events is also consistent with a quasi-static premonitory phase, or foreshock, before a critical nucleation length allows the development of a planar localization.Identifying deformation and pre-failure mechanisms preceding faulting is key for fault mechanics and for interpreting precursors to fault rupture. This study presents the results of a new and robust derivation of first motion polarity focal mechanism solutions (FMS) applied to acoustic emission (AE). FMS are solved using a least squares minimization of the fit between projected polarity measurements and the deviatoric stress field induced by dilatational (T-type), shearing (S-type), and compressional (C-type) sources. 4 × 10 cm cylindrical samples of Alzo Granite (AG, porosity <1%) and Darley Dale Sandstone (DDS, porosity ≈14%) underwent conventional triaxial tests in order to investigate the relationships between increasing confining pressure (5, 10, 20, and 40 MPa), deformation and failure mode, and role of microstructural features. Results highlight that S-type events occur in very low numbers with poor spatial correlation to fault structure. Instead, deformation is driven by a complex interplay between compactant (C-type) and dilatant (T-type) regions of deformation. C-type events are the earliest precursor related to crack nucleation and T-type events mark new cracks opening, with the onset of fracture growth characterized by periodic cycles of coalescence. For AG a single sequence is able to lead to dynamic failure, while for DDS several cycles are needed for coalescence to take place due to the competition between dilatant and compactant deforming regions induced by multiple fracture nucleation sites. The occurrence of C- and S-type events is also consistent with a quasi-static premonitory phase, or foreshock, before a critical nucleation length allows the development of a planar localization.

A three-dimensional S wave attenuation tomography of Mt. Vesuvius has been obtained with multiple measurements of coda-normalized S-wave spectra of local small magnitude earthquakes.We used 6609 waveforms, relative to 826 volcano-tectonic earthquakes, located close to the crater axis in a depth range between 1 and 4 km (below the sea level), recorded at seven 3-component digital seismic stations. We adopted a two-point ray-tracing; rays were traced in an high resolution 3-D velocity model. The spatial resolution achieved in the attenuation tomography is comparable with that of the velocity tomography (we resolve 300m side cubic cells). We statistically tested that the results are almost independent from the radiation pattern. We also applied an improvement of the ordinary spectral-slope method to both P- and S-waves, assuming that the differences between the theoretical and the experimental high frequency spectral-slope are only due to the attenuation effects. Consequently we could check the codanormalization method also comparing the S attenuation image with the P attenuation image. The images were obtained inverting the spectral data with a multiple resolution approach. Results have shown the general coincidence of low attenuation with high velocity zones. The joint interpretation of velocity and attenuation images allows us to interpret the low attenuation zone intruding toward the surface until a depth of 500m below the sea level as related to the residual part of solidified magma from the last eruption. In the depth range between −700 and −2300 images are consistent with the presence of multiple acquifer layers. No evidence of magma patches greater than the minimum cell dimension (300m) has been found. A shallow P wave attenuation anomaly (beneath the southern flank of the volcano) is consitent with the presence of gas saturated rocks. The zone characterized by the maximum seismic energy release cohincides with a high attenuation and low velocity volume, interpreted as a cracked medium.

Deception Island is one of the most active and best-documented volcanoes in Antarctica. Since its last eruption in 1970, several geophysical surveys have targeted reconstructing its magmatic systems. However, geophysics fails to reconstruct the pathways magma and fluids follow from depth to erupt at the surface. Here, novel data selection strategies and multi-frequency absorption inversions have been framed in a Geographical Information System, using all available geological (vents and faults distribution), geochemical and geophysical knowledge of the volcano. The result is the detection of these eruptive pathways. The model offers the first image of the magma and associated fluids pathways feed the 1967, 1969, and 1970 eruptions. Results suggest that future ascending paths might lead to active research bases and zones of planned helicopter rescue. The connection between seismic absorption, temperature, and fluid content makes it a promising attribute for detecting and monitoring eruptions at active calderas.

Earthquakes at Campi Flegrei have been low magnitude and sparse since 1985, denying onshore monitoring observations of their usual source for structural constraint: seismic tomography. Here we used continuous seismic records from 2011–2013 to reconstruct period-dependent Rayleigh wave group velocity maps of the volcano. The Neapolitan Yellow Tuff rim faults bound high-velocity intracrater products of historical eruptions, which act as a barrier for deep fluid migration. The anomaly with lowest group velocity is aseismic and corresponds to the portion of a fluid storage zone that was fractured during the 1984 volcanic unrest under Pozzuoli town. Earthquake locations show that fluids migrate from this area toward the Solfatara and Pisciarelli fumaroles along shallower low-velocity fractures. The aseismic anomaly is likely fed by a deep-seated offshore magmatic source. Its spatial relation with regional dynamics and observations from historical unrests mark the area as the most likely feeder pathway for fluid and magmatic inputs from depth.

The Amatrice–Visso–Norcia seismic sequence struck Central Italy across the Apenninic normal fault system in 2016. Fluids likely triggered the sequence and reduced the stability of the fault network following the first earthquake (Amatrice, Mw 6.0), with their migration nucleating the Visso (Mw 5.9) and Norcia (Mw 6.5) mainshocks. However, both spatial extent and mechanisms of fluid migration and diffusion through the network remain unclear. High fluid content, enhanced permeability, and pervasive microcracking increase seismic attenuation, but different processes contribute to different attenuation mechanisms. Here, we measured and mapped peak delay time and coda attenuation, using them as proxies of seismic scattering and absorption before and during the sequence. We observed that the structural discontinuities and lithology control the scattering losses at all frequencies, with the highest scattering delineating carbonate formations within the Gran Sasso massif. The Monti Sibillini thrust marks the strongest contrasts in scattering, indicating a barrier for northward fracture propagation. Absorption does not show any sensitivity to the presence of these main geological structures. Before the sequence, low-frequency high-absorption anomalies distribute around the NW-SE-oriented Apennine Mountain chain. During the sequence, a high-absorption anomaly develops from SSE to NNW across the seismogenic zone but remains bounded north by the Monti Sibillini thrust. We attribute this spatial expansion to the deep migration of CO2-bearing fluids across the strike of the fault network from a deep source of trapped CO2 close to the Amatrice earthquake. Fluids expand SSE-NNW primarily during the Visso sequence and then diffuse across the fault zones during the Norcia sequence.

Coda wave attenuation imaging is able to detect fluid/melt accumulation and ancient magmatic bodies in volcanoes. Here we use recently developed space-weighting sensitivity functions to invert for the spatial distributions of multifrequency coda wave attenuation (Q−1 c ), measured during the largest monitored unrest at Campi Flegrei caldera (1983–1984). High-attenuation anomalies are spatially correlated with the regions of highest structural complexities and cross faulting. They characterize deep fluid circulation in and around the aseismic roots of the 1534 A.D. Mount Nuovo eruption and fluid accumulation in the areas of highest hydrothermal hazard. Just offshore Pozzuoli, and at the highest frequency (wavelengths of ∼150 m), the main cause of ground deformation and seismicity during the unrest is an aseismic low-attenuation circular anomaly, similar in shape and nature to those produced by ancient magmatic reservoirs and active sills at other volcanoes.

Ambient noise polarizes inside fault zones, yet the spatial and temporal resolution of polarized noise on gas-bearing fluids migrating through stressed volcanic systems is unknown. Here we show that high polarization marks a transfer structure connecting the deforming centre of the caldera to open hydrothermal vents and extensional caldera-bounding faults during periods of low seismic release at Campi Flegrei caldera (Southern Italy). Fluids pressurize the Campi Flegrei hydrothermal system, migrate, and increase stress before earthquakes. The loss of polarization (depolarization) of the transfer and extensional structures maps pressurized fluids, detecting fluid migrations after seismic sequences. After recent intense seismicity (December 2019-April 2020), the transfer structure appears sealed while fluids stored in the east caldera have moved further east. Our findings show that depolarized noise has the potential to monitor fluid migrations and earthquakes at stressed volcanoes quasi-instantaneously and with minimum processing.

Passive high‐resolution attenuation tomography is used here to image the geological structure in the first upper 4 km of the shallow crust beneath the Campi Flegrei caldera, southern Italy. The inverse Q was estimated for each source‐receiver path using the coda‐normalization method (S‐waves) and the slope decay method (P‐waves and S‐waves). Inversion was performed using a multi‐resolution method, which ensures a minimum cell‐size resolution of 500 m. The study of the resolution matrix as well as the synthetic tests guarantee an optimal reproduction of the input anomalies in the center of the caldera, between 0 and 3.5 km in depth. High attenuation vertical structures are connected at the surface with the main volcanological features (e.g., the Solfatara and Mofete fumarole fields), and depict vertical Q contrast imaging important geological structures, such as the La Starza fault. These high attenuation volumes extend between the surface and a depth of about 3 km, where a hard rock layer is imaged by the sharp contrast of the quality factors. The retrieved image of the Campi Flegrei has been jointly interpreted taking into account evidence from seismological, geological, volcanological and geochemical investigations. This analysis has allowed an unprecedented view of the feeding systems in this area, and in particular it recognizes the vertically extending, high attenuation structures that correspond to gas or fluid reservoirs beneath Pozzuoli‐Solfatara, Solfatara, Mofete‐Mt. Nuovo and Agnano. This high‐attenuation system is possibly connected with the magma sill revealed at about 7 km in depth by passive travel‐time tomography.

The quantification of the seismic energy of earthquakes occurring in volcanic regions is of great importance in order to better understand the dynamics of the volcano. The amount of the released energy and its variation during seismic crises can be considered as an indicator of the source processes acting inside the volcano. In this context, the effect of the propagation in attenuative media should be considered to correct for path effects and to properly estimate the seismic energy released at the source. Moreover, in order to allow a comparison with the dynamic processes occurring in different volcanic areas, the use of magnitude scales as homogeneous as possible is strongly recommended, In this framework, new duration-based Local (Ml) and Moment (Mw) magnitude scales are obtained for the Campi Flegrei area (southern Italy), by analysing a data-set of local volcano-tectonic earthquakes. First the S-wave quality factor for the investigated area was experimentally calculated and then the distance-correction curve, logA0(r), to be used in the Richter formula Ml = logAmax − logA0(r), was numerically estimated by measuring the attenuation properties and hence propagating a synthetic S-wave-packet in the earth medium. The Local magnitude scale was normalized in order to fit the Richter formula valid for Southern California at a distance of 10 km. Ml magnitude was estimated by synthesizing Wood-Anderson seismograms and measuring the maximum amplitude. For the same data-set, Moment magnitude from S-wave distance and site corrected displacement spectra was obtained. Comparisons between Local and Moment magnitudes determined in the present paper, and the old Duration magnitude (Md) routinely used at the Istituto Nazionale di Geofisica e Vulcanologia - Osservatorio Vesuviano are presented. Moreover, relationships between Ml and Mw calculated for two reference sites are also derived.