[1] Seismic data recorded in the 2-30 s band at Stromboli Volcano, Italy, are analyzed to quantify the source mechanisms of Strombolian explosions during September 1997. To determine the source-centroid location and source mechanism, we minimize the residual error between data and synthetics calculated by the finite difference method for a point source embedded in a homogeneous elastic medium that takes topography into account. Two source centroids are identified, each representative of the distinct event types associated with explosive eruptions from two different vents. The observed waveforms are well reproduced by our inversion, and the two source centroids that best fit the data are offset 220 and 260 m beneath and $160 m northwest of the active vents. The source mechanisms include both moment-tensor and single-force components. The principal axes of the moment tensor have amplitude ratios 1:1:2, which can be interpreted as representative of a crack, if one assumes the rock matrix at the source to have a Poisson ratio n = 1/3, a value appropriate for hot rock. Both imaged cracks dip $60°to the northwest and strike northeast-southwest along a direction parallel to the elongation of the volcanic edifice and a prominent zone of structural weakness, as expressed by lineaments, dikes, and brittle structures. For our data set, the volume changes estimated from the moments are $200 m 3 for the largest explosion from each vent. Together with the volumetric source is a dominantly vertical force with a magnitude of 10 8 N, consistent with the inferred movement of the magma column perched above the source centroid in response to the piston-like rise of a slug of gas in the conduit.

Abstract. The wave fields generated by Strombolian activity are investigated using data from small-aperture seismic arrays deployed on the north flank of Stromboli and data from seismic and pressure transducers set up near the summit crater. Measurements of slowness and azimuth as a function of time clearly indicate that the sources of tremor and explosions are located beneath the summit crater at depths shallower than 200 rn with occasional bursts of energy originating from sources extending to a depth of 3 km. Slowness, azimuth, and particle motion measurements reveal a complex composition of body and surface waves associated with topography, structure, and source properties. Body waves originating at depths shallower than 200 rn dominate the wave field at frequencies of 0.5-2.5 Hz, and surface waves generated by the surficial part of the source and by scattering sources distributed around the island dominate at frequencies above 2.5 Hz. The records of tremor and explosions are both dominated by SH motion. Far-field records from explosions start with radial motion, and near-field records from those events show dominantly horizontal motion and often start with a low-frequency (1-2 Hz) precursor characterized by elliptical particle motion, followed within a few seconds by a highfrequency radial phase (1-10 Hz) accompanying the eruption of pyroclastics. The dominant component of the near-and far-field particle motions from explosions, and the timing of air and body wave phases observed in the near field, are consistent with a gaspiston mechanism operating on a shallow (<200 rn deep), vertical crack-like conduit. Models of a degassing fluid column suggest that noise emissions originating in the collective oscillations of bubbles ascending in the magma conduit may provide an adequate self-excitation mechanism for sustained tremor generation at Stromboli.

No abstract

Recent accelerometric recordings of earthquakes with moderate and intermediate magnitude (4.0 Ͻ M L Ͻ 5.9), at both local and regional distances, show a significant ground-motion amplification effect at low frequencies (0.6 Hz) in the city of L'Aquila (central Italy). The effect involves very long durations characterized by low frequencies in the coda. Starting from these observations, a series of supplementary investigations was performed in the urban area of L'Aquila by collecting and analyzing both weakmotion data from earthquakes with magnitudes ranging from 2.2 to 4.9 at distances from 20 to 105 km and ambient noise data. All the collected weak-motion data share the same characteristics as the strongmotion records and give a better image of the amplification effect in the city. In order to interpret observations in terms of the local geology, we performed 2D numerical modeling of the sedimentary basin underlying the city of L'Aquila using both finite elements and boundary elements based on a geological section derived from gravity measurements. This analysis indicates that the ground-motion amplification in the city of L'Aquila is related to the presence of a sedimentary basin, filled by lacustrine sediments, with a maximum depth of about 250 m. The combined approach to data collection and analysis used here gives useful information for risk assessment in the city of L'Aquila and can be recommended for many other urban areas that share similar characteristics.

No abstract

A small size foreshock and the two mainshocks of the Umbria Marche earthquake sequence which occurred on September 26, 1997 have been recorded by two digital 3C accelerometers located at near source distances. The close epicentral distance and azimuthal location relative to the fault orientation and geometry make these records relevant to look at the detail of the rupture kinematics. S‐wave polarizations, apparent source time duration and waveforms from strong motion records are used to constrain the location of the fracture origin point, the fault geometry, the final slip distribution, size and mechanism of the events. The final model shows that the seismic ruptures occurred along two adjacent, sub‐parallel, low angle dipping normal faults. The relative timing, location and geometry of the mainshock faults suggest the presence of a transfer zone (barrier) which has probably controlled the amplitude increase of local stress released by the first rupture at its NW edge which triggered about 9 hours later the second rupture.

The normal-faulting earthquake of 6 April 2009 in the Abruzzo Region of central Italy caused heavy losses of life and substantial damage to centuries-old buildings of significant cultural importance and to modern reinforced-concrete-framed buildings with hollow masonry infill walls. Although structural deficiencies were significant and widespread, the study of the characteristics of strong motion data from the heavily affected area indicated that the short duration of strong shaking may have spared many more damaged buildings from collapsing. It is recognized that, with this caveat of short-duration shaking, the infill walls may have played a very important role in preventing further deterioration or collapse of many buildings. It is concluded that better new or retrofit construction practices that include reinforced-concrete shear walls may prove helpful in reducing risks in such seismic areas of Italy, other Mediterranean countries, and even in United States, where there are large inventories of deficient structures.

After the 2009 April 6th Mw 6.3 L'Aquila earthquake (Central Italy) the Italian Civil Defense Department promoted the microzoning study in the ten zones in the epicentral area that suffered major damage. In this paper we present the activities and the results concerning a temporary seismic network installed in the historical L'Aquila city center indicated as "macroarea 1" in the microzoning project. Seismic data were collected to investigate the amplification effects in the city and to support the microzoning activities in verifying both geological profiles and 1D numerical modeling of the seismic response of the city. The conventional spectral approaches using both microtremor and earthquake data allowed to determine the fundamental resonance frequencies and the amplification factors within the city respectively. The spatial variability of these quantities can be related to the geological and geomorphologic characteristics of the investigated area. A comparison between the network data and the data recorded by the two strong motion instruments installed in the city was also made. This allows verifying the relative response of the accelerometric stations that recorded in the city the major events of the sequence.