Magma, faults, and gravitational loading at Mount Etna: The 2002–2003 eruptive period

Abstract

[1] Mount Etna is characterized by a complex structural setting that influences its evolution. In order to understand how the interaction between dike-forming intrusions and faulting influences the kinematics of the volcanic edifice, we developed a numerical model. It takes account of the topography, the medium heterogeneities, the gravitational loading, and the most active crustal discontinuities. A parameterization of the apparent coefficient of friction, as a function of the depth, has been considered in the range from 0.01 to 0.5. The density values used, Young's modulus, and Poisson's ratio range from about 1700 to 3200 kg/m 3 , 8.5 to 140 GPa, and 0.15 to 0.35, respectively. The model was applied to the 2002-2003 Etna eruption. The resulting deformation pattern was in agreement with the data provided by the continuous GPS stations and the seismological knowledge, obtaining displacements up to 1 m. Some discrepancies between the recorded ground deformation field and the modeled displacements allowed us to make some hypotheses on the volcano-tectonic (e.g., for the Pernicana fault) or regional-tectonic origin (Acireale-S. Alfio fault system) of the crustal discontinuities. Moreover, our results suggested that the Pernicana fault is characterized by a very low apparent coefficient of friction, less than 0.1. Furthermore, we evaluated the contribution of the gravitational loading and found that it implies a variation of about 10% of the overall deformation pattern. Our results clearly showed that the weight of the volcanic edifice acts in opposition to the magmatic intrusions. Conversely, we found that the presence of medium heterogeneities may favor the eastern flank movements toward the SE and the spreading of the summit area, playing a fundamental role in ascending magma. Finally, we investigated the shallow sliding model and found that dike-forming intrusions produce negligible displacements, less than 1 mm, along the subhorizontal detachment surface. Therefore, the gravitational loading during dike-forming intrusions is not able to trigger sliding processes along this plane.

Ground deformations and volcanic processes as imaged by CGPS data at Mt. Etna (Italy) between 2003 and 2008

“…Our modeling of the geodetic velocity field has allowed us to precisely pinpoint an area of clockwise rotation rates (Figure 7), which extends in the eastern flank, bounded approximately to the North by the North‐East Rift and the Pernicana fault system, to the North‐East by the Ripe della Naca fault scarps and to the East by the Timpe fault system. This feature is compatible with the area of clockwise rotation found by Solaro et al [2010] and with left‐lateral components revealed along the Pernicana fault system [ Azzaro , 1997, 2004; Aloisi et al , 2011b]. It is also interesting to note the good agreement between the area affected by higher seismic energy release and higher geodetic shear strain values, mainly during the inflation phases.…”

“…We used the entire CGPS network except the stations ECAN, ECNV and EPMN (Figure 1), which are located far outside the volcano edifice and that show a deformation pattern not directly linked to the volcanic dynamics but to a more broad scale tectonic deformation. Moreover, the lower eastern flank of the volcano is affected by an ESE‐ward motion whose origin is still controversial and, as mentioned in the previous paragraph, is not always directly or simply correlated to the magmatic dynamics (for a summary, see sections 6 and 7 in Aloisi et al [2011b] and reference therein). Therefore, for this modeling aimed at imaging the pressure magmatic sources exclusively, we did not take the GPS stations located on the lower eastern flank of the volcano (ELAC, ELEO, EPOZ and ESAL; Figure 1) into account.…”

Experimental study of the interplay between magmatic rift intrusion and flank instability with application to the 2001 Mount Etna eruption

“…The results of the two sets of experiments during nonintrusive periods (Figures and a, at t = 2 and 14 min, and Figures and b, at t = 18, 24, and 30 min) show no deformation within the sand ridge (e.g., listric fault formation [ Le Corvec and Walter , ]) suggesting that gravitational loading has no influence on the deformation of our model. The deep flat décollement surface being a first‐order approximation is not consistent with nature; however, as previous studies show, the décollement configuration might be more complex [ Borgia , ], possibly dipping eastward [ Palano et al ., ; Puglisi et al ., ; Aloisi et al ., ] or including more than one décollement [ Tibaldi and Groppelli , ]. The observed deformation during the nonintrusive period of our experiments is principally controlled by tectonic faulting, while the décollement surface enables the flank's lateral movement during intrusive periods accordingly to the similar results of Le Corvec and Walter [] and Walter et al .…”

An adjoint-based FEM optimization of coseismic displacements following the 2011 Tohoku earthquake: new insights for the limits of the upper plate rebound

“…As well known in literature (e.g., Cattin et al, 1999;Masterlark 2003;Aloisi et al, 2011;Hsu et al, 2011), the simple analytical model approach (homogeneous, isotropic, Poisson-solid half-space assumption) has in fact strong limitations and introduces significant displacement prediction errors (with respect to measurement uncertainties). In particular, Cattin et al (1999) found that rigidity contrasts, existing within the upper crust can increase the horizontal displacements for a given slip model by up to 40%.…”