ALBERT

Description: Volcanic edifice construction at the Earth's surface significantly modifies the stress field within the underlying crust with two main implications for caldera formation. First, tensile rupture at the Earth's surface is favored at the periphery, which enables ring fault formation. Second, edifice formation amplifies the amount of pressure decrease occurring within a magma reservoir before the eruption stops. Taking into account both of these effects, caldera formation can be initiated during a central eruption of a pre-existing volcano even when assuming elastic behaviour for the surrounding crust. Providing the roof aspect ratio is small enough, conditions for caldera formation by reservoir withdrawal can be reached whatever the reservoir shape is. However ring fault initiation is easier for laterally elongated reservoirs.

Description: At mature andesitic volcanoes, magma can reach the surface through the same path for several eruptions thus forming a volcanic conduit. Because of degassing, cooling, and crystallization, magma viscosity increase in the upper part of the conduit may induce the formation of a viscous plug. We conducted numerical simulations to quantify the deformation field caused by this plug emplacement and evolution. Stress continuity between Newtonian magma flow and elastic crust is considered. Plug emplacement causes a ground inflation correlated to a decrease of the magma discharge rate. A parametric study shows that surface displacements depend on three dimensionless numbers: the conduit aspect ratio (radius/length), the length ratio between the plug and the conduit, and the viscosity contrast between the plug and the magma column. Larger displacements are obtained for high-viscosity plugs emplaced in large aspect ratio conduits. We find that only tiltmeters or GPS located close to the vent (a few hundred meters) might record the plug emplacement. At immediate proximity of the vent, plug emplacement might even dominate the deformation signal over dome growth or magma reservoir pressurization effects. For given plug thicknesses and viscosity profiles, our model explains well the amplitude of tilt variations (from 1 to 25 μrad) measured at Montserrat and Mt. St. Helens. We also demonstrate that at Montserrat, even if most of the tilt signal is due to shear stress induced by magma flow, pressurization beneath the plug accounts for 20% of the signal.

Description: The Tungurahua volcano, in Ecuador, has been experiencing a substantial activity period since 1999, with several eruptions, including those of 2006 and 2008. We use a persistent scatterers approach to analyze a time series of ENVISAT SAR data over theperiod 2003–2009, to investigate surface deformation in the region of the volcano. We measure a continuous large scale uplift with a maximum line-of-sight displacement rate of about 8 mm/yr, which is the first evidence of a sustained inflation in the Andes for an active volcano encompassing several eruptions. We model this signal as magma emplacement in a permanent storage zone at 11.5 km below sea level, with a net inflow rate of 7 million m 3 / yr . The paroxysmal eruptions in 2006 and 2008, did not seem to disrupt this long-term signal. However, we observe significant deformation during the 2006 eruption consistent with an additional intrusion of 4.5 million m 3 of magma.

Description: Abstract Magma propagation is an unsteady process controlled by magma‐crust interaction. To provide information on its dynamics, we invert complementary ground deformation data spanning the 8 hr preceding the 26 May 2016 eruption at Piton de la Fournaise (PdF) volcano (La Réunion, France). Data are inverted using 3‐D boundary element models combined with a Monte Carlo inversion method. The final geometry of the displacement source is determined based on four interferograms spanning the whole propagation phase while the dynamics of the propagation is inferred from temporal inversion of continuous Global Navigation Satellite System (GNSS) data, using the final geometry as an a priori to constrain the source. The best modeled magma path consists in a 2,700‐m‐long sill located 800 m above sea level and connected to the eruptive fissure by a subvertical dike. The quick opening of the horizontal part of the intrusion could have been favored by limited flank sliding during the early stage of propagation. The intrusion then stalled for ∼5 hr, while pressure increased slightly, until final upward propagation and eruption. Volume budget suggests that the eruption was fed by a single batch of magma quickly disconnected from its source. The delay prior to the eruption may reflect a limited magma supply. Finally, two mechanisms, potentially acting together, might have favored the eruption: a driving role of magmatic gas and/or, as often observed at Piton de la Fournaise, an eastward flank slip.

Description: We developed an hybrid numerical model of dike propagation in two dimensions solving both for the magma trajectory and velocity as a function of the source overpressure, the magma physical properties (density and viscosity) as well as the crustal density and stress field. This model is used to characterize the influence of surface load changes on magma migration towards the surface. We confirm that surface loading induced by volcanic edifice construction, tends both to attract the magma and to reduce its velocity. In contrast, surface unloading, for instance due to caldera formation, tends to divert the magma to the periphery retarding eruption. In both cases the deflected magma may remain trapped at depth. Amplitudes of dike deflection and magma velocity variation depend on the ratio between the magma driving pressure (source overpressure as well as buoyancy) and the stress field perturbation. Our model is then applied to the July 2001 eruption of Etna, where the final dike deflection had been previously interpreted as due to the topographic load. We show that the velocity decrease observed during the last stage of the propagation can also be attributed to the local stress field. We use the dike propagation duration to estimate the magma overpressure at the dike bottom to be less than 4 MPa. This approach can be potentially used to forecast if, where and when propagating magma might reach the surface when having knowledge on the local stress field, magma physical properties and reservoir overpressure.

Description: Surges are common at all the major ice caps in Iceland. Ice masses of gigatons may shift from the upper part of the outlet glacier towards the terminus in a few months, advancing the glacier front by up to several kilometres. The advancing ice front may be up to 100 m thick, increasing the load on crustal rocks correspondingly. We use the observed change in crustal loading during a surge of the western part of the Vatnajökull ice cap, Iceland, during 1993–1995 and the corresponding elastic crustal deformation, surveyed with interferometric synthetic aperture radar, to investigate the material properties of the solid Earth in this region. Crustal subsidence due to the surge reaches ~75 mm at the edge of the Síðujökull outlet glacier. This signal is mixed with a broad uplift signal of ~12 mm yr –1 , relative to our reference area, caused by the ongoing retreat of Vatnajökull in response to climate change. We disentangle the two signals by linear inversion. Finite element modelling is used to investigate the elastic Earth response of the surge, as well as to confirm that no significant viscoelastic deformation occurred as a consequence of the surge. The modelling leads to estimates of the Young's modulus and Poisson's ratio of the underlying Earth. Comparison between the observed and modelled deformation fields is made using a Bayesian approach that yields the estimate of a probability distribution for each of the free parameters. Residuals indicate a good agreement between models and observations. One-layer elastic models result in a Young's modulus of 43.2–49.7 GPa (95 per cent confidence) and Poisson's ratio of 0–0.27, after removal of outliers. Our preferred model, with two elastic layers, provides a better fit to the whole surge signal. This model consists of a 1-km-thick upper layer with an average Young's modulus of 12.9–15.3 GPa and Poisson's ratio of 0.17, overlying a layer with an average Young's modulus of 67.3–81.9 GPa and Poisson's ratio of 0.25.

Description: The coseismic surface displacement field and slip distribution at depth due to the Kashmir earthquake (M w = 7.6, 2005) have been analysed by different authors using subpixel correlation of synthetic aperture radar (SAR) images and optical images, teleseismic analysis, GPS measurements, as well as in situ field measurements. In this paper, first, we use 23 sets of measurement from subpixel correlation of SAR images and differential interferometry to retrieve the 3-D coseismic surface displacement field. The obtained horizontal and vertical components along the fault trace are then compared, respectively, to equivalent measurements obtained from subpixel correlation of two optical ASTER images and in situ field measurements. Second, the coseismic fault geometry parameters and slip distribution at depth are estimated. In addition to the one segment slip model as reported in previous work, a two segments slip model that better fits the surface fault break is proposed. The improvement of the two segments slip model in interpreting the measured displacement field is highlighted through comparison of residuals of both slip models. Taking advantage of differential interferometry measurements that provide precise and continuous information in the far field of the fault, firstly, a wedge thrust according to Bendick et al. to the Northwest of the main rupture built on our two segments model is tested. According to the obtained results, the residual of the two segments main rupture plus wedge thrust model is slightly smaller than the residual of the two segments model to the Northwest of the Balakot–Bagh fault. Secondly, we test the sensitivity of our slip model to the presence of slip along a décollement as evidenced by Jouanne et al. through post-seismic analysis. The results indicate that the estimations of the coseismic displacement field and slip distribution in this paper are not significantly biased by such post-seismic displacement and that most coseismic displacement is located on a ~40° NE-dipping fault, as previously reported.