ALBERT

Description: We develop a model for phase separation in magma reservoirs containing a mixture of silicate melt, crystals and fluids (exsolved volatiles). The interplay between the three phases control the dynamics of phase separation and consequently the chemical and physical evolution of magma reservoirs. The model we propose is based on the 2-phase damage theory approach of Bercovici et al. [2001]; Bercovici and Ricard [2003] because it offers the leverage of considering interface (in the macroscopic limit) between phases that can deform depending on the mechanical work and phase changes taking place locally in the magma. Damage models also offer the advantage that pressure is defined uniquely to each phase and does not need to be equal among phases, which will enable us to consider, in future studies, the large capillary pressure at which fluids are mobilized in mature, crystal-rich, magma bodies. In this first analysis of three-phase compaction, we solve the 3-phase compaction equations numerically for a simple 1-D problem where we focus on the effect of fluids on the efficiency of melt-crystal separation considering the competition between viscous and buoyancy stresses only. We contrast three sets of simulations to explore the behavior of 3-phase compaction, a melt-crystal reference compaction scenario (2-phase compaction), a 3-phase scenario without phase changes and finally a 3-phase scenario with a parameterized second boiling (crystallization-induced exsolution). The simulations show a dramatic difference between two (melt-crystals) and three (melt-crystals-exsolved volatiles) compaction-driven phase separation. We find that the presence of a lighter, significantly less viscous fluid hinders melt-crystal separation.

Description: We study the October 30 th 2016 Norcia earthquake (M W 6.5) to retrieve the rupture history by jointly inverting seismograms and coseismic GPS displacements obtained by dense local networks. The adopted fault geometry consists of a main normal fault striking N155°and dipping 47° belonging to the Mt. Vettore-Mt. Bove fault system (VBFS) and a secondary fault plane striking N210° and dipping 36° to the NW. The coseismic rupture initiated on the VBFS and propagated with similar rupture velocity on both fault planes. Up-dip from the nucleation point, two main slip patches have been imaged on these fault segments, both characterized by similar peak-slip values (~3 m) and rupture times (~3 s). After the breakage of the two main slip patches, coseismic rupture further propagated southeastward along the VBFS, rupturing again the same fault portion that slipped during the August 24 th earthquake. The retrieved coseismic slip distribution is consistent with the observed surface breakages and the deformation pattern inferred from InSAR measurements. Our results show that three different fault systems were activated during the October 30 th earthquake. The composite rupture model inferred in this study provides evidences that also a deep portion of the NNE-trending section of the Olevano-Antrodoco-Sibillini (OAS) thrust broke co-seismically, implying the kinematic inversion of a thrust ramp. The obtained rupture history indicates that, in this sector of the Apennines, compressional structures inherited from past tectonics can alternatively segment boundaries of NW-trending active normal faults or break co-seismically during moderate-to-large magnitude earthquakes.

Description: The dynamic fragmentation of jointed rock blocks during rock rockslide-avalanches has been analyzed by discrete element method for a multiple rock block arrangement and simple slope geometry. The rock blocks are released along an inclined sliding plane and subsequently collide onto a flat horizontal plane at a sharp kink point. At impact, the contact force chains appear immediately at the bottom frontal corner of the rock block, and then propagate radially upwards to the top rear region. The jointed rock blocks show evident contact force concentration and discontinuity of force wave propagation near the joint, associating with high energy dissipation of granular dynamics. The corresponding force wave propagation velocity can be less than 200 m/s, which is much smaller than that of an intact rock (1316 m/s). The concentration of contact forces at the bottom leads to high rock fragmentation intensity and momentum boosts, facilitating the spreading of many fine fragments to the distal ends. However, the upper rock has very low rock fragmentation intensity but high energy dissipation due to intensive friction and damping effects, resulting in the deposition of large fragments near the slope toe. The size and shape of large fragments are closely related to the orientation and distribution of initial joints. The cumulative fragment size distribution (FSD) can be well fitted by the Weibull's distribution function, with very gentle and steep curvatures at the fine and coarse size ranges, respectively. The numerical results of FSD can match well some experimental and field observations.

Description: The rapidly increased earthquake rate in the central United States has been linked with wastewater injection. While the overall understanding appears clear at large scales, the interaction between injection and faulting at smaller scales within individual sequences is still not clear. For an earthquake sequence in central Oklahoma, we conduct finer scale analysis of the spatiotemporal evolution of seismicity, and pore pressure modeling. The pore pressure modeling suggests that nearby wells show much stronger correlation with earthquake sequence evolution. Detailed temporal analysis found correlation between earthquake rate, seismic moment and injection rates from wells in close proximity. However, the observed maximum magnitude (Mmax) is about one order of magnitude smaller than expected based on a theoretical relationship between Mmax and cumulative volume. This discrepancy may point toward additional parameters, such as fault size and stress, which influence M m a x . The lower M m a x is consistent with the truncated Gutenberg-Richter distribution observed from matched-filter detected catalog. Overall, the detailed observations suggest that it is possible to resolve relationships between individual disposal wells and induced earthquake sequences.

Description: Characterizing the 700-km-wide system of active faults on the Shan Plateau, southeast of the eastern Himalayan syntaxis (EHS), is critical to understanding the geodynamics and seismic hazard of the large region that straddles neighboring China, Myanmar, Thailand, Laos and Vietnam. Here we evaluate the fault styles and slip rates over multi-timescales, reanalyze previously-published short-term GPS velocities and evaluate slip-rate gradients to interpret the regional kinematics and geodynamics that drive the crustal motion. Relative to the Sunda plate, GPS velocities across the Shan Plateau define a broad arcuate tongue-like crustal motion with a progressively northwestward increase in sinistral shear over a distance of ~700 km followed by a decrease over the final ~100 km to the syntaxis. The cumulative GPS slip rate across the entire sinistral-slip fault system on the Shan Plateau is ~12 mm/yr. Our observations of the fault geometry, slip rates and arcuate southwesterly-directed tongue-like patterns of GPS velocities across the region suggest that the fault kinematics is characterized by a regional southwestward distributed shear across the Shan Plateau, compared to more block-like rotation and indentation north of the Red River fault. The fault geometry, kinematics and regional GPS velocities are difficulty to reconcile with regional bookshelf faulting between the Red River and Sagaing faults or localized lower crustal channel flows beneath this region. The crustal motion and fault kinematics can be driven by a combination of basal traction of a clockwise, southwestward asthenospheric flow around the EHS and gravitation or shear-driven indentation from north of the Shan Plateau.

Description: The acceleration precursor of catastrophic rupture in rock-like materials is usually characterized by a power-law relationship, but the exponent exhibits a considerable scatter in practice. In this paper, based on experiments of granites and marbles under quasi-static uniaxial and unconfined compression, it is shown that the power-law exponent varies between −1 and −1/2. Such a changeable power-law singularity can be justified by the energy criterion and a power function approximation. As the power-law exponent is close to the lowest value of −1, rocks are prone to a perfect catastrophic rupture. Furthermore, it is found that the fitted reduced power-law exponent decreases monotonically in the vicinity of a rupture point and converges to its lower limit. Therefore, the upper bound of catastrophic rupture time is constrained by the lowest value of the exponents and can be estimated in real-time. This implies that, with the increase of real-time sampling data, the predicted upper bound of catastrophic rupture time can be unceasingly improved.

Description: Relatively low rates of seismicity and fault loading have made it challenging to correlate microseismicity to mapped surface faults on the forearc of southern Vancouver Island. Here we use precise relocations of microsciesmicity integrated with existing geologic data to present the first identification of subsurface seismogenic structures associated with the Leech River fault zone (LRFZ) on southern Vancouver Island. We used the HypoDD double difference relocation method to relocate 1253 earthquakes reported by the Canadian National Seismograph Network (CNSN) catalog from 1992 to 2015. Our results reveal an ~8-10 km wide, NNE-dipping zone of seismicity representing a subsurface structure along the eastern 30 km of the terrestrial LRFZ and extending 20 km farther eastward offshore, where the fault bifurcates beneath the Juan de Fuca Strait. Using a clustering analysis, we identify secondary structures within the NNE-dipping fault zone, many of which are sub-vertical and exhibit right-lateral strike-slip focal mechanisms. We suggest that the arrangement of these near-vertical dextral secondary structures within a more general NE-dipping fault zone, located 10-15 km beneath the Leech River fault (LRF) as imaged by LITHOPROBE, may be a consequence of the reactivation of this fault system as a right-lateral structure in crust with a pre-existing NNE-dipping structural fabric. Our results provide the first confirmation of active terrestrial crustal faults on Vancouver Island using a relocation method. We suggest that slowly slipping active crustal faults, especially in regions with pre-existing foliations, may result in microseismicity along fracture arrays rather than along single planar structures.

Description: The two major explosive phases of the 22–23 April 2015 eruption of Calbuco volcano, Chile produced powerful seismicity and infrasound. The eruption was recorded on seismo-acoustic stations out to 1,540 km and on 5 stations (IS02, IS08, IS09, IS27, and IS49) of the International Monitoring System (IMS) infrasound network at distances from 1,525 to 5,122 km. The remote IMS infrasound stations provide an accurate explosion chronology consistent with the regional and local seismo-acoustic data, and with previous studies of lightning and plume observations. We use the IMS network to detect and locate the eruption signals using a brute-force, grid-search, cross-bearings approach. After incorporating azimuth deviation corrections from stratospheric cross-winds using 3D ray-tracing, the estimated source location is 172 km from true. This case study highlights the significant capability of the IMS infrasound network to provide automated detection, characterization, and timing estimates of global explosive volcanic activity. Augmenting the IMS with regional seismo-acoustic networks will dramatically enhance volcanic signal detection, reduce latency, and improve discrimination capability.

Description: : Natural water convection in subvertical fractures, fracture zones, or faults can perturb the temperature field around the fracture and enhance and focus vertical heat flow within. We investigate, by means of numerical simulation, the effects of convection in a deeply buried vertical fracture zone. Fracture zone transmissivity, defined as permeability times thickness of the permeable region, is found to be the primary control on convection style rather than fracture zone thickness or permeability alone. In an impermeable host rock, the convection-induced thermal anomaly propagates solely via conduction, diminishing away from the fracture. Convective heat flow increases with fracture transmissivity up to ca. 10 -8 m 3 , when a plateau in convective heat flow is reached, constrained by fracture size and the host rock's thermal conductivity. Permeable host rocks modify these results significantly. In a moderately permeable host rock (10 -14 m 2 ), convection in the fracture induces non-Rayleigh fluid convection, while thermal effects and fluid exchange between host rock and fracture remain moderate. If the host rock is sufficiently permeable to allow porous medium Rayleigh convection to occur (10 -13 m 2 ), the convection patterns within the fracture are overprinted by the host's convective patterns. Fluid exchange between the fracture and the rock will be significant. Our findings provide insight into how thermal anomalies in the uppermost crust may relate to locally enhanced heat flow from convection in non-outcropping fractures below. Furthermore, the results for permeable host rocks provide evidence for previously inferred hydrologic scenarios for the formation of certain hydrothermal, vein-type mineral deposits.

Description: Stress drops calculated from source spectral studies currently show larger variability than what is implied by empirical ground motion models. One of the potential origins of the inflated variability is the simplified model-fitting techniques used in most source spectral studies. This study examines a variety of model-fitting methods, and shows that the choice of method can explain some of the discrepancy. The preferred method is Bayesian hierarchical modeling, which can reduce bias, better quantify uncertainties, and allow additional effects to be resolved. Two case study earthquakes are examined, the 2016 M W 7.1 Kumamoto, Japan earthquake and a M W 5.3 aftershock of the 2016 M W 7.8 Kaikōura earthquake. By using hierarchical models, the variation of the corner frequency, f c , and the falloff rate, n , across the focal sphere can be retrieved without overfitting the data. Other methods commonly used to calculate corner frequencies may give substantial biases. In particular, if f c was calculated for the Kumamoto earthquake using an ω -square model, the obtained f c could be twice as large as a realistic value.