The Bárdarbunga caldera volcano in central Iceland collapsed from August 2014 to February 2015 during the largest eruption in Europe since 1784. An ice-filled subsidence bowl, 110 square kilometers (km2) in area and up to 65 meters (m) deep developed, while magma drained laterally for 48 km along a subterranean path and erupted as a major lava flow northeast of the volcano. Our data provide un­precedented insight into the workings of a collapsing caldera.
Collapses of caldera volcanoes are, fortunately, not very frequent, because they are often associated with very large volcanic eruptions. On the other hand, the rarity of cal­dera collapses limits insight into this major geo­logical hazard. Since the formation of Katmai caldera in 1912, during the 20fh century's largest eruption, only five caldera collapses are known to have occurred before that at Bárdarbunga. We used aircraft-based altimetry, satellite GPS, evolution of seismicity, radio-echo soundings of ice thickness, ice flow modeling, and geo-barometry to describe and analyze the evolving subsidence geometry, its underlying cause, the amount of magma erupted, the geometry of the subsurface caldera ring faults, and the moment tensor solutions of the collapse-related earthquakes.
RESULTS: After initial lateral withdrawal of magma for some days though a magma-filled fracture propagating through Earth's upper crust, preexisting ring faults under the volcano were reactivated over the period 20 to 24 August, marking the onset of collapse. On 31 August, the eruption started, and it terminated when the collapse stopped, having produced 1.5 km of basaltic lava. The subsidence of the caldera declined with time in a near-exponential man­ner, in phase with the lava flow rate. The volume of the subsidence bowl was about 1.8 km³.
Using radio-echo soundings, we find that the subglacial bedrock surface after the collapse is down-sagged, with no indications of steep fault escarpments. Using geobarometry, we determined the depth of the magma reservoir to be -12 km, and modeling of geodetic observations gives a similar result. High-precision earthquake locations and mo­ment tensor analysis of the remarkable mag­nitude M5 earthquake series are consistent with steeply dipping ring faults. Statistical analysis of seismicity reveals communication over tens of kilometers between the caldera and the dike.

| Science, 353, 6296, | doi:10.1126/science.aaf8988 |

The 2014 April 1, Mw 8.1 Iquique (Chile) earthquake struck in the Northern Chile seismic gap. With a rupture length of less than 200 km, it left unbroken large segments of the former gap. Early studies were able to model the main rupture features but results are ambiguous with respect to the role of aseismic slip and left open questions on the remaining hazard at the Northern Chile gap. A striking Observation of the 2014 earthquake has been its extensive preparation phase, with more than 1300 events with magnitude above ML 3, occurring during the 15 months preceding the main shock. Increasing seismicity rates and observed peak magnitudes accompanied the last three weeks before the main shock. Thanks to the large data sets of regional recordings, we assess the precursor activity, compare foreshocks and aftershocks and model rupture preparation and rupture effects. To tackle inversion challenges for moderate events with an asymmetric network geometry, we use full waveforms techniques to locate events, map the seismicity rate and derive source parameters, obtaining moment tensors for more than 300 events (magnitudes Mw 4.0-8.1) in the period 2013 January 1-2014 April 30. This unique data set of fore- and aftershocks is investigated to distinguish rupture process models and models of strain and stress rotation during an earthquake. Results indicate that the spatial distributions of foreshocks delineated the shallower part of the rupture areas of the main shock and its largest aftershock, well matching the spatial extension of the aftershocks cloud. Most moment tensors correspond to almost pure double couple thrust mechanisms, consistent with the slab orientation. Whereas no significant differences are observed among thrust mechanisms in different areas, nor among thrust foreshocks and aftershocks, the early aftershock sequence is characterized by the presence of normal fault mechanisms, striking parallel to the trench but dipping westward. These events likely occurred in the shallow wedge structure close to the slab interface and are consequence of the increased extensional stress in this region after the largest events. The overall stress inversion result suggests a minor stress rotation after the main shock, but a significant release of the deviatoric stress. The temporal change in the distribution of focal mechanisms can also be explained in terms of the spatial heterogeneity of the stress field: under such interpretation, the potential of a large megathrust earthquake breaking a larger segment offshore Northern Chile remains high.

| Geophys. J. Int., 204, 3, 1766-1780, | doi: 10.1093/gji/ggv544. | pdf |

Many eruptions, especially at basaltic volcanoes, occur when a magma-filled dike is released from the magma chamber and intersects the Earth’s surface. Such dikes may propagate several tens of km laterally away from the volcano, pushed by lateral stress gradients and magma chamber overpressure. But what factors contribute to arrest a propagating dike, and what is their relative contribution? Here we evaluate the factors that contributed arresting the 2000 dike at Miyakejima volcano in Japan. We find that topographic gradients played a large role but the dominant factor was continued slip or creeping of a pre-existing fault system that the dike approached during propagation. Our mechanical models reveals how interaction of magmatic intrusions and faults affects magma propagation. The model can help understanding the arrangement of rift-transform systems on Mid Ocean Ridges.

| Earth Planet. Sci. Lett., 434, 64–74, | doi.org/10.1016/j.epsl.2015.11.024 |

Strike-slip faults may be traced along thousands of kilometers, e.g., the San Andreas Fault (USA) or the North Anatolian Fault (Turkey). A closer look at such continental-scale strike faults reveals localized complexities in fault geometry, associated with fault segmentation, secondary faults and a change of related hazards. The North Anatolian Fault displays such complexities nearby the mega city Istanbul, which is a place where earthquake risks are high, but secondary processes are not well understood. In this paper, long-term persistent scatterer interferometry (PSI) analysis of synthetic aperture radar (SAR) data time series was used to precisely identify the surface deformation pattern associated with the faulting complexity at the prominent bend of the North Anatolian Fault near Istanbul city. We elaborate the relevance of local faulting activity and estimate the fault status (slip rate and locking depth) for the first time using satellite SAR interferometry (InSAR) technology. The studied NW-SE-oriented fault on land is subject to strike-slip movement at a mean slip rate of ~5.0 mm/year and a shallow locking depth of <1.0 km and thought to be directly interacting with the main fault branch, with important implications for tectonic coupling. Our results provide the first geodetic evidence on the segmentation of a major crustal fault with a structural complexity and associated multi-hazards near the inhabited regions of Istanbul, with similarities also to other major strike-slip faults that display changes in fault traces and mechanisms

| Remote Sens. , 8(10), 846; | doi:10.3390/rs8100846 | pdf |

On 16 September 2015, the MW = 8.2 Illapel megathrust earthquake ruptured the Central Chilean margin. Combining inversions of displacement measurements and seismic waveforms with high frequency (HF) teleseismic backprojection, we derive a comprehensive description of the rupture, which also predicts deep ocean tsunami wave heights. We further determine moment tensors and obtain accurate depth estimates for the aftershock sequence. The earthquake nucleated near the coast but then propagated to the north and updip, attaining a peak slip of 5–6 m. In contrast, HF seismic radiation is mostly emitted downdip of the region of intense slip and arrests earlier than the long period rupture, indicating smooth slip along the shallow plate interface in the final phase. A superficially similar earthquake in 1943 with a similar aftershock zone had a much shorter source time function, which matches the duration of HF seismic radiation in the recent event, indicating that the 1943 event lacked the shallow slip.

| Geophys. Res. Letters, 43, 2, pp. 574-583. | DOI: 10.1002/2015GL066963 | pdf |

The onset of the aftershock decay after main shocks is controversial. Physical models predict that the onset time is stress dependent, and catalog analysis shows a clear increase of the c value of the Omori-Utsu law with increasing main shock magnitude. However, earthquake catalogs are known to have variable quality and completeness levels; in particular, they miss events directly after main shocks. Thus, it has been also argued that the delayed onset of recorded aftershock activity triggered by large earthquakes is simply an artifact of the time-varying completeness. Here I utilize a recent approach describing the detection probability of earthquakes as function of the actual earthquake rate. I derive an analytical relation between apparent and true earthquake rate which only depends on the blind time of detection algorithms after the occurrence of an earthquake. This relation is tested and verified for synthetic simulations of Omori-type aftershock sequences. For a comparison, I analyze earthquake sequences occurred in Southern California and Taiwan, finding that the derived analytical decay function consistently explains the empirical aftershock activity in the catalogs. This indicates that the observed scaling of the Omori c value is mainly related to catalog incompleteness and not to any underlying physical process.

|J. Geophys. Res. Solid Earth, 121, Pages 6499–6509 | DOI: 10.1002/2016JB013319 |

Large earthquakes sometimes activate volcanoes both in the near field as well as in the far field. One possible explanation is that shaking may increase the mobility of the volcanic gases stored in magma reservoirs and conduits. Here experimentally and theoretically we investigate how sloshing, the oscillatory motion of fluids contained in a shaking tank, may affect the presence and stability of bubbies and foams, with important implications for magma conduits and reservoirs. We adopt this concept from engineering: severe earthquakes are known to induce sloshing and damage petroleum tanks. Sloshing occurs in a partially filled tank or a fully filled tank with density-stratified fluids. These conditions are met at open summit conduits or at sealed magma reservoirs where a bubbly magma layer overlays a newly injected denser magma layer. We conducted sloshing experiments by shaking a rectangular tank partially filled with liquids, bubbly fluids (foams) and fully filled with density-stratified fluids; i.e., a foam layer overlying a liquid layer. In experiments with foams, we find that foam collapse occurs for oscillations near the resonance frequency of the fluid layer. Low viscosity and large bubble size favor foam collapse during sloshing. In the layered case, the collapsed foam mixes with the underlying liquid layer. Based on scaling considerations, we constrain the conditions for the occurrence of foam collapse in natural magma reservoirs. We find that seismic waves with lower frequencies <1 Hz, usually excited by large earthquakes, can resonate with magma reservoirs whose width is >0.5 m. Strong ground motion >0.1 m/ s_1 can excite sloshing with sufficient amplitude to collapse a magma foam in an open conduit or a foam overlying basaltic magma in a closed magma reservoir. The gas released from the collapsed foam may infiltrate the rock or diffuse through pores, enhancing heat transfer, or may generate a gas slug to cause a magmatic eruption. The overturn in the magma reservoir provides new nucleation sites which may help to prepare a following/delayed eruption. Mt. Fuji erupted 49 days after the large Hoei earthquake (1707) both dacitic and basaltic magmas. The eruption might have been triggered by magma mixing through sloshing.

| doi.org/10.1016/j.jvolgeores.2016.03.010 | pdf |

This study receives international attention and was highlighted in SCIENCE magazine:
| DOI: 10.1126/science.aaf4172 |