Dr. Aranguiz-Muñoz, Rafael

The variability in obtaining estimates of tsunami inundation and runup on a near‐real‐time tsunami hazard assessment setting is evaluated. To this end, 19 different source models of the Maule Earthquake were considered as if they represented the best available knowledge an early tsunami warning system could consider. Results show that large variability can be observed in both coseismic deformation and tsunami variables such as inundated area and maximum runup. This suggests that using single source model solutions might not be appropriate unless categorical thresholds are used. Nevertheless, the tsunami forecast obtained from aggregating all source models is in good agreement with observed quantities, suggesting that the development of seismic source inversion techniques in a Bayesian framework or generating stochastic finite fault models from a reference inversion solution could be a viable way of dealing with epistemic uncertainties in the framework of nearly‐real‐time tsunami hazard mapping.

On September 16, 2015 a magnitude Mw 8.3 earthquake took place off the coast of the Coquimbo Region, Chile. Three tsunami survey teams covered approximately 700 km of the Pacific coast. The teams surveyed the area, recording 83 tsunami flow depth and runup measurements. The maximum runup was found to be 10.8 m at only one small bay, in front of the inferred tsunami source area. However, it was observed that runup in other locations rarely exceed 6 m. Tsunami runup was larger than those of the 2014 Pisagua event, despite the similar earthquake magnitude. Moreover, tsunami arrival times were found to be shorter than those of previous tsunamis along the Chilean subduction zone. Numerical simulations of the tsunami event showed a good agreement with field data, highlighting that tsunami arrival time and the spatial variation of the tsunami amplitudes were strongly influenced by the bathymetry, coastal morphology and the slip distribution of the causative earthquake.

We constructed a seismic source model for the 2015 MW 8.3 Illapel, Chile earthquake, which was carried out with the kinematic waveform inversion method adopting a novel inversion formulation that takes into account the uncertainty in the Green’s function, together with the hybrid backprojection method enabling us to track the spatiotemporal distribution of high-frequency (0.3–2.0 Hz) sources at high resolution by using globally observed teleseismic P-waveforms. A maximum slip amounted to 10.4 m in the shallow part of the seismic source region centered 72 km northwest of the epicenter and generated a following tsunami inundated along the coast. In a gross sense, the rupture front propagated almost unilaterally to northward from the hypocenter at \2 km/s, however, in detail the spatiotemporal slip distribution also showed a complex rupture propagation pattern: two up-dip rupture propagation episodes, and a secondary rupture episode may have been triggered by the strong high-frequency radiation event at the down-dip edge of the seismic source region. High-frequency sources tends to be distributed at deeper parts of the slip area, a pattern also documented in other subduction zone megathrust earthquakes that may reflect the heterogeneous distribution of fracture energy or stress drop along the fault. The weak excitation of high-frequency radiation at the termination of rupture may represent the gradual deceleration of rupture velocity at the transition zone of frictional property or stress state between the megathrust rupture zone and the swarm area.

On 1 April 2014, an earthquake with moment magnitudeMw8.2 occurred off the coast ofnorthern Chile, generating a tsunami that prompted evacuation along the Chilean coast. Here tsunamicharacteristics are analyzed through a combination of field data and numerical modeling. Despite theearthquake magnitude, the tsunami was moderate, with a relatively uniform distribution of runup, whichpeaked at 4.6 m. This is explained by a concentrated maximal slip at intermediate depth on the megathrust,resulting in a rapid decay of tsunami energy. The tsunami temporal evolution varied, with locations showingsustained tsunami energy, while others showed increased tsunami energy at different times after theearthquake. These are the result of the interaction of long period standing oscillations and trapped edgewave activity controlled by inner shelf slopes. Understanding these processes is relevant for the region,which still posses a significant tsunamigenic potential

The southern Peru and northern Chile coastal region is an active subduction zone that contains one of the most signifcant seismic gaps in the eastern Pacifc basin (~17°S–~24°S). Although the gap was partially flled by the 2014 Mw 8.1 Iquique earthquake, there is still a high seismogenic potential to release a Mw ~9 earthquake in the near future; therefore, all the near-feld coastal cities in the region face a latent tsunami threat. In this article, we propose a hybrid deterministic–stochastic multi-scenario approach to assess the current tsunami hazard level in the city of Iquique, an important commercial and industrial center of northern Chile that is home to 184,000 inhabitants. In our approach, we defned 400 stochastic, 10 deterministic and 10 homogeneous tsunamigenic earthquake scenarios, covering the entire area of the seismic gap. Based on the regional distribution of gravity anomalies and published interseismic coupling distributions, we interpreted the occurrence of four major asperities in the subduction interface of the seismic gap. The asperity pattern was used to construct a group of deterministic slip-defcit earthquake sources with seismic magnitudes ranging between Mw 8.4 and Mw 8.9. Additionally, we constructed 10 homogeneous slip scenarios to generate an inundation baseline for the tsunami hazard. Subsequently, following a stochastic scheme, we implemented a Karhunen–Loève expansion to generate 400 stochastic earthquake scenarios within the same magnitude range as the deterministic slip-defcit sources. All sources were used as earthquake scenarios to simulate the tsunami propagation and inundation by means of a non-hydrostatic model (Neowave 2D) with a classical nesting scheme for the city of Iquique. We obtained high-resolution data for fow depth, coastal surface currents and sea level elevation. The results suggest that the peak slip location and shelf resonance play an important role in the calculated coastal fow depths. The analysis of the entire set of simulated stochastic earthquake scenarios indicates that the worst-case scenario for Iquique is a Mw 8.9 earthquake. This scenario presented a tsunami arrival time of ~12 min, which is critical for the evacuation process. In addition, the maximum wave height and tsunami fow depth were found to be ~10 m and ~24 m, respectively. The observed coastal resonance processes exhibit at least three destructive tsunami wave trains. Based on historical and instrumental catalog statistics, the recurrence time of the credible worst-case earthquake scenario for Iquique (Mw 8.9) is 395 years, with a probability of occurrence of ~11.86% in the next 50 years.