Evaluación del Peligro Sísmico Para Movimientos Extremos del Terreno en Escenarios de Fuente Cercana.
Loading...
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
Este estudio aborda la reproducción y validación de simulaciones de movimientos fuertes del terreno (strong-motions) de banda ancha para grandes terremotos de megathrust (Mw > 7.5) con el fin de mejorar sus predicciones en las regiones cercanas a la fuente sísmica para un amplio rango de frecuencias (0.01 < f < 30 Hz). Se implementó una Simulación Numérica Basada en la Física (PBSs) mediante un enfoque híbrido que combina métodos cinemático y estocástico. Este modelo fue aplicado al Terremoto de Iquique de 2014 (8.2 MW), incorporando parámetros de modelos de slip co-sísmico heterogéneo, configuraciones de velocidad de la corteza y geometría de una zona de subducción. Las bajas frecuencias (f < 1 Hz) se modelaron utilizando el Método de los Elementos Finitos (3D-FEM) con un mallado en 3D detallado, simulando la ruptura de la fuente sismogénica de manera cinemática, además, se exploraron correlaciones de escalado para el tiempo duración del slip o rotura (rise time, Tr), un parámetro temporal crítico en la técnica cinemática, basándose en 45 Modelos de Ruptura de Falla Finita (FFRMs) de grandes terremotos de megathrust (Mw ≥ 7.3), estableciéndose una relación de autosimilitud para el Tr utilizando registros de strong-motions y de datos telesísmicos. Para las altas frecuencias (f > 1 Hz), se utilizó una metodología estocástica que incluye FFRMs, evaluaciones de atenuación y respuesta del sitio mediante relaciones espectrales horizontal/vertical (HVSR) para caracterizar la amplificación local. Los resultados de las historias de strong-motions híbridos (PBSs) destacan la importancia de integrar la historia de ruptura de la fuente, las condiciones reológicas y las propiedades de atenuación del terreno para reducir las incertidumbres en las evaluaciones de amenaza sísmica. El estudio muestra reducciones en los valores de la desviación estándar de los residuales intra-eventos: hasta un 10.23% en distancias cortas entre fuente-sitio (rupture-to-station), primeros ~30 a 70 km, y hasta un 53.9% en el rango de 2 a 100 Hz. Además, se observó una disminución del 50% en valores de diseño para un periodo de retorno de T = 475 años, al comparar las curvas de amenaza sísmica de datos PBSs y ergódicos. La validación del modelo híbrido aplicado al terremoto de Iquique resalta los parámetros críticos que deben considerarse para predicciones más confiables y con menor incertidumbre en futuros terremotos.
This study addresses reproducing and validating broadband strong-motions simulations for large-magnitude megathrust earthquakes (Mw > 7.5) to improve their predictions in the near field region for a broad frequency range (0.01 < f < 30 Hz). Physics-Based Numerical Simulations (PBSs), which encompasses a hybrid approach combining kinematic and stochastic methods. This model was applied to the 2014 Mw 8.2 Iquique Earthquake, incorporating parameters from heterogeneous co-seismic slip models, crustal velocity configurations and geometry of a subduction zone. Low frequencies (f < 1 Hz) were modeled using the Finite Element Method (3D-FEM) with a detailed 3D meshing, simulating the kinematic seismogenic source rupture, also, scaling correlations were explored for the rise time (Tr), a critical time parameter in the kinematic technique, based on 45 Finite Fault Rupture Models (FFRMs) of large megathrust earthquakes (Mw ≥ 7. 3), establishing a self-similarity relationship for Tr using records of strong-motions and teleseismic data. For high frequencies (f > 1 Hz), a stochastic methodology was used, which encompasses FFRMs, attenuation and site response assessments through horizontal/vertical spectral ratios (HVSRs) to characterize local amplification. Results from hybrid strong-motion histories (PBSs) highlight the importance of incorporate seismic source rupture history, rheological conditions, and ground attenuation properties to reduce the uncertainty in seismic hazard analysis. The study shows reductions in the standard deviation values of within-event residuals: to 10.23% in close rupture-to-station distances, first ~30 to 70 km, and up to 53.9% in the range of 2 to 100 Hz. In addition, a 50% decrease in design values for a return period of T = 475 years was observed when comparing the seismic hazard curves of PBSs and ergodic data. The validation of the hybrid model applied to the Iquique earthquake highlights the critical parameters that should be considered for more reliable predictions with less uncertainty in future earthquakes.
This study addresses reproducing and validating broadband strong-motions simulations for large-magnitude megathrust earthquakes (Mw > 7.5) to improve their predictions in the near field region for a broad frequency range (0.01 < f < 30 Hz). Physics-Based Numerical Simulations (PBSs), which encompasses a hybrid approach combining kinematic and stochastic methods. This model was applied to the 2014 Mw 8.2 Iquique Earthquake, incorporating parameters from heterogeneous co-seismic slip models, crustal velocity configurations and geometry of a subduction zone. Low frequencies (f < 1 Hz) were modeled using the Finite Element Method (3D-FEM) with a detailed 3D meshing, simulating the kinematic seismogenic source rupture, also, scaling correlations were explored for the rise time (Tr), a critical time parameter in the kinematic technique, based on 45 Finite Fault Rupture Models (FFRMs) of large megathrust earthquakes (Mw ≥ 7. 3), establishing a self-similarity relationship for Tr using records of strong-motions and teleseismic data. For high frequencies (f > 1 Hz), a stochastic methodology was used, which encompasses FFRMs, attenuation and site response assessments through horizontal/vertical spectral ratios (HVSRs) to characterize local amplification. Results from hybrid strong-motion histories (PBSs) highlight the importance of incorporate seismic source rupture history, rheological conditions, and ground attenuation properties to reduce the uncertainty in seismic hazard analysis. The study shows reductions in the standard deviation values of within-event residuals: to 10.23% in close rupture-to-station distances, first ~30 to 70 km, and up to 53.9% in the range of 2 to 100 Hz. In addition, a 50% decrease in design values for a return period of T = 475 years was observed when comparing the seismic hazard curves of PBSs and ergodic data. The validation of the hybrid model applied to the Iquique earthquake highlights the critical parameters that should be considered for more reliable predictions with less uncertainty in future earthquakes.
Description
Tesis presentada para optar al grado de Doctor en Ciencias Geológicas
Keywords
Ingeniería sísmica, Terremotos, Geología