Análisis de parámetros de movimiento fuerte para sismos que detonaron deslizamientos en Colombia
Cargando...
Autores
Mateus Báez, Leonardo Santos
Director
Tipo de contenido
Document language:
Español
Fecha
Título de la revista
ISSN de la revista
Título del volumen
Documentos PDF
Resumen
Al ser los deslizamientos uno de los efectos mas significativos desencadenados por un sismo, el presente trabajo busca conocer la distribución de los deslizamientos inducidos por sismos (DIS) en Colombia y su relación con las características del terreno y de la onda sísmica. De esta forma, en este estudio, se complementa el inventario existente de DIS, compilado por Rodríguez (2006) al compilar un total de 177 deslizamientos desencadenados por 56 sismos históricos, que comprenden un periodo entre 1534 y 2019. Asimismo fueron estudiados y analizados 18 parámetros de movimiento fuerte (PMF) para las señales de aceleración obtenidas en tres casos de estudio: Sismos de Armenia (1999), Quetame (2008) y Mesetas (2019). Además, con el objetivo de relacionar los PMF con las características del suelo (resistencia, cobertura vegetal, topografía, entre otros) se aplicaron dos modelos de predicción de DIS: desplazamientos de Newmark (Jibson et al., 2000) y predicción logística (Nowicki Jessee et al., 2018). Como resultado de lo anterior se comparan los resultados del modelo de desplazamiento de Newmark con el de predicción logíıstica en donde se observa coherencia entre ambos métodos. Además, los desplazamientos de Newmark mayores a 10 cm coinciden con probabilidades de ocurrencia mayores a 0.85. Finalmente, se compara el inventario de DIS del sismo de Mesetas con la predicción logística y se obtienen umbrales de probabilidad para una determinación preliminar de amenaza.
Earthquake-triggered landslides pose an enormous hazard in seismically active regions as is the Colombian territory. This study aims to understand the relationship between strong motion parameters, characteristics of the ground and the seismic wave and the distribution of historically Earthquake-triggered landslides in Colombia. A database of earthquake-induced landslides has been compiled which extends the work of Rodriguez (2006), including a total of 177 landslides triggered by 56 earthquakes, which cover the period 1530 to 2019. The acceleration records of these three different seismic events were studied through the analyzes of 18 strong motion parameters: Armenia (1999), Quetame (2008) and Mesetas (2019) earthquakes. Two prediction models of Earthquake-triggered landslides were applied to relate the strong motion parameters with the ground characteristics (strength, ground shaking, topographic slope, lithology, land cover type, a topographic index, among others): Newmark Displacement (Jibson et al, 2000) and A logistic regression (Nowicki Jessee et al., 2018). By comparing these two models it is observed that there is good correlation in the results, where a Newmark displacement bigger than 10 cm coincides with occurrence probabilities bigger than 0.85. Finally, a detailed analysis of the correlation between a comprehensive Earthquake-triggered landslide inventory of Mesetas (Meta) with a logistic regression were carried, providing an preliminary Hazard assessment tool.
Earthquake-triggered landslides pose an enormous hazard in seismically active regions as is the Colombian territory. This study aims to understand the relationship between strong motion parameters, characteristics of the ground and the seismic wave and the distribution of historically Earthquake-triggered landslides in Colombia. A database of earthquake-induced landslides has been compiled which extends the work of Rodriguez (2006), including a total of 177 landslides triggered by 56 earthquakes, which cover the period 1530 to 2019. The acceleration records of these three different seismic events were studied through the analyzes of 18 strong motion parameters: Armenia (1999), Quetame (2008) and Mesetas (2019) earthquakes. Two prediction models of Earthquake-triggered landslides were applied to relate the strong motion parameters with the ground characteristics (strength, ground shaking, topographic slope, lithology, land cover type, a topographic index, among others): Newmark Displacement (Jibson et al, 2000) and A logistic regression (Nowicki Jessee et al., 2018). By comparing these two models it is observed that there is good correlation in the results, where a Newmark displacement bigger than 10 cm coincides with occurrence probabilities bigger than 0.85. Finally, a detailed analysis of the correlation between a comprehensive Earthquake-triggered landslide inventory of Mesetas (Meta) with a logistic regression were carried, providing an preliminary Hazard assessment tool.