Book 3GSEV 2017

THIRD COLLOQUIUM OF
GEOPHYSICAL SIGNATURES IN
EARTHQUAKES AND VOLCANOES
TERCER COLOQUIO DE SEÑALES

GEOFÍSICAS DE TERREMOTOS Y
VOLCANES
ORGANIZAN:
COLABORAN:
20 -22 MARZO 2017

CONCEPCIÓN, CHILE
PROGRAMA
Lunes 20.03 – 3GSEV
17:00 -20:00 Acreditación
Martes 21.03 – 3GSEV
08:30 -09:00 Acreditación

9:00 - 09:20 Ceremonia Apertura
Plenaria Raúl Madariaga
9:20 - 10:15
Sismicidad reciente de Chile, ¿qué hemos aprendido?
10:15 – 10:40 Espacio de Discusión: Desafíos de la Geofísica en Chile
10:40 - 11:00 Coffee Break
Early Warning, Rapid Response and Monitoring Tools
11:00 - 11:20 J.C. Báez Analysis of large GNSS networks: from daily to real-time
Studying the dependence of the tsunami run-up with the source-time
11:20 - 11:40 M. Fuentes
function
Juan Fernandez as a T-Phase detector for South-American
11:40 - 12:00 M. Sáez
earthquakes
12:00 - 12:20 A. Carrillo Slow earthquakes, a concept from another perspective.
12:20 - 14:00 Lunch
Interactions between Earthquakes, Faults and Volcanoes in Subduction zone
Crustal Faults Activity inferred from GPS and its relation with the latest
14:00 - 14:20 F. García
volcanic eruptions
Numerical analysis of the role of local faults on the influence of the
14:20 - 14:40 C. Farías 2015 Illapel and 2016 Quellón earthquakes on the Chilean volcanic arc:
preliminary results
14:40 - 15:00 D. Basualto Terremotos y Reactivación Volcánica, Evidencias y Correlaciones
La desglaciación en los Andes del Sur y su impacto en sistemas
15:00 - 15:20 D. Mora
magmáticos superficiales
Geophysical Signatures of Great Earthquakes
Seismic ionospheric precursors? A study of the great 1960 and 2010
15:40 - 16:00 C. Villalobos
Chile earthquakes
Spatial distribution of coseismic slip for the 2011 Tohoku-Oki
16:00 - 16:20 F. Vera earthquake based on a seismo-geodetic perspective from GPS and W-
phase observations
Deformacion postsismica asociada al Terremoto de Tohoku-Oki
16:20 - 16:40 D. Molina M_w=9.1, interpretado por movimiento de relajacion en las interfases
de la corteza oceanica.
Heterogeneidades de la frontera manto-nucleo, observadas
16:40 - 17:00 K. Bataille
sismicamente en las zonas de sombra de los nucleos.
17:00 - 19:30 Poster Session
Three dimensional conductivity model of the Paniri-Toconce volcanic
R. Mancini
system, Chile
Control on the active crustal deformation and structural geology of
P. Köhler Laguna del Maule Volcanic Complex using seismic activity, bathymetry
and seismic profiles.
R. Madariaga Spectral properties of large recent Chilean earthquakes
FinDer performance using CSN network: a strong-motion based
S. Carrasco
algorithm for Earthquake Early
Comparison between interplate and intraplate intermediate-depth
C. Otárola
earthquakes using dynamic modeling of the seismic source
N. Krumm Identification of Tectonic Signals in GPS positional time series
Seismic tomography in the vicinity of the Arauco peninsula from the
K. Azúa
aftershocks of the Mw 8.8 Maule 2010 earthquake
Cocktail
Miércoles 22.03 – 3GSEV
Plenaria Sergio Barrientos

9:00 – 9:45
Centro Sismológico Nacional
9:45 - 10:40 Espacio de Discusión:
Geophysical Signatures at Villarrica Volcano
The origin and kinematics of lahars generated during the eruption of
11:00 - 11:20 J.L. Palma
Villarrica volcano on March 3, 2015
Feature signal extraction and classification of acoustic events at
11:20 - 11:40 E. Mardones
Villarrica Volcano using infrasound
Analysis of the deformation of Villarrica volcano using GPS and
11:40 - 12:00 G. Cid
analytical models
12:00 - 14:00 Lunch
Engineering Earthquakes: Intensities and Velocity Models

Transient and long-lived seismic velocity structure of Southern Andes
14:00 - 14:20 D. González
within latitudes 34.5°S and 37.5°S from seismic ambient noise
14:20 - 14:40 B. Potin Tomography of Chile: toward a new 3D velocity model
14:40 - 15:00 D. Inzunza Deep Rock Impacts on Earthquake Surface Intensities.
15:00 - 15:20 G. Montalva Single Station Sigma in Chile
Earthquakes at the Chilean Subduction Zone
A precision based regularization scheme for slip inversion: Application
15:40 – 16:00 F. Ortega
to the Central Andean Subduction Zone
Slip Models of the Co- and Post- seismic stages Associated to the April
16:00 - 16:20 V. Becerra 1st, 2014 Pisagua (Mw 8.2) Earthquake Constrained by GPS
Observations
16:20 - 16:40 M. Leiva Migración Sísmica asociada al terremoto de Iquique del 2014 (8.2 Mw)
Reawakening of large earthquakes in South-Central Chile: The 2016
16:40 - 17:00 S. Ruiz
Mw7.6 Chiloé event
17:00 – 19:30 Poster Session
A. Aguilar High-Frequency Scattering from Deep Earth
Correlation between seismic and infrasonic signals from Villarrica
X. González
Volcano during the last eruption crisis of 2015
Improving Volcanic Deformation monitoring using GPS Signal to Noise
G. Pérez
Ratio
Geometría de basamento en la zona de antearco de la región de Arica
D. Cabrera y Parinacota asociado a la geomorfología superficial, utilizando el
método gravimétrico
M.C. Flores Magallanes’s seismicity study
M.Medina Run-up of the 1922 earthquake gap
Palabras de Cierre
Lista de Autores y Presentaciones
Authot – TALK Mail Abstract Title

Juan Carlos Baez jcbaez@csn.uchile.cl Analysis of large GNSS networks: from daily to real-time
Daniel Basualto daniel.basualto@sernageomin.cl Terremotos y Reactivación Volcánica, Evidencias y Correlaciones
Heterogeneidades de la frontera manto-núcleo, observadas sísmicamente en las zonas de sombra
Klaus Bataille bataille@udec.cl de los núcleos.
Slip Models of the Co- and Post- seismic stages Associated to the April 1st, 2014 Pisagua (Mw 8.2)
Valeria Becerra C. valebecerra@ug.uchile.cl Earthquake Constrained by GPS Observations
Angela Cristina Carrillo
Ponce angelacarrillo@udec.cl Slow earthquakes, a concept from another perspective.
Gaspar Cid gcid@udec.cl Analysis of the deformation of Villarrica volcano using GPS and analytical models
Numerical analysis of the role of local faults on the influence of the 2015 Illapel and 2016 Quell\'on
Cristian Farías cristian.farias@uct.cl earthquakes on the Chilean volcanic arc: preliminary results
Mauricio Fuentes mauricio@dgf.uchile.cl Studying the dependence of the tsunami run-up with the source-time function
Francisco Manuel García
Araya franciscogarcia@udec.cl Crustal Faults Activity inferred from GPS and its relation with the latest volcanic eruptions
Transient and long-lived seismic velocity structure of Southern Andes within latitudes 34.5°S and
Diego González diegogonzalezvidal@gmail.com 37.5°S from seismic ambient noise
Diego Inzunza diegoinzunza@udec.cl Deep Rock Impacts on Earthquake Surface Intensities.
Mauricio Leiva maurleiva@udec.cl Migración Sísmica asociada al terremoto de Iquique del 2014 (8.2 Mw)
Edio Mardones ediomardones@udec.cl Feature signal extraction and classification of acoustic events at Villarrica Volcano using infrasound
Deformación post-sísmica asociada al Terremoto de Tohoku-Oki Mw=9.1, interpretado por
Diego Molina bataille@udec.cl movimiento de relajación en las interfases de la corteza oceánica.
Gonzalo Montalva gmontalva@udec.cl Single Station Sigma in Chile
David Mora dmora@udec.cl La desglaciación en los Andes del Sur y su impacto en sistemas magmáticos superficiales
A precision based regularization scheme for slip inversion: Application to the Central Andean
Francisco Ortega ortega.francisco@u.uchile.cl Subduction Zone
The origin and kinematics of lahars generated during the eruption of Villarrica volcano on March 3,
José Luis Palma jose@udec.cl 2015
Bertrand Potin bpotin@csn.uchile.cl Tomography of Chile: toward a new 3D velocity model
Sergio Ruiz sruiz@dgf.uchile.cl Reawakening of large earthquakes in South-Central Chile: The 2016 Mw7.6 Chiloé event
Miguel Antonio Sáez Arias msaezarias@hotmail.com Juan Fernandez as a T-Phase detector for South-American earthquakes
Spatial distribution of coseismic slip for the 2011 Tohoku-Oki earthquake based on a seismo-
Felipe Vera Sanhueza fvera.sanhueza@gmail.com geodetic perspective from GPS and W-phase observations
Carlos Ulises Villalobos
Núñez carlosvillalobos@unach.cl Seismic ionospheric precursors? A study of the great 1960 and 2010 Chile earthquakes
Authot – POSER Mail Abstract Title

Andrea Aguilar andrea.aguilar.san@gmail.com High-Frequency Scattering from Deep Earth
Seismic tomography in the vicinity of the Arauco peninsula from the aftershocks of the Mw 8.8
Kellen Azúa Quezada kazua@udec.cl Maule 2010 earthquake
Geometría de basamento en la zona de antearco de la región de Arica y Parinacota asociado a la
Daniel Cabrera cabrera.daniel.alanis@gmail.com geomorfología superficial, utilizando el método gravimétrico.
Sebastian Carrasco FinDer performance using CSN network: a strong-motion based algorithm for Earthquake Early
Morales acarrasco@csn.uchile.cl Warning
María Constanza Flores
Véliz maria.flores.veliz@ug.uchile.cl Magallanes’s seismicity study
Correlation between seismic and infrasonic signals from Villarrica Volcano during the last eruption
Ximena González xgonzalezd@udec.cl crisis of 2015.
Control on the active crustal deformation and structural geology of Laguna del Maule Volcanic
Paulina Köhler paulykohler@gmail.com Complex using seismic activity, bathymetry and seismic profiles.
Nicole Soledad Krumm
Nualart nskrumm@gmail.com Identification of Tectonic Signals in GPS positional time series
Raúl Madariaga madariag@geologie.ens.fr Spectral properties of large recent Chilean earthquakes
Renzo Antonio Mancini De
Barbieri rmancinidb@gmail.com Three dimensional conductivity model of the Paniri-Toconce volcanic system, Chile
Miguel Medina Flores miguel.medina@dgf.uchile.cl Run-up of the 1922 earthquake gap
Cristian Leonardo Otarola Comparison between interplate and intraplate intermediate-depth earthquakes using dynamic
Bravo crotarola@ug.uchile.cl modeling of the seismic source
Gustavo Pérez Gutiérrez gustavo.perez@ug.uchile.cl Improving Volcanic Deformation monitoring using GPS Signal to Noise Ratio.
Franco Vera francoevera@udec.cl Lahar generated from mixed avalanches during the eruption of Villarica volcano on March 3rd 2015
Patio de Comidas Sector Plaza Perú
Mall del Centro Lugares para almorzar y cenar
La Cocina $$
Tokio Bowl $$
Tijuana $$
Deli House $$
Verde Secreto $$$
Sumo $$$
s
Pato
o Los
sin
Ca
Campanil
uador
e Ec Auditorio “Salvador Gálvez” 3GSEV
rqu
Pa Facultad de Ingeniería
UdeC
Últimos avances en el Centro Sismológico Nacional (CSN)
Sergio Barrientos
Centro Sismológico Nacional (CSN)
Santiago, Chile
El Centro Sismológico Nacional de la Universidad de Chile (CSN), continuación del Servicio

Sismológico de la misma Universidad, inició sus operaciones en marzo de 2013. La misión
de este organismo consiste en impulsar y desarrollar los aspectos observacionales de
terremotos en el país con el fin de entregar la mejor, más completa, depurada y oportuna
información posible relacionada con la observación del proceso de terremotos.
Para ello, opera y mantiene una red de observación de los movimientos que sufre el suelo
asociados a terremotos. Utiliza dispositivos que miden desplazamiento (haciendo uso del
Sistema de Navegación Satelital, GNSS), velocidad (sismógrafos de banda ancha) y
aceleración (acelerómetros). En la actualidad existen un poco más de 100 estaciones multi-
paramétricas (incluyendo aquellas pertenecientes a redes internacionales como IPOC, IRIS
y Geoscope) desplegadas en el país, entregando datos de libre disponibilidad en tiempo
real. Este diseño incluye estaciones multi-paramétricas en Isla de Pascua y Juan Fernández.
La instalación y conexión de las estaciones de GNSS se completará durante el año 2017.
Como complemento a este equipamiento, la Oficina Nacional de Emergencia (Onemi) del
Ministerio del Interior y Seguridad Pública, en cooperación con el Ministerio de la Vivienda y
Urbanismo (Minvu) han desplegado 297 acelerógrafos, con fines ingenieriles para registrar
el movimiento fuerte asociado a grandes terremotos (umbral activación a 2% g); estos
instrumentos han sido transferidos al CSN para su operación y mantenimiento.
Este avance en la observación sismológica ha permitido registrar, de manera nunca antes

lograda, especialmente en el campo cercano, los últimos grandes terremotos que han
ocurrido en el país. Estimaciones rápidas de deslizamiento en la falla a partir de las
deformaciones observadas en superficie han permitido cuantificar y determinar las
características de los terremotos significativos. Se han establecido mapas de umbral de
detección en magnitud a partir de desplazamientos detectables en la red geodésica. En la
zona austral, por primera vez en la historia sismológica de Chile, hemos podido caracterizar
la sismicidad de menor magnitud asociada a la Falla de Magallanes.
Otros avances de importancia han sido la implementación de la estimación de magnitud y

mecanismo focal a través del método de la fase W a partir de instrumentos de banda ancha,
y al interior del CSN se ha extendido este método a datos GNSS permitiendo utilizar datos
de estaciones más cercanas que además no se saturan. Recientemente se ha
implementado y calibrado con datos locales la generación de un “Shakemap”, que permite
caracterizar rápidamente el movimiento del suelo debido a un terremoto.
Importantes esfuerzos se continúan realizando en la caracterización de sitios donde se

ubican los instrumentos de movimiento fuerte; también se ha comenzado a realizar un
monitoreo específico de la Falla San Ramón, en el borde oriental de la ciudad de Santiago.
En los aspectos de alerta temprana, nos encontramos en los inicios de un experimento que
consiste en el despliegue de estaciones de bajo costo basadas en "teléfonos inteligentes"
capaces de transmitir por sus canales de datos las señales de su sensor interno de
aceleración, junto a las generadas por un circuito especialmente diseñado para capturar las
señales GPS, lo que permite estimaciones de posición con error de unos pocos cm.
El CSN mira hacia el futuro con la tarea de mejorar continuamente sus procesos y
procedimientos entregando cada vez mejor y más oportuna información hacia las
autoridades, academia, ingenieros y público en general, dentro de los cuales la observación
en el fondo marino no puede quedar exenta, pero aún es una tarea pendiente.

Analysis of large GNSS networks: from daily to real-time

Juan Carlos Báez1 , Felipe Leyton1 , Francisco Del-Campo1 , David Pineda1 ,
Marcos Moreno2, Andres Tassara3
1- Centro Sismológico Nacional, Universidad de Chile, Chile, jcbaez@csn.uchile.cl
2- GFZ Helmholtz Centre Potsdam, German Research Centre for Geocience, Germany
3- Departamento de Ciencias de la Tierra, Universidad de Concepción, Chile
Abstract
The Chilean National Seismological Centre (CSN), have been installed and
operated, from 2013, a large GNSS network, including Pacific island and Antarctic
region, with more of 130 stations. We have implemented a schedule of processing
in two different approaches: on one hand, we have a daily solution, which are
stacking and combined to recover inter-seismic strain accumulation, seasonal and
local variation, for seismic cycle observation; these solutions are made using rapid
and definitive IGS products (15 hours to 15 days). On the other hand, we have
implemented a real time solution for rapid earthquake characterization, with 1-Hz
streams, that enable us to recover the geometry of the rupture area, assess the
maximum displacements, and W-phase analysis. We will show results from all of
these solutions. We will show our estimations for Pisagua 2014, Illapel 2015 and
Melinka 2016. In figure a) GNSS network; b) time series of station Quellón during
Melinka 2016; c) final model estimated with GNSS displacement for Illapel 2015,
daily base; d) final model estimated with kinematic results 1_hz Melinka 2016.
b
−60˚
−80˚
a
−20˚ −20˚
Others
−40˚ CSN et al −40˚
IGS
CSN
RAMSAC
−60˚
−80˚
d
c
Studying the dependence of the tsunami run-up with the source-
time function
Fuentes M(1)., Riquelme S(2)., Ruiz J(1). and Campos J(1).
(1)
Department of Geophysics, University of Chile
(2)
National Seismological Center, University of Chile
Keywords: Tsunami, run-up, Seismology.
Tsunami modeling usually assumes a non-time dependent generation. The static case is
reasonable due to fast earthquake rupture velocity compared with tsunami phase and
group velocities. Nevertheless, for slow earthquakes, as tsunami earthquakes, the time
source function may play a non-neglecting role.
A New analytical solution for the forced linear shallow water equations is obtained. This
analytical result also retrieves the case for static tsunami generation. With this formula,
time source can be included and then to verify directly a relation between rise-time, rupture
velocity and run-up.
For the simple radiation model of a box-car, we obtain that run-up decreases with rise time
and rupture velocity. Also, delays in the arrival times are observed.
With the analytical solution we also can include realistic source-time functions obtained
from seismic inversion models, which would allow modeling some of the most emblematic
tsunamis of the last decades.
Corresponding author: Mauricio Antonio Fuentes Serrano

mauricio@dgf.uchile.cl
cellphone: +569 85463404
3GSEV
Juan Fernandez as a T-Phase detector for South-American
earthquakes
Miguel Sáez (1,2), Sergio Ruiz (1) and Javier Ojeda (1)
(1) Department of Geophysics, University of Chile, Santiago, Chile
(2) Department of Geology, University of Chile, Santiago, Chile
The T-phase has been suggested as an earthquake discriminator, between explosions
and seismic events (Talandier and Okal. 2001). Here, we characterize the T- phases
generated from South-American earthquakes, using the Centro Sismológico Nacional
(CSN) broadband stations at the Chilean coast and Juan Fernandez Island. We identified
clearly the T-phases in Juan Fernandez Island stations, in the coastal stations the T-
phases were difficult to observe due to the strong land path attenuation. We use the
broad-band records filtered in the 4-7 [Hz] frequency range. We detect the T-phases for
events with magnitude between Mw 3.8 to Mw 8.3 and epicentral distance between 600-
2400 [km]. With the aim to study the potential of Juan Fernandez station as a T phase
detector we calculated the T wave flux energy (TPEF) and the amplitude-duration
discriminant (D1) using events with Mw between 5.7-8.3 that took place in 2015 and 2016
in different places of South America. Results show that most of the events have a D1<0
associated to subduction earthquakes (Talandier and Okal. 2001) while the TPEF of
interplate events associated mainly to Illapel 2015 (Mw 8.3) aftershocks follow a linear
trend with the magnitude, different to the observed for the intraplate deep events. Finally,
we suggest that the continuous T phase observations in Juan Fernandez can be used to
monitor the South-American earthquakes and its tsunamigenic potential.
Keywords: T-Phase; South American earthquake; Juan Fernandez broadband station

Slow earthquakes, a concept from another perspective.
Slow earthquake, Froude number, wave amplification.
Angela Carrillo P., Ignacia Calisto B.
University of Concepcion
Slow earthquakes are those ruptures with a predominant component in the long-period spectrum. The
duration is greater than in a normal earthquake due to its lower velocity rupture and they radiate much
less energy (Kato, 1989). The investigation of Todorovska and Trifunac in 2001 shows that the
slowness of a rupture is related with tsunami wave amplification in the context of constructive
interference when the rupture velocity is similar to the tsunami velocity.
Considering two slow earthquakes (1992 Nicaragua, 2004 Java) the produced tsunamis were
modelated from static and dynamic models using the COMCOT software expecting a wave
amplification but no unusual behavior was found. A different result was presented by the Great Chilean
earthquake of Valdivia in 1960, which was considered because of his low-frequency nature (Cifuentes
and Silver, 1989). From the source model constructed by Moreno et al. (2011) there were found larger
amplitudes on its dynamic tsunami simulation which suggest that the key factor on this phenomena is
the rupture duration time. Using the Froude number this event shows a ratio closest to 1 between -39
and -42 of latitude, which establishes this zone as susceptible to waves amplification.
The former result leads to analyze the concept of slow earthquake on this topic and define an
earthquake as slow when the Froude Number is close to 1. To improve this study the inclusion of other
events to compare with is considered. The earthquake of Tohoku-Oki in 2011 is one candidate due to
its long duration.
Figure 1: Froude Number (Fr) for Valdivia event in 1960. Fr is the ratio between the rupture and
tsunami phase velocites. If it is close to one it means that wave amplification is possible due to
constructive interference. The zone with Fr~1 corresponds to a low velocitiy zone suggested by
Cifuentes and Silver in 1989.
Crustal Faults Activity inferred from GPS and its relation with the latest
volcanic eruptions.
Keywords: geodesy, vorticity & Liquiñe-Ofqui
Francisco García, Andrés Tassara & Vicente Yáñez
Departamento Ciencias de la Tierra, Universidad De Concepción
The Southern Andes is probably one of the best places on earth to study the link between active
tectonics and volcanism. Since 2009 (the older GPS data used in this study) there have been at least
five clear eruptive events and one big megathrust earthquake (2010 8.8 Maule earthquake). The
volcanic line in the study area (between the 36º-47º S) has a strong relation with the Liquiñe-Ofqui
Fault Zone (LOFZ), a 1100 km long dextral system that accommodates the margin-parallel
component of oblique convergence. Interaction of the LOFZ with inherited structures control the
first order characteristics of the different volcanic systems, like the location and the residence time
of the magmas in the crust that directly affects the composition of volcanoes.
Volcanic eruption can be triggered by variations in the strain fields around the volcanic system.
These variations can be produced by megathrust earthquakes and also by activity of crustal faults
near or under a volcano. Geodetic data, allow us to observe the superficial expression of these
processes, and using geological information such as location of faults or volcanic systems is possible
to evaluate the role of tectonics in the triggering of a specific eruption.
This study uses continuous GPS data and trajectory models to calculate velocity and strain fields. In
order to better understand how crustal deformation can induce volcanic eruptions, the GPS time
series were fitted using a trajectory model (tm). Using these tm to evaluate the velocity of a point in
any time of the tm, and iterating a process of calculation of velocity fields, velocity gradients and
kinematic vorticity (Wk), we are for the first time capable to study strain variations not only in
latitude and longitude but also in time. The main focus of this study is to connect variations in the
crustal deformation with volcanic eruptions. Wk is very useful to study the volcano-tectonic relation
in the study area, because most of the deformation that affects the volcanic arc is accommodated
by the LOFZ. In two of the eruptions analyzed in this study, an increase in the vorticity value, prior
to the eruption, shows an change in the local strain field, passing from a elongation dominated
regime to a rotation dominated regime. The change in the way that crust accommodates strain, can
be interpreted as the reactivation of traces of the LOFZ, that “pumps” or at least make space for
magma to raise. This not only allows having a better understanding of the role that crustal
structures play on the triggering of volcanic eruptions, but also could be use as a strong tool for
volcanic monitoring.
Numerical analysis of the role of local faults on the
influence of the 2015 Illapel and 2016 Quellón
earthquakes on the chilean volcanic arc: preliminary
results
Cristian Farı́as
Departamento de Matemáticas y Fı́sica, Facultad de Ingenierı́a, Universidad Católica de Temuco.

Rudecindo Ortega 02950, Temuco, Chile.
Keywords: Earthquake-volcano interactions, Numerical simulations.
The physical mechanisms that control earthquake-volcano interactions are poorly known. This
is in a profound constrast with the improvements in both data quality and volume on this topic in the
last 15 years. As part of the project ”Cómo pueden los terremotos afectar la actividad volcánica?
Un estudio usando simulaciones numéricas y observaciones de terreno”, I present here a prelimi-
nary numerical analysis of the impact of the 2015 MW = 8,3 Illapel and 2016 MW = 7,6 Quellón
earthquakes in the chilean volcanoes, considering the geometry of their fault systems into a set of
2-D numerical simulations. One of the main features is that the geometry of the local fault systems
affects importantly both static and dynamic earthquake-induced mean stress changes. This is par-
ticularly noticeable at the Copahue volcano under the influence of the Illapel earthquake, where the
interactions between the seismic waves and the main Copahue fault turns this system into one of
the most affected by the large seismic event. Interestingly, reports from OVDAS show a swarm-like
response at this volcanic system right after the passage of the waves from the Illapel earthquake,
with the largest event being a MW = 3,1 Volcano-Tectonic event. While further studies are neces-
sary to advance in the understanding of the earthquake-volcano interactions, the results presented
here are a step forward into the improvement of post-seismic analysis of volcanic dynamics, which
can be greatly important in risk assesment programs.
Terremotos y Reactivación Volcánica, Evidencias y Correlaciones
Daniel Basualto Alarcón1*, Pablo Gonzalez Troncoso1
1
Servicio Nacional de Geología y Minería “SERNAGEOMIN”-Red Nacional de Vigilancia Volcánica”
RNVV”-Observatorio Volcanológico de los Andes del Sur “OVDAS”, Rudecindo Ortega #
03850,Temuco, Chile. *Contacto e-mail: daniel.basualto@sernageomin.cl
Palabras clave: terremoto reactivación volcánica
Resumen
El arco volcánico chileno, y en especial el Segmento de la Zona Sur (Stern 1994) ha

presentado evidencias sobre la interacción entre grandes terremotos y posterior reactivación
volcánica. Probablemente el ejemplo más claro fue la erupción del volcán Cordón Caulle reportada
el 24 de mayo de 1960, 42 horas después de generado el gran terremoto de Valdivia Mw:9,5
(Barrientos 1994; Lara 2004; Sepúlveda et al., 2005). Existen otros trabajos que abordan este tema
desde el punto de vista estadístico (Watt et al., 2005), pero con extensas ventanas temporales
(años) para lograr justificar la interacción entre estos dos fenómenos. Basándonos solo en registros
históricos, el detalle al que se puede llegar es escaso, dificultando la tarea de poder entender dicha
relación. Sin embargo, desde 2010, el monitoreo continuo de los volcanes más activos de Chile ha
generado la información necesaria para comenzar a abordar estos temas con el detalle que se
requiere. De todo el arco volcánico chileno monitoreado por OVDAS (Andes Centrales y Sur), los
volcanes Chillan (VCH), Copahue (VCO) y Villarrica (VVI) tuvieron un cambio significativo en su
actividad sísmica inmediatamente después del arribo de la onda “P” asociada con el terremoto de
Coquimbo registrado el 16/09/2015 Mw:8,4. Los cambios en el registro sísmico fueron desde un
abrupto incremento en el número de sismos Volcano-Tecónicos (650 VCO y 85 VVl en un lapso de
15 días) y de Largo Período, pasando por un incremento en los valores RSAM (Real-time Seismic
Amplitude measurement) y DR (Desplazamiento Reducido) en el tremor volcánico (Trv), asociado
además con un cambio en la frecuencia predominante del Trv (VCH y VVl). Si bien Chile posee
cerca de 100 volcanes activos a lo largo de su cordillera, que particularidad comparten estos tres
volcanes que reaccionaron al paso de las ondas sísmicas generadas por el terremoto de
Coquimbo...?. El análisis con diversas metodologías geofísicas (sismología, geodesia, geoquímica
y cámaras de vigilancia) durante varios años (2010-2017) nos ha permitido definir niveles base o
background de actividad volcánica, y con ello, lograr identificar sutiles cambios que generar
inestabilidad en los sistemas volcánicos. Bajo este criterio, fue posible establecer que estos
estratovolcanes, cuyas geoquímicas son diferentes entre sí, compartían una característica en
común. Estos tres volcanes habían registrado un cambio en el background de la actividad
volcánica meses antes del terremoto de Coquimbo. Si consideramos que los volcanes reactivados
por el terremoto se localizan a 560, 670 y 850 km de distancia del epicentro, estos se encuentran
fuera de la zona de ruptura, descartando el "estrés estático" como un posible triggering. Por otro
lado, si consideramos la importancia de las ondas elásticas provenientes de un terremoto al pasar
por un sistema volcánico activo ("estrés dinámico", Brodsky et al., 1998), estas podrían ser
capaces de reactivar un volcán que presenta una desestabilización previa. Finalmente, cabe
mencionar que si bien las correlaciones no generan causalidad por si solas, el conjunto de todas
estas evidencias sugieren que "la desestabilización de sistemas volcánicos gatillados por un
terremoto ocurren en casos donde dichos sistema volcánico se encuentre inestable", es decir por
sobre el nivel base de actividad volcánica.
La desglaciación en los Andes del Sur y su impacto en sistemas
magmáticos superficiales
David Mora y Andrés Tassara

dmora@udec.cl
Universidad de Concepción
El último máximo glaciar (18 ka A.P) en los Andes del Sur fue seguido por una rápida
desglaciación de menos de 5000 años de duración. Este período coincide con un reducido
número erupciones volcánicas piroclásticas de gran volumen, inducidas por la
descompresión del orden de decenas de megapascales de las cámaras magmáticas
contenidas en la litósfera superior. Una de las causas de la descompresión es atribuida al
abrupto ascenso del magma de los reservorios, posterior a la intrusión de magmas más
calientes en la base de las cámaras magmáticas (ej.: Ignimbrita Licán, Volcán Villarrica).
Otra causa de la descompresión puede ser asociada a las variaciones del régimen de
esfuerzo regional que ocasionan los cambios de cargas superficiales en la litósfera.
A través de modelación analítica se investiga la magnitud de la descompresión en la

litósfera que podrían ocasionar en conjunto los cambios de esfuerzos incrementales y
flexurales, resultantes de los cambios de cargas superficiales. Conjuntamente, buscamos
las distintas variables que podrían influir en el desarrollo de tales cambios de presión en el
tiempo. El modelo es aplicado al caso de la erupción de la Ignimbrita Licán, ocasionada
por una descompresión de 40 MPa ocurrida entre 600 y 1000 años posteriores al último
máximo glaciar.
Encontramos que la desglaciación por sí sola pudo contribuir con tales magnitudes de
cambios de presión. Además, las variables de primer orden que controlan los cambios de
presión en el tiempo son el espesor y el radio de la carga glaciar y la profundidad a la cual
se generan los cambios de presión, asociados a un sistema magmático superficial.
Palabras claves: descompresión, esfuerzos incrementales y flexurales.

Seismic ionospheric precursors? A study of the great 1960 and
2010 Chile earthquakes
C. U. Villalobos a, E. M. Ovalle b, M. A. Bravo c, A. J. Foppiano b

a Universidad Adventista de Chile, Casilla 7-D, Chillán, Chile. carlosvillalobos@unach.cl
b Universidad de Concepción, Casilla 160-C, Concepción, Chile.
c Universidad de Santiago de Chile, Avda. Ecuador 3493, Santiago, Chile.
Keywords: Ionospheric characteristics; TEC; Earthquake precursors.
Abstract
A study of eventual ionospheric precursors of the great 1960 and 2010 Chile
earthquakes is presented. For the first earthquake, the variations of the ionospheric
characteristics NmE, NmEs, h’E, NmF2, h’F, M3000F2 and fmin of the E and F region,
as observed by a vertical incidence ionosonde at Concepción (36.8_S; 73.0_W), are
considered. Changes associated with solar and geomagnetic activities are first filtered.
Then anomalies are identified when characteristics values are larger or smaller than
reference values (15-day running medians +/- interquartile range). In the case of the
2010 earthquake a similar analysis of NmF2 and TEC is made. TEC values for two
local GPS stations and from a regional model covering the whole of the rupture zone
are considered. Furthermore, an equivalent analysis of TEC is made for an area some
thousand km north of the rupture zone so as to identify possible anomalies not related
to the earthquake. Other statistical analysis methods used by some authors in the case
of many earthquakes in Taiwan, Japan and Russia are also considered to examine
their results consistency.
The found NmF2 results for the 1960 earthquake are consistent with some published
reports but are clearly not with others. Moreover, an evident M3000F2 diurnal-variation
pattern change from 30 April to 1 May is observed (see Figure). However, this feature
cannot unambiguously be considered as an ionospheric precursor because it could be
associated with a known seasonal variation. Found NmF2 and TEC anomalies for the
2010 earthquake cannot again be unambiguously identified as precursors.
M3000F2 observed over Concepción (36.8_S; 73.0_W) for 21 March–21 May 1960
Lahar generated from mixed avalanches during the eruption of Villarica
volcano on March 3rd 2015
Franco E. Vera 1*, José L. Palma1, Jeff Johnson2
1. Department of Earth Sciences, University of Concepción, Concepción, Chile.

2. Department of Earth Sciences, Boise State University, Idaho, USA
*francoevera@udec.cl
Key words: Mixed avalanches, ice/snow-lava interaction, drone mapping, lahars.
In the early morning of March 3rd 2015, Villarica volcano exhibited an intense eruptive
pulse of 25 minutes duration consisting of a ~1.5 km high lava fountain. The associated
explosive activity and ejection of incandescent pyroclastic material prompted the
generation of mixed avalanches of ice/snow and scoria (glassy juvenile material). These
avalanches eroded the ice on the upper flanks of the volcano and generated lahars that
descended onto the Pedregoso, Zanjón Seco and Correntoso valleys as well as towards the
SE side of the volcano. In this study we show the first drone-generated high-resolution
orthomosaic mapping of the proximal deposits of the mixed avalanches, evaluating their
interaction with ice/snow and its relevance in the generation of lahars. Several infrasound
arrays located between 3 and 10 km from the crater detected moving sources, which were
associated with the avalanches and laharic flows. These infrasound data allowed the
calculation of flow velocity. Estimation of lahar volumes from their deposits and from
measurements of ice-carved channels were combined with flow velocity in order to
estimate peak lahar discharges. Our results highlight the significance of mixed avalanches
in the generation of lahars during the most critical stage of this eruption.
Deformación postsı́smica asociada al Terremoto de Tohoku-Oki
Mw = 9.1, interpretado por movimiento de relajación en las
interfases de la corteza oceánica.
D. Molina(1) , K. Bataille(2) , F. Vera(1)
(1) Departamento Geofı́sica, Universidad de Concepción
(2) Departamento Ciencias de la Tierra, Universidad de Concepción
El terremoto de Tohoku-Oki, Japón, Mw=9.1, ocurrido el 11 de marzo de 2011, ocasionó un gi-

gantesco tsunami y los mayores desplazamientos horizontales registrados. Afortunadamente existı́a
en la zona un arreglo submarino de estaciones GPS que lograron registrar el desplazamiento durante
el terremoto y el perı́odo posterior. Con 5 años de observaciones posterior al terremoto, en más de
378 estaciones de GPS, tanto en tierra como suelo marino, proporcionan una oportunidad única de
describir los movimientos que gobiernan el proceso de subducción. Las estaciones marinas cercanas
a la fosa, registraron movimientos horizontales cosı́smicos de hasta 25 metros, algo que jamás se
habia observado. Mas aún, posterior al terremoto, estas estaciones de GPS, mostraron movimientos
en dirección opuesta al cosı́smico. Varios son los estudios (Sun et al; 2014, Watanabe et al; 2014)
que atribuyen este movimiento en superficie a una prevalencia de la deformación viscoelástica del
manto superior, sin embargo varios son los que piensan que la deformación viscoelástica tendrı́a un
efecto a mucho más largo plazo. Este trabajo busca explicar la deformación postsı́smica usando
solo un modelo de relajación de las interfases de la corteza oceánica como primera aproximación.
Lo atractivo de este modelo de deformación postsı́smica, es que atribuye como fuente de defor-
mación, no solo el movimiento tipo inverso del terremoto, sino que también la interfase inferior de
la placa subductante. A partir de esto se infiere que los movimientos que gobiernan tanto la inter-
fase superior como inferior son de carácter inverso y normal respectivamente. A través de métodos
de inversión de datos GPS, utilizando modelos homogéneos (Okada 1985) se logra cuantificar el
deslizamiento que ocurre durante los primeros 3 años de observaciones, registrando máximos de de-
splazamiento de 5 y 3 metros en las interfases superior e inferior respecticvamente. Se discutirá la
distribución de deslizamiento durante los perı́odos interı́smico, cosı́smico y postsı́smico mostrando
su rol complementario en ambas interfases.
Heterogeneidades de la frontera manto-núcleo, observadas
sı́smicamente en las zonas de sombra de los núcleos.
K. Bataille(1) , C. Sens-Schoenfelder(2) , M. Bianchi(3) , N. Carreño(4)
(1) Departamento Ciencias de la Tierra, Universidad de Concepción
(2) GeoForschungZentrum, Potsdam, Alemania.
(3) Centro de Sismologia, Universidad de Sao Paulo, Brasil.
(4) Departamento Geofı́sica, Universidad de Concepción
Los dos procesos dinámicos mas importantes en la Tierra son convección en el manto y núcleo.
Estos procesos son complejos, involucrando por un lado fuerzas que producen inestabilidades y
generando heterogeneidades, y por otro lado, fuerzas de relajación o disipación que suavizan las
heterogeneidades. Las caracterı́sticas de las zonas heterogéneas evidenciadas sı́smicamente permiten
sugerir modelos de los procesos dinámicos existentes. La primera evidencia de la presencia de
heterogenenidades en la frontera manto-núcleo (Haddon en 1972), fueron a partir de ondas (∼ 1
Hz ) en la zona de sombra del núcleo externo, como precursoras de PKP, en las distancias entre
120◦ − 143◦ . Recientemente, la oportunidad de contar con eventos sı́smicos profundos registrados
en una gran cantidad de estaciones sismológicas de calidad, permiten investigar las propiedades
de las heterogeneidades de mas pequeña escala. Reportamos la observación de ondas de alta
frecuencia (> 3 Hz) en la zona de sombra del núcleo interno, es decir para distancias > 155◦ . Estas
ondas inicialmente sugeridas como difractadas por el núcleo interno, son ondas dispersadas por la
frontera manto-núcleo. Se presentan modelos basados en los dos extremos de dispersión: débil y
fuerte, mostrando correspondencia con las observaciones.
Three dimensional conductivity model of the Paniri-Toconce
volcanic system, Chile
Renzo Mancini1,2 and Daniel Dı́az1,2
1
Departamento de Geofı́sica, Universidad de Chile
2
Centro de excelencia en geotermia de los Andes (CEGA)
1
rmancinidb@gmail.com
Abstract
The research is located in the Paniri-Tocone volcanic chain in the Central Volcanic
Zone of the Andes, in north Chile, composed by the volcanoes Paniri, Cerro de Leon,
Lavas de Chao and Toconce, which is immediately to the NW of geothermal system
El Tatio. The main goal in the study consists in determining the current behavior of
the complex, in which the oldest volcano is Toconce with an age of 1.5 My, while
Paniri and Lavas de Chao are the youngest with an age about 0.3 My. The volcanoes
don’t have historical eruptions but the date of the last eruption in the Paniri volcano is
estimated to be one hundred thousand years ago.
Thermobarometry studies made in the area show a depth for the crystallization of
lavas at about 8 km depth from Lavas de Chao, and depths greater than 24 km form
Toconce and Cerro de Leon.
In this research we measured 20 broadband magnetotelluric stations in the vicinity

of the complex. The aim is to characterize the deep conductive zones and their
relationship with magmatic bodies associated with the adjacent volcanic system. It
is expected to link the possible magmatic structures with the surface behavior of the
volcanic chain.
A dimensionality analysis was carried out, considering induction arrows, phase

tensor ellipses and calculation of geoelectric strike. Afterwards, the data has been
modeled in 3-D, using ModEM (Kelbert et al. 2014), according to the results of the
dimensionality analysis. Finally, petrological and geochemical studies will be included.
Preliminary results indicate the presence of a geothermal system to the southwest

of the complex with maximum depths about 7 km, and a posible magmatic chamber
in the N-E of the complex with depths biggest than 8 km.
Keywords: 3D inversion, Volcanoes, magnetotelluric
1
Control on the active crustal deformation and structural geology of
Laguna del Maule Volcanic Complex using seismic activity,
bathymetry and seismic profiles.
Keywords: Rhyolitic volcano system, structural geology, strike-slip fault system.
Paulina Köhler1, Andrés Tassara1, Carlos Cardona2

1
Universidad de Concepción, Departamento de Ciencias Geológicas, Chile.
2Observatorio Volcanológico Chileno, Servicio Nacional de Geología y Minería. Rudecindo Ortega
03850, Temuco, Chile.
Corresponding author: e-mail address, paulinkohler@udec.cl
Abstract
The Laguna del Maule Volcanic Complex (CVLM) is a large and dynamics active volcanic
caldera, contents a large-volume of rhyolitic deposits of which includes recent concentration
of silicic eruptions during the Holocene. The CVLM has presented a seismic activity associated
to cortical uplifting process with vertical deformation rates ~30 cm/y, whose source has been
modeled through InSAR, ranked the largest worldwide. This study comprises a geological-
structural analysis and the geometric and kinematic description of the structures found in the
CVLM, also it is considered, temporal and spatial correlating with seismic activity, bathymetry
and seismic profiles made in the CVLM. The structures observed in the CVLM were classified
into: principal and secondary faults systems. The first incumbent to two structures most
extensive in the area, with orientations NE-SW and NW-SE, both with strike-slip displacement,
dextral and sinistral, with a normal and reverse component, defined as Troncoso and Maule
Fault, respectively. The secondary structures were defined as minor structures with a less
extention in the study area, some of them generated by caldera collapse. It is concluded that
the active structure responsible for the seismic activity of VT events reported in SW vertex of
deformation source area corresponds to the trace of the active fault Troncoso, which spreads
its movement to the northeast, crossing the LDM. In turn, the volcanic complex is controlled by
compressional faults NW-SE direction (Maule Fault), associated with the location of more
differentiated magmatism and rhyolitic volcanism explosive high volume. Therefore, the
structural geometry of CVLM, consists of a strike-slip dextral-sinistral fault system and,
associated structures in extensional domains of the main fault (Troncoso Fault), with incipient
development of a releasing bend in the center of the basin, which is bounded by normal faults
orthogonal to the main fault, inflation generated by the boiler associated with magmatic site.
Spectral properties of large recent Chilean earthquakes
Raul Madariaga (1), Gianina Meneses (1), Sergio Ruiz (2) and Fabian Bonilla (3)
(1) Laboratoire de Géologie ENS, Paris France
(2) Departamento de Geofisica, Universidad de Chile, Santiago, Chile
(3) Ifsttar, Ecole de Ponts et Chaussées, Marne La Vallée, France
It is widely accepted in seismology that displacement spectra of earthquakes is well explained by

the so-called omega squared model. In this model the Fourier spectrum of ground displacement is
flat at low frequencies and then decays as omega squared beyond the corner frequency. This is not
the case for the very large earthquales in Chile (Mw>7.5). The accelerograms recorded close to
these earthquakes show a more complex behaviour and, in particular, they all have broad
intermediate frequency bands where the spectra decays as the inverse of omega. This was observed
for the first time after the Tocopilla earthquake of 15 November 2007 on the accelerograms of the
Plate Boundary Observatory by Lancieri et al (2011). Thanks to the new digital instrumentation
available in Chile it is now possible to record on scale both accelerograms and GPS records. F.
Leyton and J.C. Baez from CSN have published a large collection of accelerograms integrated to
ground velocity and compared them to GPS records differentiated to obtain ground velocity. Within
the common spectral band between 0.025 and 0.3 Hz these records are identical. This gives us
confidence that accelerogram records in Chile can be used to study seismic radiation for frequencies
down to 0.025 Hertz and even lower for some events. The omega minus one slope is observed in
almost all the accelerograms recorded by the Iquique earthquake of 1 April 2014 near the Chilean
coast. The earthquake that occurred in Southern Chiloé on 25 December 2016 confirmed these
observations and provided a simple explanation : the omega minus one slope is due to finite steps in
displacement at the recording sites. These finite steps are due to the permanent ground displacement
produced by large earthquakes. We are currently using wavelet transforms to identify the origin of
these finite steps in the seismograms. The simplest explanation, that still has to be proven, is that
high frequency energy radiated by large finite fault events is depleted in favor of permanent
displacements observed with GPS and integrated accelerograms.
FinDer performance using CSN network: a strong-motion based
algorithm for Earthquake Early Warning
Sebastián Carrasco(a)*, Maren Böse(b)
(a) National Seismological Center, University of Chile

(b) Swiss Seismological Service, Swiss Federal Institute of Technology Zurich
*acarrasco@csn.uchile.cl
Keywords: strong ground-motions, earthquake early warning.
Rapid knowledge of seismic source parameters and accurate ground-motion predictions are
key to Earthquake Early Warning (EEW), which aims to warn the community before shaking initiates. In
recent decades, a variety of EEW algorithms have been proposed, including FinDer (Finite-Fault
Rupture Detector). In contrast to other algorithms, FinDer estimates fault-rupture extent in real-time
(assuming a line-source), which is important for ground-motion predictions for EEW in large
earthquakes, as these motions are strongly controlled by rupture-to-site rather than hypocentral
distances. FinDer uses observed Peak Ground Acceleration (PGA) values from a local seismic network
and compares these amplitudes with a set of precomputed templates based on a specific Ground
Motion Prediction Equation (GMPE). In this way, FinDer determines the location, length, and strike of a
line-source that best matches the current ground-motion observations. Estimates are regularly updated,
e.g. every second, while rupture is continuing. Here, we evaluate the FinDer performance through off-
line tests of the largest earthquakes in Chile during the last four years, using acceleration records from
C1, C and CX real-time networks from National Seismological Center (CSN), as well as from new non-
real-time National Accelerograph Network (RNA). In general, we observe good rupture length and strike
estimations, while station density plays a critical role. In the current CSN real-time network, FinDer
could provide quite accurate and robust estimations of source dimensions and expected ground-
motions around one minute after event origin, which is too late for cities near the epicenter, but could
warn more distant cities, where shaking in large earthquakes is still strong.
Figure: Location, strike, rupture length and magnitude estimations for 2015 M8.4 Illapel earthquake
from PGA values observed in CSN real-time networks at 60 seconds after origin time.
Comparison between interplate and intraplate intermediate-depth
earthquakes using dynamic modeling of the seismic source
C. Otarola¹, S. Ruiz¹, C. Herrera², R. Madariaga3 and C. Siegel¹
¹Department of Geophysics, University of Chile, Santiago, Chile

² University of Victoria, Victoria, Canada
3 Ecole Normale Superieure, Paris, France.
ABSTRACT:
Earthquakes dynamic models take into consideration stress-strain states and the friction
laws governing the rupture of earthquakes. The slip on the fault has to be considered as a
consequence of the stress conditions and the strength of the material in the focal region.
We perform dynamic source inversions of five interplate earthquakes, the magnitudes
around Mw 6.5, and hypocentral depth between 20 km to 40 km. The studied events are
one foreshock and four aftershocks of the Iquique 2014 Mw 8.2 earthquake. We perform a
comparison of the dynamic parameters between these events and three intraplate
intermediate-depth earthquakes occurred in Northern Chile, Japan and Argentina between
60 km to 250 km depth (Michilla 2007, Mw 6.7; Iwate 2008, Mw 6.8 and Jujuy 2015, Mw
6.7). The results show that the stress drop and the fracture energy rate for interplate
earthquakes are smaller than the intraplate intermediate-depth earthquakes. We conclude
that our results are consistent with that found in other studies. Finally, we speculate that the
higher stress drop for intraplate intermediate-depth events could explain the major
destructiveness observed in these events.
KEYWORDS: Intraplate and interplate earthquakes; Dynamic inversion; Stress drop.

Identification of Tectonic Signals in GPS positional time series
Nicole Krumm Nualart1, Francisco Ortega Culaciati1,2, Diana Comte1,2
(1) Departamento de Geofísica, FCFM, Universidad de Chile. (2) Advance Mining Technology Center
(AMTC), FCFM, Universidad de Chile.
In the last decade, technological advances in space geodesy have enhanced the quality and
quantity of crustal deformation observations. In particular, GNSS network instruments have
allowed to precisely measure crustal displacements, which may contain signals related to
different physical phenomena that are under the geoscience interest. Particularly, high seismic
activity zones such as Chile and Japan, contain GNSS instruments, whose observations are of
extreme importance to understand the physical processes associated to the occurrence and
generation of great earthquakes. To understand these processes, it is necessary to develop an
efficient and robust methodology to assess the crustal displacements featured in the seismic
cycle, considering all of the deformation sources, and fitting a physical model described for the
time series containing those displacements. Inspired in the work of Riel et al. [2014], we pro-
pose a methodology for sparse representation of geodetic time series, that will allow the
separation of its constituent signals, by identifying from a vast set of basis functions, the
minimum amount of functions required to accurately represent the geodetic time series.
Keywords: Tectonic signals, GPS, crustal deformation.

Seismic tomography in the vicinity of the Arauco peninsula from
the aftershocks of the Mw 8.8 Maule 2010 earthquake
Kellen Azúa1, Matthew Miller1, Andreas Rietbrock2, Stephen Hicks2.
Departamento de Geofísica, Universidad de Concepción1, School of Environmental Sciences,
University of Liverpool2.
Kazua@udec.cl
Maule earthquake, seismic tomography, Arauco peninsula.

The Arauco peninsula (37.2-38.5° S) is located at the overlap between the southern part of
the slip caused by the Mw 8.8 Maule 2010 earthquake and the northern section of slip
produced by the Mw 9.5 Valdivia 1960 earthquake. According to historical earthquake
records, the Arauco peninsula is identified as a possible barrier to propagation of
megathrust ruptures causing seismic segmentation along the Nazca subduction zone.
Data from a large number seismic stations, from a community of international
organisations, and Ocean Bottom Seismometers (OBS) allow high-resolution seismic
tomography images of Vp and Vp / Vs up to the trench, providing new information on the
properties of the subduction zone, seismic velocities, presence of fluids and sediments.
The velocity model obtained allows the determination of the main characteristics of the
subduction zone. Below the coastline, anomalies of high Vp (~ 7.5 km/s) have been found
on the subducted plate north of 38° S. High values (> 2 ) of Vp / Vs have been found in the
marine arc at ~37.9° S, which coincides with the subduction of the Mocha fracture in an
área of low co-seismic slip. It is possible that high Vp / Vs ratios control the up-slip limit,
and the increase of P-wave velocity (~ 7.5 km/s) control the down-slip limit, of the Maule
2010 earthquake rupture.
The origin and kinematics of lahars generated during the eruption of
Villarrica volcano on March 3, 2015
Jose Luis Palma(1), Jeffrey B.Johnson(2), Oscar Valderrama(3), Nayaret Flores(1), Edio
Mardones(4), Franco Vera(1)
(1) Department of Earth Sciences, University of Concepción, Biobio, Chile

(2) Department of Earth Sciences, Boise State University, Idaho, USA
(3) Volcano Observatory of the Southern Andes (OVDAS) SERNAGEOMIN, Chile
(4) PhD program in Geological Sciences, University of Concepción, Biobio, Chile.
On March 3, 2015, Villarrica volcano exhibited a ~1.5 km high lava fountain that generated rapid
avalanches of hot scoria mixed with snow and ice. Some of these avalanches eroded the ice cap,
created channels and added meltwater to the flows which quickly developed into lahars. The timing
of these events were captured by webcam and manual photographs as well as direct visual
observations. Using acoustic arrays we measured lahar velocities of up to 38 m/s near the source
for a lahar that descended in the Correntoso-Turbio drainage. Despite the acoustic interference of
the lava fountain and strombolian activity, we were also able to distinguish different lahars and flow
pulses travelling down into the Correntoso valley (NW). These observations are consistent with
local seismic records and analysis of the deposits left by these flows within 10 km of the crater.
This presentation shows the capabilities of acoustic (infrasound) arrays to detect lahars near the
source and to measure flow speed, which can be used to estimate the time of their arrival to
populated areas. In addition, using acoustic data to analyse the flow path, flow pulses (or waves)
and their speed - i.e. the kinematics of the lahars - is a valuable contribution to the understanding
of the origin and characteristics of lahars, as is demonstrated by the data collected regarding the
lahars generated at Villarrica volcano.
Keywords: Lahars, infrasound, Villarrica volcano

Feature signal extraction and classification of acoustic events at
Villarrica Volcano using infrasound.
Edio Mardones1, José Luis Palma2
1Phd program in Geological Sciences, University of Concepcion. Concepción, Chile
(ediomardones@udec.cl)
2Department of earth Sciences, University of Concepcion. Concepción, Chile
Keywords: Infrasound, clustering
Infrasound is a powerful tool to study and monitor the dynamics of explosive events in active
volcanoes. Appropriate pattern recognition techniques on acoustic signals are useful to
identify physical properties associated with explosive activity. We created clusters with the
intrinsic characteristics of infrasound waves using neural networks in order to analyze their
temporal evolution through time. To create the clusters we selected a group of relevant
features that are extracted from individual events, including waveform, frequency content,
energy ratio and amplitude decay. We show that the different clusters obtained reflect the
changes in the characteristics of the volcanic activity observed before the eruption of
Villarrica volcano of March 3, 2015. This type of analysis allows us to relate specific activity
patterns of the volcano with the acoustic signal, which for purposes of monitoring future
activity can be very useful.
Figure 1: Explosive events selected for 4 days prior to eruption with changes in frequency and waveform.
Analysis of the deformation of Villarrica volcano using GPS and
analytical models
Authors: Gaspar Cid Contreras, José Luis Palma y María Loreto Córdova
Affiliation: Universidad de Concepción
Mail: gcid@udec.cl
Keywords: Villarrica Volcano, GPS, Analytical model.
Recent progress in geodetic techniques, have improved data availability for volcanic
deformation analysis. These data used in conjunction with analytical models, allow
estimation of possible magmatic sources and their main parameters on active volcanoes.
In this study, Villarrica Volcano flank displacements and its potential origins are investigated,
using analytical techniques and geodetic tools, emphasizing on the post – eruptive period
related to the eruption occurred on March 3, 2015. For this purpose, a descriptive analysis
of geodetic displacements is carried out, resulting on a temporal segmentation of the data.
Forward and inverse analytical models are also applied for the post – eruptive period and a
magmatic source is proposed as solution.
The most possible source of deformation consists of a dike (prolate spheroid sensu stricto)
of 5 km radius, 5 km deep, 8 km to the SE of the crater and oriented N60°W/26°SE. This
dike would represent a magmatic conduit with an estimated volume increase of 5.85 x 106
m3 between April and May of 2015. A forward model is applied from this solution to obtain
surface displacement field maps.
Previous studies based on seismic data, exhibit a partial coincidence between the volcano
inner structure, volcano – tectonic (VT) earthquake locations and the dike proposed as a
solution in this study.
From the magmatic conduit geometry and flank displacements observed between 2012 and
2015, a structural/tectonic control of the deformation is inferred. These structures could
correspond to Liquiñe – Ofqui Fault Zone (LOFZ) and Mocha – Villarrica Fault Zone (MVFZ).
With the proposed background, a geodetic arrangement of stations is suggested that would
allow a better observation of the Villarrica Volcano flanks deformation.
Modified from Córdova

et al., 2015
Location of VT earthquakes (red dots) observed between March and June, 2015. The
modelled magmatic source location is also represented with a black line.
Transient and long-lived seismic velocity structure of
Southern Andes within latitudes 34.5◦S and 37.5◦S from
seismic ambient noise
Diego M. González Vidal ∗1 , Anne Obermann †2 , Klaus Bataille ‡1
, Stephen A.
Miller §3 , and Matteo Lupi ¶4
1
Department of Earth Sciences, University of Concepción, Concepción, Chile.
2
Swiss Seismological Service, ETH, Zürich, Switzerland.
3
Centre d’Hydrogéologie et de Géothermie, University of Neuchatel, Neuchatel, Switzerland.
4
Department of Earth Sciences, University of Geneva, Geneva, Switzerland.
The southern part of central Chile is associated with the oblique convergence of Nazca plate
beneath the South American plate with a velocity of ∼ 66 mm/yr. This produces the development
of a volcanic arc subparallel to the subduction trench, which includes numerous stratovolcanoes and
hundreds of minor eruptive centers. Also, the region is affected by frequent megathrust earthquakes
(e.g. Concepción 1835, Valparaiso 1906, Talca 1928, Chillán 1939, Valdivia 1960, Maule 2010) and
several seismological experiments were conducted to obtain a detailed image of the subduction interface
and upper crust involved in such deformation.
Here we present a regional-scale Rayleigh surface-wave tomography from seismic ambient noise
that highligths the three-dimensional shear-wave velocity structure at crustal depths. This study is
the first attempt of a regional-scale Rayleigh surface-wave ambient noise tomography that highlights
the three-dimensional shear-wave velocity structure of the region comprended from 34.5◦ S to 37.5◦ S of
the SVZ. We found that the seismic velocity anomalies are commonly related to the long-term crustal
structure attached to the gravity field. By the other hand, we note a strong correlation between
the presence of low-velocity anomalies and the extensional seismic sequence near 34.5◦ S after Maule
2010 earthquake. We interpret these velocity anomaly as a transient phenomena affecting the seismic
structure in the upper crust that could be related with the circulation of fluids accompanying the
aftershock sequence. In this scenario, ambient noise tomography is a effective method to investigate
the high-resolution velocity structure of the crust that can be applied to understand the transient and
long-lived processes ocurring in volcanic environments.
Keywords: Ambient noise tomography, Shear-wave velocity structure, Andean South Volcanic
Zone
∗
email: diegogonzalezv@udec.cl
†
email: anne.obermann@sed.ethz.ch
‡
email: bataille@udec.cl
§
email: stephen.miller@unine.ch
¶
email: matteo.lupi@unige.ch
1
Tomography of Chile: toward a new 3D velocity model
Keywords: travel-times tomography, velocity model, lithosphere
Bertrand Potin1,2 , Bernard Valette3 , Sebastián Araujo2,4 , Vadim Monteiller5

1
CSN, universidad de Chile, Santiago (bpotin@csn.uchile.cl),
2
ISTerre, Université Grenoble-Alpes, Grenoble, France,
3
ISTerre, IRD, Université Savoie-Mont-Blanc, Le-Bourget-du-Lac, France,
4
Universidad Politécnica Salesiana, Quito, Ecuador,
5
LMA, CNRS, Mareille, France.
In spite of advances made during the last two decades, most routine locations of earthquakes use
tabular models made of constant-velocity layers. These models are particularly ill-suited when one
addresses subduction zones and mountain ranges with their forelands, beneath which the depth
to the Moho — the major discontinuity in the lithosphere — can vary considerably. Besides, a
crust where the seismic velocity varies continuously, and not across sometimes arbitrary first-order
discontinuities, provides a much more stable model in which focal depths and focal mechanisms
can be better ascertained.
Accurate velocity models are generally carried out in restricted areas, from specific data-sets often
acquired during temporary experiments. Such models are intended to improve knowledge of the
deep structures in these regions of interest, therefore, they are not well-suited to be used by
observatories for earthquake locations in much larger areas. Large-scale models generally do not
achieve the same level of resolution, as both numbers of data and model parameters are restricted,
mainly due to stability issues in the inversion process.
We developed the INSIGHT algorithm in order to achieve large-scale tomography by inversion of
local earthquakes P- and S-waves arrival-times. This code is based on the law of large numbers and
allows us to manage large geographic domains and data-sets of more than one million data. Using
observations provided by permanent and temporary networks, we model the deep structures of
orogenes where seismic velocities vary continuously. This method was implemented in the Western
Alps, in Indonesia and in Ecuador, and achieved high-resolution models in which deep-structures
such as deep-thrusts, Moho and slab topography, magma chambers, or volcanoes structures were
recognizable. Shortly, this method will be implemented in Chile to carry out a large scale model
intended to improve earthquakes locations.
0 5 5.2 5
7
5.6 5.4 51
6
4.6 4.8
51
29
5.8
5.
5.8
29
5.6 6 73 1˚
73
5 5.8
5.4 6
10 5. 6 6
5.8 1 40
6
6 118 0˚
118
6 7. 95
6.2
95
15 6 5.8 6 7 2 162
Depth (km)
4
18
6.6
6.
7 29 51
6.4
29
8
51
20 5.8
-1˚
6
73
73
162
200
206
25
7.6
6.2
140
14
0
30 118
6
118
184
-2˚
95
95
35 6.4
7.2
6.2
6.6
7.4
6.2
7 29 51
29
40
51
7
73 -3˚
140
6.4 6.6
73
8 6.87
7.2
7.4
7.6
7.8
6.4
118
6.6
45 8 76
7
6.8
.86427
95
7.6
50 -4˚
-120 -100 -80 -60 -40 -20 0 -82˚ -81˚ -80˚ -79˚ -78˚ -77˚
x (km)
Figure 1: Two examples of results obtained with the INSIGHT algorithm. Left: cross section in
the Southern Alps, where several thrusts are revealed by ”high-on-low” velocity structures. The
resolution do not exceed a few kilometers. Right: Slab topography underneath Ecuador, deduced
from velocity anomalies and earthquakes relocation.
Deep Rock Impacts on Earthquake Surface Intensities.
Diego A. Inzunza1, Gonzalo A. Montalva2.
(1) MSc. Student, University of Concepcion, Chile, diegoinzunza@udec.cl

(2) Department of Civil Engineering, Faculty of Engineering, University of Concepcion, Chile,
gmontalva@udec.cl
Earthquake surface intensities are the main objective of much of the research done in earthquake
engineering, and its relevance has been identified as one the priorities for national development
in several countries. With economic growth come slender structures which include taller buildings,
bridges, and wind farms. Many of these new and increasingly abundant structures have structural
periods in the order of two seconds.
We present evidence of a site where shear-wave velocity (Vs) structure was calculated down to
500 meters. The first 100 meters are composed by soft sediments, and the underlaying rock is
composed by two different units up to that depth. The rock units have a shear-wave velocity of
approximately 1000 (m/s) for the shallower unit, and 3700 (m/s) for the deeper one.
Site amplification, accounting for much of the surface intensity during an earthquake, is largely
controlled by the impedance contrast. This is by the ratio of the rocks density and Vs to the density
and Vs of the overlaying sediments. Hence, the relevance of the rock characteristics is
paramount. Moreover, in the case of Concepción, Chile, we find that this double impedance
contrast has a strong effect on site amplification. Frequencies in the order of 0.5 Hz would,
according to our analysis, be controlled by the 1000 (m/s) rock layer between 100 and 460 m.
This finding matches the observed surface intensities recorded during the 2010 Maule
earthquake.
Keywords: Shear-wave velocity structure, Impedance contrast, site effect.

SINGLE STATION SIGMA IN CHILE
G. Montalva (1), N. Bastías (2)

(1)Associate Professor, Civil Engineering Dept., Universidad de Concepción, Concepción, Chile. gmontalva@udec.cl
(2)Partner, GENSIS Geotechnical Earthquake Engineering. nbastias@gensis.cl
Abstract
Single station sigma allows for a more realistic estimation of expected seismic demand. The use
of single-station sigma in a non-ergodic Probabilistic Seismic Hazard Analysis framework avoids
the double counting of uncertainties and allows for the more rigorous incorporation of site and
path effects into hazard estimations. We present a ground motion prediction model (GMPE)
developed from a catalogue that includes Chilean interplate and inslab events. The model is used
to evaluate the quantity of uncertainty that can be attributed to source, path, site effects, and what
can be treated as pure aleatory uncertainty. Our results show remarkable agreement with other
studies in some components of the overall uncertainty, but differences in other components, which
shed light into the areas that could be advanced in the future. Site characterization and prediction
of Vs30 through proxies represent a key challenge and where leap improvement can be achieved.
Keywords: Single-station sigma, ground motion prediction model, Chile.

A precision based regularization scheme for slip inversion:
Application to the Central Andean Subduction Zone
Francisco Ortega-Culaciati (1,2), Valeria Becerra (1), Mark Simons (3), Marcos Moreno (4),
Javier Ruiz (1), Diana Comte (1,2), Eduardo Contreras-Reyes (1), Andrei Maksymowicz (1),
Anne Socquet (5), Jorge Jara (5), and Daniel Carrizo (2) 
(1)
Universidad de Chile, Facultad de Ciencias Físicas y Matemáticas, Departamento de
Geofísica, Santiago, Chile (ortega.francisco@u.uchile.cl), (2) Advanced Mining
Technology Center, Facultad de Ciencias Físicas y Matemáticas, Departamento de
Geofísica, Santiago, Chile, (3) Seismological Laboratory, California Institute of
Technology, Pasadena, California 91125, USA., (4) Helmholtz Centre Potsdam, GFZ
German Research Centre for Geosciences, Potsdam, Germany., (5) ISTerre, Université
Grenoble Alpes, Grenoble, France.
Keywords: Fault slip inversion, Subduction Zone, Earthquake source
Abstract
Imaging subsurface slip behavior from surface observations is essential to increase our level of
understanding of the kinematics and physical processes controlling earthquake and tsunami
occurrence. As the estimation procedure is an inherently ill-posed problem, the adopted inverse
methodology to obtain such estimates, particularly the form of the a priori information, plays a key
role in this learning process. There are two general end member approaches to estimate the
distribution of slip on a fault that deals with the inherent instability of the inverse problem: An
unregularized, computationally expensive, fully Bayesian MCMC approach and a much more
expedient but biased optimization approach using some form of regularized least squares. We
focus our efforts in the latter approach. On the regularized inversion, the chosen form of a priori
information, or regularization scheme, will introduce a bias on fault slip estimates that needs to
be well understood to be able to achieve rigorous interpretation of the obtained slip values.
Here we discuss the effects that the a priori information implied by commonly used regularization
schemes has on slip estimates of fault behavior. Also, we propose a novel regularization scheme,
based on the precision of the model parameters in the unregularized problem, that accounts for
the spatial variability of the constraints provided by the observations (typically onland). The
proposed regularization scheme improves the stability and resolution of the inferred slip
distributions of fault behavior.
We illustrate our findings by analyzing synthetic cases in the Central Andean subduction
megathrust, and apply the methodology to perform an analysis of the seismic cycle in the region.
Slip Models of the Co- and Post- seismic stages Associated to the
April 1st, 2014 Pisagua (Mw 8.2) Earthquake Constrained by GPS
Observations
Valeria Becerra C., Francisco Ortega-Culaciati
Departamento de Geofísica, Universidad de Chile, Santiago, Chile
Keywords: Northern Chile, GPS, sparse solutions
On April 1, 2014, at 23:46:45 UTC a great earthquake (Mw 8.2) occurred near Pisagua-Chile, at
the segment of the subduction zone megathrust where the 1877 mega-earthquake (Mw 8.8 - 9.0)
defined a seismic gap that did not rupture for 118 years. Within the past two decades, crustal
deformation has been extensively monitored through campaign and continuous GPS networks in
northern Chile and southern Peru. Thus, providing constraints on the spatial and temporal
evolution of crustal deformation before, during and after the recent Pisagua Mw 8.2 earthquake.
We estimate co-seismic crustal deformation signals from 5-minute positional time series obtained
for the day of the main earthquake, and for the day of the mainly aftershock. Also, we obtain time
series of daily estimates of position for all the available continuous GPS data, from which we
isolate the post-seismic deformation signal, through finding sparse solutions for a family of base
functions defined by normalized b-splines integrals, centered at the co-seismic instant and with
different characteristic decay times. These functions better represent the transient behavior of the
post-seismic decay, and this method allows to find the post-seismic displacements without the
need to know the earthquake decay time.
We present our inferences of co- and post- seismic slip associated with the Pisagua earthquake
and compare it with current models of interseismic coupling. We analyze our results in the
context of the seismotectonics of the region in order to better characterize the mechanical
properties at the subduction megathrust and discuss its implications for earthquake potential in
the surrounding region.
Migración Sísmica asociada al terremoto de Iquique del 2014 (8.2 Mw)
Keywords: Migración, Sísmica, Iquique
Afiliación: Universidad de Concepción.
Nombre: Mauricio Leiva Bustamante
%Mediante el cálculo de los esfuerzos de Coulomb se caracteriza las zonas más probables de actividad
sísmica durante el enjambre sísmico asociado al terremoto de Iquique del 1 de Abril del 2014 (Mw = 8,2).
Considerando la acumulación de esfuerzos debido a fuentes sísmicas y asísmicas, el 67% de los
eventos se ubican en sitios donde la acumulación de esfuerzos de Coulomb es positivo. Los datos
cubren desde el 27 de Junio del 2005 hasta el 1 de Abril del 2014 y desde los 18° hasta los 21° de latitud
sur, y desde los 69 hasta los 72° de longitud Oeste. Son utilizados tres catálogos, (a) del Centro
Sismilógico Nacinal (CSN), que contiene la localización de los eventos, (b) del Harvard CMT que
contiene los mecanismos focales de la mayor parte de los eventos, y (c) del GeoForchungZentrum
(GFZ) que contiene los mecanismos focales de algunos eventos no incluidos en el catálogo del Harvard
CMT. Se simula la acumulación de stress durante el intersísmico al suponer movimientos en dos planos
de falla, con la geometría de la subducción en la zona de estudio. Se asume que la subducción
involucra movimiento inverso en la interfase superior y movimiento normal en la interfase inferior. La
interfase superior corresponde a la zona de contacto entre las placas Sudamericana y de Nazca. La
interfase inferior corresponde al límite de la respuesta frágil con la superficie dúctil del manto. Se
aproxima de manera similar el Evento Lento reportado por Ruiz et.al 2014, como una segunda fuente
asísmica de esfuerzos, que involucra movimiento únicamente en la interfase superior.
Reawakening of large earthquakes in South-Central Chile: The
2016 Mw7.6 Chiloé event
S. Ruiz1, M. Moreno2, D. Melnick3, F. del Campo4, P. Poli5, J.C. Baez4, F. Leyton4, R.
Madariaga6
1
Geophysics Department. Facultad de Ciencias Físicas y Matemáticas, Universidad de
Chile, Chile.
2
Deutsches GeoForschungsZentrum – GFZ, Section Lithosphere Dynamics,
Telegrafenberg, 14473 Potsdam, Germany
3
Instituto de Ciencias de la Tierra, TAQUACH, Universidad Austral, Chile.
4
Centro Sismológico Nacional, Facultad de Ciencias Físicas y Matemáticas, Universidad
de Chile, Chile.
5
Massachusetts Institute of Technology, USA.
6
Ecole Normale Superieure, Paris, France.
Corresponding author: Sergio Ruiz (sruiz@dgf.uchile.cl)
Abstract
On 25 December 2016, the Mw 7.6 Chiloé earthquake broke a small plate-boundary

asperity in the Central-South Chile midway along the 1960 Mw 9.5 Valdivia rupture zone.
To get insight about the decadal deformation trend, and its relationship with the Chiloé
earthquake, we combine geodetic, teleseismic and regional seismological data. The GPS
velocities increased at a continental-scale after the 2010 Maule earthquake, causing an
apparently abrupt end to the viscoelastic mantle relaxation following the 1960 earthquake
as well as a local increase in the degree of plate locking. The Chiloé earthquake occurred
within the region of increased locking, breaking a ~15-km-radius circular patch at~30 km
depth, located near the bottom of the seismogenic zone. We propose that the Chiloé
earthquake is a first sign of the seismic reawakening of the Valdivia segment, in response
to the interaction between postseismic viscoelastic relaxation and super-interseismic
locking of adjacent seismotectonic segments.
Keywords: Earthquakes - postseismic - seismic source

HIGH-FREQUENCY SCATTERING FROM
DEEP EARTH.
A. Aguilar1; K. Bataille2
March 9, 2017
1. email: andaguilar@udec.cl Sernageomin, Chile. Physics, Geophysics, Uni-

versity of Concepción, Concepcion, Chile.
2. Earth Sciences, University of Concepción, Concepción, Chile.
1 Abstract
The distribution of heterogeneities in inside the Earth, reflects the dynamic pro-
cesses that act on it. Large earthquakes radiate enough high-frequency waves,
above 3 Hz, for example the Bolivia earthquake (1994), causing coda waves
which are observable at all distances. These coda waves reflects the small-scale
dynamic processes within the Earth which are important to understand. The
origin of these scattering be discussed in light of the distribution of scatterers.
We model the coda of these waves using single-scattering theory, assuming scat-
terers distributed separately through the inner core, outer core, mantle, CMB
and ICB. Codas can be observed for all phases P, PKP, PKIKP, Pdiff, etc.,
that interact with the inner Earth, Broadband data from the M = 8.2 deep
(h = 630km) Bolivia earthquake, recorded at the distance range from 114◦ to
159◦ contain an important contribution of scattered PKP waves in the frequency
band 3 − 10Hz. These waves reflect the interaction with heterogeneities dis-
tributed through the mantle and core. For this model we consider the IASPEI
velocity model. We find that the data is consistent with the distribution of het-
erogeneities in the Inner Core Boundary and heterogeneities in the Core Mantle
Boundary to explain the coda.
1
Abstract Submission (Poster)
Correlation between seismic and infrasonic signals from Villarrica Volcano

during the last eruption crisis of 2015.
keywords: villarrica volcano,infrasound,seismic
Authors: Ximena González(1), José L. Palma(2)

Affiliation: (1) Departamento de Geofísica, Universidad de Concepción (2) Departamento de Ciencias de la
Tierra, Universidad de Concepción.
Villarrica Volcano (39.41°S, 71.94°W) is one of the most active volcanoes in the Southern Andes. It
shows a continuous outgassing activity through an open vent that holds a lava lake. The
characteristics of its activity allow the use the sound produced in the infrasonic (< 20 Hz) band as a
way to monitor the explosive and outgassing activity.
In 2014, a 6-element short-aperture infrasound array was installed ~8 km from the vent in order to
characterize the volcanic activity. This array has been recording the volcanic activity continuously until
today. These data were complemented with seismicity recorded by the observatory network
("Observatorio Volcanológico de los Andes del Sur" - OVDAS/SERNAGEOMIN) in order to analyze
the changes in amplitude through the 2014-2015/04 period, using the RMS (Root-mean-square) and
VASR (Volcanic-Acoustic Seismic Ratio) methods. In addition, this study considered a brief analysis of
the changes in signal amplitude and frequency content related to noise caused by bad weather. Our
results show that it is possible to separate the high-amplitude signal related to atmospheric noise from
the signal of volcanic origin. The correlation of both time series show that in the long-term (months)
there is a general good correlation but in a time-scale of weeks and days both series can exhibit either
similar or different trends. These results are currently being analyzed using visual observations
(webcams, photos and videos) in order to interpret possible causes of the variations in amplitude of
the seismicity and infrasound. Examples of these data during different phases of activity are provided.
Improving Volcanic Deformation monitoring using GPS Signal to
Noise Ratio.
Gustavo Pérez (1), Loreto Córdova (2), Cristian Mardones (2) and Daniel Díaz (1).
(1) Departamento de Geofísica, Facultad de Ciencias Físicas y Matemáticas, Universidad de Chile,
Santiago, Chile.
(2) Observatorio Volcanológico de los Andes del Sur (OVDAS), SERNAGEOMIN, Temuco, Chile.
Volcanic deformation is an important phenomenon to evaluate the potential eruption in

active eruptive systems, that’s why most of volcano observatories watch deformation data
from GPS in real time. However GPS data can contain outliers. For instances, the snow,
commonly present in many volcanoes, may cause this GPS data outlier. The Signal to Noise
Ratio (SNR) is the ratio of the signal power sent by satellite and the noise power of the
receptor. It has been showed the utility of the SNR for scientific surveys, but is usually use
to find outliers in GPS data. GPS data outliers are important to detect false information from
the volcanic deformation, which could be crucial in critical moments.
We used the analysis of SNR, using the SNR_CODE developed by Kristine Larson, to
identify snow outliers in deformation data from GPS stations in Villarrica and Llaima
volcanoes, both presenting a high level of activity during the last decades, in a populated
area of the lake district of the Araucania region. The results indicate some periods of low
SNR during the winter. In one GPS station the very low SNR was related with antenna
covered by snow, making deformation data useless during a long time period. In other GPS
station, some specific anomalies can be related with low SNR and not to changes in the
volcanic behavior, so these periods could be filtered from the GPS series. SNR analysis has
proved to be a useful tool to check the GPS deformation data on Chilean volcanoes.
Geometría de basamento en la zona de antearco de la región de
Arica y Parinacota asociado a la geomorfología superficial,
utilizando el método gravimétrico.
Daniel Cabrera A., Andrei Maksymowicz J., Daniel Diaz A.
Departamento de Geofísica, Facultad de Ciencias Físicas y Matemáticas, Universidad de Chile.
Keyboards: Arica y Parinacota, gravimetría, tectónica.
El extremo norte de Chile es una zona árida ubicada en la parte central de los Andes
Centrales. En las cercanías de la ciudad de Arica (18°28’30’’S, 70°18’52’’O) se pueden
distinguir las siguientes unidades morfoestructurales ubicadas de Oeste a Este: Cordillera
de la Costa, Depresión Central, Cordillera Occidental con el arco volcánico actual y Altiplano
(Muñoz y Charrier, 1996). La deformación que provoca el alzamiento del Altiplano es
predominantemente Neogeno (Isacks, 1988, McQuarrie et al., 2005) y se puede ver su
manifestación a través de flexuras y fallas existentes en el área estudiada. En la
Precordillera la deformación afecta principalmente a la formacion Oxaya. En la Depresión
Central la deformación se expresa por medio de flexuras y fallas menores que afectan la
formación El Diablo (García et al., 2002). Al sur de la región se encuentra la flexura
Humayani formada por el movimiento de la falla ciega inversa subvertical Taltape (García
et al., 2002). Ambas fallas sitúan basamento mesozoico sobre la formacion Azapa lo que
indica fallamiento de edad Oligocena (García et al., 2002).
Para comprender el verdadero alcance de estas deformaciones se ha desarrollado un
estudio gravimétrico en la región de Arica y Parinacota con el cual se busca relacionar las
deformaciones observadas en las secuencias oligocenas – neogenas con deformaciones
en el basamento mesozoico – paleoceno. El método gravimétrico permite distinguir cambios
de densidad a lo largo de la zona estudiada, en esta ocasión se espera observar
manifestaciones en la señal de la interfaz basamento/relleno sedimentario y volcánico-
sedimentario en función del contraste de densidades existente entre estos, así como la
existencia y alcance de fallas maestras no observadas en superficie. Para constreñir la
información geológica y el modelo obtenido por el método gravimétrico se utilizará un perfil
sísmico de reflexión realizado en la misma ubicación del perfil gravimétrico.
Magallanes’s seismicity study
María Constanza Flores1*, Paula Manríquez2, Sebastián Carrasco2
1 Departamento de Geofísica, Universidad de Chile.

2 Centro Sismológico Nacional, Universidad de Chile.
* maria.flores.veliz@ug.uchile.cl
Keywords: local seismicity, Magallanes-Fagnano fault system.
Monitoring of the seismicity in Magallanes has been an ongoing challenge to the scientific
community due to the underling geographic features in the region. It is for this reason that
we performed a study in Magallanes to characterize it’s seismicity between 2015 and 2016,
making use of Centro Sismológico Nacional permanent network.
Detection of earthquakes was performed using a trigger search algorithm implemented in
Python, producing a preliminary catalog which was then manually improved. Afterwards the
events where localized using Seiscomp3 (Weber et al., 2007), from which we obtained a
169 earthquake catalog with local magnitudes ranging from 1 to 5.6 (Ml) roughly. A thorough
analysis of this catalog was carried out studying variables such as azimuthal gap value,
localization errors, number of phases used, among others, all of this to appraise said catalog
and also to assess the spatial distribution of the actual network. Also, with the idea of a
future waveform modeling, the signal to noise ratio (SNR) of the main events found was
calculated.
Some events that stand out inside the final catalog are: Microseismicity in the vicinity of
Puerto Natales, significant earthquakes (Ml>4) associated to the subduction of the Antarctic
plate under the Scotia plate, and finally a group of events that may be related to the
Magallanes-Fagnano fault system, which would reveal the active nature of this feature and
also of the region as a whole. Furthermore, the analysis of the catalog variables suggests
an enhancement of the network spatial distribution, so more reliable solutions can be
obtained and also to improve monitoring in the area.
Figure: Identified seismicity in the Magallanes region during 2015 and 2016.
Run-up of the 1922 earthquake gap
Miguel Medina1*, Matías Mocanu2, Sebastián Riquelme3, Mauricio Fuentes2
1 Departamento de Geofísica, Universidad de Chile.
2 Departamento de Geología, Universidad de Chile.
3 Centro Sismológico Nacional, Universidad de Chile.
* miguel.medina@dgf.uchile.cl
Keywords: Near-field tsunami, Stochastic sources, Subduction zone.
Great subduction earthquakes are prone to produce extensive tsunamis that can cause
destruction and fatalities. In Chile there’s a broad history of destruction caused by near-field
tsunamis, even recently, such as the 2010 Maule earthquake and the 2015 Mw 8.3 Illapel
earthquake. In this regard, there’s some regions in Chile with great tsunamigenic potential,
one of them being the seismic gap defined by the rupture area of the 1922 earthquake.
In order to mitigate damages in the coast, tsunami modeling can be performed, but without
the inclusion of source complexity, uniform earthquakes will always underestimate the
tsunami extent in the near-field. With that in mind, we generate random realistic earthquake
scenarios following a k2 distribution (Mai & Beroza, 2002) to assess the tsunami hazard in
the 1922 earthquake region.
Here we use the subduction interface complexity, following Slab1.0 (Hayes et al. 2012), in
order to define the earthquake source geometry, that is later used to generate 90 random k2
earthquake scenarios of magnitude Mw 8.8, which we estimate to be the maximum
magnitude for an earthquake in the region. For each event, the tsunami numerical modeling
is carried out with initial ocean conditions which are assumed to be identical to the vertical
deformation of the bottom of the ocean, but we also include the horizontal contribution.
The results obtained show a run-up envelope which exhibits a great variability among the
Chilean coast, ranging from a few meters to run-ups greater than 20 meters for some
scenarios. This helps to characterize areas more likely to be affected in a potential tsunami.
Figure: Left: Earthquake source geometry. Right: Tsunami run-up envelope obtained.

Book 3GSEV 2017

Uploaded by

Copyright:

Available Formats

Book 3GSEV 2017

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Book 3GSEV 2017

Uploaded by

Copyright:

Available Formats

THIRD COLLOQUIUM OF

TERCER COLOQUIO DE SEÑALES

20 -22 MARZO 2017

17:00 -20:00 Acreditación

Martes 21.03 – 3GSEV

08:30 -09:00 Acreditación

Miércoles 22.03 – 3GSEV

Plenaria Sergio Barrientos

Engineering Earthquakes: Intensities and Velocity Models

Authot – TALK Mail Abstract Title

Authot – POSER Mail Abstract Title

El Centro Sismológico Nacional de la Universidad de Chile (CSN), continuación del Servicio

Este avance en la observación sismológica ha permitido registrar, de manera nunca antes

Otros avances de importancia han sido la implementación de la estimación de magnitud y

Importantes esfuerzos se continúan realizando en la caracterización de sitios donde se

Analysis of large GNSS networks: from daily to real-time

Keywords: Tsunami, run-up, Seismology.

Corresponding author: Mauricio Antonio Fuentes Serrano

Juan Fernandez as a T-Phase detector for South-American

(1) Department of Geophysics, University of Chile, Santiago, Chile

(2) Department of Geology, University of Chile, Santiago, Chile

The T-phase has been suggested as an earthquake discriminator, between explosions

generated from South-American earthquakes, using the Centro Sismológico Nacional

monitor the South-American earthquakes and its tsunamigenic potential.

Keywords: T-Phase; South American earthquake; Juan Fernandez broadband station

Slow earthquake, Froude number, wave amplification.

Angela Carrillo P., Ignacia Calisto B.

Keywords: geodesy, vorticity & Liquiñe-Ofqui

Francisco García, Andrés Tassara & Vicente Yáñez

Departamento Ciencias de la Tierra, Universidad De Concepción

Departamento de Matemáticas y Fı́sica, Facultad de Ingenierı́a, Universidad Católica de Temuco.

Keywords: Earthquake-volcano interactions, Numerical simulations.

Palabras clave: terremoto reactivación volcánica

El arco volcánico chileno, y en especial el Segmento de la Zona Sur (Stern 1994) ha

David Mora y Andrés Tassara

A través de modelación analítica se investiga la magnitud de la descompresión en la

Palabras claves: descompresión, esfuerzos incrementales y flexurales.

C. U. Villalobos a, E. M. Ovalle b, M. A. Bravo c, A. J. Foppiano b

Keywords: Ionospheric characteristics; TEC; Earthquake precursors.

Franco E. Vera 1*, José L. Palma1, Jeff Johnson2

1. Department of Earth Sciences, University of Concepción, Concepción, Chile.

Key words: Mixed avalanches, ice/snow-lava interaction, drone mapping, lahars.

El terremoto de Tohoku-Oki, Japón, Mw=9.1, ocurrido el 11 de marzo de 2011, ocasionó un gi-

Renzo Mancini1,2 and Daniel Dı́az1,2

In this research we measured 20 broadband magnetotelluric stations in the vicinity

A dimensionality analysis was carried out, considering induction arrows, phase

Preliminary results indicate the presence of a geothermal system to the southwest

Keywords: 3D inversion, Volcanoes, magnetotelluric

Keywords: Rhyolitic volcano system, structural geology, strike-slip fault system.

Paulina Köhler1, Andrés Tassara1, Carlos Cardona2

It is widely accepted in seismology that displacement spectra of earthquakes is well explained by

(a) National Seismological Center, University of Chile

Keywords: strong ground-motions, earthquake early warning.

¹Department of Geophysics, University of Chile, Santiago, Chile

KEYWORDS: Intraplate and interplate earthquakes; Dynamic inversion; Stress drop.

Keywords: Tectonic signals, GPS, crustal deformation.

Maule earthquake, seismic tomography, Arauco peninsula.

(1) Department of Earth Sciences, University of Concepción, Biobio, Chile

Keywords: Lahars, infrasound, Villarrica volcano