Magma recharge patterns control eruption styles and magnitudes at Popocatépetl volcano (Mexico)

Diffusion chronometry has produced petrological evidence that magma recharge in mafic to intermediate systems can trigger volcanic eruptions within weeks to months. However, less is known about longer-term recharge frequencies and durations priming magma reservoirs for eruptions. We use Fe-Mg diffusion modeling in orthopyroxene to show that the duration, frequency, and timing of pre-eruptive recharge at Popocatépetl volcano (Mexico) vary systematically with eruption style and magnitude. Effusive eruptions are preceded by 9–13 yr of increased recharge activity, compared to 15–100 yr for explosive eruptions. Explosive eruptions also record a higher number of individual recharge episodes priming the plumbing system. The largest explosive eruptions are further distinguished by an ~1 yr recharge hiatus directly prior to eruption. Our results offer valuable context for the interpretation of ongoing activity at Popocatépetl, and seeking similar correlations at other arc volcanoes may advance eruption forecasting by including constraints on potential eruption size and style.

A unified probabilistic framework for volcanic hazard and eruption forecasting

The main purpose of this article is to emphasize the importance of clarifying the probabilistic framework adopted for volcanic hazard and eruption forecasting. Eruption forecasting and volcanic hazard analysis seek to quantify the deep uncertainties that pervade the modeling of pre-, sin-, and post-eruptive processes. These uncertainties can be differentiated into three fundamental types: (1) the natural variability of volcanic systems, usually represented as stochastic processes with parameterized distributions (aleatory variability); (2) the uncertainty in our knowledge of how volcanic systems operate and evolve, often represented as subjective probabilities based on expert opinion (epistemic uncertainty); and (3) the possibility that our forecasts are wrong owing to behaviors of volcanic processes about which we are completely ignorant and, hence, cannot quantify in terms of probabilities (ontological error). Here we put forward a probabilistic framework for hazard analysis recently proposed by Marzocchi and Jordan (2014), which unifies the treatment of all three types of uncertainty. Within this framework, an eruption forecasting or a volcanic hazard model is said to be complete only if it (a) fully characterizes the epistemic uncertainties in the model's representation of aleatory variability and (b) can be unconditionally tested (in principle) against observations to identify ontological errors. Unconditional testability, which is the key to model validation, hinges on an experimental concept that characterizes hazard events in terms of exchangeable data sequences with well-defined frequencies. We illustrate the application of this unified probabilistic framework by describing experimental concepts for the forecasting of tephra fall from Campi Flegrei. Eventually, this example may serve as a guide for the application of the same probabilistic framework to other natural hazards.

Volcano monitoring and hazard assessments in Chile

Volcanism in Chile occurs in a variety of tectonic settings but mostly in the context of oceanic-continental plate collision, including 92 potentially active volcanoes. There have been more than 30 documented eruptions in the last few centuries. The Servicio Nacional de Geología y Minería (SERNAGEOMIN) is a statutory agency of the Government of Chile responsible for volcano monitoring and hazard assessments across the country. After the impacts derived from volcanic activity at the end of the 20th century, SERNAGEOMIN created the Volcano Hazards Program and the Observatorio Volcanológico de Los Andes del Sur (OVDAS). Despite this effort, most volcanoes in Chile remained unmonitored. In 2008, the aftermath of the eruption of Chaitén led to a nationwide program in order to improve eruption forecasting, development of early warning capabilities and our state of readiness for volcanic impacts through hazard assessments. In the last decade responses to volcanic crises have been indubitably successful providing technical advice before and during volcanic eruptions. El volcanismo en Chile ocurre en una amplia variedad de regímenes tectónicos, aunque principalmente en el contexto de la colisión de placas. Alrededor de 92 volcanes son considerados potencialmente activos y más de 30 presentan actividad histórica documentada en los últimos siglos. El Servicio Nacional de Geología y Minería (SERNAGEOMIN) es la agencia gubernamental responsable de la evaluación de peligros y monitoreo de la actividad volcánica en el país. Como consecuencia de los impactos derivados de las erupciones volcánicas ocurridas hacia finales del siglo pasado, SERNAGEOMIN creó el Programa de Riesgo Volcánico y el Observatorio Volcanológico de los Andes del Sur (OVDAS). No obstante, a pesar de este esfuerzo la mayoría de los volcanes en Chile se mantenían sin monitoreo. Luego de los impactos derivados de la erupción del volcán Chaitén en 2008, un nuevo programa nacional fue creado con el fin de fortalecer la vigilancia y la evaluación de los peligros volcánicos en el país. En la última década, la respuesta a crisis volcánicas ha sido exitosa, proporcionando apoyo técnico en forma previa y durante erupciones.

Temporal Changes in Seismic Anisotropy as a New Eruption Forecasting Tool

<p>The orientation of crustal anisotropy changed by ~80 degrees in association with the 1995/96 eruption of Mt. Ruapehu volcano, New Zealand. This change occurred with a confidence level of more than 99.9%, and affects an area with a radius of at least 5 km around the summit. It provides the basis for a new monitoring technique and possibly for future mid-term eruption forecasting at volcanoes. Three deployments of seismometers were conducted on Mt. Ruapehu in 1994, 1998 and 2002. The fast anisotropic direction was measured by a semi-automatic algorithm, using the method of shear wave splitting. Prior to the eruption, a strong trend for the fast anisotropic direction was found to be around NW-SE, which is approximately perpendicular to the regional main stress direction. This deployment was followed by a moderate phreatomagmatic eruption in 1995/96, which ejected material with an overall volume of around 0.02-0.05 km3. Splitting results from a deployment after the eruption (1998) suggested that the fast anisotropic direction for deep earthquakes (>55 km) has changed by around 80 degrees, becoming parallel to the regional stress field. Shallow earthquakes (<35 km) also show this behaviour, but with more scatter of the fast directions. Another deployment (2002) covered the exact station locations of both the 1994 and the 1998 deployments and indicates further changes. Fast directions of deep events remain rotated by 80 degrees compared to the pre-eruption direction, whereas a realignment of the shallow events towards the pre-eruption direction is observed. The interpretation is that prior to the eruption, a pressurised magma dike system overprinted the regional stress field, generating a local stress field and therefore altering the fast anisotropic direction via preferred crack alignment. Numerical modelling suggests that the stress drop during the eruption was sufficient to change the local stress direction back to the regional trend, which was then observed in the 1998 experiment. A refilling and pressurising magma dike system is responsible for the newly observed realignment of the fast directions for the shallow events, but is not yet strong enough to rotate the deeper events with their longer delay times and lower frequencies. These effects provide a new method for volcano monitoring at Mt. Ruapehu and possibly at other volcanoes on Earth. They might, after further work, serve as a tool for eruption forecasting at Mt. Ruapehu or elsewhere. It is therefore proposed that changes in anisotropy around other volcanoes be investigated.</p>

Temporal Changes in Seismic Anisotropy as a New Eruption Forecasting Tool

<p>The orientation of crustal anisotropy changed by ~80 degrees in association with the 1995/96 eruption of Mt. Ruapehu volcano, New Zealand. This change occurred with a confidence level of more than 99.9%, and affects an area with a radius of at least 5 km around the summit. It provides the basis for a new monitoring technique and possibly for future mid-term eruption forecasting at volcanoes. Three deployments of seismometers were conducted on Mt. Ruapehu in 1994, 1998 and 2002. The fast anisotropic direction was measured by a semi-automatic algorithm, using the method of shear wave splitting. Prior to the eruption, a strong trend for the fast anisotropic direction was found to be around NW-SE, which is approximately perpendicular to the regional main stress direction. This deployment was followed by a moderate phreatomagmatic eruption in 1995/96, which ejected material with an overall volume of around 0.02-0.05 km3. Splitting results from a deployment after the eruption (1998) suggested that the fast anisotropic direction for deep earthquakes (>55 km) has changed by around 80 degrees, becoming parallel to the regional stress field. Shallow earthquakes (<35 km) also show this behaviour, but with more scatter of the fast directions. Another deployment (2002) covered the exact station locations of both the 1994 and the 1998 deployments and indicates further changes. Fast directions of deep events remain rotated by 80 degrees compared to the pre-eruption direction, whereas a realignment of the shallow events towards the pre-eruption direction is observed. The interpretation is that prior to the eruption, a pressurised magma dike system overprinted the regional stress field, generating a local stress field and therefore altering the fast anisotropic direction via preferred crack alignment. Numerical modelling suggests that the stress drop during the eruption was sufficient to change the local stress direction back to the regional trend, which was then observed in the 1998 experiment. A refilling and pressurising magma dike system is responsible for the newly observed realignment of the fast directions for the shallow events, but is not yet strong enough to rotate the deeper events with their longer delay times and lower frequencies. These effects provide a new method for volcano monitoring at Mt. Ruapehu and possibly at other volcanoes on Earth. They might, after further work, serve as a tool for eruption forecasting at Mt. Ruapehu or elsewhere. It is therefore proposed that changes in anisotropy around other volcanoes be investigated.</p>

A Unified Probabilistic Framework for Volcanic Hazard and Eruption Forecasting

The main purpose of this article is to emphasize the importance of clarifying the probabilistic framework adopted for volcanic hazard and eruption forecasting. Eruption forecasting and volcanic hazard analysis seeks to quantify the deep uncertainties that pervade the modeling of pre-, sin- and post-eruptive processes. These uncertainties can be differentiated into three fundamental types: (1) the natural variability of volcanic systems, usually represented as stochastic processes with parameterized distributions (aleatory variability); (2) the uncertainty in our knowledge of how volcanic systems operate and evolve, often represented as subjective probabilities based on expert opinion (epistemic uncertainty); and (3) the possibility that our forecasts are wrong owing to behaviors of volcanic processes about which we are completely ignorant and, hence, cannot quantify in terms of probabilities (ontological error). Here we put forward a probabilistic framework for hazard analysis recently proposed by Marzocchi &amp; Jordan (2014), which unifies the treatment of all three types of uncertainty. Within this framework, an eruption forecasting or a volcanic hazard model is said to be complete only if it (a) fully characterizes the epistemic uncertainties in the model's representation of aleatory variability and (b) can be unconditionally tested (in principle) against observations to identify ontological errors. Unconditional testability, which is the key to model validation, hinges on an experimental concept that characterizes hazard events in terms of exchangeable data sequences with well-defined frequencies. We illustrate the application of this unified probabilistic framework by describing experimental concepts for the forecasting of tephra fall from Campi Flegrei. Eventually, this example may serve as a guide for the application of the same probabilistic framework to other natural hazards.

Use of Local Seismic Network in Analysis of Volcano-Tectonic (VT) Events Preceding the 2017 Agung Volcano Eruption (Bali, Indonesia)

This study provides an attempt to analyze the pre-eruptive seismicity events for volcano eruption forecasting. After more than 50 years of slumber, Agung volcano on Bali Island erupted explosively, starting on November 21, 2017. The eruption was preceded by almost 2 months of significant increase of recorded seismicity, herein defined as “seismic crisis.” Our study provides the first analysis of VT events using data from eight local seismic stations deployed by the Center for Volcanology and Geological Hazard Mitigation of Indonesia (CVGHM) to monitor the Agung Volcano activity. In total, 2,726 Volcano-Tectonic (VT) events, with 13,023 P waves and 11,823 S wave phases, were successfully identified between October 18 and November 30, 2017. We increased the accuracy of the hypocenter locations of these VT events using a double-difference (DD) relative relocation and a new velocity model appropriate to the subsurface geological conditions of Agung volcano. We found two types of seismicity during the recording period that represent the VT events relating to fracture network reactivation due to stress changes (during the seismic crisis) and magma intrusion (after the seismic crisis). The characteristics of each event type are discussed in terms of Vp/Vs values, phase delay times, seismic cluster shapes, and waveform similarity. We interpret that the upward migrating magma reached a barrier (probably a stiff layer) which prohibited further ascent. Consequently, magma pressurized the zone above the magma chamber and beneath the barrier, reactivated the fracture zone between Agung and Batur volcanoes, and caused the seismic crisis since September 2017. In early November 2017, the barrier was finally intruded, and magma and seismicity propagated toward the Agung summit. This reconstruction provides a better depth constraint as to the previous conceptual models and explains the long delay (∼10 weeks) between the onset of the seismic crisis and the eruption. The distinction between the fracture reactivation and magma intrusion VT events observed in this study is significant for eruption forecasting and understanding the subsurface structure of the magmatic system. Based on the results obtained in this study, we emphasize the importance of prompt analysis (location and basic seismic characteristics) of the seismic crisis preceding the Agung eruption.

Reconciling the location of lava domes and eruption centers in Paleocene-Eocene calderas in northern Chile

&lt;p&gt;In the Atacama Desert, at the Precordillera of northern Chile, a series of Paleocene-Eocene caldera deposits and ring-faults are exceptionally well-preserved&lt;sup&gt;1&lt;/sup&gt;. Here we aim to build on previous mapping efforts to consider the location, timing and style of pre, syn and post caldera volcanism in the region. We focus on the partially nested caldera complexes of Lomas Bayas and El Durazno&lt;sup&gt;2,3&lt;/sup&gt; where deposits record several stages of caldera evolution (pre-collapse, collapse/intra-caldera and extra-caldera, resurgence and post-collapse eruptive deposits). The pre-caldera basement is a thick sequence of early Paleocene mafic lavas&lt;sup&gt;4, 5&lt;/sup&gt;. The caldera complex formed between around 63 and 54 Ma&lt;sup&gt;4, 5&lt;/sup&gt;. Both calderas constitute subcircular structures approximately 13 km in diameter and are cut by several NNW to NNE-trending felsic dikes which are spatially related to felsic domes interpreted as resulting from post caldera formation unrest&lt;sup&gt;1,&lt;/sup&gt;&lt;sup&gt;4&lt;/sup&gt;. These calderas have been interpreted as part of the Carrizalillo megacaldera complex&lt;sup&gt;2 &lt;/sup&gt;. We combine field observations, such as the attitude of dikes, as well as information on their dimension and composition, the size, location and composition of domes and lava flows, as well as the evidence of the regional stress field operating during the caldera evolution from measurements of fault kinematics. This data will be used as the input to finite element method models to investigate the effect of nested caldera geometry, ring-faults and crustal heterogeneities on the location of domes and eruptive centers generated during caldera unrest. The results will be potentially useful for constraining models of eruption forecasting during periods of unrest in calderas and ore deposition models which have been shown to be linked to caldera structure and magma emplacement.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;References&lt;/strong&gt;&lt;/p&gt;&lt;p&gt;&lt;sup&gt;1 &lt;/sup&gt;Rivera, O. and Falc&amp;#243;n, M. (2000). Calderas tipo colapso-resurgentes del Terciario inferior en la Pre-Cordillera de la Regi&amp;#243;n de Atacama: Emplazamiento de complejos volcano-plut&amp;#243;nicos en las cuencas volcano-tect&amp;#243;nicas extensionales Hornitos y Indio Muerto: IX Congreso Geol&amp;#243;gico Chileno, v. 2.&amp;#160;Soc. Geol. de Chile, Puerto Varas.&lt;/p&gt;&lt;p&gt;&lt;sup&gt;2 &lt;/sup&gt;Rivera, O., and Mpodozis, C. (1994). La megacaldera Carrizalillo y sus calderas anidadas: Volcanismo sinextensional Cret&amp;#225;cico Superior-Terciario inferior en la Precordillera de Copiap&amp;#243;, paper presented at VII Congreso Geol&amp;#243;gico Chileno. Acad. de Cienc. del Inst. Chilecol. de Geol. de Chile, Concepci&amp;#243;n.&lt;/p&gt;&lt;p&gt;&lt;sup&gt;3 &lt;/sup&gt;Rivera, O. (1992). El complejo volcano-plut&amp;#243;nico Paleoceno-Eoceno del Cerro Durazno Alto: las calderas El Durazno y Lomas Bayas, Regi&amp;#243;n de Atacama, Chile. Tesis Departamento de Geolog&amp;#237;a, Universidad de Chile, 242. (Unpublished).&lt;/p&gt;&lt;p&gt;&lt;sup&gt;4 &lt;/sup&gt;Ar&amp;#233;valo, C. (2005). Carta Los Loros, Regi&amp;#243;n de Atacama. Servicio Nacional de Geolog&amp;#237;a y Miner&amp;#237;a, Carta Geol&amp;#243;gica de Chile, 92, 1(100.000), 53 p.&lt;/p&gt;&lt;p&gt;&lt;sup&gt;5 &lt;/sup&gt;Iriarte, S., Ar&amp;#233;valo, C., Mpodozis, C. (1999). Mapa Geol&amp;#243;gico de la Hoja La Guardia, Regi&amp;#243;n de Atacama. Servicio Nacional de Geolog&amp;#237;a y Miner&amp;#237;a. Mapas Geol&amp;#243;gicos, 13, 1(100.000).&lt;/p&gt;