Search

SUMMARY Seven years after the beginning of a massive wastewater injection project in eastern Colombia, local earthquake activity increased significantly. The field operator and the Colombian Geological Survey immediately reinforced the monitoring of the area. Our analysis of the temporal evolution of the seismic and injection data together with our knowledge of the geological parameters of the region indicate that the surge of seismicity is being induced by the re-injection of produced water into the same three producing reservoirs. Earthquake activity began on known faults once disposal rates had reached a threshold of ∼2 × 106 m3 of water per month. The average reservoir pressure had remained constant at 7.6 MPa after several years of production, sustained by a large, active aquifer. Surface injection pressures in the seismically active areas remain below 8.3 MPa, a value large enough to activate some of the faults. Since faults are mapped throughout the region and many do not have seismicity on them, we conclude that the existence of known faults is not the only control on whether earthquakes are generated. Stress conditions of these faults are open to future studies. Earthquakes are primarily found in four clusters, located near faults mapped by the operator. The hypocentres reveal vertical planes with orientations consistent with focal mechanisms of these events. Stress inversion of the focal mechanisms gives a maximum compression in the direction ENE-WSW, which is in agreement with borehole breakout measurements. Since the focal mechanisms of the earthquakes are consistent with the tectonic stress regime, we can conclude that the seismicity is resulting from the activation of critically stressed faults. Slip was progressive and seismic activity reached a peak before declining to few events per month. The decline in seismicity suggests that most of the stress has been relieved on the main faults. The magnitude of a large majority of the recorded earthquakes was lower than 4, as the pore pressure disturbance did not reach the mapped large faults whose activation might have resulted in larger magnitude earthquakes. Our study shows that a good knowledge of the local fault network and conditions of stress is of paramount importance when planning a massive water disposal program. These earthquakes indicate that while faults provide an opportunity to dispose produced water at an economically attractive volume–pressure ratio, the possibility of induced seismicity must also be considered.

International audience; The 2015 reactivation of the Cotopaxi volcano urges us to understand the complex eruptive dynamics of Cotopaxi for better management of a potential major crisis in the near future. Cotopaxi has commonly transitioned from andesitic eruptions of strombolian style (lava flows and scoria ballistics) or nuées ardentes (pyroclastic flows and ash falls) to highly explosive rhyolitic ignimbrites (pumiceous pyroclastic flows), which entail drastically different risks. To better interpret geophysical and geochemical signals, Cotopaxi magma storage conditions were determined via existing phase-equilibrium experiments that used starting materials chemically close to the Cotopaxi andesites and rhyolites. The results suggest that Cotopaxi's most mafic andesites (last erupted products) can be stored over a large range of depth from ~7 km to ≥16 km below the summit (pressure from ~200 to ≥400 MPa), 1000 °C, NNO +2, and contain 4.5–6.0±0.7 wt% H2O dissolved in the melt in equilibrium with ~30–40% phenocrysts of plagioclase, two pyroxenes, and Fe-Ti oxides. These mafic andesites sometimes evolve towards more silicic andesites by cooling to 950 °C. Rhyolitic magmas are stored at 200–300 MPa (i.e. ~7–11 km below the summit), 750 °C, NNO +2, and contain ~6–8 wt% H2O dissolved in a nearly aphyric melt (

Understanding the plumbing system structure below volcanoes and the storage conditions (temperature, pressure, volatile content and oxygen fugacity) of erupted magmas is of paramount importance for eruption forecasting and understanding of the factors controlling eruptive dynamics. Phase equilibria experiments have been performed on a Tungurahua andesite (Ecuador) to shed light on the magmatic conditions that led to the July-August 2006 eruptions and the parameters that controlled the eruptive dynamics. Crystallization experiments were performed on a representative August 2006 mafic andesite product between 950 and 1025A degrees C, at 100, 200 and 400 MPa and NNO + 1 and NNO + 2 (where NNO is nickel-nickel oxide buffer), and water mole fractions in the fluid (XH2O) from 0 center dot 3 to 1 (water-saturation). Comparison of the natural phenocryst assemblage, proportions and phenocryst compositions with our experimental data indicates that the natural andesite experienced two levels of ponding prior to the eruption. During the first step, the magma was stored at 400 MPa (15-16 km), 1000A degrees C, and contained c. 6 wt % dissolved H2O. In the second step, the magma rose to a confining pressure of 200 MPa (8-10 km), where subsequent cooling (to 975A degrees C) and water-degassing of the magma led to the crystallization of reversely zoned rims on pre-existing phenocrysts. The combination of these processes induced oxidation of the system and overpressure of the reservoir, triggering the July 2006 eruption. The injection of a new, hot, volatile-rich andesitic magma from 15-16 km into the 200 MPa reservoir shortly before the eruption was responsible for the August 2006 explosive event. Our results highlight the complexity of the Tungurahua plumbing system in which different magmatic reservoirs can coexist and interact in time and are the main controlling factors of the eruptive dynamics.

This paper presents the results of 7 years (Aug. 1999-Oct. 2006) of SO2 gas measurements during the ongoing eruption of Tungurahua volcano, Ecuador. From 2004 onwards, the operation of scanning spectrometers has furnished high temporal resolution measurements of SO2 flux, enabling this dataset to be correlated with other datasets, including seismicity. The emission rate of SO2 during this period ranges from less than 100 to 35,000 tonnes/day (t d(-1)) with a mean daily emission rate of 1458 t d(-1) and a standard deviation of +/- 2026 t d(-1). Average daily emissions during inferred explosive phases are about 1.75 times greater than during passive degassing intervals. The total amount of sulfur emitted since 1999 is estimated as at least 1.91 Mt, mostly injected into the troposphere and carried westwards from the volcano. Our observations suggest that the rate of passive degassing at Tungurahua requires SO2 exsolution of an andesitic magma volume that is two orders of magnitude larger than expected for the amount of erupted magma. Two possible, and not mutually exclusive, mechanisms are considered here to explain this excess degassing: gas flow through a permeable stagnantmagma-filled conduit and gas escape from convective magma overturning in the conduit. We have found that real-time gas monitoring contributes significantly to better eruption forecasting at Tungurahua, because it has provided improved understanding of underlying physical mechanisms of magma ascent and eruption.

Renewed seismic activity of Cotopaxi, Ecuador, began in January 2001 with the increased number of long-period (LP) events, followed by a swarm of volcano-tectonic (VT) earthquakes in November 2001. In late June 2002, the activity of very-long-period (VLP) (2 s) events accompanying LP (0.5–1 s) signals began beneath the volcano. The VLP waveform was characterized by an impulsive signature, which was accompanied by the LP signal showing non-harmonic oscillations. We observed temporal changes of both the VLP and LP signals from the beginning until September 2003: The VLP signal gradually disappeared and the LP signal characterized by decaying harmonic oscillations became dominant. Assuming possible source geometries, we applied a waveform inversion method to the observed waveforms of the largest VLP event. Our inversion and particle motion analyses point to volumetric changes of a sub-vertical crack as the VLP source, which is located at a depth of 2–3 km beneath the northeastern flank. The spectral analysis of the decaying harmonic oscillations of LP events shows frequencies between 2.0 and 3.5 Hz, with quality factors significantly above 100. The increased VT activity and deformation data suggest an intrusion of magma beneath the volcano. A release of gases with small magma particles may have repetitively occurred due to the pressurization, which was caused by sustained bubble growth at the magma ceiling. The released particle-laden gases opened a crack above the magma system and triggered the resonance of the crack. We interpret the VLP and LP events as the gas-release process and the resonance of the crack, respectively.

Tungurahua Volcano in the Ecuadorian Andes is a large andesitic stratovolcano (5023 m) that has been erupting mildly since 1999. We studied the three-dimensional (3-D) P-wave velocity (Vp) structure beneath the volcano down to 5 km below the summit. We inverted 1708 P-wave first-arrival times from 263 volcano-tectonic (VT) earthquakes recorded by 5 to 10 short-period vertical seismic stations on the volcano from August 1999 to May 2003. A tomographic inversion method was used to image the velocity structure, in which first-arrival times were calculated with a finite-difference method. The Root Mean Square of the arrival time residuals (RMS) was reduced by 43% after running 10 iterations from the initial RMS of 0.15 s. The relocated hypocenters in our model are tightly clustered along a vertical structure at depths between sea level and the summit crater. A high-velocity zone exists above the central base of the volcano under the vertically aligned hypocenters, and may be interpreted as the source zone for recharge of the shallow magmatic system. High-velocity zones are also identified under the lower northeastern and southern flanks of the edifice. The southern high-velocity anomaly lies close to the surface and is connected to the high-velocity zone in the central base of the volcano, a feature suggesting an old lateral dike system. Except for these high-velocity zones in the central, northern, and southern flanks, the volcanic edifice is composed of low-velocity materials down to a depth of 2 km above sea level. These low-velocity zones correlate with young unconsolidated deposits, and older highly fractured and/or altered volcanic materials.

[1] The sustained heat and gas output from Erebus volcano reflects a regime of magma convection that we investigate here using a bi-phase (melt and crystals), fluid dynamical model. Following validity and verification tests of the model, we carried out four single-phase and three bi-phase numerical 30-year- simulations, in an idealized 2D geometry representing a lava lake cooled from above and a reservoir heated from below that are linked by a 4-to-10–m-diameter conduit. We tested the effects of crystals on convection while changing conduit size and the system boundaries from closed to open. Neglecting crystal settling yields only a limited number of features, i.e., (i) the formation of a central instability, (ii) the average temperature evolution, and (iii) the average velocity range of the surface flow motion. Bi-phase simulations show that while crystals are quite efficiently transported by the liquid phase a small decoupling reflecting their large size (5 cm) results in settling. This leads to more complex circulation patterns and enhances the vigor of fluid motion. A sufficiently large conduit sustains convection and retains 6 and 20% of crystals in suspension, for a closed and open system, respectively. Model outputs do not yet correspond well with field observations of Erebus lava lake (e.g., real surface velocities are much faster than those modeled), suggesting that exsolved volatiles are an important source of buoyancy.

[1] We systematically used two approaches to analyze broadband seismic signals for monitoring active volcanoes: one is waveform inversion of very-long-period (VLP) signals assuming possible source mechanisms; the other is a source location method of long-period (LP) events and tremor using their amplitudes. The deterministic approach of the waveform inversion is useful to constrain the source mechanism and location but is basically only applicable to VLP signals with periods longer than a few seconds. The source location method assumes isotropic radiation of S waves and uses seismic amplitudes corrected for site amplifications. This simple approach provides reasonable source locations for various seismic signals such as a VLP event accompanying LP signals, an explosion event, and tremor associated with lahars and pyroclastic flows observed at five or fewer stations. Our results indicate that a frequency band of about 5–12 Hz and a Q factor of about 60 are appropriate for the determination of the source locations. In this frequency band the assumption of isotropic radiation may become valid because of the path effect caused by the scattering of seismic waves. The source location method may be categorized as a stochastic approach based on the nature of scattering waves. Systematic use of these two approaches provides a way to better utilize broadband seismic signals observed at a limited number of stations for improved monitoring of active volcanoes.