Search

SUMMARY Seismic source amplitudes determined by using the amplitudes of high-frequency (5−10 Hz) tremor signals generated by sustained explosive eruptions have been shown to be related to eruption plume height by power-law and exponential relations and to eruption volume flux by a proportional relation. We further examined these relations and extended this source quantification approach to investigate eruption duration by using the envelope width, defined by the ratio of the cumulative source amplitude to the source amplitude. We first confirmed that the relationship between source amplitude and plume height proposed by a previous study holds for small eruptions at Nevado del Ruiz (Colombia), although slight modifications were required. We then showed that the relations of envelope width with source amplitude and with cumulative source amplitude of eruption tremor associated with sub-Plinian eruptions at Kirishima (Japan) and Tungurahua (Ecuador) could be described by a power law. The source amplitude functions of these tremors were characterized by three periods and could be approximated by a trapezoidal shape. A power-law function fitted to the relation between eruption volume and eruption duration obtained from these relations was similar to that estimated by fitting a power-law function to previously reported eruption volume and duration data of well-documented silicic and andesitic eruptions. Our results suggest that eruption duration may systematically vary with eruption volume when the conduit is stably open during the second period of the trapezoidal source amplitude function. This study demonstrated that source amplitudes can be used for real-time predictions of both plume height and eruption duration, which in turn may be used to estimate ashfall distributions and tephra transport for local residents and aviation operations.

An analysis of seismicity of Valle Medio del Magdalena (VMM) Basin, Colombia was done over the period between 2014 and 2017 with the aim to obtain a baseline of the seismic activity in this region. Thirteen portable broad band seismic stations were deployed during the study period and were complemented by 15 permanent broad band seismic stations. Hypocenter distribution and focal mechanism analyses were performed. Hypocenter determination was performed for 7073 earthquakes with local magnitudes between 0.1 and 5.7. Those earthquakes were located at crustal as well as subduction depths. The former were related to fault systems that evidenced seismic activity, such as the Bucaramanga fault system and the Cimitarra, Curumani and Arrugas faults. The latter delineated the shape of the subduction slab of the Caribbean plate beneath South American plate, which exhibited changes in the dip angle between 29° and 39° probably associated to deep ruptures. Inversion of focal mechanisms obtained by first motion P-wave analysis for crustal earthquakes showed that the VMM region was under a regional strike-slip stress regime.

Analysis of seismic data is the most important method for volcano monitoring. Such data typically consists in digital signals acquired with an arrangement of triaxial seismic sensors which are strategically deployed on the volcano and its surrounding areas. Very rich measurements of the underlying phenomena are obtained from the arrangement because each sensor, through their corresponding three sensing axes, uninterruptedly acquires data at a relatively high sampling rate. Such an uninterrupted acquisition, however, turns manual classification of seismic signals into an inefficient and often error-prone task. As a solution, several systems for automated classification of seismic-volcanic signals have been proposed. All these systems, however, are limited to the usage of only one direction of acquisition; typically the vertical one. In this paper we make a step forward, exploring the potential benefit of using information from the three axes of the signals gathered by a single sensor. Integration is performed by classifier combining techniques, applied at different levels: this permits to take into account in the classification all the three orthogonal orientations —vertical, East-West and North-South— of the phenomenon. Preliminary experimental results on a set of volcanic signals gathered at Nevado del Ruiz volcano in Colombia confirmed the richness of this information.

Nevado del Ruiz Volcano (NRV) had a phreatomagmatic eruption in 1985. The eruption partially melted the volcano’s ice cap leading to floods and lahars flowing down to nearby towns, which killed at least 25,000 people. This event has raised particular importance of monitoring activity including small eruptions at ice-capped high-altitude volcanoes. However, the high altitude makes it difficult to maintain monitoring stations near the summit crater. Moreover, the visibility of the summit area is frequently prevented by clouds. In this paper, we report the results of a feasibility study for detecting thermal anomalies and small eruptions using satellite thermal remote sensing and ground-based infrasound technique. We newly included South and Central America to the observation areas of the near-real-time monitoring system of the active volcanoes, which uses infrared images from satellites. We also operated three infrasound stations in the distances of 4–6 km from the active crater. Each of the stations consisted of a pair of infrasound sensors, and a cross-correlation technique was applied. The thermal and infrasound data acquisition started in August 2015 and December 2016, respectively, and recorded the recent dome-forming activity of NRV. We proposed parameters representing the visibility of the thermal anomalies and infrasound signals. These parameters are useful for monitoring because the severe weather condition at NRV frequently prevents signal detections. We discussed the detected thermal anomalies and infrasound signals in comparison with their visibilities and the changes in the volcanic activity of NRV reported by the local observatory. The thermal anomaly and infrasound detections were consistent with the high eruptive activity occurring at the NRV from October 2015 to May 2017 and its subsequent decline. Within the active period, there were breaks in the detections of thermal anomaly and infrasound. The visibility analyses allowed us to interpret the breaks as a result of bad weather conditions and to distinguish them from the confirmed low-activity periods after May 2017.

Se llevo a cabo un estudio de la relacion entre la sismicidad de fractura y las emisiones de radón (222Rn) en suelo en la region volcanica de los Coconucos, usando tres estaciones isotopicas ubicadas en el area de influencia de fallas activas en el periodo 2006-2013. Para este estudio se definieron anomalias de gas radon como aquellos valores de concentración por encima del promedio mas la mitad de la desviacion estandar. Estos valores se correlacionaron con sismos de magnitudes locales (ML) mayores a 2,8, y se llego a la conclusion de que valores por encima de 2300 pCi/L estuvieron asociados a la ocurrencia posterior de estos sismos en un radio de 35 km del centro de la red de estaciones. El rango de tiempo entre la deteccion de la anomalia y la ocurrencia del sismo (ML > 2,8) vario entre 3 y 84 dias, con un promedio de 31 dias. De un total de 15 sismos ocurridos en el periodo de estudio con ML > 2,8, trece fueron asociados con anomalías de gas radon, para una probabilidad de deteccion del 87 %. Esto permite establecer que la medición de concentracion de gas radon en el suelo, usando una distribucion apropiada de estaciones, es una herramienta util para el pronostico de sismicidad posterior en la zona de los Coconucos.

En el volcán Puracé se realizó un estudio de eventos sísmicos tipo tornillo durante el periodo comprendido entre 2000 y 2012. Con el fin de caracterizar la fuente de generación y la naturaleza de los fluidos que interactúan en las grietas internas del volcán Puracé, el factor Q del resonador y las frecuencias dominantes fueron calculados utilizando el método Sompi. Encontramos que los valores del parámetro Q del resonador variaron entre 100 y 400 en promedio, y que en algunos periodos (2010) el Q del resonador alcanzó valores de hasta 1200. Las frecuencias complejas presentaron diferentes rangos (1,5-13 Hz), predominando en promedio los 6 Hz. Fue posible observar variaciones temporales tanto en el parámetro Q del resonador como en frecuencias dominantes, que pudieron estar asociadas a cambios en el contenido del fluido, desde más gaseoso hasta una mezcla de gas y material particulado (azufre mineralizado), y a cambios en los tamaños de las grietas, de 30 a 50 m, que se estima están ubicadas a una profundidad de entre 300 y 900 m del fondo del cráter. Proponemos un modelo conceptual en el cual el fluido que genera los eventos tornillo está asociado a la actividad magmática del volcán Puracé, que libera pulsos de gas que interactúan con el sistema hidrotermal y con capas superficiales de zonas mineralizadas con azufre nativo, que a su vez se mezcla con el gas proveniente del magma, lo cual causa la aparición de sismos tipo tornillo y las variaciones temporales que se presentan en los valores del parámetro Q del resonador y el contenido de las frecuencias dominantes.

Se definió la primera fórmula de magnitud local para la zona del Paletará (ZP) a partir de una inversión de amplitudes de componentes horizontales de estaciones de banda ancha y corto periodo, simuladas en un sismograma Wood-Anderson, para sismos VT y tectónicos locales con profundidades menores a 37 km y distancias hipocentrales menores a 220 km. Se invirtieron 628 sismos, con un total de 2848 amplitudes, en dieciocho estaciones sísmicas. La función de corrección por distancia encontrada para la ZP fue −log(Ao) = 1,3541 log(r/17) + 0,001639(r − 17) + 2, donde log (Ao) es la corrección por distancia y r es la distancia hipocentral. La fórmula de magnitud para la ZP queda definida así: ML = log(Amm) + 1,3541 log(r/17) + 0,001639(r − 17) 2 + S, o alternativamente, según la formulación Iaspei: ML = log(Anm) + 1,3541 log(r) + 0,001639r − 2,37 + S Donde Amm es la amplitud pico en milímetros simulada en un sensor Wood-Anderson, Anm es la amplitud pico en nanómetros simulada en un sensor Wood-Anderson, con una amplificación estática de 1, y S es la corrección por estación debida a la amplificación de cada sitio.

A local Magnitude (ML) scale for Valle Medio del Magdalena (VMM) region was defined by using 514 high quality earthquakes located at VMM area and inversion of 2797 amplitude values of horizontal components of 17 stations seismic broad band stations, simulated in a Wood-Anderson seismograph. The derived local magnitude scale for VMM region was: M L = log ( A ) + 1.3744 ∗ l o g ( r ) + 0.0014776 ∗ r − 2.397 + S Where A is the zero-to-peak amplitude in nm in horizontal components, r is the hypocentral distance in km, and S is the station correction. Higher values of ML were obtained for VMM region compared with those obtained with the current formula used for ML determination, and with California formula. With this new scale ML values are adjusted to local conditions beneath VMM region leading to more realistic ML values. Moreover, with this new ML scale the seismicity caused by tectonic or fracking activity at VMM region can be monitored more accurately.

The Nevado del Ruiz volcano is considered one of the most active volcanoes in Colombia, which can potentially threaten approximately 600,000 inhabitants. The existence of a glacier and several streams channelling in some main rivers, flowing downslope, increases the risk for the population living on the flank of the volcano in case of unrest, because of the generation of lahars and mudflows. Indeed, during the November 1985 subplinian eruption, a lahar generated by the sudden melting of the glacier killed twenty thousand people in the town of Armero. Moreover, the involvement of the local hydrothermal system has produced in the past phreatic and phreatomagmatic activity, as occurred in 1989. Therefore, the physico-chemical conditions of the hydrothermal system as well as its contribution to the shallow thermal groundwater and freshwater in terms of enthalpy and chemicals require a close monitoring. The phase of unrest occurred since 2010 and culminated with an eruption in 2012, after several years of relative stability, still maintains a moderate alert, as required by the high seismicity and SO 2 degassing. In October 2013, a sampling campaign has been performed on thermal springs and stream water, located at 2600–5000 m of elevation on the slope of Nevado del Ruiz, analyzed for water chemistry and stable isotopes. Some of these waters are typically steam-heated (low pH and high sulfate content) by the vapour probably separating from a zoned hydrothermal system. By applying a model of steam-heating, based on mass and enthalpy balances, we have estimated the mass rate of hydrothermal steam discharging in the different springs. The composition of the hottest thermal spring (Botero Londono) is probably representative of a marginal part of the hydrothermal system, having a temperature of 250 °C and low salinity (Cl ~ 1500 mg/l), which suggest, along with the retrieved isotope composition, a chiefly meteoric origin. The vapour discharged at the steam vent “Nereidas” (3600 m asl) is hypothesized to be separated from a high-temperature hydrothermal system. Based on its composition and on literature data on fluid inclusions, we have retrieved the P-T-X conditions of the deep hydrothermal system, as well as its pH and f O 2 . The vapour feeding Nereidas would separate from a biphasic hydrothermal system characterized by the following parameters: t = 315 °C, P = 15 MPa, NaCl = 10 wt%, CO 2 = 5 wt%, and similar proportion between liquid and vapour. Considering also the equilibria involving S-bearing gases and HCl, pH would approach the value of 1.5 while f O 2 would correspond to the FeO-Fe 2 O 3 buffer. Chlorine content is estimated at 10,300 mg/l. Changes in the magmatic input into the hydrothermal system could modify its degree of vapourization and/or P-T-X conditions, thus inducing corresponding variations in vapour discharges and thermal waters. These findings, paralleled by contemporary measurements of water flow rates, could give significant clues on risk evaluation.

Cerro Machin volcano in Colombia recently (2009–2013) exhibited new seismic activity. A geochemical characterization of emitted fluids was carried out based on the chemical and isotopic compositions of fumaroles and thermal springs discharged in this volcanic area with the aim of improving the volcano monitoring program. The isotopic composition of fumaroles located in the upper part of the dome was characterized by depleted δD and δ 18 O relative to the meteoric recharge fluids, which is due to a separation of the liquid phase in the lower part of the dome. A clear magmatic He signature was found in the fumaroles, with an increased magmatic contribution observed in the discharged fluids during 2011–2013 corroborating the increase in recorded seismic activity. Coupling the geochemical and seismic information has highlighted a clear awakening of Cerro Machin volcanic activity, indicating that this volcano is still active deserve an improvement of volcanic monitoring activity by the scientific community.

We performed repeated three-dimensional seismic tomography in 2009–2018 for the area influenced by the intersection of the Silvia–Pijao (Buesaco) and Cauca–Almaguer fault systems beneath Galeras volcano, Colombia. We analyzed two periods, 2009–2012 and 2013–2018, which were characterized by explosive eruptive activity and relative quiescence at Galeras, respectively. We used P- and S-wave arrival times of 2848 (2009–2012) and 2289 (2013–2018) earthquakes that occurred beneath the volcano. P- and S-wave velocity (Vp and Vs, respectively) anomalies and their changes between the two periods were interpreted to be associated with magma movement in the volcanic plumbing system, which was controlled by the intersection of the faults. Two possible pathways of magma ascent and accumulation were identified in the fault systems. One, located SW of the active crater, consisted of areas with high Vp/Vs values (>1.9) at depths of −2 to 4 km below sea level, and the other located NE of the active crater with high and low Vp/Vs values. The low-Vp/Vs (

A detailed 3D velocity tomography for P and S-waves and P to S ratios (Vp/Vs) was performed for the NE margin of Colombia, called the Valle Medio del Magdalena Basin (VMM) (Middle Magdalena Valley basin). A total of 21,265 local earthquakes, were used to obtain the 3D velocity structure of that region and surroundings. Low Vp and Vs values were found beneath the VMM basin at depths up to 15 km. Moreover, high Vp/Vs values were found beneath the VMM from depths of 2–10 km. On the other hand, the subducting slab portion observed beneath the VMM showed low and high Vp/Vs regions from depths of 20–100 km. This subduction slab is interpreted as corresponding to the Caribbean (CP) and the Nazca Plate (NP). The slab seems to be segmented in two main different blocks, one exhibiting low-Vp/Vs ratios (high-Vp and Vs), at depth >80 km associated to the Caribbean Plate with a dip angle of ∼34°, and one exhibiting high-Vp/Vs ratios (low-Vp and Vs), at depths >80 km associated with an eastern and northernmost block of the Nazca Plate, with a dip angle of ∼50°. The limit of both plates is located at 6.6°N. The Bucaramanga seismic nest (high-Vp and Vs, and low Vp/Vs) is interpreted as the collision of the northernmost block of the NP and the CP, in a region with mantle-rich material and high-Mg content. The limit of the northernmost block of the NP and the southern wide block of the same plate shifted 200 km to the west is the Caldas tear, which is proposed as located inside the NP, not the boundary between the CP and the NP. This tearing of the NP is associated with Quaternary volcanism lined up in an E-W direction, such as the current magmatic active Paipa-Iza volcanic Complex, and the Villamaria-Termales and Samana monogenetic volcanic fields. This finding indicates that the active volcanism in Colombia is extended to ∼6°N, one degree further north than previously assumed.

The timing and processes that govern the end of volcanic eruptions are not yet fully understood, and there currently exists no systematic definition for the end of a volcanic eruption. Currently, end of eruption is established either by generic criteria (typically 90 days after the end of visual signals of eruption) or criteria specific to a given volcano. We explore the application of supervised machine learning classification methods: Support Vector Machine, Logistic Regression, Random Forest and Gaussian Process Classifiers and define a decisiveness index D to evaluate the consistency of the classifications obtained by these models. We apply these methods to seismic time series from two volcanoes chosen because they display contrasting styles of eruption: Telica (Nicaragua) and Nevado del Ruiz (Colombia). We find that, for both volcanic systems, the end-date we obtain by classification of seismic data is 2–4 months later than end-dates defined by the last occurrence of visual eruption (such as ash emission). This finding is in agreement with previous, general definitions of eruption end and is consistent across models. Our classifications have a higher correspondence of eruptive activity with visual activity than with database records of eruption start and end. We analyze the relative importance of the different features of seismic activity used in our models (e.g. peak event amplitude, daily event counts) and find little consistency between the two volcanic systems in terms of the most important features which determine whether activity is eruptive or non-eruptive. These initial results look promising and our approach may offer a robust tool to help determine when an eruption has ended in the absence of visual confirmation.

Distinguishing among the different seismic volcanic patterns is still one of the most important and labor-intensive tasks for volcano monitoring. This task could be lightened and made free from subjective bias by using automatic classification techniques. In this context, a core but often overlooked issue is the choice of an appropriate representation of the data to be classified. Recently, it has been suggested that using a relative representation (i.e. proximities, namely dissimilarities on pairs of objects) instead of an absolute one (i.e. features, namely measurements on single objects) is advantageous to exploit the relational information contained in the dissimilarities to derive highly discriminant vector spaces, where any classifier can be used. According to that motivation, this paper investigates the suitability of a dynamic time warping (DTW) dissimilarity-based vector representation for the classification of seismic patterns. Results show the usefulness of such a representation in the seismic pattern classification scenario, including analyses of potential benefits from recent advances in the dissimilarity-based paradigm such as the proper selection of representation sets and the combination of different dissimilarity representations that might be available for the same data. HighlightsA representation, based on the DTW measure, is proposed for seismic classification.Recent advances of the dissimilarity based representation are investigated for DTW.Experiments with large scope dataset confirm the suitability of the DTW-space.The proposed space, when derived from spectrograms, is the best representation.Selecting small representation sets reduces the number of required DTW comparisons.

Se realizó un estudio sísmico de la estructura interna para onda P en el Volcán Galeras (VG), utilizando tomografía sísmica tridimensional. Fueron invertidos 2.400 sismos de alta calidad, con magnitudes 5%) se encuentra cubriendo todo el edificio volcánico y hasta una profundidad de 4 km; la otra zona se encuentra a partir de 10 a 12 km de profundidad, que se asociaría con el basamento. La zona de alta velocidad de O a 4 km de profundidad se asocia en superficie a los depósitos de lavas, y posiblemente al trazo de las fallas de Buesaco, Aranda y Romeral; en profundidad, hacia el oeste del volcán Galeras, se aumenta aún más la perturbación de la velocidad y estaría representando la cubierta de un cuerpo magmático antiguo en proceso de enfriamiento, cristalización y desgasificación. Igualmente, se encontró una zona de baja velocidad al este del cráter activo, la que se extiende hasta 5 km de profundidad, aproximadamente. Esta zona se asocia a un cuerpo magmático remanente y en estado más fluido. Las zonas de baja velocidad de gran extensión, encontradas en este trabajo, hacen suponer que el Volcán Galeras posee una extensa cámara magmática con volúmenes que alcanzarían 20 o 25 km3.

A seismic tomographic study was performed for Nevado del Ruiz Volcano (NRV) using more than 1500 high-quality local and regional. Three low P wave velocity (low- V P ) and low S wave velocity (low- V S ) zones were found; one low- V S zone at depths 2–4 km located beneath the volcano; a second low- V P and low- V S zone at depths 5–10 km located beneath the crater, elongated and dipping to the E–SE; and a third low- V P and low- V S zone at 10 to ∼12 km farther to the east. These three low- V P and low- V S zones are believed to be the location of heat sources. A high-velocity zone for both P and S waves was found at shallow depths (0∼5 km) around the active crater. The upper part (0∼2 km depth) of the high- V P and high- V S zone was characterized by low- V P / V S ratios ( V P / V S ratios (>1.80). The low- V P / V S zone is correlated with a steam-dominated geothermal system. The high- V P / V S is interpreted as an intrusive body of magmatic origin which includes partial melting zones associated with low- V S anomalies. A small low- V P zone in which long-period (LP) earthquakes were clustered was found to the southwest of the volcano. Based on the data obtained with the tomography in combination with seismicity, geochemistry, geology and gravimetry, we suggest a model for the seismic activity of NRV. Volcano-tectonic (VT) earthquakes that occur very often in swarm-like patterns located in several clusters around the volcano seem to be due to changes in stress produced by the passing of fluids and/or gas through many small cracks. A fault and a caldera-like structure separate the VT swarms located to the west of the volcano from the source of LP earthquakes.

International audience; The geochemical behaviour of Rare Earth Elements, Zr and Hf was investigated in the thermal waters of Nevado del Ruiz volcano system. A wide range of pH, between 1.0 and 8.8, characterizes these fluids. The acidic waters are sulphate dominated with different Cl/SO4 ratios. The important role of the pH and the ionic complexes for the distribution of REE, Zr and Hf in the aqueous phase was evidenced. The pH rules the precipitation of authigenic Fe and Al oxyhydroxides producing changes in REE, Zr, Hf amounts and strong anomalies of Cerium. The precipitation of alunite and jarosite removes LREE from the solution, changing the REE distribution in acidic waters.Y–Ho and Zr–Hf (twin pairs) have a different behaviour in strong acidic waters with respect to the water with pH near-neutral. Yttrium and Ho behave as Zr and Hf in waters with pH near neutral-to-neutral, showing super-chondritic ratios. The twin pairs showed to be sensitive to the co-precipitation and/or adsorption onto the surface of authigenic particulate (Fe-, Al-oxyhydroxides), suggesting an enhanced scavenging of Ho and Hf with respect to Y and Zr, leading to superchondritic values. In acidic waters, a different behaviour of twin pairs occurs with chondritic Y/Ho ratios and sub-chondritic Zr/Hf ratios. For the first time, Zr and Hf were investigated in natural acidic fluids to understand the behaviour of these elements in extreme acidic conditions and different major anion chemistry. Zr/Hf molar ratio changes from 4.75 to 49.29 in water with pH < 3.6. In strong acidic waters the fractionation of Zr and Hf was recognized as function of major anion contents (Cl and SO4), suggesting the formation of complexes leading to sub-chondritic Zr/Hf molar ratios.

The coda Q has been calculated for Nevado del Ruiz Volcano, Colombia (NRV) from 1985 to 1999 by using a single scattering model. During this period, the inverse of Q (Q–1 proportional to attenuation) exhibited a long-term decrease with time, as well as shorter-term variations related to the volcanic activity. Q–1 increased prior to volcanic crises and decreased afterward. Based on these observations, a seismic warning criterion has been developed. The parameters (frequency band, size of moving average window, and threshold levels) necessary to evidence clear and significant short-term changes in Q–1 have been investigated and appropriated values are proposed. We suggest a phenomenological model with three stages for the short-term temporal changes in Q–1 at NRV. Firstly, Q–1 increases before a volcanic crises because of accumulation of gas and/or liquid, which decreases the aspect ratio of fluid pockets and increases the fractional volume of fluid in the rocks and the pore aspect ratio. Secondly, Q–1 starts to decrease during the crises by the discharging of fluids such as gas, water, etc. from the volcano. Finally, Q–1 becomes more stable after the crisis at a lower value because of the degassing and/or increasing of rigidity of the medium because of the long-term crystallization and cooling processes. Q–1 seems to be a promising monitoring tool at NRV. It is possible that the observed temporal changes of Q–1, combined with other parameters, may help to predict with greater accuracy a volcanic crisis at NRV.

Resumen Se realizo un mapeo tridimensional del valor-b para la sismicidad volcano-tectonica del Volcan Nevado del Ruiz (VNR) para el periodo comprendido entre 1985 y 2002. Se pudieron definir cuatro zonas anomalas (altos valores) de b en el VNR, localizadas, unaal oeste del crater activo cerca al crater la Olleta (b=1.3-1.8), asociada con un sistema hidrotermal en superficie (z=1Km) y una zona de alimentacion magmatica en profundidad (z=6Km); otra zona (b=1.4), ubicada bajo el crater activo Arenas (z=1Km) asociada con un sector de acumulacion de vapor y fluidos que harian parte del sistema hidrotermal del VNR; otra zona (b=1.7), localizada al sur del crater activo (Z=6Km) asociada posiblemente a la localizacion de una de las camaras magmaticas del VNR; y otra zona (b=1.7), ubicada al NE del crater activo (z=3Km) cerca al crater la Pirana, asociada tambien a una zona de alimentacion magmatica. Estas zonas son coincidentes con las halladas en otros estudios usando tomografia sismica para el VNR, al igual que con estudios petrologicos y geoquimicos. Palabras clave: valor-b, mapeo 3D, sismicidad, distribucion frecuencia-magnitud de sismos Abstract Three-dimensional mapping of b-value (frequency-magnitude distribution) was done for Nevado del Ruiz Volcano (NRV), Colombia for the period 1985-2002.Four high b-value zones were detected. The first zone (b=1.3 -1.8), located to the west of the active crater (Arenas), close to the Olleta crater. This zone was associated with a hydrothermal system at surface (z=1Km),andwith a zone ofmagma supply atdepth(z >6Km). The second zone (b=1.4), located beneath the Arenas crater (z=1Km) was associated with a zone of fluid and vapor accumulation related with the hydrothermal system of NRV. The third zone (b=1.7), located to the south of Arenas crater (z=6Km) was associated with one of the magmatic chambers of NRV; and the forth zone (b=1.7), located to theNE of the Areas crater (z=3Km) near to the Pirana crater, was associated with another magma supply zone. Those zones are in agreement with studies of seismic tomography, geochemistry and petrology done previouslyat NRV. Keywords: b-value, 3d mapping, seismicity, earthquake frequency-magnitude distribution

Spectral analyses for volcano-tectonic earthquakes were carried out at Nevado del Ruiz Volcano (NRV) for the period 1985–1996 for several earthquake swarms around the volcano, named North, East, West, South and Crater swarm zones. Important spectral peaks for each earthquake swarm zone were found by counting the number of spectra that had the same spectral peaks. Each swarm zone showed some characteristic peaks, which could help to differentiate between them; however, the most important peaks were similar for all the zones. These results suggest that the earthquake swarms at NRV were influenced directly by the source (activity of the volcano) and could also be influenced by the site effect. Some temporal changes were observed in spectral parameters such as a change in the frequency contents in almost all the swarm zones, and the frequency of the P-waves in the West earthquake swarm zone. Before the eruptions on November 13, 1985 and September 1, 1989, P-waves showed low frequencies (1–2 Hz) at the West earthquake swarm. After the eruptions, the frequencies of P increased (2–4 Hz). This fact showed that changes (decreasing of frequencies) in the spectra of P-waves at the West earthquake swarm could help in the monitoring of volcanic activity at NRV. This swarm zone seems to be related directly with the most important volcanic crises that have occurred. This suggests that the West swarm zone should be monitored in more detail in the future.

Coda Q–1 was calculated at Nevado del Ruiz Volcano (NRV) before and after two phreatomagmatic eruptions (November 1985, September 1989) and for a period of stability (May 1987) using a functional form for coda derived from a single scattering model (Sato 1977). Substantial changes were found before and after the eruptions. The highest value of Q–1 was found during the November 1985 eruption, an intermediate value for the September 1989 eruption, and the lowest value for May 1987. It seems that the changes in coda Q–1 at NRV have a still-unknown relationship with the volume or magnitude of the eruption. A relatively strong frequency dependence was found for all periods. Also Q–1 clearly changed with time, suggesting that the scattering was strong for the eruption of November 1985 and decreased for the eruption of September 1989, and that the intrinsic absorption probably increased. This suggests the possibility that crystallization is taking place in the NRV magma. The clear change of coda Q–1 before and after the eruptions at NRV also suggests the possibility that coda Q–1 is a premonitory tool of activity at this volcano.

The analysis and classification of seismic patterns, which are typically registered as digital signals, can be used to monitor and understand the underlying geophysical phenomena beneath the volcanoes. In recent years, there has been an increasing interest in the development of automated systems for labeling those signals according to a number of pre-defined volcanic, tectonic and environmental classes. The first and crucial stage in the design of such systems is the definition or adoption of an appropriate representation of the raw seismic signals, in such a way that the subsequent stage —classification— is made easier or more accurate. This paper describes and discusses the most common representations that have been applied in the literature on classification of seismic-volcanic signals; namely, time-frequency features and cepstral coefficients. A comparative study of them is performed in terms of two criteria: (i) the leave-one-out nearest neighbor error, which provides a parameterless measure of the discriminative representational power and (ii) a visual examination of the representational quality via a scatter plot of the best three selected features.

Resumen Se desarrollo un sistema que permite monitorear remotamente movimientos de remocion en masa, usando sensores de diferentes caracteristicas, mediante un esquema multi-parametrico. En este caso se usaron sensores tipo geofono, termocupla, pluviometro, higrometro, e inclinometro. La informacion capturada se digitaliza en campo usando una tarjeta digitalizadora disenada en este trabajo, con la caracteristica de variabilidad en la rata de muestreo en cada canal de adquisicion. Mediante un sistema informatico, la informacion es procesada, analizada usando un sistema experto basado en reglas y arboles de decisiones, y posteriormente almacenada en una base de datos. El sistema permite la georeferenciacion de la informacion de los sensores y puede ser desplegada al estilo de un Sistema de Informacion Geografica basico. Mediante un esquema de alertas, usando un sistema experto estilo arbol de decisiones (Decision Tree), el sistema envia informacion sobre el estado de actividad del deslizamiento de tierra, permanentemente. Se realizaron pruebas piloto en un movimiento por remocion en masa activo, ubicado en el municipio de Manizales (Caldas, Colombia), durante un mes. Los resultados muestran una respuesta adecuada del sistema ante la actividad del deslizamiento, evidenciando su uso para monitorear otros deslizamientos de tierra en regiones pobladas. Palabras clave Movimientos por remocion en masa, sistema monitoreo multiparametro, sistema experto. Abstract A remote system to monitoring active landslides was developed, by using a multi-parameter device, middle-low cost sensors, and an expert system based on rules. The system consists of six (6) analogue sensors: geophone, hygrometer, rain gauge, thermocouple, tiltmeter. All the sensors are connected to a A/D converter board of variable rate sampling, which stores, packs and transmit digitally the information to a remote device (PC or computer server). At the remote device, an expert system based on rules by means a Decision Tree (DT), and developed in open source code, processes the data and analyses the information, given a state or activity level of the landslide continuously. Information is stored in a data base. A technique of changes in slope of cumulative normalized curves of the data set for each sensor and a combination of them, is used as a one of the parameters to evaluate the level of activity of the landslide. The system was installed in an active landslide at Manizales city (2100 masl), located in the central part of Colombia, at the Andean zone. The system collected data for 1 month. Preliminary results showed that the system is able to detect temporal changes in several parameters, allowing to have an idea of the level of activity of the monitored landslide. The system can be useful for monitoring zones with very active landslides and with population living near it, in order to alert promptly in case of a possible strong mass movement. Keywords Landslide, multi-parameter remote monitoring system, expert system.

An analysis of coda Q ( Q c ), intrinsic Q ( Q i ), scattering Q ( Q s ) and total Q ( Q t ), was made at Nevado del Ruiz Volcano (NRV) for local earthquakes recorded during 1992–1994. Coda Q has been calculated for the first time at Nevado del Ruiz Volcano (NRV) using the single-scattering model proposed by Aki and Chouet (1975). Estimations of Q c were made for five frequency bands centred at 1.5, 3.0, 6.0, 12.0 and 24.0 Hz. The estimated Q c varies from around 60 at 1.5 Hz to around 980 at 24 Hz, depending on the frequency with Q c = 30 ± 1.2 F 1.03±0.001 . in general, Q c values at NRV are rather similar to those obtained in other volcanic areas in the world. A steady trend in Q c -1 was found from January 1992 to January 1994. This variation is coincident with the stability in the volcanic activity since the last phreato-magmatic eruption on 1 September 1989. In the middle of April 1994, a new activity started at NRV and a small volcanic crisis occurred. Continuous increasing was observed in Q c -1 around two months before the crisis. This suggests a possibility that the temporal variation of Q c can be useful for monitoring the volcanic activity at NRV. Q i and Q s were separated from Q t by using a multiple-scattering model called “Multiple Lapse Time Window Analysis”, MLTWA (Fehler et al., 1992). Estimations for Q i , Q s and Q t were made for four frequency bands, 3.0, 6.0, 12.0 and 24.0 Hz. Q i ; values were found to be near to those of Q c for frequencies higher than 6.0 Hz. For low frequencies Q s values were near to Q c values. This can suggest that the intrinsic absorption could be the predominant attenuation phenomenon in the NRV area at frequencies higher than 6 Hz and the scattering phenomenon is predominant for low frequencies; therefore the change in Q c at high frequencies is considered to reflect variations in the inner conditions of the volcano, such as changes in temperature, movement of fluids, and so on.

Classifying seismic signals into their corresponding types of volcanic earthquakes is among the most important tasks for monitoring volcano activity. Such a duty must be routinely conducted—in a daily basis— and implies, therefore, a significant workload for the personnel. The discipline of pattern recognition (PR) provides volcanic seismology practitioners with theories and methods to design classification systems and, together with digital signal processing (DSP) techniques, has given rise to promising and challenging opportunities for the automated identification of volcanic earthquakes.

Multi-way data analysis is a multivariate data analysis technique having a wide application in some fields. Nevertheless, the development of classification tools for this type of representation is incipient yet. In this paper we study the dissimilarity representation for the classification of three-way data, as dissimilarities allow the representation of multi-dimensional objects in a natural way. As an example, the classification of seismic volcanic events is used. It is shown that in this application classification based on 2D spectrograms, dissimilarities perform better than on 1D spectral features.

The dissimilarity representation has demonstrated advantages in the solution of classification problems. Meanwhile, the representation of objects by multi-dimensional arrays is necessary in many research areas. However, the development of proper classification tools that take the multi-way structure into account is incipient. This paper introduces the use of the dissimilarity representation as a tool for classifying three-way data, as dissimilarities allow the representation of multidimensional objects in a natural way. As an example, the classification of three-way seismic volcanic data is used. A comparison is made between dissimilarity measures used in different representations of the three-way data. 2D dissimilarity measures for three-way data can be useful.