Currently, video observation systems are actively used for volcano activity monitoring. Video cameras allow us to remotely assess the state of a dangerous natural object and to detect thermal anomalies if technical capabilities are available. However, continuous use of visible band cameras instead of special tools (for example, thermal cameras), produces large number of images, that require the application of special algorithms both for preliminary filtering out the images with area of interest hidden due to weather or illumination conditions, and for volcano activity detection. Existing algorithms use preselected regions of interest in the frame for analysis. This region could be changed occasionally to observe events in a specific area of the volcano. It is a problem to set it in advance and keep it up to date, especially for an observation network with multiple cameras. The accumulated perennial archives of images with documented eruptions allow us to use modern deep learning technologies for whole frame analysis to solve the specified task. The article presents the development of algorithms to classify volcano images produced by video observation systems. The focus is on developing the algorithms to create a labelled dataset from an unstructured archive using existing and authors proposed techniques. The developed solution was tested using the archive of the video observation system for the volcanoes of Kamchatka, in particular the observation data for the Klyuchevskoy volcano. The tests show the high efficiency of the use of convolutional neural networks in volcano image classification, and the accuracy of classification achieved 91%. The resulting dataset consisting of 15,000 images and labelled in three classes of scenes is the first dataset of this kind of Kamchatka volcanoes. It can be used to develop systems for monitoring other stratovolcanoes that occupy most of the video frame.

A key problem of any video volcano surveillance network is an inconsistent quality and information value of the images obtained. To timely analyze the incoming data, they should be pre-filtered. Additionally, due to the continuous network operation and low shooting intervals, an operative visual analysis of the shots stream is quite difficult and requires the application of various computer algorithms. The article considers the parametric algorithms of image analysis developed by the authors for processing the shots of the volcanoes of Kamchatka. They allow automatically filtering the image flow generated by the surveillance network, highlighting those significant shots that will be further analyzed by volcanologists. A retrospective processing of the full image archive with the methods suggested helps to get a data set, labeled with different classes, for future neural network training.

KUDRIAVY volcano on Iturup island in the Kuril arc is an active calc-alkaline volcano. It has not erupted this century; its current volcanic activity is characterized by hot (up to 910oC) gas jets which have been stable for at least 30 years. The composition of the gaseous emissions is typical of high-temperature fumaroles, but we report here the discovery of unusual subsurface sublimates associated with one gas jet—a sulphide mineral containing rhenium as the only cation. To our knowledge, this is the first reported occurrence of a pure rhenium mineral. The concentration of rhen-ium in the fumarole gas is only 2–10 p.p.b., so the condensation of pure rhenium sulphide from this gas requires both enrichment of rhenium by eight orders of magnitude and remarkable selectivity. Rhenium is generally believed to exist in only trace amounts at the Earth's surface, but our findings demonstrate that it can be readily mobilized, dispersed and concentrated by degassing magmas.

A high-temperature (up to 940°C) fumarolic activity at Kudriavy volcano had been studied during 1990–1999. The maximum gas temperatures of the fumaroles were measured in 1992 as 940°C, then gradually decreased with time and reached to 907°C in 1999. Gas composition of the high-temperature fumarole became enriched in H2O and depleted in other gas components, in particular in CO2. Hydrogen isotopic compositions of the high-temperature fumarolic gases were gradually depleted in deuterium. The gradual and continuous decrease in temperature and changes in gas composition observed during the last 10-year suggest that a magmatic melt have been degassing in a relatively steady-state manner from a single magma chamber. The detail investigations in 1998 and 1999 revealed short-term changes in gas composition characterized by sporadic increases in H2, CO2, and Stotal after intense precipitations. Small-scale eruptions occurred on October 7, 1999 at the summit. The ratios of major gas components (C/S, C/Cl, S/Cl, C/F, S/F, and Cl/F) significantly increased just prior to the eruption. The eruption at the Kudriavy volcano in 1999 was likely a phreatic eruption as a result of the intense precipitations after unusually long dry period. Meteoric water penetrated into the hot zone of volcano edifice and rapidly boiled causing the eruption.

We present the results of time-dependent local earthquake tomography for the Kluchevskoy group of volcanoes in Kamchatka, Russia. We consider the time period from 1999 to 2009, which covers several stages of activity of Kluchevskoy and Bezymianny volcanoes. The results are supported by synthetic tests that recover a common 3D model based on data corresponding to different time windows. Throughout the period, we observe a robust feature below 25 km depth with anomalously high Vp/Vs values (up to 2.2). We interpret this feature as a channel bringing deep mantle materials with high fluid and melt content to the bottom of the crust. This mantle channel directly or indirectly determines the activity of all volcanoes of the Kluchevskoy group. In the crust, we model complex structure that varies over time. During the pre-eruptive period, we detected two levels of potential magma storage: one in the middle crust at 10–12 km depth and one close to the surface just below Kluchevskoy volcano. In 2005, a year of powerful eruptions of Kluchevskoy and Besymiyanny volcanoes, we observe a general increase in Vp/Vs throughout the crust. In the relaxation period following the eruption, the Vp/Vs values are generally low, and no strong anomalous zones in the crust are observed. We propose that very rapid variations in Vp/Vs are most likely due to abrupt changes in the stress and deformation states, which cause fracturing and the active transport of fluids. These fluids drive more fracturing in a positive feedback system that ultimately leads to eruption. We envision the magma reservoirs beneath the Kluchevskoy group as sponge-structured volumes that may quickly change the content of the molten phases as fluids pulse rapidly through the system.

Avacha group includes two active and potentially dangerous volcanoes, Avachinsky and Koryaksky, located close to Petropavlovsk-Kamchatsky, the main city of Kamchatka. We present the results of two independent seismic studies of shallow crustal structures beneath the Avacha group based on passive and active source observations. The first study is based on the analysis of continuous recording by 11 seismic stations installed over the Avacha group in 2012 and 7 permanent stations in the same region. We present a series of 2D Rayleigh-wave group velocity maps based on correlation of ambient noise, that were then converted into 3D distribution of shear wave velocity. The second work was based on the reprocessing of an active source deep seismic sounding profile across the Avachinsky volcano that was shot in 1982–1984. We made the analysis of travel times of refracted waves using a 2D tomography inversion. The resulting seismic models appear to be consistent with each other and show clear low-velocity zone to the SW of the Avachinsky volcano and high velocity structures to NE. These observations also agree with the existing gravity and magnetotelluric measurements. Based on the obtained seismic models we identify two large buried calderas and large lava flows that are thought to be related to a series of large eruption episodes of Avachinsky occurred within the last 30,000 years.