Search

Relevance. The science of studying the causes of earthquakes is rapidly developing. Each cause of an earthquake can be considered a precursor of an earthquake and using these precursors, predictive models can be built. To date, there are quite a few earthquake prediction models, which allow you to analyze these models to improve accuracy, that is, apply forecast data with higher probabilities. Aim. Analyze forecasting models and, based on the substantiation of existing forecasts, classify them into “necessary” and “sufficient” models, and define these terms. And also, to determine the algorithms for planning further actions to obtain much better forecasting models. It is “necessary” to develop algorithms that bring the “necessary” model to the “sufficient” one and vice versa. “necessary” forecasting models are models whose set of forecasts always includes a set of actually occurring events, and “sufficient” forecasting models are models whose forecasts always come true. The research methodology is to process the existing large data structures that are specified for further use in our algorithm. To calculate the probability of forecast accuracy, an algorithm with “parallel data” – “parallel probability” is used, which allows you to select those pairs of forecasting models (or triples, quadruples, etc.), whose “joint” probability of forecast accuracy gives a much better result than separately. Results were the formation of an author’s approach to processing earthquake forecast models and obtaining a generalized model that gives forecasts with a higher probability due to the use of statistics from already existing forecast models and their further observation. Algorithms have been defined for a) when to analyze all available required models and obtain one best model by combining the appropriate number of required models and b) when to combine enough models closest to guessing all predictions so that their number is less than in other unions. Also exists an algorithm that determines the study to be carried out after the occurrence of each event – the calculation of the probability of justification for individual models, as well as paired and triple models. The problem of using these algorithms in a specific area – earthquake prediction is discussed, and the results of the work of the author’s algorithm are shown.

The Kyushu Island, as well as whole Japanese archipelago, is equipped with dense GPS network (GEONET). It allows us to track the movements and deformations of the earth’s surface over long-term time intervals. In this study, based on daily determinations of the coordinates of GPS stations, analysis has been made on long-term trends in the accumulation of movements and deformations in large areas of the Kyushu Island before the series of April 14–16, 2016 Kumamoto earthquakes to identify deformation precursors and locked, immobile fault zones. The study of the seismic deformation process was performed using the data from 70 continuous permanent GPS stations for the period 2009–2016.The movement and deformation features found characterize the kinematics of the axial zone of the southwestern part of the island arc of the Japanese archipelago. The combination of coseismic compression and uplift in the center of the formed triad of deformation extrema and the consistency between subsidence and extensions at its edges demonstrate the mechanism of growth of the central region of the island arc under compression and the role of volcanism. Of particular interest in the context of the development of movements and deformations during the generation of the Kumamoto earthquakes is the behavior of the minimum displacement moduli of the GNSS sites. Analysis of their kinematics shows the formation of a zone of minimum displacements, in which subsequent strong seismic events were localized. It is shown that rather dense and extensive GNSS networks allow observing and studying the seismic-deformation process at the stages of seismic generation, discharge and relaxation, thereby providing an empirical basis for the development of models for predicting large seismic events.

Introduction. Determinations of (234U/238U) in groundwater samples are used for monitoring current deformations in active faults (parentheses denote activity ratio units). The cyclic equilibrium of activity ratio 234U/238U≈≈(234U/238U)≈γ≈1 corresponds to the atomic ratio ≈5.47×10–5. This parameter may vary due to higher contents of 234U nuclide in groundwater as a result of rock deformation. This effect discovered by P.I. Chalov and V.V. Cherdyntsev was described in [Cherdyntsev, 1969, 1973; Chalov, 1975; Chalov et al., 1990; Faure, 1989]. In 1970s and 1980s, only quite laborious methods were available for measuring uranium isotopic ratios. Today it is possible to determine concentrations and isotopic ration of uranium by express analytical techniques using inductively coupled plasma mass spectrometry (ICP‐MS) [Halicz et al., 2000; Shen et al., 2002; Cizdziel et al., 2005; Chebykin et al., 2007]. Sets of samples canbe efficiently analysed by ICP‐MS, and regularly collected uranium isotope values can be systematized at a new quality level for the purposes of earthquake prediction. In this study of (234U/238U) in groundwater at the Kultuk polygon, we selected stations of the highest sensitivity, which can ensure proper monitoring of the tectonic activity of the Obruchev and Main Sayan faults. These two faults that limit the Sharyzhalgai block of the crystalline basement of the Siberian craton in the south are conjugated in the territory of the Kultuk polygon (Fig 1). Forty sets of samples taken from 27 June 2012 to 28 January 2014 were analysed, and data on 170 samples are discussed in this paper.Methods. Isotope compositions of uranium and strontium were determined by methods described in [Chebykin et al., 2007; Pin et al., 1992] with modifications. Analyses of uranium by ISP‐MS technique were performed using an Agilent 7500ce quadrapole mass spectrometer of the Ultramicroanalysis Collective Use Centre; analyses of strontium were done using a Finnigan MAT 262 mass spectrometer of the Baikal Analytical Centre for Collective Use. A natural uranium isotope standard (GSO 7521‐99, Ural Electrochemical Plant, Novouralsk, Russia) and a strontium isotope standard (NBS 987) were used for quality control of the measurements.Results. The Kultuk polygon occupies large valleys of the Kultuchnaya, Angasolka, Talaya rivers and small valleys of the Medlyanka and Vorotny streams. The erosion basis of these valleys corresponds to the surface of Lake Baikal. In the valleys, there are several testing sites, including Staraya Angasolka, Slyudyanka, Vorotny, and Medlyanka. In the Kultuchnaya river valley, there are two sites, Tigunchikha and Verbny. Another two sites, Shkolny and Zemlyanichny, are located on slopes where no permanent water streams are available (Fig. 2). Measured U concentrations and(234U/238U) in water from the sites of the Kultuk polygon are placed in Table 1.Analysis and discussion of results. In water from an active fault, (234U/238U) depends on current deformation. The higher is the strain that causes fracturing, the higher is (234U/238U). The isotope composition of Sr sufficiently depends on the chemical weathering of rocks. The primary composition may be preserved in central parts of rock minerals and is detectable after preliminary treatment of an altered rock by HCl [Rasskazov et al., 2012]. In general, isotoperatios of U and Sr in groundwater and surface water depend on the composition of host rocks, weathering, and alkalinity. Dissolved uranium migrates as uranyl‐ion (UO22+) characterised by its highest degree of oxidation (+6). Reduced forms of U(+4) are practically water‐insoluble. Therefore, an indirect assessment of oxidation‐reduction properties of the medium can be based on uranium concentrations. For the Kultuk polygon, surface water with low (234U/238U) is divided by uranium content into two groups, with anomalously low (below 0.009 mkg/l), and medium (~0.5 mkg/l) concentrations of uranium (components from the Medlyanka river and Kultuchnaya river, respectively). The U abundances reflect relatively reduced conditions in group 1 and more oxidized in group 2. The higher (234U/238U) in the surface water with intermediate concentrations of uranium (0.009–0.500 mkg/l) may indicate the admixture of a groundwater component (Fig. 3). Figure 4 shows relations between surface water and groundwater components in the Kultuk polygon in terms of U content. In Figure 5, the field of data points of U and Sr isotope ratios in groundwater from the Kultuk polygon is contoured by curved lines that meet with each other at compositions corresponding to the end members E (87Sr/86Sr=0.7205, 234U/238U=1.0) and NE (87Sr/86Sr=0.70534, 234U/238U=3.3). Uranium ratios of the former and the latter components show equilibrium and the most nonequilibrium compositions, respectively. The field is characteristic of water samples from the rocks of the southern suture zone of the Siberian craton. Shift of the data points of water from stations 26 and 1310 to the right of this data field (i.e. with relative increasing 87Sr/86Sr) is due to lateral transition from the rocks of the suture zone to the Archean rocks of the Sharyzhalgai block (Fig. 6). The isotope systematics of uranium and strontium in the strongly nonequilibrium uranium segment is supplemented by the systematics of uranium in (234U/238U) vs. 1/U diagram (Fig. 7). The U composition in water from station 40 reflects a combination of processes that take place at station 27 (i.e. in the central part of the deformation system) and at station 38 (i.e. at its periphery). Approximately equal contents of uranium at the three above‐mentioned stations may reflect similar oxidization levels of the medium. In the Southern Baikal basin, the Irkutsk Seismic Station recorded an earthquake of class 11.2 on 08 January 2013 [Map…, 2013]. The earthquake epicentre was located near Listvyanka settlement (51.85° N, 105°16 E), at a distance of ~100 km from Kultuk settlement eastward of the Obruchev fault. On 24 April 2013, an earthquake of class 10 took place near Kultuk settlement. Another seismic event occurred on 07 June 2013 (Fig. 8). During the monitoring period, nine maximums and ten minimums of (234U/238U) were recorded at station 9, i.e. nine full cycles can be identified (Table 2). At station 9, amplitudes of the cycles exceed the measurement error by a factor of 2 to 4. In Fig. 9, at the curve showing temporal variations of (234U/238U) in water from station 9, deviations from similar curves for stations 11 and 8 are not marked. Curves of temporal variations of (234U/238U) in water from stations 40 and 27 are shown in Figure 10. At the first station (diagram а), there were three time intervals of monitoring: (1) 12 April 2013 to 04 July 2013, (2) 04 July 2013 to 21 October 2013, and (3) 21 October 2013 to 17 January 2014. The initial and final intervals are similar and show an abrupt decline of the curve with a clearly manifested drop of (234U/238U) in the middle part, a minimum and subsequent rise of the curve. The time interval between the compared periods of observation lasted 5–6 months. This middle interval marked a rapid increase of the average values of (234U/238U) in the range from 2.34 to 2.47 activity units with the average rate of about 0.2 units per year. In the curve of station 27, there is also a downward segment with a drop, a minimum and subsequent rise of the curve, which is partly coincident in time with the initial segments for station 40. Correlation in time is revealed between the earthquakes that occurred in the Kultuk polygon and the drops in the curves for the above‐mentioned stations. Considering the shape of the final segment of the curve based on observations at station 40, it could be expected that the drop in the downward curve should have been associated with earthquakes. However, no earthquakes took place. In this regard, attention should be paid to the fact that a concurrent drop lacks in the curve for station 27. This suggests that an earthquake would happen only in a case of co‐seismic (234U/238U) drops at both stations. Seismic processes are controlled by triggers that provide the synchronization effect. Self‐organization processes may be the cause of its manifestation. Intervals of synchronization of oscillations (similar to foreshock activation) are indicators of the unstable state of a seismic region [Sobolev et al., 2005]. Similar information of the transition to the pre‐seismogenic state can be obtained by analysing variations of (234U/238U) in water from active faults. In the initial monitoring stage, the deformation system of the Kultuk polygon (stations 8, 9 and 11) developed slowly, 110–170 days per cycle. The first indicators of the pre‐seismogenic state in the polygon were observed as a coincidence of the minimums in the cycles of all the stations on 16 March 2013. The first seismic event took place on 24 April 2013, i.e. 39 days after all the maximums coincided. In the period of the pre‐seismogenic state, relatively short cycles were manifested. The second seismic event occurred on 07 June 2013. It was reflected by the coincidence of the minimums of short cycles at stations 8, 9 and 40 (Fig. 11). The entire monitoring period at the Kultuk polygon can be divided into two time intervals starting from (1) 10 July 2012, and (2) 07 August 2013. The first time interval includes the preparation and occurrence of seismic events of class 10 in the polygon. In the second time interval, the deformation system was further developed, and a new seismogenic state became possible. The time interval from 10 July 2013 to 07 August 2013 includes three stages starting from (1) 10 July 2012, (2) 10 January 2013, and (3) 12 April 2013 (Fig. 12). Higher strain values along the line from station 8 to station 9 were accompanied by the occurrence of deformation along the line from station 40 to station 47 (submeridional direction at 14°), which resulted in the synchronization of (234U/238U) at these stations (Fig. 13). At the background of the chaotic state of the monitoring system of the Kultuk polygon, it is possible to distinguish sequential self‐organization phases from а to г as evidenced by the azimuthal synchronization of the stations. The spatial development of the recorded processes was represented the sequential seismogenic activation of the western termination of the Obruchev fault (Fig. 14). From the analyses of temporal variations of U concentrations (Fig. 15), we infer that the dynamics of uranium ingress into water was different at stations 9 and 8. In the initial monitoring stage, the background extremely high values of (234U/238U) and concentrations of uranium were inconsistent at stations 9 and 8. Later on, at station 9, episodes of the high mobility of uranium from the deformation zone alternated with episodes of the high mobility of uranium from the oxidation zone. At station 8, in the period from 26 October 2012 to 04 July 2013, uranium impulses took place occasionally in the deformation zone, and a few were combined with earthquakes of class 9 or 10. From 07 August 2013, the above‐mentioned impulses were replaced by uranium impulses from the oxidation zone. At this stage, an anomalous ingress of uranium was recorded.Conclusion. To validate the system of monitoring stations in the Kultuk polygon for earthquake prediction, spatial variations of (234U/238U) both in groundwater and surface water were studied. On sites of the tectonically stable areas, it was found that components of the surface runoff had admixtures of ground water components from the nearsurface water sources. On sites located at active faults, surface runoff components had admixtures of groundwater components from the deformation zone and oxidation zone. On sites located at active faults whereat permanent water streams lacked, the components from the deformation zone contained admixture of near‐surface ground water. The Sr–U‐isotopic systematics of groundwater at the Kultuk polygon was validated. Stations with high (234U/238U) (2.0–3.3activity units) and low 87Sr/86Sr (0.705341–0.712927) were selected for monitoring that lasted from 27 June 2012 to 28 January 2014. It was observed that (234U/238U) fluctuated in time, the duration of cycles and amplitudes of (234U/238U) fluctuations were variable, and the cycles of (234U/238U) in water were synchronized in the lines of the monitoring stations in the sublatitudinal and submeridional direction at the time intervals when seismic shocks occurred at the Kultuk polygon. The basic scenario of (234U/238U) variations in groundwater, recorded in the Kultuk polygon during the monitoring session, was examined in connection with the seismogenic activation of the western termination of the Obruchev fault. The SSE termination of the Main Sayan fault did not reveal any evidence on current tectonic deformations. The scenario of the reactivating Obruchev fault can be used for prediction of potential earthquakes in the Southern Baikal basin.

Reviewed are aspects of modern geodynamics and methods of mapping of geodynamic processes which have been developed since 1960s in the Institute of the Earth's Crust, Siberian Branch of RAS (specifically since 1980s by the Laboratory of Tectonophysics and now jointly with the Laboratory of Recent Geodynamics). Achievements and prospects of the studies are discussed. The publication is devoted to the 35th anniversary of the Laboratory of Tectonophysics and its main achievements in geodynamics.