INTRODUCTION

The studies of paleostresses using the method of cataclastic analysis of the Greater Caucasus meganticlinorium show significant reorientations of the principal stress axes near the Anapa and Dzhiginka fault zones (Marinin and Sim, 2015). Similar results were obtained for the Northwestern Caucasus (Marinin, 2013). Other publications also show a change in the stress state of the geological medium over time (Rebetskii and Alekseev, 2014). However, the reasons for such changes have not been considered up to the present. This publication shows that one of the factors for an abrupt and short-term change in the stress state is the impact of strong and catastrophic earthquakes.

In order to investigate the time variation in the Racha-Dzhava seismic-zone tectonic stress field, we studied the averaged focal mechanisms of weak earthquakes. The method for determining averaged focal mechanisms is that a set of the first arrivals of signs of body waves from various earthquakes is used. In this case, it is believed that all the arrivals belong to one focus. This method is especially effective if a small number of recording stations is available (Aki, 1966; Zverev et al., 1976; Misharina and Solonenko, 1977). A summation of the first arrival signs is possible, provided that the seismic process is associated with a long-term shift along faults and fractures and its duration is from several months to several years (Kocharyan, 2016).

We used the averaged mechanisms not only to track the changes in the directions of compression vectors in time, but also to distinguish time intervals within which the energy released by local earthquakes is summed up. In our opinion, the results allow us to have insight into what changes the directions of compression forces and what causes the seismic process activity both for the Racha-Dzhava seismic zone and for the region of the Caucasus as a whole.

STUDY AREA

The study area is an underthrust zone where the continental microplates of the Southern and Northern Caucasus interact, leading to the tectonic stratification of alpine formations into various south-verging allochthonous and parautochthonous thrust strata. Large linearly stretched tectonic units correspond to the axis of the alpine marginal sea basin. The alpine cover was compressed in the underthrust zone and is shifted southward. During the continental stage of alpine tectogenesis, intense lateral compression has been caused by the Arabian Plate indentation into the structures of the southern margin of Eurasia. The modern geodynamic activity is confirmed by GPS data, according to which the Southern Caucasian block moves to the north at a rate of up to 29 mm/yr. To compare, the North Caucasian microplate motion rate is as low as 0–6 mm/yr (Babaev et al., 2017).

The main reasons for seismic activity in the Caucasus region are, first, compression under the influence of the indenting Arabian Plate and, second, stresses emerging from the differences in topological levels of tectonic blocks (Kangarli et al., 2018). Such a difference in motion rate produces a differentiated stress field and additional conditions for local seismicity to occur. The earthquake foci are confined to the intersections between fractures and faults, as well as to the planes of deep tectonic faults with lateral displacements along unstable contacts between tectonic units. Based on the focal mechanisms of seismic events, chiefly subvertical planes of normal dips and strike-slips were determined. Moreover, it was found that the earthquake foci, as a rule, are confined to the intersections between fractures trending along the “main Caucasian” and “cross-Caucasian” directions.

In order to analyze the time variation in focal mechanisms of the Caucasus earthquake, we selected the Racha-Dzhava seismic zone (Fig. 1), where low-energy local seismic events are concentrated. These events are confined to the intersection of two transverse (NE‑striking) faults, namely, the Tskhinvali–Kazbek deep fault in the east and the Rioni–Ossetian fault in the west, with the perpendicular faults that separate the Greater Caucasus folded system (from west to east) into the Svaneti, Racha, and Kartli transverse segments.

Fig. 1.
figure 1

Seismic cluster of the Racha-Dzhava seismic zone (marked with a rectangle) on the map showing the distribution of earthquake epicenters in the Caucasus, after (Burmin et al., 2019).

Full size image

In 1991, the Racha-Dzhava earthquake with M = 7.2 and epicentral shaking intensity of VIII, which appeared to be the strongest in the history of seismological observations in the Caucasus, occurred in this zone, being accompanied by a large number of aftershocks (Aref’ev et al., 1993; Papalashvili et al., 1997; Aref’ev, 2003; Belousov, 2009; Tatevossian and Aptekman, 2011; Vakarchuk et al., 2013; Burmin, 2016). The mainshock had an oblique (reverse-dip) thrust mechanism along a NW-striking plane gently inclined to the northeast. According to the distribution of surface dislocations and epicenters of the aftershocks, the source of the Racha-Dzhava earthquake spatially coincides with the Racha–Lechkhumi segment of the Kakheti–Lechkhumi suture zone, which is the largest tectonic unit on the Southern slope of the Greater Caucasus. The Racha–Lechkhumi suture zone separates the Dzirula ledge, corresponding to the middle massif of the Georgian block, on south, from the folded zone of the Southern slope of the Greater Caucasus on north (Rogozhin, 2009).

STUDY METHOD

There are several methods for studying the stress state of the geomedium from the set of weak earthquakes. C. Richter (1958) proposed determining focal mechanisms by the distribution of the first arrival signs of P-waves observed at one station from multiple earthquake foci surrounding this station.

Another approach (Aki, 1966; Zverev et al., 1976; Misharina and Solonenko, 1977), which is used in the present study, implies the analysis of stresses by determining focal mechanisms of earthquakes from the distribution of the first arrival signs of P-waves, which were recorded at several stations from groups of earthquakes united on a territorial basis. It is also assumed that the first arrival signs correspond to one earthquake focus.

INITIAL DATA

In order to construct the compression vectors, we used averaged focal mechanisms obtained from the first arrival signs of P-waves from weak earthquakes, which had been recorded in the Racha-Dzhava zone by the seismological agencies of the Russian Academy of Sciences, Armenia, Georgia, and Azerbaijan. The hypocenters of earthquakes for the period of 1971–2016 were taken for processing; the coordinates of hypocenters were obtained in (Burmin et al., 2019). Unfortunately, signs of the first arrivals are not provided for all events in the seismological bulletins; in this respect, we had to exclude significant periods of time from consideration.

The time intervals for constructing the averaged focal mechanisms were distinguished by changing the azimuthal distribution of the first arrival signs relative to the fault planes of the previous averaged focal mechanism (Shumlyanskaya and Burmin, 2016). Thus, 39 spatiotemporal intervals were identified, for which 18 averaged focal mechanisms were constructed. Table 1 shows the time intervals for constructing the averaged focal mechanisms.

Table 1. Time intervals for constructing the averaged focal mechanisms of earthquakes of the Racha-Dzhava zone
Full size table

RESULTS

Table 2 presents the averaged focal mechanisms constructed for 18 selected time intervals from Table 1. In order to construct the earthquake focal mechanisms, 3893 events, for which the first arrivals signs of P-waves had been provided in the bulletins, were used. All foci were located in the Earth’s crust.

Table 2.   Time changes of averaged focal mechanisms of earthquakes in the Racha-Dzhava zone for the time period of 1971–2011
Full size table

During the interval under consideration (1971–2011), the focal mechanism type (pure strike-slip) has not changed. In addition, the average total energy released per month was calculated for each time interval (Fig. 2).

Fig. 2.
figure 2

Coincidence of the changes in direction of compression vector (1) in time with strong and catastrophic earthquakes (2) in comparison with monthly total energy Eav released during local earthquakes within the considered time interval (see Tables 1, 3).

Full size image

The data given in Table 2 show that the obtained directions of the compression vector coincide with the distribution of vectors of horizontal stress components for the Caucasus region (Babaev et al., 2017), according to which the western and central parts of the Greater Caucasus suffer a NE–NW shortening. Despite the occurrence of the strong earthquakes in the immediate vicinity of the Racha-Dzhava zone, this direction demonstrated minor changes with time. However, the considered time intervals demonstrate a correlation between the sharp change in the direction of the compression vector (up to 90° in the counterclockwise direction) and distant strong and catastrophic earthquakes with Mw ≥ 8 after 2007. The list of these earthquakes is given in Table 3.

Table 3.   Strongest earthquakes in the world, 1971–2011 (https://earthquake.usgs.gov/earthquakes/)
Full size table

Let us compare the averaged focal mechanisms of local earthquakes with those of the mainshock and aftershocks of the Racha earthquake of April 29, 1991, which is the strongest in the region with Mw = 6.9 (based on the data from the International Seismological Centre).

The mainshock and all aftershocks had focal mechanisms of a reverse-dip type. The reverse-dip occurred along a NW-striking (azimuth 300°) plane gently (35°) dipping to the northeast. This mechanism closely matches the idea of the main tendencies of the Greater Caucasus geological evolution, namely, underthrusting of the Dzirula massif basement beneath the anticlinorium of the Greater Caucasus Range (Aref’ev, 2003).

We compared the directions of compression vectors obtained from the averaged focal mechanisms (see Fig. 2) with the directions of stress axes of the 1991 Racha earthquake (Fig. 3). We can see that until 2007 the direction of compression vector by the averaged focal mechanisms coincides with the direction of stresses that caused the mainshock of the Racha earthquake. It confirms the long-term regional character of underthrusting of the Dzirula massif basement beneath the Greater Caucasus Range, which is associated with the northward indentation of the Arabian Plate (Rogozhin, 2009).

Fig. 3.
figure 3

(a) Versions of the mainshock mechanism of the 1991 Racha earthquake; (b) mechanisms of its aftershocks and the stress tensor, after (Aref’ev, 2003).

Full size image

CONCLUSIONS

The results of studying the time variation in the direction of compression vectors in the Racha-Dzhava seismic zone for the time period of 1971–2011 show that the directions azimuths of these compressional stresses are changing with time. Compression forces are produced by the general compression setting throughout the entire continental stage of alpine tectogenesis, which was caused by indentation of the Arabian Plate into the structures of the southern margin of Eurasia.

Unfortunately, we do not have a complete picture of how focal mechanisms of earthquakes of this region changed, since the data on first arrival signs of P-waves are fragmentary and incomplete. According to the data, two periods with different characters of the compression vectors can be distinguished. The time division between these periods is year 2007, after which the largest number of distant earthquakes with a magnitude of more than 8 occurred.

Before 2007, the compression vectors vary in the SW–NE direction within the range of 45°. After 2007, a change in the behavior of compression stress directions was observed: the southeastern and northwestern directions appeared and they were perpendicular to the directions obtained from all the averaged focal mechanisms in the previous time interval. A sharp change in the behavior coincides with an increase in the total energy released by local seismic events (the trend can be seen in Fig. 2).

Our study results show that there are several sources determining the changes in geodynamic setting of the Racha-Dzhava zone. At the global level, the geodynamic setting persists for long periods of time comparable with geological eras in duration, and that is reflected in the directions of compression vectors. At the local level, the character of changes is affected by local factors such as the topology of geological layers, the sizes of tectonic blocks, and the compositions of these blocks.

After 2007, the activation of local seismicity was observed, indicating an increase in the instability of the region under consideration. This instability creates the conditions for short-term changes in the directions of vectors determining the stress regime of the studied region under the influence of distant strong earthquakes. The energy of these earthquakes is so large that local factors in the region stop governing.

The results of this study are interesting for understanding some aspects of local seismic activity, and the causes of changes in the stress field and the activity of seismic process of the Racha-Dzhava seismic zone, which is part of the Caucasus system.