Introduction

The Vrancea area (Romania) is known as a zone with frequent crustal and intermediate (60–180 km in depth) seismic activity and represent one of the most active seismic zones in Europe (Fig. 1) (Knapp et al. 2005; Marmureanu et al. 2016a, b, 2017). It is located in the South-Eastern corner of the Carpathians arc, at the contact between the East-European plate and the Intra-Alpine and Moesian subplates (Radulian et al. 2000; Marmureanu et al. 2016a, b).

Fig. 1
figure 1

(Left) Seismic events location between 1940 and 2020 along Romanian territory (minimap). The location of the analyzed events epicenters on the geological map of central and eastern Romania (red contour in the minimap; Maţenco et al. 2007). Hypocentral depths are indicated with the color scale while their sizes vary with their magnitude. (Right) The pre-1940 seismicity depth distribution within the 3D thermal geometry of the Vrancea slab (exemple from Ferrand and Manea 2021)

Full size image

From the seismogenic point of view, the complexity of the seismic processes of the Vrancea area remains far from being fully understood. The Vrancea region is a puzzling part of the Carpathian–Pannonian system due to its unique intra-continental situation as it marks the last part of the subduction/collision process along this region (Ismail-Zadeh et al. 2012; Martin et al. 2006). The hypocenters of the earthquakes are limited to a small epicentral region of approximately 30 × 70 km2 lateral and 160 km vertical extent (Knapp et al. 2005) within a remnant slab of the subducted oceanic Tethyan lithosphere (Ismail-Zadeh et al. 2012; Manea et al. 2022; Ferrand and Manea 2021). The distribution of the hypocenters within a 3D thermal model can be seen in Fig. 1 (right). The intense seismicity at intermediate depths in the Vrancea region is either the evidence of the current dehydration of an oceanic slab (Ferrand and Manea 2021) and/or due to an increase in the stress applied to this volume (Ismail-Zadeh et al. 2000). For this reason, large mantle seismic events are frequent in the region and over four earthquakes with moment magnitudes (Mw) > 7 occurred in the last century (Craiu et al. 2016a, b; Cioflan et al. 2022; Manea et al. 2020) which produced significant damage in Romania and neighboring countries (Kronrod et al. 2013; Cioflan et al. 2016; 2022).

The specific features of this seismic area (earthquake energy, macroseismic area and epicenter distribution) are comparable to the seismic foci from Hindu Kush Mountains-Pamir area close to the collision zone between the Indian and Eurasian plates, in central Asia and Bucaramanga region in the Andes of Colombia, South America (Ismail-Zadeh et al. 2012).

In the last decades, the number of focal mechanisms available for Vrancea intermediate-depth has increased significantly because of extended observation time, improvement in the seismic network and more elaborate calculation procedures (Radulian et al. 2019, Craiu et al. 2012). Many studies on calculation or analysis earthquakes focal mechanism were previously performed for Romania: analysis of the seismic activity in the Vrancea intermediate-depth source region during the period 2010–2015 (Craiu et al. 2016a, b); seismic activity and focal mechanisms analysis for different seismic zones of Romania 2010–2014 (Craiu et al. 2015b, a); the source mechanism of the seismic events recorded in the adjacent zones of Vrancea (Craiu et al. 2016a, b, 2018); the catalog published by Radulian et al. (2002) for the earthquakes that occurred before 1997; the mechanisms of several low-magnitude earthquakes of Vrancea region and surroundings, determined by high-frequency waveform inversion in Ardeleanu (2018) and Ardeleanu and Raileanu (2011). The fault plane solutions computed for the (1998–2012) was published by Radulian et al. (2018), the present paper being an extension and an improvement in previous studies. The focal mechanism of earthquakes is essential in the study of seismotectonic. They are illustrating the relationship between earthquakes and their causative faults. As such, they provide very useful information about tectonic activity in the study area.

In this study, we determined the fault plane solutions (FPS) of 437 seismic events that occurred starting in January 2005, in the Vrancea region and their magnitudes range between 2.7 and 5.6 Mw, according with Romplus catalog (Mihaela et al. 2022), (see Figs. 1 and 2). Besides this, the temporal and spatial variations of regional stress field orientation were determined from the inversion of fault kinematic and earthquake focal mechanism data. This newly focal mechanisms catalog can be used to improve understanding of the contemporary seismotectonic processes and contribute to a better constraint of the source variability for the seismic hazard evaluations.

Method and data acquisition

Earthquake focal mechanisms

As part of the upgrade of the Romanian Seismic Network during the last decade, earthquake detection has increased since the number of seismic stations installed has also known an impressive rate, therefore we have obtained much more fault plane solutions.

We processed earthquake data recorded by the National Institute for Earth Physics from Romania (NIEP) from 2005 to 2020. During this period, NIEP had undergone a steady upgrade from digital strong-motion K2—composed of 44 strong motion sensors, a telemetry network consisting of 14 short-period sensors, and a seismic array composed of 10 seismic sensors (9 short-period sensors and a 3 component broadband sensors), in 2005 to 117 velocity stations (broadband and short-period) and 163 strong motion stations (EpiSensor-2 g full scale) (Fig. 2) by the end of 2020 (Marmureanu et al. 2021). The stations are distributed to cover the entire territory with more stations located around the Vrancea seismic zone.

Fig. 2
figure 2

Location of the analyzed Vrancea intermediate-depth earthquake epicenters (time window between 2005 and 2020; red dots) and of the seismic stations (black triangles) on a geological sketch map. See Asch (2003) for a detailed legend of the geological map

Full size image

More than 4000 local earthquakes were recorded by NIEP from January 1, 2005 to December 31, 2020, in the Vrancea subcrustal/intermediate-depth region, with magnitudes Mw ranging between 2.1 and 5.6 (ROMPLUS catalog—www.infp.ro). Of these, the fault plane solutions could be calculated for 437 seismic events, with magnitude Mw starting from 2.7 to 5.6.

In this study, we used a relocated data set that was obtained by manually picking and location (Hypocenter code/Seisan software (Havskov et al. 2020). Errors in P-wave first-motion observations can occur due to station polarity reversal or incorrect selection of direct P-arrival picks because of a low signal-to-noise ratio. Following the careful analysis of the waveforms (3 components/station) all the stations for which the minimum quality conditions were not met, were eliminated (the signal-to-noise ratio greater than 2.5 and the recorded P-waves were unfiltered). For these reasons, to constrain our focal mechanism solutions by reducing the number of discrepancies, we proof-checked the dataset for possible reversal polarities and we filtered the polarities based on the weight assigned to the P-onset (discrete classed from 0, best, to 3, worse) keeping only polarities for picks with weights 0 and 1 and rejecting those with weight 2 and 3. The 1-D velocity model used for event locations and focal plane solutions was the one obtained by Koulakov et al. (2010).

The fault plane solutions have been obtained on the basis of the first motion of P-wave polarities using FOCMEC code developed by Snoke et al. (1984), distributed as part of the FOCMEC package (http://www.geol.vt.edu/outreach/vtso/focmec) and incorporated in SEISAN software (Havskov and Ottemöller, 2020).

The solutions estimated from P-wave polarities were obtained by using no less than 10 observations, this, in the first part of the catalog, where the number of stations was small, following that the number of polarities increased, with improving the number of stations installed in the Romanian Seismic Network, therefore its confidence level is considerable.

In order to obtain a high degree of confidence for the fault plane solutions, the input parameters were manually processed (phase picking, weight and polarity, location, etc.). The code makes a grid search and finds how many polarities fit each possible solution. All solutions with less than a given number of wrong polarities within given error limits are then written out and plotted. Then the preferred solution is selected, which is stored in the output file or the database.

We used the R software (R Core Team 2021) and RFOC library (Lees 2018), which includes subroutines to represent and classify all the focal mechanisms in our catalog. In order to represent focal mechanism data, Frohlich and Apperson (1992) proposed a diagram to visualize focal mechanism data as a function of the rupture type. This classification is not accurate, presenting distortion issues (Frohlich 2001), and Kagan (2005) used the Kaverina et al. (1996) projection to avoid them. The latter projection is applied in our study. The fault type was divided by seven types of double couple classification: (N: normal; N-SS: normal-strike-slip; SS-N: strike-slip-normal; SS: strike-slip; SS-R: strike-slip-reverse; R-SS: reverse-strike-slip; R: reverse) alike to the conceptual geological characterization of faults. To furnish thorough data and facilitate easy insertion in different databases, we also provided the classification adopted by the World Stress Map project (Zoback 1992). This classified the events into five types (N, N-SS, SS, R-SS, R), and those not fitting in any of the five types were placed in the unknown category (Table 1 in Appendix).

Tectonic stress inversion

The stress regime based on the inversion of 437 earthquake focal mechanisms obtained during this study has been determined by the Win-Tensor program package (Delvaux and Sperner 2003). The inversion is especially supported by the idea that the stress field is uniform and invariant in space and time, and slip on a plane occurs in the direction of the maximum resolved shear stress (Bott 1959; Delvaux and Barth 2010). The fault-slip data are inverted to get the parameters of the reduced stress tensor. These four parameters are the principal stress axes σ1, σ2 and σ3, where σ1 > σ2 > σ3 and the stress ratio R = σ2 − σ3/σ1 − σ3, (0 < R < 1), which defines the magnitude of σ2 relative to the magnitudes of σ1 and σ3 (Delvaux and Sperner 2003).

According to Delvaux and Sperner (2003), the stress regime depends on the orientation of the principal stress axes and the shape of the stress ellipsoid (R): extensional when σ1 is vertical, strike-slip when σ2 is vertical, and compressional when σ3 is vertical. In order to express numerically the stress regime, we used the stress regime index R′, determined based on the stress ratio R and the most vertical stress axis (σ1, σ2 or σ3). This is defined as: R’ = R for normal faulting regime (σ1 is vertical), R′ = 2 − R for strike-slip regimes (σ2 is vertical) and R′ = 2 + R for thrust faulting regimes (σ3 is vertical) (Delvaux and Sperner 2003).

The horizontal principal stress direction SHmax (maximum horizontal stress axis) or Shmin (minimum horizontal stress axis) are computed according to the method of Lund and Townend (2007) and follow the guidelines and terminology used in the World Stress Map Project (available online at http://www.world-stress-map.org/).

Results and discussion

Seismicity and focal mechanism retrieval and distribution

We computed a catalog of focal mechanisms, with strike, dip, and rake for both nodal planes of each mechanism, P and T axis (azimuth and plunge), following the convention by Aki and Richards (1980). The minimum and maximum horizontal stress (Shmin and SHmax), stress regime index, the style of faulting, polarities number and polarities errors are also presented in Table 1 in Appendix.

The new catalog of focal mechanisms obtained for earthquakes recorded in Vrancea intermediate-depth area comprises 437 solutions, for events with magnitudes Mw starting from 2.7 to 5.6 and interval depth 60–162 km, that occurred from 2005 to 2020 (Table 1 in Appendix). A magnitude Mw threshold of 2.7 was chosen, considering the values that provided well-constrained fault plane solutions, on the basis of the network configuration and the resulting available polarities.

From 2005 to 2020 have been recorded a large series of significant intermediate-depth earthquakes in Vrancea seismic area. The substantial improvements in the Romanian Seismic Network (especially in the number of the real-time seismic stations) and a greater azimuthal coverage makes it possible a better understanding of the fault plane mechanisms and a better-constrained solution for the depth-intermediate Vrancea earthquakes. At the same time, the quality of the recorded data has been improved allowing a better location of the seismic events. Therefore, in 2005 the minimum magnitude to obtain a focal mechanism solution was around 3.8, using 10 good polarities, and in 2018, the minimum magnitude is 2.7, using 40 good polarities. Figure 3 shows the difference in the number of stations used for an earthquake of the same magnitude Mw = 4.1, the first in 2005-with 15 polarities and the second in 2020 with 79 polarities.

Fig. 3
figure 3

Focal mechanism determination and distribution of P-wave polarities for 2 seismic events recorded on October 21, 2005 and March 11, 2020, with the same magnitude Mw = 4.1

Full size image

The magnitude for intermediate-depth Vrancea earthquakes depends on seismic attenuation, and for that reason (strong attenuation in the Carpathian Arc) the seismic signal amplitude is diminished (Oth et al. 2008), which makes it difficult for us to detect earthquakes with Mw < 2.0, and also the capacity of the seismic source to generate frequent events with Mw > 7.0 (Radulian et al. 2000). Most of the seismic events recorded in this area have an average magnitude of Mw = 3.5 (Radulian et al. 2000).

In Fig. 4 are plotted the hypocenters of the seismic events with focal mechanisms solutions available from the Vrancea region, which is characterized by a puzzling intermediate-depth high seismicity body (Ismail-Zadeh et al. 2012; Martin et al. 2006) with the stronger seismic activity concentrated between 60 and 160 km depth. This indicates the almost vertical pattern of the intermediate-depth seismicity, which is very well correlated with the seismic tomography of Vrancea (Petrescu et al. 2021, Maţenco et al. 2007, Ismail-Zadeh et al. 2000).

Fig. 4
figure 4

Hypocenter distribution of the focal mechanisms determined in this study (2005–2020) (represented by red dots)

Full size image

The spatial distribution of the focal mechanisms that was obtained in our study is shown in Fig. 5. To observe the local changes in the distribution of the focal mechanisms with depth, we divided them into different depth intervals: 60–80 km, 80–110 km, 110–130 km and 130–160 km. We can distinguish two domains of depth depending on the number of events that occurred in the analyzed period (60–110 km and 110–160 km) in Fig. 5, as well as on the account of the energy released by these earthquakes in the same period (Trifu et al. 1994; Radulian et al. 2002).

Fig. 5
figure 5

Spatial distribution of focal mechanisms from Vrancea intermediate-depth, between 2005 and 2020. Focal mechanism faulting styles and the selected number of focal mechanisms on each depth interval are also presented. The type of focal mechanisms is displayed in legend, with different colors corresponding to different types of faulting according to the standard classification (Kaverina et al. 1996). Colored areas show compressional quadrants

Full size image

The spatial distribution (hypocenters) of the focal mechanisms is in perfect agreement with the geometry of the Vrancea zone, oriented NE–SW, parallel with the Carpathian Arc (Radulian et al. 2000; Ismail-Zadeh et al. 2012; Ferrand and Manea 2021).

Focal mechanisms classification and interpretation

To analyze the type of faulting, we have classified earthquakes with focal mechanisms as normal (N), normal-strike-slip (N-SS), strike-slip-normal (SS-N), strike-slip (SS), strike-slip-reverse (SS-R), reverse–strike-slip (R-SS), reverse (R) faulting based on the theory described in the methodology. The resulting diagram incorporating those 7 fields classification to the Kaverina projection is shown in Fig. 6A. The average style of faulting by depth thresholds and in agreement with the World Stress Map project (Zoback 1992) is shown in Fig. 6B.

Fig. 6
figure 6

Focal mechanisms classification. A Triangle diagram for displaying the distribution of double couple focal mechanism, according to the standard classification (Kaverina et al. 1996). B Diagram depth distribution of the focal mechanism data. The number of events is plotted on different depth intervals and the colors code is described in the legend (NF- normal fault, NS- normal strike-slip fault, SS- strike-slip fault, RF- reverse fault, RS- reverse strike-slip fault, UF- unidentified fault), by Zoback (1992)

Full size image

For the Vrancea seismic zone in most of the cases (75%), the fault plane solutions show a reverse fault mechanism, followed by a reverse fault with an important strike-slip component (12.6%). Normal or normal—strike-slip faults are observed only for 17 events (3.8%) with magnitudes Mw ≤ 4.0, located in the deeper part of the seismogenic zone (h > 110 km). Several mechanisms exhibit a strike-slip significant component, 28 events (6.4%). At 9 events, the type of faults could not be identified (Fig. 6).

Tectonic regime stress inversion

For the intermediate-depth earthquakes recorded in Vrancea, from the nodal plane orientation point of view, we can distinguish two main earthquakes categories. The first one contains the majority of earthquakes that are mainly characterized by the nodal planes oriented NE–SW, parallel with the Carpathian Arc and with the strike angle distribution in perfect agreement with the geometry of the Vrancea zone (Fig. 7) and with previous studies (Craiu et al. 2015a, b; Radulian et al. 2000; Tugui and Craiu 2008). Notably, this geometric alignment corresponds to the system of rupture faults for the most important Vrancea earthquakes (Onescu and Trifu 1987; Radulian et al. 2002). The second category that characterizes the Vrancea intermediate-depth earthquakes is the fault plane solutions in which the nodal planes are oriented NW–SE, perpendicular to the Carpathian Arc.

Fig. 7
figure 7

Angular diagrams for strike, dip and rake angles of the nodal planes for investigated Vrancea earthquakes

Full size image

For the majority of events, the slip is strongly concentrated around 90°direction, indicating a clear reverse faulting mechanism on steeply dipping faults. For the other predominant part of the events, the slip angle indicates the pronounced dip-slip faulting, but with an increase of a second strike-slip component (Fig. 7), (Radulian et al. 2019).

An alternative way of indicating the spatial distribution of earthquake focal sphere data is to plot the density distributed contours of P- and T-axes as they vary spatially (Fig. 8). This represents a measure of the variation of stress orientation in an active tectonic region.

Fig. 8
figure 8

Stereographic projections of the compression (P), extension (T) axes and combined of focal mechanisms for the period 2005–2020. The dots indicate individual focal mechanism solutions. The colored background indicates the smoothed density distribution of all solutions

Full size image

As can be seen in Fig. 8, the principal T axis tends to be almost vertical, while the principal P axis tends to be horizontal, with a predominant direction oriented perpendicular to the Carpathians Arc.

While individual focal mechanism solutions are the primary focus of this study (Table 1 in Appendix), it is worthwhile describing the overall spatial structure of the seismotectonic deformation field, as inferred from the analysis of our catalog. Indeed, we can make the assumption that the earthquake focal mechanisms reflect the state of stress of the studied areas, which implies that at a first approximation, the P-axes and T-axes correspond to the principal stress axes σ1 and σ3, respectively (Figs. 8 and 9).

Fig. 9
figure 9

A Stress inversion of the focal mechanisms data based on Table 1 in Appendix. The results are represented by the orientation of the 3 principal stress axes: inclination/direction (a red dot surrounded by a circle for σ1, a triangle for σ2 and a square for σ3). The related SHmax and Shmin orientations are represented by large arrows outside the stereogram (SHmax—blue arrows, Shmin—green arrows). Their type, length and color symbolize the horizontal deviatoric stress magnitude relative to the isotropic stress (σi) and are in function of the stress regime and the stress ratio R = σ2 − σ3/σ1 − σ3. The strike direction of the fault plane solutions (gray rose diagram), the stress ratio-R and the selected number of focal mechanisms on each depth interval is also presented. B Stability of stress regime index (R′)

Full size image

Figure 9 shows the well-known Right Dihedron method (Delvaux 2012), a graphical method for the determination of the range of possible orientations of σ1 and σ3, stress axes in fault analysis.

For the seismic source delineated in the Vrancea intermediate-depth area, the stress field has almost a pure compressional stress regime (σ3 vertical), with maximum horizontal stress (SHmax) oriented in NW–SE direction. The horizontal stress magnitude is not very well constrained due to the radial compressional character of the stress tensor. The R ratio is 0.81 (close to 1), which means that the relative magnitudes of the σ2 and σ1 stress axes are almost similar. As these axes are subhorizontal for a compressional regime, SHmax (~ σ1) and Shmin (~ σ2) are not very well differentiated.

The results of stress inversions of intermediate-depth earthquakes, SHmax orientation and thrust fault regime with a radial component (R′ = 2.87) are consistent with the observed geological Vrancea setting and the results of previous studies (Petrescu et al. 2021; Oncescu and Trifu 1987).

The most encountered fault plane solution striking in the NE–SW direction and dipping toward NW (Figs. 7 and 9) is considered a rupture plane, in agreement with the spatial distribution of earthquakes and source directivity analysis (Radulian et al. 2000; Tugui and Craiu 2008; Oncescu and Trifu 1987). At the same time, a smaller number of less significant earthquakes have the rupture plane oriented almost perpendicular to the previous rupture plane and to NE–SW alignment of epicentral distribution of earthquakes.

Another aspect of the stress patterns obtained from formal stress inversion of fault plane solutions reveals numerous characteristics of the current stress field that are not reproduced by large-scale numerical models.

Conclusions

We determined the latest focal mechanisms catalog of 437 earthquakes that occurred in the Vrancea intermediate zone, from January 2005 to December 2020 using first-motion polarities, processed manually, applying the FOCMEC method (Table 1 in Appendix). We also determined the stress field via focal mechanism inversion, (the main stress axes, maximum horizontal stress) and the tectonic regime for the seismogenic body delimited by the Vrancea area.

To determine a refined catalog with high-quality focal mechanisms, we used selection criteria based on parameters that are sensitive to the variation of data quality and the data coverage on the focal sphere. Nodal plane uncertainty is correlated with the number of observations in which the more observations the smaller range of uncertainty.

The new result shows that in the subcrustal Vrancea source, reverse faults are predominant, regardless of depth or magnitude range. The stress field obtained is pure compressional with a radial component (R′ = 2.87). The maximum horizontal stress (SHmax) obtained for Vrancea intermediate-depth is mainly NW–SE oriented, with in good agreement with the geometry Vrancea slab and other study (Petrescu et al. 2021; Radulian et al. 2018). Other tectonic regimes as normal and strike-slip faults are also present in the studied dataset, but they are retrieved for a reduced number of events.

It is well known that Vrancea is one of the most active seismic zones in Europe, being characterized by intermediate-depth seismicity with frequent earthquakes of magnitude MW ≥ 7. Intermediate-depth earthquakes (60–160 km), analyzed in this paper, constitute a part of a small subcrustal body-oriented NE–SW, which represents a relic slab of the subducted lithosphere (Ismail-Zadeh et al. 2000), with the axes of compression and tension close to the horizontal and vertical directions, respectively. Vrancea region is still a major source of seismic hazard (Cioflan et al. 2022), and latest catalog of focal mechanisms obtained in this study will help to implement the new research about the seismic source and seismotectonic characterization.