Ductile Versus Brittle Tectonics in the Anatolian–Aegean–Balkan System

Abstract

It is hypothesized that the present tectonic setting of the Anatolian, Aegean and Balkan regions has been deeply influenced by the different deformation styles of the inner and outer belts which constituted the Oligocene Tethyan system. Stressed by the Arabian indenter, this buoyant structure has undergone a westward escape and strong bending. The available evidence suggests that in the Plio–Pleistocene time frame, the inner metamorphic core mainly deformed without undergoing major fragmentations, whereas the orogenic belts which flanked that core (Pontides, Balkanides, Dinarides and Hellenides) behaved as mainly brittle structures, undergoing marked fractures and fragmentations. This view can plausibly explain the formation of the Eastern (Crete–Rhodes) and Western (Peloponnesus) Hellenic Arcs, the peculiar time-space features of the Cretan basins, the development of the Cyprus Arc, the North Aegean strike-slip fault system, the southward escapes of the Antalya and Peloponnesus wedges and the complex tectonic setting in the Balkan zone. These tectonic processes have mostly developed since the late Late Miocene, in response to the collision of the Tethyan belt with the Adriatic continental domain, which accelerated the southward bending of the Anatolian and Aegean sectors, at the expense of the Levantine and Ionian oceanic domains. The proposed interpretation may help us to understand the connection between the ongoing tectonic processes and the spatio-temporal distribution of major earthquakes, increasing the chances of estimating the long-term seismic hazard in the study area. In particular, it is suggested that seismic activity in the Serbo–Macedonian zone may be favored by the post-seismic relaxation that develops after seismic crises in the Epirus thrust front and inhibited/delayed by the activations of the North Anatolian fault system.

1. Introduction

The geodynamics of the complex tectonic evolution of the eastern Mediterranean region (Figure 1) is still subject to debate in the relevant literature. The aspect that stimulates most discussions is the development of extensional zones in a region stressed by the convergence of the confining plates (Nubia, Arabia and Eurasia). The most cited interpretation suggests that the Aegean extensional zone has been generated by the gravitational sinking of the Ionian slab (slab-pull mechanism, e.g., [1,2,3,4,5,6,7]). This hypothesis has been considerably encouraged by the velocity field delineated by geodetic observations (e.g., [8,9]), which shows a faster motion of the Aegean zone (30–40 mm/y) with respect to the Anatolian wedge (10–15 mm/y). However, this view does not consider that the present kinematics can be a transient phase of the post-seismic relaxation triggered by the strong seismic sequence that developed along the North Anatolian fault since 1939. The tentative simulation, by numerical modelling, of that perturbation [10,11] shows that the expected present velocity field is very similar to the one defined by the geodetic data in the Anatolian–Aegean region. Furthermore, it must be pointed out that the almost uniform velocity field delineated by geodetic data in the Aegean area (e.g., [8,9]) is not compatible with the extensional regimes that generated the Cretan basins. The kinematic citations given in the text are always with reference to Eurasia.

Other authors [12,13,14,15,16,17,18,19,20,21], on the other hand, suggest that a dominant role in the evolution of the eastern Mediterranean region has been played by the westward extrusion of the Anatolian wedge and by the development of extrusion processes. This choice is supported by an accurate analysis of the compatibility of the proposed interpretation with the observed deformation pattern and of the major difficulties that the slab-pull model can encounter to explain such a pattern [17,18,19,20,21,22,23]. The main objectives of this work can be synthetized as follows. One aim is to provide further arguments in support of our previous interpretations of the study area, discussing in greater detail the compatibility between the proposed driving forces and the observed deformation pattern in the Balkan zones. In particular, it is pointed out that the evolution of the Anatolian–Aegean system since the Late Miocene–Early Pliocene can be interpreted as the effect of two parallel belts that deform by different styles. The inner metamorphic chain behaves as a mainly ductile structure, undergoing only minor fragmentation, whereas the outer orogenic belts behave as mainly brittle structures, undergoing major fragmentations. This view is supported by the fact that seismicity is mainly related to the outer chains.

Another objective of this work is to provide a tentative explanation for the apparently ambiguous evidence on the strain regimes recognized in the Serbo–Macedonian zone on the basis of geological and geodetic data. Subsequently, some hypotheses are advanced about the possible relation between the proposed tectonic setting of the Anatolian and Aegean regions and the spatio-temporal distribution of major earthquakes since 1600. Some considerations are also devoted to the comparison of the expected effects of the major seismic sequence that developed along the North Anatolian fault since 1939 with the seismicity patterns in the Aegean and Balkan zones.

2. Geodynamics and Neogene Tectonic Evolution

The orogenic system that in the Oligocene separated the African and European domains (Iberian–Alpine and Tethyan belts, Figure 2a) was generated by the consumption of the Northern Neo Tethys domain since the Cretaceous and by the accretion of continental fragments interposed between small oceanic basins (e.g., [24,25,26,27,28]). The consumption and partial obduction of the oceanic domains generated ophiolitic belts (e.g., [29,30,31,32]), while underthrusting processes at continental collision boundaries generated the main metamorphic core [33,34,35,36,37,38,39,40]. As suggested by a number of authors (e.g., [26,34,41,42]), that orogenic system (Tethyan belt) was constituted by parallel chains: an inner core (Figure 2), consisting of oceanic remnants and crystalline massifs (Pelagonian–Cyclades–Anatolian), and two outer accretionary chains, one with European affinity (Carpathians, Balkanides and Pontides) and one with African affinity (Dinarides and Hellenides). The geometry of these belts shown in Figure 2 is only indicative.

Stressed by the strong compressional context and uplift created by the Arabian indenter in the Late Oligocene, the Anatolian Tethyan belt decoupled from the European domain by a shear zone predating the NAF, i.e., the SE–NW Trans Anatolian decoupling fault system (e.g., [45,46,47]) and underwent a long NW-ward displacement (Figure 2b). This process may have been determined by the fact that in such structural/tectonic settings, the consumption of the oceanic and thinned continental domains in the Magura zone [48] was the most convenient shortening process. The migration path of that buoyant orogenic material was guided by the corridor that existed between the continental European domain (Moesia) and the continental Adriatic (e.g., [49]). The consumption of the Magura domain was produced by the opposite rotations of the Alkapa and Tisza blocks (Figure 2b). Back arc extension developed in the wake of these diverging blocks, forming the Pannonian basin. When the above trench zone sutured in response to the complete consumption of the thinned domains, around the Early–Middle Miocene ([50,51], Figure 2b), the deformation pattern of the Tethyan belt changed drastically, since the most convenient shortening process became the consumption of the Ionian and Levantine oceanic domains. This result was achieved by the SW-ward bendings of the Anatolian–Aegean Tethyan belt (Figure 2c, [19,20]).

This evolutionary phase lasted until the collision of the Aegean Tethyan belt with the continental Adriatic domain. Initially, this collision caused shortening and uplift in both the subsequent fold and thrust structures, occasionally affected by the inversion of previous normal faults in the Ionian domain during the Tortonian–late Messinian interval (e.g., [52]) and by thrusting and uplift in the western Aegean zone (Central Greece, [53,54]). Once the most thinned domains in the collision zone were consumed, the westward displacement of the extruding Anatolian wedge was accommodated by further bending of the Tethyan belt (Figure 3).

We advance the hypothesis that the very complex distribution of tectonic processes that developed during the Pliocene–Quaternary evolution in the study area (Figure 3b) was mainly conditioned by the different rheological behaviors of the inner Tethyan belt [55,56], which may have deformed as a mainly ductile medium, and the surrounding orogenic belts, which deformed undergoing major fragmentations, indicating a mainly brittle character. The ductile behavior can most reliably be inferred for the inner metamorphic belt (violet in Figure 1), while the pink belt, being only partially formed by metamorphic material (Ophiolites units), may deform with an intermediate style (between brittle and ductile). The computation of rheological profiles in the central and eastern Mediterranean area [55] indicates that the brittle upper crustal layer may reach 10 to 20 km, depending on the geothermal state. However, it is difficult to understand if this value may be taken as representative of the Plio–Quaternary situation.

The reliability of the above interpretation is checked by taking into account the strain regimes that the deformation of the inner core is expected to induce in the surrounding orogenic belts and analyzing their compatibility with the observed deformation patterns.

A major example of the supposed ductile-brittle behavior may be recognized in the formation of the western (Peloponnesus) and eastern (Crete–Rhodes) Hellenic arcs (Figure 3a). Stressed by the convergence between the Anatolian block and the Adriatic plate, the inner belt (Cycladic arc) underwent southward bowing, without significant fragmentations, whereas the outer brittle belt, being affected by intense belt-parallel extensional stresses in its most curved zone, broke in two sectors, forming the western and eastern Hellenic arcs (Figure 3b). After separation, these two arcs progressively released their previous horizontal flexure, by clockwise (Peloponnesus) and counterclockwise (Crete–Rhodes) rotations. The consequent divergence between these belt sectors and the Cyclades massif (moderately deformed) induced the extension that formed the western Cretan basin, in the Early–Middle Pliocene (e.g., [57,58,59]). This interpretation can account for the almost triangular shape of that basin and for the timing of that event.

After its detachment from the eastern Hellenic Arc, the Peloponnesus wedge underwent E-W compression, driven by the convergence between the inner core (Cyclades massif) and Adria (Figure 3c, e.g., [60,61]). This shortening was accommodated by uplift and thrusting in the Epirus zone, while the compressional deformation pattern was more complex in the Peloponnesus wedge (Figure 3c). The difference was mainly caused by the SSW–NNE orientation of the southernmost Adria border, which created an asymmetric compressional mechanism. In such a context, the Aegean Tethyan belt underwent a clockwise bending (Figure 3c), which induced belt-parallel extensional stresses at its eastern side. This led to the formation of some E–W to SE–NW troughs, such as the Corinth and Ambracique, in the Peloponnesus sector, and the troughs that now separate the Trapezona, Parnassus and Eubea land structures, in the Greek inner belt (Figure 1). The effects of this mechanism were emphasized by the southward extrusion of the Peloponnesus slats [62,63]. The fact that the western guide of this extrusion (the Cephalonia fault) was shorter than the eastern one (inner Aegean Tethyan belt) produced an oblique escape, involving clockwise rotations of curved slats in the Peloponnesus (Figure 3c).

Another major example of ductile versus brittle tectonics is given by the development of the Cyprus arc [64,65,66,67]. The detachment of this belt sector from the main Anatolian body was caused by the strong bending it underwent at the outer side of the main central Anatolian curved sector, similarly to what happened in the Hellenic Arc. This detachment led to the formation of the Antalia and Cilicia–Adana basins (e.g., [68,69]).

The E–W to SSE–NNW convergence between the Anatolian wedge, on one side, and the Rhodope massif and the Adriatic continental domain, on the other side, (Figure 3b,c) induced E–W compression in the northwestern Anatolian corner and in the northern–central Aegean zone (e.g., [54,57,70,71,72]). Contemporaneously, these zones were stressed by N–S to NNW–SSE extension, induced by the southward bending of the inner Aegean core (Figure 3c). The combined effect of these deformations led to the formation of a system of dextral transtensional faults in the northern Aegean and northwestern Anatolian zones [3,73,74,75,76].

The present slab under the Aegean zone is identified by intermediate earthquakes, reaching a depth of about 130–140 km [77,78]. Another evidence about the oceanic subducted lithosphere under the Aegean area is given by the calc-alkaline and high-K calc-alkaline geochemical signature of the volcanic arc located along the southern margin of the Cyclades Arc (Figure 3c, [79]). A very interesting feature of this magmatic arc is the fact that its western and central—eastern sectors developed in rather different times. The western part (Aegina, Methana and the oldest volcanic rocks of Milos and Santorini) formed in the Early Pliocene, while the central and eastern sectors (Milos, Santorini, Nisyros, e.g., [78,80,81] and references therein) date back to the Middle–Late Pleistocene.

Since these times correspond well to the formations of the western Cretan basin (Early Pliocene) and the central–eastern Cretan basin (mid to late Quaternary), one may suppose that all magmas were generated at the same time (i.e., when the slab reached the opportune depth), but their uprise was only allowed by the formation of transtensional fault systems in the upper crust of the western and eastern Cretan basins. The possible connection between pull-apart transtensional features and volcanic activity is suggested by what happened in many other zones, such as the Roman and Campanian volcanism in the Italian region [82,83] and elsewhere (e.g., [74,84,85,86,87,88,89,90]).

The westward displacement of the central Anatolian body and the contemporaneous SE-ward bending of the western Anatolia metamorphic belt (Figure 3) induced E–W compression in the interposed zone. This mechanism may explain the SW-ward escape of the Antalya wedge, guided by the Fethiye–Burdur fault. Since at that time the Antalya zone was still connected with the eastern Hellenic arc, the above extrusion also involved the Crete–Rhodes sector, accelerating its SW-ward motion. Around the early Pleistocene, this mechanism led to the collision of the Crete–Rhodes sector with the Lybian promontory. Resisted by this obstacle, the southern part of Crete (Figure 3c) slowed down, whereas the main Crete slat was only deviated, assuming the present E–W orientation. The relative motion between main Crete and the resisted southern slats was allowed by the formation of the Pliny and Strabo faults (Figure 3). The clockwise rotation and westward displacement of the Cretan slat can explain the formation of the eastern Cretan basin and the interruption of the land continuity with Karpathos and Rhodes [57,62,91].

Pushed by the Tethyan belt, the Northern Hellenides belt (Epirus) was forced to migrate westward, overthrusting the Adriatic domain (Figure 3). The decoupling of this belt sector from the southern Dinarides was accommodated by the formation of dextral shear faults, such as the Scutari–Pec and Vlore–Elbasan, and by the clockwise rotation of the Albanides [92,93].

The Neogene–Quaternary tectonic setting in the Balkan zone (Pelagonian, Vardar, Serbo–Macedonian and Rhodope structures) is mostly recognized as related to S–N to SW–NE extension (e.g., [3,94,95,96,97,98,99]), but the driving mechanism of this strain regime is still subject to debate. Some authors suggest a connection with the extensional process that formed the Aegean zone (e.g., [3]).

In our opinion, the Plio–Quaternary deformation pattern in the SMM zone can be explained as an effect of two driving mechanisms, both induced by the westward displacement and deformation of the Tethyan belt. One is the westward push of northern Anatolia on the Rhodope massif. The zone where this push is exerted corresponds to the Ganos–Gelibolu thrust zone, where Pliocene compressional deformations are recognized ([100], Figure 3). This interaction causes a roughly NW-ward displacement, along with a clockwise rotation, of the Rhodope massif (Figure 3). The consequent divergence between this massif and the European domain (Moesia) induced the dextral transtensional regime that formed the upper Thrace trough. (Figure 3, [3,19,101]). If this was the only driving mechanism, one could expect compressional deformation in the zone stressed by the Rhodope massif, i.e., the Serbo–Macedonian massif (SMM). To explain why this last zone is instead characterized by S–N extension, one must consider the effects of the other driving mechanism that induces E–W extension in the SMM, i.e., the roughly westward displacement of the Pelagonian–Vardar units, as a part of the Tethyan belt (Figure 3c). In fact, the relative motion between the Rhodope massif (moving roughly NW-ward) and the Pelagonian–Vardar belt (moving roughly west- to WNW-ward) induces a S–N extension in the interposed zone, i.e., the SSM.

Another obstacle against a full understanding of tectonics in this zone is given by the fact that geodetic observations in the Simitli graben (SMM) indicate compressional deformations and uplift (Pirin horst) [102], in contrast to the extensional regime indicated by geological evidence [3,94,95,96,97,98,99]. We suggest that this puzzling problem can find a plausible solution by supposing that the present shortening recognized in the Simitli graben is a transient phase of the stress field perturbation that was triggered by the strong earthquakes that occurred along the NAF from 1939 to 1999. Such a sequence allowed a remarkable westward displacement of northern Anatolia [103], increasing its push on the Rhodope massif and consequently compression in the SMM zone. Such processes, propagated with velocities controlled by post-seismic relaxation, are expected to reach the SMM zone after several decades from the triggering events [10,11]. Thus, it is reasonable to suppose that the compressional stress field involved by the above event is still affecting the SMM zone. This hypothesis is supported by the monitoring of the strain rate at some points in the Simitli graben in the time interval 1980–2010 (Struma and Krupnic faults, [102]), which show that the 1999 strong Izmit earthquake (M = 7.4), along the western NAF, determined a remarkable increase in shortening in the observation points.

Since it is widely recognized that an increase in compression on a fault zone tends to inhibit or at least delay the occurrence of major earthquakes, the present compressional stress field in the Simitli graben could indicate a low probability of major earthquakes in that zone. This possibility can find important support from the fact that since 1940 (during the NAF seismic sequence), only very few earthquakes (with M < 6) occurred in the SMM zone (Figure 4).

The distribution of major earthquakes shown in Figure 4 can also support another basic concept of seismotectonics, i.e., the hypothesis that an increase in extensional stresses on a fault is expected to favor seismic activity. The most convincing evidence in this regard is given by the fact that the strongest earthquakes in the SMM zone just occurred during or after major seismic crises in the Epirus zone (Figure 4), which are supposed to induce E–W extension in the SMM zone [19].

3. Main Seismic Zones

The distribution of major earthquakes that occurred since 1600 (Figure 5a) can provide an idea about the location of the seismogenic faults in the study area. Focal mechanisms and geological data allow the recognition of the strain fields in the main seismic zones (Figure 5b).

In the Albania zone, seismicity is associated with SE–NW compression (Albanides thrust fronts), segmented by two main NE–SW dextral transversal faults, i.e., the Scutari–Pec and Vlora–Elbasan [92,125,126,127,128].

In the Epirus zone, NW–SE to S–N thrust faults dominate, allowing the belt units to overthrust the southern Adriatic domain (e.g., [129,130,131]).

In the Cephalonia fault zone, dextral transpressional focal mechanisms, associated with a significant uplift, are related to the overthrusting of the Peloponnesus wedge on the Adriatic plate (e.g., [132,133,134]). This fault system is interpreted as the surface expression of a tear fault which decouples the Hellenic slab from the Adriatic continental margin that underthrusts the Northern Hellenides (e.g., [135]).

The central part of Greece is undergoing a roughly S–N extensional regime, mainly evidenced by focal mechanisms in the Corinth, Ambracique and Thessaly grabens [60,81,136,137,138]. This regime is compatible with the southward extrusion of the Peloponnesus wedge, in response to E–W compression [19].

The Northern Aegean and Northwestern Anatolian regions are cut by a system of dextral NE–SW transtensional faults [139,140], with a progressive increase in the extensional component from east to west (e.g., [136]). In the northern Aegean, those faults bound morphological depressions, such as the Sporades, Skiros and Ikaria. The north Aegean fault system is connected with the North Anatolian fault through the Ganos transpressional lineament and the Marmara pull-apart trough [15,19,141,142,143]. The northwestern Anatolian zone is characterized by prominent E–W grabens (e.g., [144]). The stress field is dominated by N–S extension, compatible with the source of the recent 2020 Samos earthquake (M = 7.0, [145] and references therein).

In the SMM zone, most authors (e.g., [3,96,97,98,99,146,147]) suggest that the main faults were generated by N–S extension. In the upper Thrace graben, along the northern side of the Rhodope massif, N–S to NE–SW extension is recognized (e.g., [3,101]). The focal mechanism of the strong 1928 destructive earthquake (Plovdiv, Bulgaria, M = 6.8 and 7.1) indicates a WNW–ESE oriented normal fault with a significant dextral component [148].

In SW Anatolia, a NE–SW shear zone, known as Fethiye–Burdur fault (Figure 1) has been identified along the western side of the Antalya wedge (e.g., [149]). Strong earthquakes in the southernmost sector of this fault, near the Rhodes basin (e.g., 1957, M = 6.8, 7.2; 2012, M = 6.2), indicate NE-trending sinistral strike-slip mechanisms [150].

In the Eastern Hellenic arc, both GPS observation and earthquakes focal mechanisms suggest NE–SW sinistral transpressional deformation on the outer side (Pliny and Strabo trenches) and prevalent extensional regime on the internal side, with orientations of the extensional principal axis progressively varying from NW–SE in eastern Crete to E–W in the Cretan and Karpathos basins [81,151,152].

The North Anatolian fault (NAF) is mainly characterized by right-lateral strike-slip deformation. Slip rates based on GPS data (20–25 mm/y (e.g., [8]) are significantly greater than the rates inferred from geological evidence (6.5 mm/y in the eastern sector, to 17–19 mm/y in the western sector, e.g., [15,153]). The western NAF divides into two or three sub-parallel strands [15,154], where source mechanisms indicate transpressional (e.g., Ganos fault, [100,155]) or transtensional (e.g., Edremit gulf, [156,157]) regimes. It can be noted that earthquakes with M ≥ 7.5 only occurred in the eastern NAF [158].

The Eastern Anatolian fault (EAF) is dominated by left-lateral strike-slip motion along the main strand, which extends about 700 km in N60° E direction (e.g., [159]). Towards SW, it connects with the N35° E Amanos fault and then with the N–S Dead Sea fault. Sub-parallel branches split from EAF, giving rise to restraining and releasing (pull-apart) bends, with reverse and normal components in the relative strike slip solutions [160]. Two major earthquakes occurred in 2023 (M = 7.8, 7.7), one in the main strand, almost at the intersection with the Amanos fault, and the other in the northern, E–W oriented strand. Another strong event (M = 6.7) occurred in 2020, northeast of the above bifurcation [161]. Geological and geomorphological studies indicate EAF slip rates decreasing from the northeastern (10 mm/y), to the southwestern (4 mm/y) sectors, in agreement with geodetic data ([160] and references therein).

4. What Can We Learn from the Post-1939 Seismic Sequence Along the NAF?

This work suggests that the Neogene tectonic activity in the Aegean zone has been driven by the westward extrusion of the Anatolian wedge. To check if this context is still going on, it is useful to investigate the possible tectonic implications of the seismic sequence that activated the entire NAF since 1939. For instance, one can try to understand the possible connection between the time pattern of seismicity in the Aegean zone and the expected effects of the above seismic sequence (Figure 6a,b). To make this check, one must consider that the strain perturbation triggered by that sequence travelled with velocities controlled by the rheological properties of the structures involved, as suggested by the concepts of post-seismic relaxation [162,163]. In this regard, we take into account the results obtained by [10,11], who tried to reproduce the effects of the post-seismic relaxation induced by a sudden westward displacement of Anatolia. The resulting time pattern of the induced strain energy rate in the central Aegean zone (Figure 6c) shows the highest values in the time interval 1950–1970. The fact that during the same phase the number of major earthquakes (M > 5.5) in that zone was higher than in the other time intervals (Figure 6b) suggests that seismotectonic activity in the Aegean area may be mainly conditioned by the westward accelerations of Anatolia. If the extensional seismotectonic activity was mainly driven by slab-pull forces, the effects of the post-1939 seismic sequence would have attenuated rather than increased seismic activity in the Aegean zone. This is because a SW-ward motion of the northern Aegean zone (induced by post-1939 relaxation) would have reduced the divergence of this zone from the retreating Hellenic arc.

Another significant aspect of the NAF seismic sequence is that fault activations propagated westward over time. This trend indicates that the previous strain accumulation mainly developed in the eastern Anatolian zone, as implied by the geodynamic interpretation proposed here. Conversely, the slab-pull mechanism implies that strain accumulation would mainly develop in the southern Aegean zone in response to the SW-ward retreat of the Hellenic arc [164].

A further piece of evidence that may provide insights into the tectonic context of the study area is the fact that subcrustal seismicity in the Aegean zone (Figure 7) has undergone a considerable slowdown since 1940. Considering that this time interval mainly corresponds to when the most intense perturbation of the velocity field induced by post-1939 seismic sequence reached the southern Aegean zone, one could suppose that the expected progressive SW-ward displacement of the Aegean zone had increased friction at the Hellenic trench faults, slowing down subduction and then mitigating the related seismicity.

5. Possible Connections Between the Seismic Activations of the Main Faults Bordering the Arabian and Eastern Anatolian Structures

Driven by the spreading of the Red Sea and the Aden gulf, the Arabian plate tries to move roughly northward. This motion is accommodated by shortening processes at the collision zone with Eurasia. At the northeastern front (Zagros), the shortening is accommodated by underthrusting, uplift and lateral escape of masses in a relatively wide zone (e.g., [164,165]). At the northwestern front, the shortening is accommodated by uplift and the westward escape of the Anatolian wedge. The decoupling of Arabia from Nubia is mainly allowed by strong earthquakes along the Dead Sea Fault (DSF). The activities in the above seismic zones are connected to each other, as the decoupling earthquakes in one fault zone may increase stress at the other, still blocked, faults. To tentatively evaluate the reliability of this hypothesis, we have considered the distribution of major earthquakes that have occurred since 1600 in the northern Arabian and eastern Anatolian (EAF and NAF) zones (Figure 8). To evidence the main seismic phases at the above fault systems, we have considered four time intervals (Figure 8c–f).

In the long first phase (1600–1871, Figure 8c), seismic activity mostly occurred along the DSF and the Zagros thrust front. The transcurrent slips along various sectors of DSF allowed the western side of the Arabian indenter to decouple from Nubia, while the activation of several thrust faults in the Zagros zone accommodated a northward displacement of the Arabian indenter. One may expect that these partial accelerations of the indenter increased stress at the sectors of the northern Arabian front that were still blocked. In particular, this phenomenon may have increased shear stress at the EAF and NAF. This hypothesis may explain what happened in the next phases, when four strong earthquakes (1874 M = 7.1; 1875 M = 6.7; 1893 M = 7.1; 1905 M = 6.8) occurred along the EAF (Figure 8d) and then, since 1939, a series of major shocks have activated the entire NAF: 1939 M = 7.9; 1942 M = 7.0; 1943 M = 7.6; 1944 M = 7.4; 1957 M = 7.0; 1967 M = 7.1; 1999 M = 7.4 (Figure 8e).

It is reasonable to think that the exceptional decoupling that accompanied the seismic sequence along the NAF allowed a marked westward displacement of the Anatolian wedge, causing a strong increase in shear stress at the EAF. This hypothesis could explain why three strong earthquakes recently occurred at that fault (2020 M = 6.8, 2023 M = 7.8, 7.7, Figure 8f). It can finally be noted that these last shocks only occurred at sectors of the EAF that were not activated by the major 1874–1905 seismic crisis [166].

6. Conclusions

The complex Neogene tectonic evolution of the Anatolian–Aegean–Balkan system is explained as an effect of the westward migration and deformation that the Tethyan belt has undergone in response to the indentation of the Arabian indenter. The fact that the inner core of the belt has undergone strong bending without significant interruptions suggests a mainly ductile rheological behavior of that structure. Conversely, the two orogenic belts that flanked the inner core underwent major fragmentations (in the zones of maximum bending), indicating a mainly brittle character. Most of the tectonic processes took place since the late Miocene, after the collision of the Tethyan belt with the Adriatic domain, which accelerated the bending of that system. The most evident examples of ductile versus brittle tectonics are given by the formation of the Cyprus and Aegean arcs, where the inner cores underwent a marked bending with no fragmentation, whereas the outer orogenic belts were affected by major breaks in the zones of strongest curvature. The proposed interpretation may provide plausible explanations for the evolution of the Peloponnesus and Creta–Rhodes belt sectors, the formation of the Cretan and Cilicia–Adana basins and the complex set of tectonic processes in the northern Aegean and Balkan regions.

In particular, the tectonic setting in the Serbo–Macedonian massif (mainly the Struma zone) is explained as an effect of two simultaneous driving mechanisms. The roughly NW-ward motion and clockwise rotation of the Rhodope massif (stressed by northwestern Anatolia), induces a compressional regime in the Serbo–Macedonian zone, while the westward motion of the Aegean Tethyan belt (Epirus) induces extension in the same zone. In the long term, the combined effect of these two mechanisms produced a S–N extension in the SMM zone, as indicated by geological data. In the short term, the SMM zone can be affected by compressional or tensional deformations, depending on the previous seismic activity in the NAF or the Epirus thrust front, respectively. After intense seismic crises in the NAF, the consequent westward displacement of Anatolia stresses the Rhodope massif, inducing a compressional regime in the SMM zone. This effect propagates with velocities controlled by the rheological nature of the structures involved. The results of numerical modellings suggest that such perturbation can affect the Balkan zones for several decades. This hypothesis can explain why, during the activation of the NAF since the 1939, seismic activity in the SMM zone was rather low (Figure 4). The above view can also help to understand why a compressional strain field is now recognized, by geodetic data, in the Struma and Krupnic fault zones, where geological evidence indicates an extensional regime. When instead, strong shocks occur in the Epirus thrust front, the consequent westward displacement of the western Aegean Tethyan belt temporarily (decades) induces an extensional regime in the SMM zone. This can explain why the most intense seismic crises in the SMM zone occurred after or during the most intense seismic periods in the Epirus (Figure 4).

The time pattern of the annual number of major earthquakes (M > 5.5) in the Aegean zone during the last two centuries (1800–2024, Figure 6) shows the highest values in the time interval 1950–1980, when that zone was undergoing the highest effects of the post-seismic relaxation triggered by the strong NAF seismic sequence since 1939. This time correspondence may support the hypothesis that tectonic activity in the Aegean zone is mainly influenced by the westward displacement of Anatolia.

The distribution of major earthquakes since 1600 (Figure 8) in the Arabian and eastern Anatolian zones may provide insights into how the activations of the main fault systems are connected to each other. In particular, this evidence indicates that seismic activity mainly occurred along the northern Arabian front and the Dead Sea fault system for a long time interval, during which no major earthquakes occurred at the EAF and NAF. This evidence may suggest that at present, after the last strong seismic crises in the NAF (1939–1999) and EAF (1874–1905 and 2023), the probability of major seismic crises along these fault systems may be significantly low for a long time interval. However, one must be aware that evaluating the statistical significance of seismicity patterns inferred from relatively short seismic histories (in zones characterized by long recurrence times) may be very difficult. Furthermore, we can hardly understand if the strain accumulated in the Anatolian wedge has been completely released by the last seismic crises at the NAF and EAF faults. Thus, it seems opportune to monitor the ongoing strain rate in the most dangerous seismic zones through geodetic and geophysical observations.