Source modelling and stress transfer scenarios of the October 30, 2020 Samos earthquake: seismotectonic implications

Chia sẻ: Tần Mộc Phong | Ngày: | Loại File: PDF | Số trang:22

Thêm vào BST

Báo xấu

17
lượt xem 0
download

Download Vui lòng tải xuống để xem tài liệu đầy đủ

On October 30, 2020, a strong earthquake (Mw6.6–7.0) occurred offshore, just north of Samos Island, causing life losses, injuries and damages, especially on the Turkish side. The broader area is characterized by a complex geodynamic setting with both rich seismic history and numerous active faults of different direction and kinematics. The first aim of this study is to define the seismic source of the mainshock, based on seismological and geodetic data (GPS measurements and originally processed GNSS records), as well as our field observations on Samos Island few days after the mainshock. The integration of this information leads to a N-dipping normal fault (Kaystrios fault) that controls the central-northern coast of Samos Island.

Chủ đề:

Bình luận(0) Đăng nhập để gửi bình luận!

Lưu

Nội dung Text: Source modelling and stress transfer scenarios of the October 30, 2020 Samos earthquake: seismotectonic implications

Turkish Journal of Earth Sciences Turkish J Earth Sci (2021) 30: 699-717 http://journals.tubitak.gov.tr/earth/ © TÜBİTAK Research Article doi:10.3906/yer-2107-25 Source modelling and stress transfer scenarios of the October 30, 2020 Samos earthquake: seismotectonic implications 1,2, 3 4 4 Sotiris SBORAS *, Ilias LAZOS , Stylianos BITHARIS , Christos PIKRIDAS , 1 4 3 3 Dimitris GALANAKIS , Aristeidis FOTIOU , Alexandros CHATZIPETROS , Spyros PAVLIDES 1 Hellenic Survey of Geology & Mineral Exploration (HSGME), Attica, Greece 2 Institute of Geodynamics, National Observatory of Athens, Lofos Nymfon, Athens, Greece 3 Aristotle University of Thessaloniki, School of Geology, University Campus, Thessaloniki, Greece 4 Aristotle University of Thessaloniki, School of Rural and Surveying Engineering, University Campus, Thessaloniki, Greece Received: 26.07.2021 Accepted/Published Online: 25.10.2021 Final Version: 30.10.2021 Abstract: On October 30, 2020, a strong earthquake (Mw6.6–7.0) occurred offshore, just north of Samos Island, causing life losses, injuries and damages, especially on the Turkish side. The broader area is characterized by a complex geodynamic setting with both rich seismic history and numerous active faults of different direction and kinematics. The first aim of this study is to define the seismic source of the mainshock, based on seismological and geodetic data (GPS measurements and originally processed GNSS records), as well as our field observations on Samos Island few days after the mainshock. The integration of this information leads to a N-dipping normal fault (Kaystrios fault) that controls the central-northern coast of Samos Island. We modelled the seismic source and calculated the theoretical dislocation (using the Okada formulae) on the surrounding GPS/GNSS stations, comparing it with the measured values. The results are very encouraging, especially on the station installed on Samos Island, giving confidence to our source model. We then used our seismic source to study the spatiotemporal evolution of the aftershock sequence by exploiting published seismological data (focal mechanisms and two seismic catalogues, one of which with relocated hypocentres) and our calculated Coulomb static stress changes caused by the mainshock. This comparison suggests that more faults than the Kaystrios fault were involved in the aftershock sequence. In order to investigate possible triggering and/or delay scenarios of the mainshock on nearby faults, the Coulomb stress changes are also studied showing various results according to each receiver fault. Key words: 2020 Samos seismic sequence, seismotectonics, fault modelling, dislocation, Coulomb stress changes, Aegean 1. Introduction This event took place in a geodynamically and A strong and destructive earthquake occurred on tectonically complex area, while the seismic source lies October 30, 2020 (11:51 UTC) in the eastern part of offshore, preventing any direct observation. Although the Aegean Sea region, between Samos Island (Greece) and various first published focal mechanisms suggest E-W- the western coastal area of Turkey, while the epicentre striking normal faulting, with only this information it is is located within Kuşadası Bay. According to several hard to decide whether the fault plane dips to the north or research centres, the earthquake was rather shallow south. The aftershock sequence evolution can reveal many (approximately 10 km) with a magnitude (Mw) ranging aspects of the tectonic setting in the epicentral area, while between 6.6 and 7.0 and was followed by a significant the recorded GPS dislocations on stations installed in the tsunami that hit Samos and other Eastern Aegean Sea broader area can contribute to the identification of the Greek Islands (Triantafyllou et al., 2021). Primary the source parameters. Not only can the modelling of the source mainshock, and secondary the aftershock sequence, help us better comprehend qualitatively the mechanism caused extensive damages and human losses especially which produced the mainshock, but also to develop stress in the broader İzmir (Smyrna) region where hundreds transfer scenarios in order to (i) examine whether other of deaths were recorded; on Samos Island, two students nearby structures participated in the aftershock evolution, were killed and damages were much more limited. The and (ii) study the stress changes on the numerous faults financial and social consequences of this seismic event that have been recognised in the broader area. Our source are still under evaluation. modelling is based on published seismological data, GPS * Correspondence: sboras@noa.gr 699 This work is licensed under a Creative Commons Attribution 4.0 International License.
SBORAS et al. / Turkish J Earth Sci measurements with specific process of permanent GNSS strain pattern estimations for the broader Aegean (e.g., data, and original field observations of ground deformation Floyd et al., 2010; England et al., 2016) suggest that in phenomena on Samos Island. our study area extension prevails in a roughly NNE-SSW direction, whereas contraction is almost negligible. 2. Geodynamic and tectonic setting The broader epicentral area is characterized by two 2.1. The broader Aegean region major fault systems based on geometry and kinematics, The Aegean–West Anatolia region constitutes one of the both compatible with the regional extensional field: most tectonically active regions globally, characterized by normal E-W-striking and shear NE-SW-striking faults. the westward tectonic escape of the Aegean microplate The normal E-W-striking fault system is responsible as the Anatolian pushes towards the west. Aegean’s for the great E-W-trending tectonically controlled valleys microplate motion occurs along two major lithospheric- in Western Turkey, such as the Büyük Menderes, Küçük scale structures: the Hellenic Arc and Subduction zone Menderes and Gediz. These valleys are considered to have (HAS), and the westward prolongation of the North been formed by either extensional graben-style high-angle Anatolian Fault (NAF) in the Aegean Sea. The HAS normal faults (e.g., Cohen et al., 1995; Hakyemez et al., represents the Nubian-Aegean plate convergence which 1999; Bozkurt and Sözbilir, 2004), or detachment faults causes the Nubian plate to subduct under the Aegean, (supradetachment basins) due to the exhumation of the a process started in the Late Cretaceous (e.g., Jolivet et Menderes Metamorphic Core Complex (MMCC; e.g., al., 2003; 2013; Van Hinsbergen et al., 2005; Brun and Dilek et al., 2009; Jolivet et al., 2013). This type of faulting Faccena, 2008; Jolivet and Brun, 2010) and continuing continues westwards in the Aegean Sea and is likely affecting Samos Island. until today with the southward retreat and roll-back of the Shear faulting with NE-SW orientation also dominates subducting slab which leaves room for crustal extension the broader Karaburun area also leaving its imprint on and volcanism in the back-arc area, with increased velocity the relief by forming rhomboidal basins in combination since Miocene (e.g., Jolivet et al., 2013; 2018; Schmid et al., with the E-W normal faults. Various theories exist for 2020; and references therein). On the other hand, other the formation of these basins (summarized by Bozkurt, authors (e.g., Le Pichon and Angelier 1979; Kokkalas 2003; Erkül et al., 2005). According to some authors (e.g., et al., 2006; Searle and Lamont 2021) suggest that the Okay et al., 1991; Ring et al., 1999; 2017; Uzel et al., 2013; Hellenic subduction zone was likely formed no earlier Westerweel et al., 2020), in the broader epicentral area, the than the early Miocene. During the above process and major NE-SW-striking fault zones (Priene-Sazlı, Sığacık, due to the westward motion of the Anatolian microplate, and Tuzla; Figure 1c) are part of a wide transfer zone the NAF appeared in the Late Miocene and extended (characterized as “wrench corridor” by Ring et al., 1999), even westwards in the North Aegean Sea in two strands: known as the İzmir-Balıkesir Transfer Zone (İBTZ), which the northern one, consisting of the North Aegean Trough is the result of differential extension between the Aegean (NAT) to the east and the North Aegean Basin (NAB) and Menderes Massif that started in the early Miocene to the west, and the southern one forming the Central (e.g., Ring et al., 2017). In similar orientation, but possibly Aegean Trough (CAT, aka Skyros Basin/Trough) (e.g., with less pure strike-slip kinematics, is the Fourni (aka Maley and Johnson, 1971; Barka, 1992). The tectonic Karlovasi-Fourni) Fault Zone, just west of Samos and activity of these structures is expressed by the numerous Fourni islands coastline. recorded (instrumentally or historically) strong seismic The area of Samos Island is also marked by faults of occurrences, causing destructive consequences. both kinds of orientation. Two E-W-striking fault zones 2.2. The western Anatolia have been recognized: the Kaystrios Fault Zone is an The geodynamic complexity of the study area is evident offshore active tectonic structure that runs along and by the regional focal mechanisms (e.g., Eyidoğan, 1988; controls the northern coastline of the island (Pavlides et Kiratzi and Louvari, 2003; RCMT catalogue, Pondrelli et al., 2009; Sboras, 2011; Chatzipetros et al., 2013; Caputo al., 2011), which reveal normal, strike-slip and transitional and Pavlides, 2013). On April 2, 1996 a moderate (Mw5.3) oblique-slip kinematics. The direction of extension (T-axis) earthquake occurred very near to the fault with quasi- has a dominating NNE-SSW azimuth with some events pure normal, E-W-striking focal mechanism (Kiratzi and showing a roughly N-S azimuth. GPS measurements Louvari, 2003; Figure 1b). The microseismic investigation (e.g., Nyst and Thatcher, 2004; Aktug et al., 2009; Çırmık of Tan et al. (2014) revealed a cluster offshore of the central et al., 2017), as well as palaeomagnetic surveys (e.g., van northern coast with a spatial distribution that depicts a Hinsbergen et al., 2010; Uzel et al., 2017), suggest that the N-dipping fault. This fault is also the candidate fault of the wider İzmir area is shifted approximately 25 mm/a in a 2020 mainshock. The Pythagorio Fault Zone is the second SSW direction, according to a stable Eurasia. GPS-based E-W-striking normal tectonic structure that crosses the 700
SBORAS et al. / Turkish J Earth Sci Figure 1. (a) Historic and instrumental seismicity by combining the catalogues of Papazachos and Papazachou (2003), for the period between 550 BC - 1964 AD, and IG-NOA for the period 1964 – Today. Yellow star is the epicentre of the October 30, 2020 Samos event (KOERI) and dark lines the major faults by GreDaSS. Yellow star and dark lines same as (a). (b) Focal mechanisms after the catalogues of Kiratzi and Louvari (2003) with black beachballs and RCMT with red beachballs. Beachball size is analogue to magnitude. (c) The 2020 sequence of the October 30, 2020 earthquake of Samos until November 30 (KOERI seismic catalogue). Moment tensor solutions are from KOERI moment tensor catalogue for the mainshock (yellow) and Altunel and Pinar (2021) (blue). (d) Relocated events of Samos 2020 sequence (data from Cetin et al., 2020). central part of the island from east to west (Mountrakis 2.3. Samos Island et al., 2003; Chatzipetros et al., 2013; Caputo and Pavlides, Samos is a mountainous island dominated by three 2013). This fault zone consists of two major segments, the mountains; the western Kerketeas Mt (1434 m), the Pythagorio Fault to the east and the Kerketeas Fault to the Ampelos Mt (1153 m) in the central part and the low west; nevertheless, both fault segments could be probably mountainous area of the eastern part. It consists of breached. alpidic high-pressure metamorphic rocks, ophiolites and 701
SBORAS et al. / Turkish J Earth Sci limestone, postalpidic Miocene-Pliocene sediments and extremely steep seafloor NW of the islands, which forms a Tertiary volcanic rocks. Neogene-Quaternary basins have submarine depression at more than 1000 m depth (Figure been formed on the metamorphic basement and filed 1c). It is believed to be a longer than 25 km long WNW– with lacustrine and fluviatile sediments (Theodoropoulos, ENE striking fault, dipping to the NNW (Chatzipetros et 1979a, b). The basins were initially formed by late alpidic al., 2013; Caputo and Pavlides, 2013). low-angle detachments when the exhumation of the The 1992 seismic events, north of Samos, are associated Kerketeas tectonic window appeared (Ring et al., 1999; with a right lateral, strike-slip, which is possibly related Kumerics et al., 2005). Boronkay and Doutsos (1994) and to the corresponding NE-SW strike-slip fault, identified Kokkalas et al. (2006) had a different opinion suggesting in the field in the NW edge of Samos (Karlovasi-Fourni that the mountainous area of Samos was uplifted from fault). The dominant E-W trending faults correspond to steep right-lateral shear zones with thrust component the western termination of a large rupture zone which due to transpressional regime which was active during also has an E-W direction and extends further inland for the Miocene and ended 7 Ma ago; this stage was followed several tens of kilometres. by transtension, affecting only the northern parts of the basins, which gradually turned into the broader back- 3. Prior seismicity arc extension. Neotectonic normal faults control the Samos broader region has a long seismic history, including geomorphology of Samos Island, as they deform a complex several tens of strong earthquakes (M ≥ 6.0) and three Quaternary horst block in the eastern Aegean Sea. Some major ones (M ≥ 7.0) (Figures 1a and 1b). Ten events have important strike-slip faults trending NE-SW to ENE-WSE damaged the island during 19th and 20th centuries with with right-lateral sense of movement, while the NW-SE six of them occurring as a seismic cluster between 1865 to NNW-SSE are left-lateral. (Mountrakis et al., 2003; and 1904. The oldest event (496 BC, M 6.0) of the wider Chatzipetros et al., 2013; Caputo and Pavlides, 2013; Ring area occurred in the offshore area between Chios Island et al., 2017). (Greece) and Çeşme (Turkey). According to Papazachos The most significant on land active faults are: The and Papazachou (2003) the strongest event occurred in Pythagorion normal fault segment with its westward 1653 in the wider Aydin region (east of Izmir) (Figure prolongation, i.e. the Marathokambos segment, moderately 1a); Ambraseys and Jackson (1998), however, place this dipping toward SSW, running the central part of the island earthquake far more to the east. This earthquake killed form the eastern to the western coast, affecting Pliocene approximately 2500 people (Papazachos and Papazachou, sediments, while the recent reactivations of the fault affect 2003; Ambraseys, 2009). A recent seismic event also worth the tectonic breccia of Late Pleistocene age. The Karlovasi mentioning is the Söke earthquake (1955, M 6.9), causing oblique-slip fault, part of the larger Karlovasi-Fourni zone the death of 25 people. that continues offshore, trend NE-SW, dip 75°/NW, right The October 2005 seismic sequence (three mainshocks: lateral strike-slip with normal component, has a detectable October 17, Mw5.4 and Mw5.8; October 20, Mw5.9) that 12 km length along the NW coast of the island and occurred in the Sığacık bay revealed the existence of presents some very impressive, polished slickensides on the NE-SW-striking sinistral strike-slip Sığacık fault the metamorphic rocks. It is characterized by steep scarp (Benetatos et al., 2006; Aktar et al., 2007). slopes and Pleistocene fault scree, which have been affected by younger (recent) reactivations of the fault. Its recent 4. The 2020 seismic sequence activity is also proved by recent seismological data, that 4.1. Seismological information is the distribution of low magnitude earthquakes offshore, The October 30, 2020 mainshock (Mw6.9–7.0) occurred ca. as well as by the seismic activity in 2005 near the Turkish 10 km north of Samos Island at a depth of 13.1 km (KOERI coast in a parallel direction to the fault (Mountrakis et al., catalogue) or 8.2 km according to the relocation by the 2003). Seismological Laboratory of the National and Kapodistrian The Kokkari-Vathy fault (trend WNW-ESE, dip 50°/ University of Athens (SL-NKUA) (Cetin et al., 2020). All NNE, normal) in the northeastern coastal part of the published focal mechanisms suggest E-W-striking pure island, caused an impressively tilting (almost vertical) of normal faulting. Based on the KOERI seismic catalogue, the Pliocene marly limestone. Such geological evidence the aftershock sequence is characterized by 2 events of ML suggest that the fault WNW-ESE trending and the E-W ≥ 5.0 (ML5.2 few hours after the mainshock and ML5.0 on westward prolongation is an active structure parallel to the next day) and 54 events of 4.0 ≤ ML < 5.0 in the next the offshore seismogenic one of the latest earthquake. one week. The focal mechanisms according to Altunel and The offshore North Samos fault, called Kaystrios fault, Pinar (2021) for the strongest events of the sequence also is a rather longer structure, possibly extending north of reveal normal faulting with strike ranging from E-W to Ikaria Island as inferred from the linear coastline and the ENE-WSW (Figure 1c). 702
SBORAS et al. / Turkish J Earth Sci The spatiotemporal evolution of the sequence (Figure the same period of November 3–10, the gap between 2; KOERI seismic catalogue) shows a great concentration clusters A and B is indistinctive. A diachronic gradual of hypocentres (cluster A) in the epicentral area of the extension of cluster A toward the NE is also observed. The mainshock. At the same time, a much smaller one is maximum depth of the sequence reached ca. 36 km. observed (cluster B) west of cluster A with a quasi- Similar situation is depicted, more clearly, by the distinctive gap in-between. This cluster (B) persists on spatiotemporal analysis of the 989 relocated events (Figure showing during all time periods, while the gap from 3; Cetin et al., 2020), which have an average relative cluster A started to fill in since the next day (October 31). horizontal and depth error less than 1 km. Indeed, during Since October 31, a cluster (cluster C) hesitantly started to the first day, only cluster “1” (corresponding to B in Figure develop immediately SE of the epicentral cluster A which 2) is formed along with the epicentral cluster (Figure 3a). A afterwards was rather enriched and expanded. During the gap between cluster “1” and the epicentral cluster (similar period November 3–10, a poor, less concentrated cluster to the one in Figure 2) is notable and has been marked by (cluster D) appears even further to the west, north of Cetin et al. (2020). During the next days and until one week Ikaria Island, and just a few hypocentres also occurred after the mainshock, the gap between cluster “1” and the immediately south of the Chios-Çeşme fault. These two epicentral cluster starts to fill in with new events (Figure occasions also continued in the next period (November 3b). At the same time, cluster “2” (corresponding to C in 11–20), with cluster D significantly diminishing. During Figure 2) started to form (Figure 3b). The much western Figure 2. Spatiotemporal evolution of the 2020 Samos aftershock sequence (KOERI seismic catalogue). Starting from the top left, the hypocentral distribution of the sequence is shown during October 30, 31, November 1, 2, 3–10 and 11–20. Since day 1, a second cluster, besides the one of the mainshock, is formed just to the west. During the period November 3–10, a third poor and scattered cluster occurs even further to the west, north of Ikaria Island. A gradual extension of the hypocentres toward the NE is also observed. 703
SBORAS et al. / Turkish J Earth Sci Figure 3. Spatial distribution of the relocated Samos 2020 aftershock sequence, including 989 events with best locations: a) 1st day, b) 1st week, c) 1st month (modified from Cetin et al., 2020). 704
SBORAS et al. / Turkish J Earth Sci cluster “4” (corresponding to D in Figure 2) started to form were observed by other researchers as well (Mavroulis et after the first week, as well as some scattered events to the al., 2021). We also recorded earthquake-triggered slope east (group “5”) just by the Turkish coast. The maximum failures (rockfalls and landslides) blocking the local road depth of the relocated hypocentres is 15.8 km, with the network, most located on the footwall of Kaystrios fault. majority (56.5%) occurring at depths between 6 and 10 km The slope failure near Avlakia (Figures 4a and 4e) occurred with a peak (16.3%) between 8 and 9 km. on a steep slope of a NW-SE orientated low ridge consisting Other aftershock epicentral relocations followed of steeply NW-dipping Miocene marls. Although we had afterwards (Papadimitriou et al., 2020; Foumelis et al., no access to the southwestern side of the ridge, there is a 2021; Karakostas et al., 2021), all ending up, more or less, strong suspicion that the valley that follows is tectonically in the same clusters recognition. controlled. This fault can be either related to the Kokkari- The moment tensor solutions of the sequence (Figure Avlakia fault shown in Figure 4a, or it can be considered a 1c; Altunel and Pinar, 2021, after Cetin et al., 2020) reveal strand of the offshore Kaystrios fault. the domination of E-W-striking pure to quasi-pure normal According to other research teams, a liquefaction faulting in the epicentral cluster, with some NE-SW- / phenomenon was observed on the shore, near Vathy NW-SE-striking, strike-slip exceptions implying a partial (Figure 4a; Vadaloukas et al., 2020) and several tsunami reactivation of the adjacent NW-SE-striking Karlovasi- occurrences at various places along the island’s coastline Fourni fault. In cluster B (Figure 2), or else cluster “1” in (Figure 4a; Triantafyllou et al., 2021). Uplift markers were Figure 3a, the moment tensors show mixed fault kinematics found all along the coast of Samos Island with maximum of quasi-normal (sometimes with significant right-lateral values observed on the northwestern coast and especially component) and right-lateral strike-slip (sometimes with at the Aghios Isidoros cove (30 cm uplift) (Figure 4a; minor reverse component). The four focal mechanisms in Evelpidou et al., 2021; Mavroulis et al., 2021). cluster C (Figure 2), or else cluster “2” in Figure 3b, are Regarding the western Turkey coast, tsunami of right-lateral strike-slip kinematics, mostly with a slight occurrences were observed in the wider Çeşme region normal component. The best fault candidate of the two (Yalciner et al., 2020; Aksoy, 2021), mainly documented nodal planes is the NE-SW-striking one, perhaps a strand by extended inundation phenomena, with a maximum of the Pythagorio Fault (Figure 1c). estimated distance of 320 m (Yalciner et al., 2020). 4.2. Geological information Moreover, in the same area, seismic soil liquefaction and The 2020 Samos seismic sequence was accompanied with induced ground failures are also recorded (Cetin et al., several ground deformation phenomena on Samos Island. 2020). After arriving on the island few days after the mainshock, 4.3. Geodetic information ground ruptures were observed in Aghios Nikolaos area 4.3.1. GPS/GNSS data analysis (northern part of the island), where the E-W-trending, The satellite geodetic methods, relying on GPS/GNSS almost linear coastline ends (Figure 4). The ruptures, all data, are significant for the analysis and estimations of of dilatational character, show horizontal and vertical co-seismic displacements, caused by strong earthquakes. offset of several centimetres. We tracked them in the fields, Our analysis includes 38 GPS/GNSS sites; 10 of them are on the roads and on slopes. They demonstrate a gradual located close to the seismic event. Data analysis was based change in their strike, starting from N35°E to the NNW on 30-s daily GPS/GNSS observations, characterized by 10° and reaching N70°E ca. 500 m to the SSW. This direction elevation cut-off angle for a short time-period, before and is similar to the Fourni fault (Figures 1 and 4a) which runs after the earthquake occurrence, while they were retrieved along the coastline of the northwestern part of the island. from different permanent networks, such as HermesNet These ruptures could be either directly related to the of Auth (Fotiou et al., 2010), HxGn-SmartNet Greece, Fourni fault, or they can be considered as minor effects of a Uranus, EPN/Euref and Turkish National Permanent RTK transition zone between the Kaystrios offshore fault and the Network. In order to estimate the co-seismic displacements Fourni fault. A long fissure was also observed in an active 6-days data before (24–29 October 2020) and 4-days data landslide near Kontakaiika village. The village is built on after (31 October–03 November 2020) the seismic event conglomerates saturated with water, causing a preexisting, were processed. The process was executed, implementing but active landslide with ground deformation phenomena the scientific GAMIT/GLOBK package (Herring et al., such as fissures. Most of these fissures were further 2015) in a three-step approach. expanded due to the ground shake of the 2020 sequence. In Initially, a network positioning approach is followed, Aghios Nikolaos area, local residents described a tsunami in which the baselines between all observed stations are occurrence, as well as whirlpools near the coast which also simultaneously estimated; this method is also applied rotated whitish fine-grained material probably discharged for estimating satellite orbits and Earth orientation from the sea bottom. Ground ruptures in the same area parameters (EOP) from GNSS reference stations. In our 705
SBORAS et al. / Turkish J Earth Sci Figure 4. (a) Topographic map of Samos Island showing the on land active and potentially active faults after Mountrakis et al. (2003; 2006) and Chatzipetros et al. (2013), reassessed in this study, and the coseismic effects of the 2020 Samos earthquake: the observed ground ruptures and the landslide fissure are from this study, the tsunami locations are from Triantafyllou et al. (2021), and the liquefaction phenomenon near Vathy is after Vadaloukas et al. (2020). (b) Phtotomosaic after the Hellenic Cadastre showing the ground ruptures observed in this area. The location is located on the map (a). (c, d) Ground ruptures. In (c) a cement block was ruptured as well in a greenhouse. (d) Photo taken by an UAV, showing the earthquake -induced slope failure. processing schema, we utilize precise orbits from the part of the Island and nearest to earthquake epicentre; the International GNSS Service (IGS) and absolute calibration estimated value is 38.58 cm and it is considered as a rather values from IGS tables, in order to model the receiver high value. The next step includes the calculation of the and satellite antenna phase centre variation. Then, the average position per topocentric component (N, E, Up) for individual loosely constrained estimates are imposed and the period (before and after the earthquake), as well as the the reference frame definition is implemented, using inner differences between these datasets. The extracted results of constraints to coordinates and their velocities in 9 IGS the closest to the earthquake epicentre stations are shown stations. The daily position estimates were applied in the in Table 1. current realization of the ITRF-NNR frame (ITRF2014). The horizontal co-seismic displacements are shown Finally, we estimate the coseismic displacements, based on in Figure 5. It worth mentioning that the daily positions the analysis of the GPS/GNSS daily position time-series. results show high accuracy, with lower uncertainties values For the purposes of this study, the spatial displacement in the horizontal and vertical components (±0.5 and ±2 vectors were estimated for the referred ECEF system, in mm, respectively), highlighting the advanced processing which SAMOS permanent station is located at the NW scenario of the geodetic data. 706
SBORAS et al. / Turkish J Earth Sci Table 1. The GPS stations to which the measured and modelled values of the 2020 Samos earthquake correspond. GPS displacement values Modelled displacement values Station Longitude Latitude Easting Northing Upwarding Easting Northing Upwarding (cm) (cm) (cm) (cm) (cm) (cm) SAMO 26.705 37.793 –5.90 –36.90 8.40 –4.98 –36.63 13.41 IZMI 27.082 38.395 1.60 3.40 0.20 0.57 3.08 –0.14 CESM 26.373 38.304 –1.20 5.20 0.20 –0.38 1.35 0.14 CHIO 26.127 38.368 –0.40 2.00 0.50 –0.19 0.62 0.14 DIDI 27.269 37.372 –0.10 –0.80 0.40 0.82 –2.74 0.13 IKAR 26.224 37.628 –1.20 –3.30 0.50 –0.65 –0.95 0.46 LERO 26.855 37.136 0.00 –2.00 –0.80 –0.13 –2.43 –0.15 XIOS 26.136 38.367 –0.50 2.10 0.10 –0.19 0.64 0.14 KALY 26.962 36.962 0.50 –1.20 –0.40 0.02 –1.57 –0.13 AYD1 27.838 37.841 0.00 0.30 0.10 –0.09 0.05 0.37 Figure 5. Measured and modelled horizontal coseismic displacements for the nearest GNSS stations to EQ epicentre (see also Table 1). The modelled 2020 Samos mainshock source (see also Table 2) is represented by the yellowish rectangular (vertical projection of the fault plane on the map and red thick line its upper part). 707
SBORAS et al. / Turkish J Earth Sci Several InSAR images have been also published mostly along the main epicentral cluster. Four different (Papadimitriou et al., 2020; Foumelis et al., 2021; Ganas fault models, based on four different focal mechanisms, et al., 2021; Mavroulis et al., 2021) showing clear uplift are proposed by Foumelis et al. (2021). All four models of the north-northwestern coast of Samos Island and an diverge rather significantly from the quasi-E-W direction abrupt change from uplift to subsidence near the central- proposed in our and the other two models. The authors northern coast, implying the possible emergence of the modelled the surficial displacement by the four fault fault rupture, or a strand of it, on land. In other words, it models and compared it with their InSAR images. Various is strongly supported that the rupture reached the surface rupture (fault) models have been also proposed by Akkar and that was not very far from the coastline. et al. (2021), based on various published moment tensor solutions and fault models, taking into consideration 5. Fault modelling other seismotectonic attributes. A seismologically induced 5.1. The 2020 Samos seismic source fault model has been proposed by Vallianatos and Pavlou The modelling of the mainshock-fault (seismic source) (2021) and the USGS (https://earthquake.usgs.gov/ is mostly based on the available seismological data. Its earthquakes/eventpage/us7000c7y0/executive, visited on location is set by combining its morphological imprint July 18, 2021). In fact, the latter proposed a south dipping along the northern coastline of Samos Island, the finite model. The closest fault model to ours is the one hypocentre (both horizontal location and depth), and the proposed by Chousianitis and Konca (2021). The authors aftershock spatial distribution of both catalogues (KOERI; used joint slip inversion to reach to their model, which Cetin et al., 2020). Although primary ground ruptures differs slightly in geometry and dimensions. It is obvious were not observed on land, based on the earthquake that numerous fault models can be proposed depending magnitude and the hypocentral depth, the rupture should on the methodology applied. Our model, combining have reached the seafloor, thus 0 is set as minimum fault seismological, geodetic and geological information, aims depth. Dimensions (Length and Width) are calculated by at giving another perspective from most of the other ones the scaling relationships of Wells and Coppersmith (1994) above, which show significant differences in most of their of Magnitude versus Length and Magnitude versus Width. parameters, The big resemblance of our model with the Geometry (strike and dip), as well as kinematics (rake), are one of Chousianitis and Konca (2021), each using different obtained from the KOERI focal mechanism. approach of modelling, gives more confidence to ours. We tested our seismic source by comparing the 5.2. Modelling of surrounding faults modelled dislocation components, after applying Okada’s In order to better calculate the stress changes on each (1992) dislocation solution formulae, with the measured surrounding fault, or other similar ones, in the area of GPS dislocations. According to Table 1 and Figure 5, the interest, we also proceeded with fault modelling. Twelve results show almost identical horizontal resultant direction (12) significant faults are recognized in the area according in almost all cases with some magnitude discrepancies. to the available literature (Figure 1, Table 2). Some of Nevertheless, at the station on Samos Island (SAMO) both these faults have been associated with past earthquakes, direction and magnitude of the horizontal resultant are very most of them with rather high confidence. For the similar, adding confidence to our model. The recordings of modelling we used all available information from existing the distant stations contain many imponderables due to active fault databases and map-series (e.g., Mountrakis the heterogeneous medium between the source and each et al., 2006; Emre and Özalp, 2011; Emre et al., 2011; station (the Okada model applies on a uniform elastic half- Duman et al., 2011; Caputo and Pavlides, 2013; Emre et space). al., 2018), published morphotectonic-neotectonic and Other proposed fault models place the fault plane more seismotectonic investigations, and focal mechanisms from to the west. The model of Ganas et al. (2020) is based only databases (e.g., RCMT, Pondrelli et al., 2011) and articles on geodetic data inversion, suggesting a minimum fault (e.g., McKenzie, 1972; Kiratzi and Louvari, 2003). In Table depth at 0.9 km (blind fault) with fault plane dimensions 2, all parametric information and associated references are of 40 ± 3 km length and 15 ± 3 km width. A strike of 276° shown for the fault models that are used in the Coulomb and a dip of 37° are the rest geometric attributes. The stress calculations. It must be mentioned that maximum model of Karakostas et al. (2021) is based on seismological expected magnitude is estimated by the rupture surface data and specifically on slip inversion. Both models area (SA) versus magnitude scaling relationships of Wells extend in the gap formed between clusters A and B of and Coppersmith (1994). Following this approach for the KOERI’s spatial distribution (Figure 2), or between main earthquake-related faults, besides the Priene-Sazlı Fault, epicentral cluster and cluster “1” of Cetin’s et al. (2020) all the others are considered to have higher potential than spatial distribution (Figure 3), whereas our model extends the past recorded events. 708
SBORAS et al. / Turkish J Earth Sci Table 2. Seismic sources/fault models and their parametric information in the broader epicentral area of the 2020 Samos earthquake. Reference keys: 1: GreDaSS; Sboras (2011); 2: Emre et al. (2018); 3: Chatzipetros et al. (2013); 4: McKenzie (1972); 5: Kiratzi and Louvari (2003); 6: Benetatos et al. (2006); 7: Altinok et al. (2005); 8: Mountrakis et al. (2003; 2006); 9: Pavlides et al. (2009); 10: Karakaisis et al. (2010); 11: Altunel (1998); 12: Yolsal-Çevikbilen et al. (2014); 13: Mozafari et al. (2019); 14: Topal (2019); 15: Gürer et al. (2001); 16: Yönlü et al. (2010); 17: Sümer et al. (2013); 18: Ocakoğlu et al. (2005); 19: Genç et al. (2001); 20: Sözbilir et al. (2011); 21: Uzel et al. (2013); 22: Uzel et al. (2012); 23: Aksu et al. (1987); 24: Hancock and Barka (1987); 25: Stewart and Hancock (1988; 1991); 26: Tan et al. (2014). Top Max Length Width Strike Rake Fault name depth Dip (°) expected Associated EQ References (km) (km) (°) (°) (km) magnitude 30/10/2020, Mw6.9-7.0 Kaystrios 39.0 19.5 0.0 272 55 –93 6.9 See main text 2/4/1996, Mw5.3 (?) Karlovasi-Fourni 30.0 10.0 0.0 225 50 –140 6.5 1, 8, 9 Pythagorio 32.0 17.0 0.0 96 45 –90 6.7 1, 8, 9, 10, 26 Ikaria 23.0 13.5 0.0 234 55 –130 6.5 1, 3, 9 Kuşadası (Yavansu) 23.0 13.5 0.0 76 60 –110 6.5 1, 2, 24, 25 İncirliova 32.0 17.0 0.0 100 60 –85 6.7 1, 2, 24, 25 1, 2, 4, 10, 11, 13, Priene-Sazlı 33.0 17.5 0.0 55 51 –133 6.8 16/7/1955, Mw6.8 14, 15, 16, 17 Chios-Çeşme 39.0 19.5 0.0 92 48 –92 6.9 3/4/1881, Mw6.5 1, 7, 9, 18 Karaburun 35.0 11.0 0.0 3 80 180 6.6 1, 2, 18, 21, 23 Sığacık 38.0 11.0 0.0 228 79 –171 6.7 17/10/2005, Mw5.8 1, 6, 12, 21 Tuzla 33.0 10.5 0.0 53 77 166 6.6 6/11/1992, Mw6.0 1, 2, 10, 18, 19, 21 Karareis 19.0 11.5 0.0 116 60 –90 6.3 6/4/1969, Mw5.9 1, 4, 21 İzmir 22.0 13.0 0.0 257 65 –110 6.4 1, 2, 20, 21, 22 6. Static stress changes October 30 – November 30, and of the relocated catalogue The Coulomb failure criterion (Δσf = Δτ + μ΄ ⋅ Δσn, where of Cetin et al. (2020), for the time period October 30 – Δτs is the shear stress change on the failure plane, μ΄ is the December 1. In Figure 6, we calculated the Coulomb stress friction coefficient and Δσn is the normal stress change) changes at a depth of 9 km for both cases, which is near to can be used to calculate the static stress changes on the relocated hypocentral depth of the mainshock (Cetin surrounding faults after the rupture of a seismic source et al., 2020). For receiver faults, we used the attributes and evaluate trigger or delay effects. The Coulomb stress (geometry and kinematics) of the seismic source. changes are highly sensitive to the parameterization of the Although the KOERI seismic catalogue is richer, the source-fault and dependent on three specific parameters spatial distribution is rather poor in accuracy (as already (strike, dip and rake) of the receiver-fault. This means expected), forming an artificial high concentration that the more well-defined the seismic fault is, the more lineament at the depth of 5 km (Figure 6, top). However, confident the Coulomb stress change results are. the bulk of the hypocentres is gathered around the fault In this paper, we study the effects of (i) the 2020 Samos plane within the stress-drop (blue) area created by the stress mainshock effects in the aftershock sequence (in-sequence relief after the rupture. The two clusters (B and C according effects) using faults of similar geometry and kinematics as to Figure 2, or “1” and “2” in Figure 3, respectively) are receiver faults, and (ii) the 2020 Samos mainshock effects situated in different stress regimes: Cluster B/“1” is located on the surrounding faults (postsequence effects). in a stress-rise area. However, as it will be discussed below, 6.1. In-sequence effects this cluster can be also ascribed to the adjacent Karlovasi- To better understand the spatiotemporal aftershock Fourni Fault and be associated with the ground ruptures evolution and the clusters formation of the 2020 Samos that were observed near Agios Nikolaos area (Figures sequence in terms of stress changes, we used both the 4a–4d). The other cluster (C/“2”) is probably produced by data of the KOERI seismic catalogue, for the time period a minor strike-slip fault, as mentioned before. Thus, with 709
SBORAS et al. / Turkish J Earth Sci Figure 6. Coulomb static stress changes for the 2020 Samos aftershock sequence investigation (in-sequence effects) using the seismic catalogue of KOERI (top) and the relocated catalogue of Cetin et al. (2020) (bottom). The mainshock rupture model (seismic source) is the red rectangular (vertical projection of the fault plane on the map) and the green line is its upper part. Read main text for further explanations. 710
SBORAS et al. / Turkish J Earth Sci such a different receiver fault, the specific Coulomb stress slip, E-W-striking faulting, while the aftershock focal change calculation does not apply for this case. mechanisms not only revealed normal faulting, but The relocated hypocentral distribution in the stress strike-slip faulting as well. A distinctive cluster west of the change field (Figure 6, bottom) gives a much better picture epicentral area (B/“1” in Figures 2 and 3a, respectively) of the aftershock sequence development. The hypocentres consists of focal mechanisms with both kinematics, around the mainshock fault plane are denser concentrated whereas an adjoining cluster to the SE (C/“2” in Figures and they are all located in the stress-drop zone where the 2 and 3b, respectively) consists mostly of strike-slip rupture of the fault plane took place. The interpretation mechanisms. of the two aforementioned clusters in the previous case In order to define the source of the mainshock we remains the same. However, the hypocentres that fill in the combined all available seismological, geological and gap between cluster B/“1” and the epicentral cluster (A) geodetic data, ending up to an emerging, N-dipping, occur in the highest stress-rise and can be ascribed to a normal dip-slip fault, just offshore and along the northern possible western extension of the main fault rupture. coast of Samos Island, compared with other fault models 6.2. Postsequence effects (Akkar et al., 2021; Foumelis et al., 2021; Ganas et al., The state of stress after the 2020 rupture depends on the 2020; Karakostas et al., 2021, Vallianatos and Pavlou, 2021; attributes (geometry and kinematics) of the receiver fault. USGS), which are either located further westwards, and/ For this reason, the Coulomb stress change was calculated or diverging significantly from the quasi-E-W orientation for each recognized fault of the total twelve in the broader parallel to the northern coastline, or considered as a blind epicentral area (Table 2; Figure 7). For a down in-depth fault (Ganas et al., 2020). We modelled the dislocations visualisation of the receiver faults position, we modelled on the nearby GPS/GNSS stations by applying the Okada the faults as mentioned previously. The calculation depth (1992) dislocation solution formulae and compared them of the horizontal sections varies, based on the maximum with the measured ones (Figure 5). The horizontal vectors depth that each receiver fault reaches. We set an appropriate coincide very satisfactorily on the very near SAMO station depth in order to cross the deepest part of each receiver (Samos Island) in both direction and magnitude. On fault, where a possible rupture nucleation is expected to most of the other stations, the direction fits better than occur. the magnitude besides İzmir where magnitude is closer, The results (Figure 7) show that three faults (İncirliova, but direction deflects more. These discrepancies at the Priene-Sazlı and Karareis faults) are not or negligibly distant stations can be justified by the fact that modelled affected by any stress change. The Karlovasi-Fourni fault, dislocations are calculated in a homogeneous elastic half- which is immediately west of the Kaystrios source fault, space (a condition never existing in nature) and that the lies entirely in the stress-rise (red) area. In this area, the aftershock sequence, which contributes to the cumulative aftershock cluster “B”/“1”, with the mixed nearly normal deformation of the area, does not participate in the and strike-slip moment tensor solutions, also lies (Figures calculations. 1–3). Other faults that are located in stress-rise (red) Two cases of Coulomb static stress changes were areas are the Kuşadası (Yavansu) and the Ikaria faults considered in our study: the effects of the mainshock (the latter in a rather fade red area), while the Chios- rupture (i) during the aftershock sequence (in-sequence Çeşme fault undergoes stress increase only towards the effects) on similar faults (Figure 6), and (ii) on twelve (12) deepest eastern part of the fault plane, approximately surrounding faults (postsequence effects) which are also where some aftershocks occurred during the 2 weeks in modelled for better visualisation (Figure 7). In case (i), early-mid November (Figure 2). For the rest of the faults stress-rise areas can explain the occurrence of cluster B/“1” (Pythagorio, Karaburun, Sığacık, Tuzla and İzmir) stress as a western extension of the ruptured fault (Figure 6), also drop is observed. supported by the normal, E-W-striking focal mechanisms 7. Summary and conclusion (Figure 1). On the other hand, the same cluster can be The October 30, 2020 (Mw6.9–7.0) Samos earthquake, explained by the partial reactivation of the adjacent besides the extended damages, life losses and injuries, NE-SW-striking Karlovasi-Fourni fault (matching the produced several secondary coseismic ground effects, corresponding strike-slip focal mechanisms in Figure 1), such as ground ruptures, rockfalls and limited liquefaction which demonstrates stress increase in case (ii) (Figure 7), phenomena. GPS/GNSS data (Figure 5) demonstrated and which probably produced the ground ruptures near coseismic divergence between Samos Island and the whole Agios Nikolaos (Figures 4a–4d). Thus, cluster B/“1” must region across Chios Island, Karaburun peninsula and be the result of the partial reactivation of two, very near to İzmir (Smyrna). The aftershock spatiotemporal evolution each other, faults: the western part of the Kaystrios fault, (Figures 2 and 3) revealed a complex pattern with few which did not follow the main rupture of the October 30 clusters, some of which occurred delayed. The mainshock mainshock, and the northeast most part of the Karlovasi- focal mechanisms identified almost pure normal dip- Fourni fault. 711
SBORAS et al. / Turkish J Earth Sci Figure 7. Coulomb static stress changes in horizontal and vertical projections after the 2020 Samos mainshock for the twelve surrounding faults (Table 2). The mainshock rupture model (seismic source) is the red rectangular (vertical projection of the fault plane on the map) and the green line is its upper part. The modelled receiver fault is each time represented by a black rectangular and the others in dashed grey. Blue dashed horizontal line in profiles is the depth of the corresponding horizontal section. Read main text for further explanations. 712
SBORAS et al. / Turkish J Earth Sci Cluster C/“2” (Figures 2 and 3b, respectively) is a matching the results of Chousianitis and Konca (2021). more problematic case. The focal mechanisms (Figure 1) The other faults show either stress-drop or negligible indicate the occurrence of a (preferably) NE-SW-striking stress change, matching again the results of Chousianitis strike-slip fault which could be an eastern strand of the and Konca (2021), with the only exception the Karaburun Pythagorio fault. At the same time, the cluster is entirely fault which in the results of the latter authors, a low and located in a stress-drop area in the in-sequence effects partial stress rise is demonstrated. It must be noted that case. Thus, it cannot be considered as part of a wider zone triggering or delaying effects on nearby faults due to static around Kaystrios fault, but an individual fault. Even if we stress changes is a relative information which depends on consider it as a fault of similar geometry and kinematics the state of stress in which every fault is (maturity stage) with either the Kuşadası (Yavansu) or the Priene-Sazlı fault, and cannot be used to predict the time of the next rupture. the location of the cluster remains in the stress-drop area. In other words, stress rise or drop can shorten or lengthen A similar scenario was performed by Karakostas et al. the next rupture in the undetermined future assuming that (2021), who calculated the static stress changes pattern the crustal deformation rate remains stable and no other for comparing it with the aftershock spatial distribution. earthquake will occur. Since their calculations are based on a different fault model and aftershock spatial distribution, there is no common Acknowledgments ground for comparing their results with ours. On the other The authors would like to thank all the GPS/GNSS hand, great similarities can be observed with the results data providers of cGPS Permanent Networks, HxGn- of Chousianitis and Konca (2021), who also show that the SmartNet Greece, Uranus and HermesNet. Authors aftershock sequence is exclusively constrained in stress CP and SB acknowledge support of this work by the rise areas. However, using the optimum oriented normal project “HELPOS – Hellenic System for Lithosphere receiver-faults in their calculations, instead of ours which Monitoring” (MIS 5002697) which is implemented under involves receiver faults identical to the slipped one (source the Action “Reinforcement of the Research and Innovation fault), Chousianitis and Konca (2021) show that the cluster Infrastructure”, funded by the Operational Programme C/“2” (Figures 2 and 3b, respectively) falls entirely in a “Competitiveness, Entrepreneurship and Innovation” stress-rise area, in contrast with our results which show (NSRF 2014-2020) and cofinanced by Greece and the that the respective cluster lies in a stress-drop area and, European Union (European Regional Development Fund). thus, implying a triggered reactivation of a yet unknown HSMGE’s geologist Kostas Kontodimos is thanked for his smaller fault. contribution in the fieldwork. The guest editor R. Caputo Concerning the postsequence effects and triggering and two anonymous reviewers are thanked for improving scenarios on the other surrounding faults, notably the the manuscript. The application of the Okada model and Karlovasi-Fourni and the Kuşadası (Yavansu) faults, and the Coulomb failure criterion were performed with the less the Ikaria and Chios-Çeşme faults (the latter shows Coulomb v3.4 software (Toda et al., 2005; Lin and Stein, diverse stress change along its strike), show stress increase, 2004). References Akkar S, Çağlar NM, Kale Ö, Yazgan U, Sandıkkaya MA (2021). Aktug B, Nocquet JM, Cingöz A, Parsons B, Erkan Y et al. (2009). Impact of rupture-plane uncertainty on earthquake hazard: Deformation of western Turkey from a combination of observations from the 30 october 2020 Samos earthquake. permanent and campaign GPS data: Limits to block-like Bulletin of Earthquake Engineering 19 (7): 2739-2761. doi: behavior. Journal of Geophysical Research 114 (B10). doi: 10.1007/s10518-021-01099-9 10.1029/2008JB006000 Aksoy (2021). Post-event field observations in the İzmir–Sığacık Altinok Y, Alpar B, Özer N, Gazioglu C (2005). 1881 and 1949 village for the tsunami of the 30 October 2020 Samos (Greece) earthquakes at the Chios-Cesme Strait (Aegean Sea) and M w 6.9 earthquake. Acta Geophysica 1. their relation to tsunamis. Natural Hazards and Earth System Sciences 5 (5): 717-725. doi: 10.5194/nhess-5-717-2005 Aksu A, Piper D, Konuk T (1987). Late Quaternary tectonic and sedimentary history of outer Izmir and Candarli bays, western Altunel (1998). Evidence for damaging historical earthquakes at Turkey. Marine Geology 76: 89-104. doi: 10.1016/0025- Priene, Western Turkey. Turkish Journal of Earth Sciences 7: 3227(87)90019-3 25. Aktar M, Karabulut H, Özalaybey S, Childs D (2007). A conjugate Altunel E, Pinar A (2021). Tectonic implications of the Mw 6.8, 30 strike-slip fault system within the extensional tectonics of October 2020 Kuşadası Gulf earthquake in the frame of active Western Turkey. Geophysical Journal International 171 (3): faults of Western Turkey. Turkish Journal of Earth Sciences 30 1363-1375. doi: 10.1111/j.1365-246X.2007.03598.x (4): 436-448. doi: 10.3906/yer-2011-6 713
SBORAS et al. / Turkish J Earth Sci Ambraseys N (2009). Earthquakes in the Mediterranean and Middle Dilek Y, Altunkaynak Ş, Öner Z (2009). Syn-extensional granitoids in East.. doi: 10.1017/CBO9781139195430 the Menderes core complex and the late Cenozoic extensional Ambraseys, Jackson (1998). Faulting associated with historical and tectonics of the Aegean province. Geological Society, London, recent earthquakes in the Eastern Mediterranean region. Special Publications 321 (1): 197-223. doi: 10.1144/SP321.10 Geophysical Journal International 133 (2): 390-406. doi: Duman TY, Emre Ö, Özalp S, Elmaci H (2011). 1:250,000 scale Active 10.1046/j.1365-246X.1998.00508.x Fault Map Series of Turkey, “Aydin” (NJ35-11) Quadrangle, Barka (1992). The North Anatolian fault zone. Annales Tectonicae General Directorate of Mineral Research and Exploration, 6: 164. Ankara, Turkey.Emre (2011). Urla” (NJ35-6). 1 (250). Benetatos C, Kiratzi A, Ganas A, Ziazia M, Plessa A et al. (2006). Emre Ö, Özalp S (2011). 1:250,000 scale Active Fault Map Series of Strike-slip motions in the Gulf of Siğaçik (western Turkey): Turkey, “Urla” (NJ35-6) Quadrangle, General Directorate of Properties of the 17 October 2005 earthquake seismic Mineral Research and Exploration, Ankara, Turkey. sequence. Tectonophysics 426 (3-4): 263-279. doi: 10.1016/j. Emre Ö, Özalp S, Duman TY (2011). 1:250,000 scale Active Fault tecto.2006.08.003 Map Series of Turkey, “Izmir” (NJ35-7) Quadrangle, General Boronkay K, Doutsos T (1994). Transpression and transtension Directorate of Mineral Research and Exploration, Ankara, within different structural levels in the central Aegean Turkey. region. Journal of Structural Geology 16 (11): 1555-1573. doi: Emre Ö, Duman TY, Özalp S, Şaroğlu F, Olgun Ş et al. (2018). Active 10.1016/0191-8141(94)90033-7 fault database of Turkey. Bulletin of Earthquake Engineering Bozkurt E (2003). Origin of NE-trending basins in western Turkey. 16 (8): 3229-3275. doi: 10.1007/s10518-016-0041-2 Geodinamica Acta 16 (2-6): 61-81. doi: 10.1016/S0985- England P, Houseman G, Nocquet J (2016). Constraints from GPS 3111(03)00002-0 measurements on the dynamics of deformation in Anatolia Bozkurt E, Sözbilir H (2004). Tectonic evolution of the Gediz and the Aegean. Journal of Geophysical Research: Solid Earth Graben: field evidence for an episodic, two-stage extension 121 (12): 8888-8916. doi: 10.1002/2016JB013382 in western Turkey. Geological Magazine 141 (1): 63-79. doi: Erkül F, Helvaci C, Sözbilir H (2005). Evidence for two episodes of 10.1017/S0016756803008379 volcanism in the Bigadiç borate basin and tectonic implications Brun J, Faccenna C (2008). Exhumation of high-pressure rocks for western Turkey. Geological Journal 40 (5): 545-570. doi: driven by slab rollback. Earth and Planetary Science Letters 10.1002/gj.1026 272 (1-2): 1-7. doi: 10.1016/j.epsl.2008.02.038 EPN/Euref: https://www.epncb.oma.be/ Pavlides (2013). The Greek Database of Seismogenic Sources (GreDaSS), version 2.0.0: A compilation of potential Evelpidou N, Karkani A, Kampolis I (2021). Relative sea level seismogenic sources (Mw > 5.5) in the Aegean Region.. doi: changes and morphotectonic ımplications triggered by the 10.15160/unife/gredass/0200 Samos earthquake of 30th October 2020. Journal of Marine Science and Engineering 9 (1): 40. doi: 10.3390/jmse9010040 Cetin (2020). Seismological and Engineering Effects of the M 7.0 Samos Island (Aegean Sea) Earthquake. Geotechnical Extreme Eyidoğan H (1988). Rates of crustal deformation in western Turkey Events Reconnaissance Association: Report Geer-069. doi: as deduced from major earthquakes. Tectonophysics 148 (1-2): 10.18118/G6H088 83-92. doi: 10.1016/0040-1951(88)90162-X Chatzipetros A, Kiratzi A, Sboras S, Zouros N, Pavlides S Floyd MA, Billiris H, Paradissis D, Veis G, Avallone A et al. (2010). A (2013). Active faulting in the north-eastern Aegean Sea new velocity field for Greece: Implications for the kinematics Islands. Tectonophysics 597-598: 106-122. doi: 10.1016/j. and dynamics of the Aegean. Journal of Geophysical Research tecto.2012.11.026 115 (B10). doi: 10.1029/2009JB007040 Chousianitis K, Konca AO (2021). Rupture Process of the 2020 Fotiou (2010). The Hermes GNSS NtripCaster of AUTh. 69 (1): 35. Mw 7.0 Samos Earthquake and its Effect on Surrounding Foumelis (2021). On rapid multidisciplinary response aspects for Active Faults. Geophysical Research Letters 48 (14). doi: Samos 2020 M7. 0 earthquake. Acta Geophysica: 1. 10.1029/2021GL094162 Ganas (2021). Co-seismic and post-seismic deformation, field Çırmık A, Pamukçu O, Gönenç T, Kahveci M, Şalk M et al. (2017). observations and fault model of the 30 October 2020 Mw= 7.0 Examination of the kinematic structures in İzmir (Western Samos earthquake. Aegean Sea. Acta Geophysica: 1. Anatolia) with repeated GPS observations (2009, 2010 and 2011). Journal of African Earth Sciences 126: 1-12. doi: Genç CŞ, Altunkaynak Ş, Karacık Z, Yazman M, Yılmaz Y (2001). 10.1016/j.jafrearsci.2016.11.020 The Çubukludağ graben, south of İzmir: its tectonic significance in the Neogene geological evolution of the Cohen HA, Dart CJ, Akyüz HS, Barka A (1995). Syn-rift sedimentation western Anatolia. Geodinamica Acta 14 (1-3): 45-55. doi: and structural development of the Gediz and Büyük Menderes 10.1080/09853111.2001.11432434 graben, western Turkey. Journal of The Geological Society 152 (4): 629-638. doi: 10.1144/gsjgs.152.4.0629 GreDaSS: http://gredass.unife.it/ 714
SBORAS et al. / Turkish J Earth Sci Gürer O (2001). Neogene basin development around Söke-Kusadası Pichon XL, Angelier J (1979). The hellenic arc and trench system: A (western Anatolia) and its bearing on tectonic development key to the neotectonic evolution of the eastern mediterranean of the Aegean region. Geodinamica Acta 14 (1-3): 57-69. doi: area. Tectonophysics 60 (1-2): 1-42. doi: 10.1016/0040- 10.1016/S0985-3111(00)01059-7 1951(79)90131-8 Yavuz Hakyemez H, Erkal T, Göktas F (1999). Late Quaternary Lin J, Stein RS (2004). Stress triggering in thrust and subduction evolution of the Gediz and Büyük Menderes grabens, Western earthquakes and stress interaction between the southern Anatolia, Turkey. Quaternary Science Reviews 18 (4-5): 549- San Andreas and nearby thrust and strike-slip faults. 554. doi: 10.1016/S0277-3791(98)00096-1 Journal of Geophysical Research: Solid Earth 109 (B2). doi: Hancock P, Barka A (1987). Kinematic indicators on active normal 10.1029/2003JB002607 faults in Western Turkey. Journal of Structural Geology 9 (5-6): Maley (1971). Morphology and structure of the Aegean Sea. 18: 109. 573-584. doi: 10.1016/0191-8141(87)90142-8 Mckenzie D (1972). Active Tectonics of the Mediterranean Region. Herring TA, King RW, Floyd MA, McClusky SC (2015). Introduction Geophysical Journal International 30 (2): 109-185. doi: to GAMIT/GLOBK, Release 10.6, Mass. Inst. of Technol, 10.1111/j.1365-246X.1972.tb02351.x Cambridge. Mountrakis D, Kilias A, Vavliakis E, Psilovikos A, Thomaidou E HxGn-SmartNet Greece: https://hxgnsmartnet.com/el-gr (2003). Neotectonic map of Samos Island (Aegean Sea, Greece): Jolivet L, Brun J (2008). Cenozoic geodynamic evolution of the Implication of Geographical Information Systems in the Aegean. International Journal of Earth Sciences 99 (1): 109-138. geological mapping. In: 4th European Congress on Regional doi: 10.1007/s00531-008-0366-4 Geoscientific Cartography and Information Systems: 11. Jolivet L (2003). Subduction tectonics and exhumation of high- Mountrakis D, Kilias A, Vavliakis E, Psilovikos A, Karakaisis G, pressure metamorphic rocks in the Mediterranean orogens. Papazachos C et al. (2006). Neotectonic Map of Greece American Journal of Science 303 (5): 353-409. doi: 10.2475/ series, “Samos” sheet, 2 1:75,000 scale maps with pamphlet, ajs.303.5.353 Earthquake Planning and Protection Organization. Jolivet L, Faccenna C, Huet B, Labrousse L, Le Pourhiet L et al. Mozafari N, Tikhomirov D, Sumer Ö, Özkaymak Ç, Uzel B et al. (2013). Aegean tectonics: Strain localisation, slab tearing and (2019). Dating of active normal fault scarps in the Büyük trench retreat. Tectonophysics 597-598: 1-33. doi: 10.1016/j. Menderes Graben (western Anatolia) and its implications for tecto.2012.06.011 seismic history. Quaternary Science Reviews 220: 111-123. doi: Jolivet L, Menant A, Clerc C, Sternai P, Bellahsen N et al. (2018). 10.1016/j.quascirev.2019.07.002 Extensional crustal tectonics and crust-mantle coupling, a view Nyst M, Thatcher W (2004). New constraints on the active tectonic from the geological record. Earth-science Reviews 185: 1187- deformation of the Aegean. Journal of Geophysical Research: 1209. doi: 10.1016/j.earscirev.2018.09.010 Solid Earth 109 (B11). doi: 10.1029/2003JB002830 Karakaisis G, Papazachos C, Scordilis E (2017). SEISMIC Sources and Ocakoğlu N, Demirbağ E, Kuşçu İ (2005). Neotectonic structures maın seısmıc faults ın the aegean and surroundıng area. Bulletin in İzmir Gulf and surrounding regions (western Turkey): of The Geological Society of Greece 43 (4): 2026. doi: 10.12681/ Evidences of strike-slip faulting with compression in the bgsg.11393 Aegean extensional regime. Marine Geology 219 (2-3): 155- Karakostas (2021). Seismotectonic implications of the 2020 Samos, 171. doi: 10.1016/j.margeo.2005.06.004 Greece, M w 7.0 mainshock based on high-resolution aftershock Okada Y (1992). Internal deformation due to shear and tensile relocation and source slip model. Acta Geophysica: 1. faults in a half-space. Bulletin of The Seismological Society of Kiratzi A (2003). Focal mechanisms of shallow earthquakes in the America 82 (2): 1018-1040. doi: 10.1785/BSSA0820021018 Aegean Sea and the surrounding lands determined by waveform Okay (1991). Geology and tectonic evolution of the Biga Peninsula, modelling: a new database. Journal of Geodynamics 36 (1-2): northwest Turkey. Bull. - Tech. Univ. Istanbul 44: 191. 251-274. doi: 10.1016/S0264-3707(03)00050-4 Papadimitriou P, Kapetanidis V, Karakonstantis A, Spingos I, KOERI moment tensor catalogue: http://www.koeri.boun.edu.tr/ Kassaras I et al. (2020). First Results on the Mw=6.9 Samos sismo/2/moment-tensor-solutions/ Earthquake of 30 October 2020. Bulletin of The Geological KOERI seismic catalogue: http://www.koeri.boun.edu.tr/sismo/2/ Society of Greece 56 (1): 251. doi: 10.12681/bgsg.25359 latest-earthquakes/map/ Papazachos B, Papazachou C (2003). The earthquakes of Greece, 3rd Kokkalas (2006). Postcollisional contractional and extensional Edition. ed. Ziti Publications. deformation in the Aegean region. Geological Society of Pavlides S, Tsapanos T, Zouros N, Koravos G, Chatzipetros A (2009). America 409 (06): 97. doi: 10.1130/2006.2409(06 Using active fault data for assessing seismic hazard: a case Kumerics C, Ring U, Brichau S, Glodny J, Monié P (2005). The study from NE Aegean Sea, Greece, in: 17th International extensional Messaria shear zone and associated brittle Conference on Soil Mechanics and Geotechnical Engineering, detachment faults, Aegean Sea, Greece. Journal of The Geological Earthquake Geotechnical Engineering Satellite Conference. Society 162 (4): 701-721. doi: 10.1144/0016-764904-041 pp. 1–14. 715
SBORAS et al. / Turkish J Earth Sci Pondrelli S, Salimbeni S, Morelli A, Ekström G, Postpischl L et al. Topal S (2019). Evaluation of relative tectonic activity along the (2011). European–mediterranean regional centroid moment Priene-Sazlı Fault (Söke Basin, southwest Anatolia): Insights tensor catalog: solutions for 2005–2008. Physics of The Earth from geomorphic indices and drainage analysis. Journal of and Planetary Interiors 185 (3-4): 74-81. doi: 10.1016/j. Mountain Science 16 (4): 909-923. doi: 10.1007/s11629-018- pepi.2011.01.007 5274-x Ring U, Gessner K, Güngör T, Passchier CW (1999). The Menderes Triantafyllou I, Gogou M, Mavroulis S, Lekkas E, Papadopoulos massif of western Turkey and the cycladic massif in the GA et al. (2021). The tsunami caused by the 30 October aegean—do they really correlate?. Journal of The Geological 2020 Samos (Aegean Sea) Mw7.0 Earthquake: hydrodynamic Society 156 (1): 3-6. doi: 10.1144/gsjgs.156.1.0003 features, source properties and ımpact assessment from post- event field survey and video records. Journal of Marine Science Ring U, Gessner K, Thomson S (2017). Variations in fault-slip data and Engineering 9 (1): 68. doi: 10.3390/jmse9010068 and cooling history reveal corridor of heterogeneous backarc extension in the eastern Aegean Sea region. Tectonophysics Turkish National Permanent RTK Network: http://www.tusaga-aktif. 700-701: 108-130. doi: 10.1016/j.tecto.2017.02.013 gov.tr/ Sboras (2011). The Greek Database of Seismogenic Sources: Uranus: http://www.uranus.gr/ seismotectonic implications for North Greece. University of Uzel B, Sözbilir H, Özkaymak Ç, Kaymakcı N, Langereis CG Ferrara. (2013). Structural evidence for strike-slip deformation in Schmid SM, Fügenschuh B, Kounov A, Maţenco L, Nievergelt P et the İzmir–Balıkesir transfer zone and consequences for late al. (2020). Tectonic units of the Alpine collision zone between Cenozoic evolution of western Anatolia (Turkey). Journal of Eastern Alps and western Turkey. Gondwana Research 78: 308- Geodynamics 65: 94-116. doi: 10.1016/j.jog.2012.06.009 374. doi: 10.1016/j.gr.2019.07.005 Uzel (2012). Neotectonic evolution of an actively growing Searle (2021). Compressional origin of the Aegean Orogeny, Greece. superimposed basin in western Anatolia: the ınner bay of Geoscience Frontiers, In Press. İzmir, Turkey. Turkish Journal of Earth Sciences 21: 439-471. Sözbilir H, Sarι B, Uzel B, Sümer Ö, Akkiraz S (2010). Tectonic Uzel B, Sümer Ö, Özkaptan M, Özkaymak Ç, Kuiper K et al. implications of transtensional supradetachment basin (2017). Palaeomagnetic and geochronological evidence for a development in an extension-parallel transfer zone: the major middle Miocene unconformity in Söke Basin (western Kocaçay Basin, western Anatolia, Turkey. Basin Research 23 Anatolia) and its tectonic implications for the Aegean region. (4): 423-448. doi: 10.1111/j.1365-2117.2010.00496.x Journal of The Geological Society 174 (4): 721-740. doi: 10.1144/jgs2016-006 Stewart I, Hancock P (1991). Scales of structural heterogeneity within neotectonic normal fault zones in the Aegean region. Journal Vadaloukas G, Vintzilaiou E, Ganas A, Giarlelis C, Ziotopoulou K et of Structural Geology 13 (2): 191-204. doi: 10.1016/0191- al. (2020). Samos earthquake, 30 October 2020 ‐ Preliminary 8141(91)90066-R Report. Hellenic Association of Earthquake Engineering, 65 p. Available at: https://www.eltam.org/images/nltr/ Stewart IS, Hancock PL (2007). Normal fault zone evolution and newsletters/20201125/etam_report_samos2020earthquake. fault scarp degradation in the Aegean region. Basin Research 1 pdf (3): 139-153. doi: 10.1111/j.1365-2117.1988.tb00011.x Van Hinsbergen DJJ, Hafkenscheid E, Spakman W, Meulenkamp J, Sümer Ö, İnci U, Sözbilir H (2013). Tectonic evolution of the Söke Wortel R (2005). Nappe stacking resulting from subduction of Basin: Extension-dominated transtensional basin formation oceanic and continental lithosphere below. Greece. Geology 33 in western part of the Büyük Menderes Graben, Western (4): 325. doi: 10.1130/G20878.1 Anatolia, Turkey. Journal of Geodynamics 65: 148-175. doi: Van Hinsbergen DJJ, Dekkers MJ, Bozkurt E, Koopman M (2010). 10.1016/j.jog.2012.06.005 Exhumation with a twist: Paleomagnetic constraints on the Tan DC, Zhao JL, Liu XF, Fan YY, Liu J et al. (2014). Features of evolution of the Menderes metamorphic core complex, western regional variations of the spontaneous field. Chinese Journal of Turkey. Tectonics 29 (3). doi: 10.1029/2009TC002596 Geophysics 57 (3): 318-331. doi: 10.1002/cjg2.20106 Wells (1994). New Empirical Relationships among Magnitude, Theodoropoulos D (1979a). Geological Map of Greece, 1:50,000 Rupture Length, Rupture Width, Rupture Area, and Surface scale, “Samos” sheet (2 maps), Insitute of Geological and Displacement. Bulletin of the Seismological Society of America Mining Research (IGMR), Athens. 84: 974. Theodoropoulos D (1979b). Geological Map of Greece, 1:50,000 Westerweel J, Licht A, Cogné N, Roperch P, Dupont‐nivet G et al. scale, “Samos” sheet (2 maps), Insitute of Geological and (2020). Burma terrane collision and northward ındentation Mining Research (IGMR), Athens. in the eastern Himalayas recorded in the eocene‐miocene Toda S (2005). Forecasting the evolution of seismicity in chindwin basin (Myanmar). Tectonics 39 (10). doi: southern California: Animations built on earthquake stress 10.1029/2020TC006413 transfer. Journal of Geophysical Research 110 (B5). doi: Yalciner (2020). Izmir-Samos earthquake and tsunami; post-tsunami 10.1029/2004JB003415 field survey preliminary results. The 30 October 11. 716
SBORAS et al. / Turkish J Earth Sci Yolsal-çevikbilen S, Taymaz T, Helvacı C (2014). Earthquake Yönlü Ö, Altunel E, Karabacak V, Akyüz S, Yalçıner Ç (2010). Offset mechanisms in the Gulfs of Gökova, Sığacık, Kuşadası, and the archaeological relics in the western part of the Büyük Menderes Simav Region (western Turkey): Neotectonics, seismotectonics graben (western Turkey) and their tectonic implications. In: and geodynamic implications. Tectonophysics 635: 100-124. Ancient Earthquakes. Geological Society of America 10 (21): doi: 10.1016/j.tecto.2014.05.001 269. doi: 10.1130/2010.2471(21 717
SBORAS et al. / Turkish J Earth Sci Annex The following diagrams shows the coordinates time series for the selected permanent GNSS stations for three days before and four days after the seismic event. 1