The Erzincan earthquake of 13 March 1992 in Eastern Turkey

The 13 March 1992 Erzincan earthquake, M=6.8, occurred in the eastern half of the Erzincan basin. The largest aftershock took place near Pülümür on 15 March 1992. No clear surface breaks were observed, although teleseismic studies suggested that it was a strike-slip earthquake striking parallel to the North Anatolian fault, with a focus of approximately 10±2 km depth, 30 km rupture length, 95 cm of slip, and a 1.16×10²⁶ dyn.cm seismic moment. The aftershock distribution concentrated at an area of the intersection between the North Anatolian fault and the Ovacik fault. These results indicate that the previously suggested seismic gap along the North Anatolian fault, east of Erzincan, still remains unruptured.     

The Roermond earthquake of 13 April 1992, the Netherlands: geological aspects

The epicentre of the Roermond earthquake is located near the western boundary fault of the Roer valley trough, one of the deepest and the most active in the Quaternary part of the Lower Rhine graben. The Late Pleistocene and Holocene activity of the trough is manifested by offsets of the main (Mindel) and the lower (Riss and Late Pleistocene) terraces along the boundary faults.
Surface fractures have been observed in an area of more than 50 km²: 2.5–3.5 km to the north of the town of Roermond, at 0.8 km to the south of the village of Herkenbosch and in the southeastern part of the village of Montfort. Three types of ruptures were differentiated: scarps up to 50 cm high along open fractures near the Maas River; open fractures (continued by scarps in some places) and open fractures with a liquefaction of the Quaternary alluvium sands. The last type is predominant. The location of the ruptures depends on the landscape and water-table of the region. While they could be produced solely by hydraulic shock during the earthquake (increased by the wet spring season), the majority of the ruptures strike N50°W ± 10°, i.e. parallel to the main trough faults, or N55°E ± 10°, along 'neotectonic lines', parallel to the Maas valley and deduced from space imagery. Thus, the ruptures could be the secondary surficial effect of the earthquake, linked indirectly with the active tectonics of the region.     

28 March 1970 Gediz earthquake fault,western Turkey: palaeoseismology and tectonic significance

On 28 March 1970, an unexpected and destructive earthquake (Ms = 7.2) originated along the Erdo?mu? fault (EF), which forms the southern margin of the modern Erdo?mu?–Yenigediz graben in the central part of the Ak?ehir–Simav fault system. The EF is a N-dipping normal fault, ～12 km long, generally E–W-trending, and characterized by a minor right-lateral strike–slip component. To determine its past activity, a palaeoseismological exploratory trenching study was conducted. Two trenches (EFT-1 and EFT-2) were excavated on the ground surface rupture of the 1970 Gediz earthquake near Erdo?mu? village. Based on the relative displacement between units observed and mapped in EFT-1, at least three events were identified. Two events were also identified in EFT-2. Only one of the events in EFT-1 can be dated via ¹⁴C. The estimated recurrence interval on the EF is ～910 ± 40 years.     

The Roer Valley Graben earthquake of 13 April 1992 and its seismotectonic setting

A moderately sized pure, normal dip-slip earthquake occurred in the Roer Valley Graben (RVG) near Roermond, The Netherlands on 13 April 1992 at 1h 20m UTC. This contribution presents an overview of the locations, fault-plane solutions and magnitudes obtained for the mainshock and the aftershocks by the different scientific groups involved in their analysis. The observed maximum intensity of VII is compared with that of other earthquakes in the region to illustrate the relatively low level of damage caused by the mainshock.
Using SH and Lg waves recorded at seven local and regional broadband stations, we determine a seismic moment of 1.4 × 10¹⁷ Nm, a static stress drop of 9.7 MPa and an average displacement of 33 cm over a rupture surface of approximately 11 km².
The seismotectonics of the region extending from the RVG to the city of Liège including the western part of the Rhenish Massif (WRM) and the eastern part of the Brabant Massif (EBM) is analysed based on the Roermond earthquake studies and data collected since 1985 by the Belgian seismic network. The geographical distribution of focal mechanism reveals four different seismotectonic regimes in this area. From stress tensor inversion we find that ₃ coincides with the minimum horizontal stress component in the RVG, the WRM and possibly in the EBM, while in the Liège region ₃ is approximately vertical. The minimum horizontal stress component shows a 30° rotation to the north in the WRM and the Liège region and possibly 50° in the EBM when compared with the minimum horizontal stress component in the RVG.     

The 20 March 1992 South Aegean,Greece, earthquake (Ms= 5.3): possible anomalous effects

The earthquake (M_s= 5.3) of 20 March 1992 and its aftershocks, which occurred near the volcanic island complex of Milos, South Aegean, Greece, are studied on the basis of filed observations and instrumental data. The mainshock caused some building damage, the maximum intensity of VI+ (MM) being assigned to Triovasalos, Milos. Ground cracks, liquefaction in soil, landslides and rockfalls were observed in Milos. Liquefaction took place at an apparently anomalously long epicentral distance (D= 12 km) and is associated with unusually small earthquake magnitude. Abnormal animal behaviour was reported no longer than twelve hours before the mainshock. The b-value (= 1.02) of the G–R relation for the aftershock sequence, the exponentially decreasing number of aftershocks with time, and the difference (= 0.5) in magnitude between the mainshock and its largest aftershock imply that the origin of these earthquakes is tectonic and not associated with the volcanic field of Milos.     

Geomorphic evidence for active tectonic deformation in the coastal part of Eastern Black Sea,Eastern Pontides,Turkey

The Eastern Pontides (EP), which is the under transpressional deformation zone, is an active mountain belt that has been rising rapidly since the Cenozoic era because of the Arabian-Eurasian convergence. Morphometric studies have been performed to investigate the tectonic activity of this region and better understand the characteristics of the faults geomorphologically; the faults control the mountain fronts in the drainage basin of the EP. The results show the Hypsometric Curve (HC)-Hypsometric Integral (0.37-HI-0.67), Basin-Shaped Analysis (1.2-B_s-7), Valley-Floor-Width to Height-Ratio (0.4-V_f-1.2) and Asymmetry Factor (35-AF-81) applied to 46 drainage basins together with 9 tectonically controlled geomorphic indices (1.2-S_mf-1.5) and a Stream Length Gradient (30-S_L-120) indicate that the EP is tectonically active, and when the areas are evaluated according to S_mf and V_f analyses, the tectonic level is relatively high. According to our conceptual model for the uplifting of the EP, with respect to field studies and morphometric analysis, (i) the EP is the active deformation zone and has a “push-up” geometry in conjunction with the North Anatolian Fault; (ii) the EP is progressively uplifting at a rate of more than 0.5 mm/yr in along with the thrust faults of the Black Sea Fault (BSF) and Borjomi-Kazbegi Fault (BKF).     

Pre-eruptive conditions revealed by mega- and pheno-cryst compositions from the Quaternary Erzincan Volcanics, Eastern Turkey: Insights into the magma processes

Quaternary Erzincan Volcanics (QEVs) from the Erzincan Basin consist of mega- and pheno-cryst-bearing high-K calc-alkaline dome lavas. Fourteen nearly phenocrystic domes, with a range of basaltic-andesite, andesite, dacite and rhyolite compositions, were emplaced in the North Anatolian Fault Zone. The emplacement ages yielded by the unspiked K–Ar technique range from 102 to 140 ka. The andesitic domes (each less than 3 km in diameter) contain amphibole megacrysts. Amphibole compositions show a linear variation from ferro-edenite, edenite to pargasite from rhyolite to andesite. Pargasitic amphibole megacrysts scattered into the groundmass are very similar in composition to the microlites. All plagioclases are 53 mol%. Oscillation types are An₃₂₋₅₀ whose variations range from 10 to 16 mol% An and have 10–150 μm in thickness. Pre-eruptive conditions, calculated from mega- and pheno-cryst composition, using pyroxene and two oxide thermometers and the Al-in-hornblende barometer, ranged from 918 to 837 °C and 6.6 to 4.3 kbar for andesitic magma, 824–755 °C and 4.6–4.2 kbar for dacitic magma to 803–692 °C and 4.3–3.9 kbar for rhyolitic magma, which correspond to a depth of >10 km for storage region of the crust. The fO₂ values vary from −14.25 to −15.35 log units which are plotted just below nickel–nickel oxide (NNO) buffers. The systematic decrease in thermobarometric results from andesite to rhyolite is consistent with a single magma reservoir moving upward through the crust followed by fractional crystallization. Textural and compositional relationships of mega- and pheno-crystic phases suggest that magma mixing, fluid input to the reservoir and fractional crystallization processes, with a small amount crustal contamination play key role in evolution of the QEVs.     

Assessment of liquefaction potential of Erzincan Province and its vicinity,Turkey

One of the most important causes of damages after the earthquakes is the soil liquefaction. Liquefaction can be defined as temporary loss in strength of saturated sandy and silty deposits under transient and cyclic loadings due to excess pore water pressure. This study includes determination of liquefaction potential in Erzincan city center and its vicinity. Due to the proximity of the North Anatolian Fault Zone, in a probable earthquake, Erzincan Province is thought to be affected. In this context, the earthquake scenarios were produced using the empirical expressions. Liquefaction potential for different earthquake magnitudes was determined. These earthquake magnitudes were selected as 6.0, 6.5, 7.0, 7.5, respectively. Liquefaction potential was investigated using standard penetration test (SPT) data. The first stage of the study, 63 boreholes in different locations was drilled and SPT was performed. Disturbed and undisturbed soil samples were taken from these boreholes. Laboratory testing was performed to determine physical properties of soil samples, and liquefaction potential analyses were examined using three methods, namely Seed and Idriss (J Soil Mech Found Div ASCE 97(9):1249–1273, 1971), Tokimatsu and Yoshimi (Soil Found 23(4):56–74, 1983), Iwasaki et al. (Soil liquefaction potential evaluation with use of the simplified procedure. International Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics, St. Louis, pp 209–214, 1981). In order to complete liquefaction analysis within a short time, MATLAB program was prepared. Liquefaction potential analyses were carried out with the MATLAB program. At the final stage of this study, liquefaction potential maps were prepared for different earthquake magnitudes. The expected results will be shared with the local authorities and important engineering remedial measurements will be proposed to prevent further life losses and to mitigate property losses.     

Evaluation of seismic hazard at Roermond, The Netherlands: A comparison of results after the 13 April 1992 earthquake

Mean return periods, RP, for the site of Roermond, The Netherlands, as calculated by different methods, are compared, and its quality evaluated by a simple two-tail test of hypothesis. Results show that RP values by the EGO-method are statistically more likely. They can be considered, despite their broad 90% probability intervals, and for the site and data used, more reliable, since the Roermond earthquake was not an unusual or surprising event for the Lower Rhine Embayment area, where earthquakes of comparable size have occurred since the 18th century.     

Tethyan evolution of Turkey: A plate tectonic approach

The Tethyan evolution of Turkey may be divided into two main phases, namely a Palaeo-Tethyan and a Neo-Tethyan, although they partly overlap in time. The Palaeo-Tethyan evolution was governed by the main south-dipping (present geographic orientation) subduction zone of Palaeo-Tethys beneath northern Turkey during the Permo-Liassic interval. During the Permian the entire present area of Turkey constituted a part of the northern margin of Gondwana-Land. A marginal basin opened above the subduction zone and disrupted this margin during the early Triassic. In this paper it is called the Karakaya marginal sea, which was already closed by earliest Jurassic times because early Jurassic sediments unconformably overlie its deformed lithologies. The present eastern Mediterranean and its easterly continuation into the Bitlis and Zagros oceans began opening mainly during the Carnian—Norian interval. This opening marked the birth of Neo-Tethys behind the Cimmerian continent which, at that time, started to separate from northern Gondwana-Land. During the early Jurassic the Cimmerian continent internally disintegrated behind the Palaeo-Tethyan arc constituting its northern margin and gave birth to the northern branch of Neo-Tethys. The northern branch of Neo-Tethys included the Intra-Pontide, Izmir—Ankara, and the Inner Tauride oceans. With the closure of Palaeo-Tethys during the medial Jurassic only two oceanic areas were left in Turkey: the multi-armed northern and the relatively simpler southern branches of Neo-Tethys. The northern branch separated the Anatolide—Tauride platform with its long appendage, the Bitlis—Pötürge fragment from Eurasia, whereas the southern one separated them from the main body of Gondwana-Land. The Intra-Pontide and the Izmir—Ankara oceans isolated a small Sakarya continent within the northern branch, which may represent an easterly continuation of the Paikon Ridge of the Vardar Zone in Macedonia. The Anatolide-Tauride platform itself constituted the easterly continuation of the Apulian platform that had remained attached to Africa through Sicily. The Neo-Tethyan oceans reached their maximum size during the early Cretaceous in Turkey and their contraction began during the early late Cretaceous. Both oceans were eliminated mainly by north-dipping subduction, beneath the Eurasian, Sakaryan, and the Anatolide- Tauride margins. Subduction beneath the Eurasian margin formed a marginal basin, the present Black Sea and its westerly prolongation into the Srednogorie province of the Balkanides, during the medial to late Cretaceous. This resulted in the isolation of a Rhodope—Pontide fragment (essentially an island arc) south of the southern margin of Eurasia. Late Cretaceous is also a time of widespread ophiolite obduction in Turkey, when the Bozkir ophiolite nappe was obducted onto the northern margin of the Anatolide—Tauride platform. Two other ophiolite nappes were emplaced onto the Bitlis—Pötürge fragment and onto the northern margin of the Arabian platform respectively. This last event occurred as a result of the collision of the Bitlis—Pötürge fragment with Arabia. Shortly after this collision during the Campanian—Maastrichtian, a subduction zone began consuming the floor of the Inner Tauride ocean just to the north of the Bitlis—Pötürge fragment producing the arc lithologies of the Yüksekova complex. During the Maastrichtian—Middle Eocene interval a marginal basin complex, the Maden and the Çüngüş basins began opening above this subduction zone, disrupting the ophiolite-laden Bitlis—Pötürge fragment. The Anatolide-Tauride platform collided with the Pontide arc system (Rhodope—Pontide fragment plus the Sakarya continent that collided with the former during the latest Cretaceous along the Intra Pontide suture) during the early to late Eocene interval. This collision resulted in the large-scale south-vergent internal imbrication of the platform that produced the far travelled nappe systems of the Taurides, and buried beneath these, the metamorphic axis of Anatolia, the Anatolides. The Maden basin closed during the early late Eocene by north-dipping subduction, synthetic to the Inner-Tauride subduction zone that had switched from south-dipping subduction beneath the Bitlis—Pötürge fragment to north dipping subduction beneath the Anatolide—Tauride platform during the later Palaeocene. Finally, the terminal collision of Arabia with Eurasia in eastern Turkey eliminated the Çüngüş basin as well and created the present tectonic regime of Turkey by pushing a considerable piece of it eastwards along the two newly-generated transform faults, namely those of North and East Anatolia. Much of the present eastern Anatolia is underlain by an extensive mélange prism that accumulated during the late Cretaceous—late Eocene interval north and east of the Bitlis—Pötürge fragment.     

Petrochemical constraints on the models of Cretaceous-Eocene tectonic evolution of the Eastern Pontic chain (Turkey)

Sixteen selected samples from the Upper Cretaceous volcanic belt of the Eastern Pontids have been analysed for major elements, Rb, Sr and Zr. On the basis of the K₂O versus SiO₂ distribution, two groups of rocks have been distinguished, one with calc-alkaline affinity and a second group with shoshonitic character. The calc-alkaline rocks have porphyritic texture with clinopyroxene, plagioclase and orthopyroxene as phenocryst and in the groundmass. The orthopyroxene is lacking in the shoshonites where plagioclase, clinopyroxene and, in the more evolved terms, amphibole and biotite are the main phenocryst minerals. The shoshonitic rocks have higher $\text{K}{}_2\text{O}\text{Na}{}_2\text{O}$ ratio, K₂O, P₂O₅ and Rb, contents with respect to the calc-alkaline samples. The TiO₂ content is invariably low, never exceeding approximately 1%. The occurrence of volcanic rocks ranging in composition from calc-alkaline to shoshonitic in the Upper Cretaceous volcanic belt of the Eastern Pontids suggests that the Upper Cretaceous volcanic cycle reached its mature stage before the onset of the Eocene calc-alkaline volcanism which is believed to be neither genetically nor tectonically related with the Upper Cretaceous volcanism.     

Seismic damage assessment based on regional synthetic ground motion dataset: a case study for Erzincan,Turkey

Estimation of seismic losses is a fundamental step in risk mitigation in urban regions. Structural damage patterns depend on the regional seismic properties and the local building vulnerability. In this study, a framework for seismic damage estimation is proposed where the local building fragilities are modeled based on a set of simulated ground motions in the region of interest. For this purpose, first, ground motion records are simulated for a set of scenario events using stochastic finite-fault methodology. Then, existing building stock is classified into specific building types represented with equivalent single-degree-of-freedom models. The response statistics of these models are evaluated through nonlinear time history analysis with the simulated ground motions. Fragility curves for the classified structural types are derived and discussed. The study area is Erzincan (Turkey), which is located on a pull-apart basin underlain by soft sediments in the conjunction of three active faults as right-lateral North Anatolian Fault, left-lateral North East Anatolian Fault, and left-lateral Ovacik Fault. Erzincan city center experienced devastating earthquakes in the past including the December 27, 1939 (Ms = 8.0) and the March 13, 1992 (Mw?=?6.6) events. The application of the proposed method is performed to estimate the spatial distribution of the damage after the 1992 event. The estimated results are compared against the corresponding observed damage levels yielding a reasonable match in between. After the validation exercise, a potential scenario event of Mw?=?7.0 is simulated in the study region. The corresponding damage distribution indicates a significant risk within the urban area.     

The 1 October 1995 Dinar earthquake, SW Turkey

The Dinar earthquake (M_s= 6.1, USGS-PDE) of 1 October 1995 occurred on the NWSE-trending Dinar Fault. The earthquake is associated with a 10-km-long surface rupture with predominantly normal faulting. The mainshock was preceded by a series of foreshocks that started 6 days before the mainshock and included two M_d = 4.5 events. The mainshock source mechanism derived from the inversion of broad-band P waves revealed that two sub-events occurred on a NW-SE trending normal fault with a small strike-slip component. According to the source model estimated in this study, the first rupture started at a depth of about 8 km and reached to a depth of about 12 km propagating north-west. The total seismic moment found from the inversion of P waveforms is 2.0 times 10 ¹⁸ Nm. The seismic moment of the second sub-event was about four times larger than the first one. Field observations, GPS measurements and slip vector obtained from the inversion of broad-band P waveforms suggest that the NW-SE trending Dinar Fault is due to the internal deformation of SW Anatolia moving south-westwards.     

The role of thrust ramp reactivation in pull-apart mechanism of the Erzincan basin,North Anatolian Fault Zone,Turkey

The new field data obtained from the southwestern margin of the Erzincan pull-apart basin located on the eastern segment of North Anatolian Fault Zone indicate that the opening of the basin is not only controlled by pull-apart mechanism but also by a lateral ramp structure associated with SSE-NNW Late Miocene thrusting along the Sivas Basin. The fault bordering the southwestern margin of the basin is the lateral part of the Karada thrust that is the roof thrust of the Sivas fold-thrust system, rather than a segment of the North Anatolian Fault Zone. The Erzincan basin was nucleated as a lateral ramp basin during the Pliocene on the lateral ramp-related folds and expanded by the pull-apart opening mechanism between two segments of the North Anatolian Fault Zone. The WSW-ENE pull-apart opening of the basin was recorded by the Pliocene lacustrine-fluvial sediments and Quaternary volcanics as listric normal faulting.     

An objective test of earthquake migrations in Turkey

An objective method for the examination of supposed migration of earthquake epicenters is discussed and tested for the Turkey region.     

The Rukwa earthquake of 13 December 1910 in East Africa

We reappraised the Rukwa earthquake of 1910 in the East African Rift System. With a magnitude of 7.4, no earthquake in East Africa is known to be larger and it is rivalled only by the recent earthquake in southern Sudan on 20 May 1990. More than 80 per cent of the moment release in the Rift during the last 110 years is due to the Rukwa earthquake and its aftershocks that occurred between the Tanganyka and Nyasa systems. In spite of its large magnitude, the Rukwa earthquake caused very little damage to local types of dwellings and no loss of life. With increasing development of urban areas with modern types of houses, there is now increasing risk in East Africa from major earthquakes.     

Quaternary tectonic faulting in the Eastern United States

Paleoseismological study of geologic features thought to result from Quaternary tectonic faulting can characterize the frequencies and sizes of large prehistoric and historical earthquakes, thereby improving the accuracy and precision of seismic-hazard assessments. Greater accuracy and precision can reduce the likelihood of both underprotection and unnecessary design and construction costs. Published studies proposed Quaternary tectonic faulting at 31 faults, folds, seismic zones, and fields of earthquake-induced liquefaction phenomena in the Appalachian Mountains and Coastal Plain. Of the 31 features, seven are of known origin. Four of the seven have nontectonic origins and the other three features are liquefaction fields caused by moderate to large historical and Holocene earthquakes in coastal South Carolina, including Charleston; the Central Virginia Seismic Zone; and the Newbury, Massachusetts, area. However, the causal faults of the three liquefaction fields remain unclear. Charleston has the highest hazard because of large Holocene earthquakes in that area, but the hazard is highly uncertain because the earthquakes are uncertainly located.Of the 31 features, the remaining 24 are of uncertain origin. They require additional work before they can be clearly attributed either to Quaternary tectonic faulting or to nontectonic causes. Of these 24, 14 features, most of them faults, have little or no published geologic evidence of Quaternary tectonic faulting that could indicate the likely occurrence of earthquakes larger than those observed historically. Three more features of the 24 were suggested to have had Quaternary tectonic faulting, but paleoseismological and other studies of them found no evidence of large prehistoric earthquakes. The final seven features of uncertain origin require further examination because all seven are in or near urban areas. They are the Moodus Seismic Zone (Hartford, Connecticut), Dobbs Ferry fault zone and Mosholu fault (New York City), Lancaster Seismic Zone and the epicenter of the shallow Cacoosing Valley earthquake (Lancaster and Reading, Pennsylvania), Kingston fault (central New Jersey between New York and Philadelphia), and Everona fault-Mountain Run fault zone (Washington, D.C., and Arlington and Alexandria, Virginia).     

Probabilistic evaluation of observed earthquake damage data in Turkey

Empirical, theoretical or hybrid methods can be used for the vulnerability analysis of structures to evaluate the seismic damage data and to obtain probability damage matrices. The information on observed structural damage after earthquakes has critical importance to drive empirical vulnerability methods. The purpose of this paper is to evaluate the damage distributions based on the data observed in Erzincan-1992, Dinar-1995 and Kocaeli-1999 earthquakes in Turkey utilizing two probability models—Modified Binomial Distribution (MBiD) and Modified Beta Distribution (MBeD). Based on these analyses, it was possible (a) to compare the advantages and limitations of the two probability models with respect to their capabilities in modelling the observed damage distributions; (b) to evaluate the damage assessment for reinforced concrete and masonry buildings in Turkey based on these models; (c) to model the damage distribution of different sub-groups such as buildings with different number of storeys or soil conditions according to the both models. The results indicate that (a) MBeD is more suitable than the MBiD to model the observed damage data for both reinforced concrete and masonry buildings in Turkey; (b) the sub-groups with lower number of stories are located in the lower intensity levels, while the sub-groups with higher number of stories depending on local site condition are concentrated in the higher intensity levels, thus site conditions should also be considered in the assessment of the intensity levels; (c) the detailed local models decrease the uncertainties of loss estimation since the damage distribution of sub-groups can be more accurately modelled compared to the general damage distribution models.