Earthquake hazard in Marmara Region, Turkey

Earthquake hazard in the Marmara Region, Turkey has been investigated using time-independent probabilistic (simple Poissonian) and time-dependent probabilistic (renewal) models. The study culminated in hazard maps of the Marmara Region depicting peak ground acceleration (PGA) and spectral accelerations (SA)'s at 0.2 and 1 s periods corresponding to 10 and 2% probabilities of exceedance in 50 yrs. The historical seismicity, the tectonic models and the known slip rates along the faults constitute the main data used in the assignment. Based on recent findings it has been possible to provide a fault segmentation model for the Marmara Sea. For the main Marmara Fault this model essentially identifies fault segments for different structural, tectonic and geometrical features and historical earthquake occurrences. The damage distribution and pattern of the historical earthquakes have been carefully correlated with this fault segmentation model. The inter-event time period between characteristic earthquakes in these segments is consistently estimated by dividing the seismic slip estimated from the earthquake catalog by the GPS-derived slip rate of 22±3 mm/yr. The remaining segments in the eastern and southern Marmara region are also identified using recent geological, geophysical studies and historical earthquakes. The model assumes that seismic energy along the segments is released by characteristic earthquakes. For the probabilistic studies characteristic earthquake based recurrence relationships are used. Assuming normal distribution of inter-arrival times of characteristic earthquakes, the ‘mean recurrence time’, ‘covariance’ and the ‘time since last earthquake’ are developed for each segment. For the renewal model, the conditional probability for each fault segment is calculated from the mean recurrence interval of the characteristic earthquake, the elapsed time since the last major earthquake and the exposure period. The probabilities are conditional since they change as a function of the time elapsed since the last earthquake. For the background earthquake activity, a spatially smoothed seismicity is determined for each cell of a grid composed of cells of size 0.005°×0.005°. The ground motions are determined for soft rock (NEHRP B/C boundary) conditions. Western US-based attenuation relationships are utilized, since they show a good correlation with the attenuation characteristics of ground motion in the Marmara region. The possibility, that an event ruptures several fault segments (i.e. cascading), is also taken into account and investigated by two possible models of cascading. Differences between Poissonian and renewal models, and also the effect of cascading have been discussed with the help of PGA ratio maps.     

1971 San Fernando and 1994 Northridge, California, earthquakes: did the zones with severely damaged buildings reoccur?

This paper compares the distribution of damage from the San Fernando, 1971, and Northridge, 1994, earthquakes. Both events had similar size, occurred on blind thrust faults beneath the densely populated San Fernando Valley of the Los Angeles metropolitan area, and hence offer a rare opportunity to compare the effects of the two earthquakes. In a previous study of the distribution of red-tagged (‘unsafe’) buildings and of breaks in the water distribution system caused by the Northridge earthquake, the authors discovered that buildings were damaged less where the soil response was not linear (as indicated by the breaks in the water pipes), except in localized areas of very severe shaking (peak ground velocity exceeding 150 cm/s). The study in this paper shows that the same applies to the damage caused by the San Fernando earthquake, and that the areas with severely damaged buildings (so called ‘gray zones’) for both earthquakes overlapped. This reoccurrence of damage within the same area is interpreted to result from some specific properties of local soil and geology. These properties are not fully understood at present, but should be explored to provide a basis for a new tool for forecasting microzonation maps, and reducing future seismic hazard.     

Estimating shaking-induced casualties and building damage for global earthquake events: a proposed modelling approach

Recent earthquakes such as the Haiti earthquake of 12 January 2010 and the Qinghai earthquake on 14 April 2010 have highlighted the importance of rapid estimation of casualties after the event for humanitarian response. Both of these events resulted in surprisingly high death tolls, casualties and survivors made homeless. In the Mw = 7.0 Haiti earthquake, over 200,000 people perished with more than 300,000 reported injuries and 2 million made homeless. The Mw = 6.9 earthquake in Qinghai resulted in over 2,000 deaths with a further 11,000 people with serious or moderate injuries and 100,000 people have been left homeless in this mountainous region of China. In such events relief efforts can be significantly benefitted by the availability of rapid estimation and mapping of expected casualties. This paper contributes to ongoing global efforts to estimate probable earthquake casualties very rapidly after an earthquake has taken place. The analysis uses the assembled empirical damage and casualty data in the Cambridge Earthquake Impacts Database (CEQID) and explores data by event and across events to test the relationships of building and fatality distributions to the main explanatory variables of building type, building damage level and earthquake intensity. The prototype global casualty estimation model described here uses a semi-empirical approach that estimates damage rates for different classes of buildings present in the local building stock, and then relates fatality rates to the damage rates of each class of buildings. This approach accounts for the effect of the very different types of buildings (by climatic zone, urban or rural location, culture, income level etc), on casualties. The resulting casualty parameters were tested against the overall casualty data from several historical earthquakes in CEQID; a reasonable fit was found.     

Spectral amplitudes of strong earthquake accelerations recorded in buildings

Fourier spectrum amplitudes of horizontal and vertical earthquake accelerations recorded at the foundation levels of 57 buildings in the Los Angeles metropolitan area have been used to study the dependence of spectral amplitudes on the building foundation sizes. Comparison of these amplitudes with those predicted by empirical models for scaling ‘free field’ Fourier amplitude spectra does not indicate any significant dependence of the spectral amplitudes on the size of the foundation. Third degree polynomials have been employed to smooth the spectra of the accelerations recorded inside the buildings and their coefficients have been examined as functions of the foundation plan dimensions. These results also indicate no significant dependence of the spectral amplitudes on the foundation dimensions. A qualitative analysis of the spectral amplitudes for possible effects caused by the phenomena associated with soil-structure interaction indicates that the Fourier spectra of the recorded accelerations may experience some amplification as the relative ‘density’ of the foundation-structure system increases.     

The impact of epistemic uncertainty on an earthquake loss model

Models capable of estimating losses in future earthquakes are of fundamental importance for emergency planners, for the insurance and reinsurance industries, and for code drafters. Constructing a loss model for a city, region or country involves compiling databases of earthquake activity, ground conditions, attenuation equations, building stock and infrastructure exposure, and vulnerability characteristics of the exposed inventory, all of which have large associated uncertainties. Many of these uncertainties can be classified as epistemic, implying—at least in theory—that they can be reduced by acquiring additional data or improved understanding of the physical processes. The effort and cost involved in refining the definition of each component of a loss model can be very large, for which reason it is useful to identify the relative impact on the calculated losses due to variations in these components. A mechanically sound displacement‐based approach to loss estimation is applied to a test case of buildings along the northern side of the Sea of Marmara in Turkey. Systematic variations of the parameters defining the demand (ground motion) and the capacity (vulnerability) are used to identify the relative impacts on the resulting losses, from which it is found that the influence of the epistemic uncertainty in the capacity is larger than that of the demand for a single earthquake scenario. Thus, the importance of earthquake loss models which allow the capacity parameters to be customized to the study area under consideration is highlighted. Copyright © 2005 John Wiley & Sons, Ltd.     

Nonlinear soil response as a natural passive isolation mechanism. Paper II. The 1933, Long Beach, California earthquake

The areas that experienced large strains and differential motions in the soil (indicated by breaks in the water and gas pipe distribution systems) and the areas with severely damaged buildings showed remarkable separation during the March 10, 1933, Long Beach, California earthquake. With analogous results for the 1994 Northridge, California earthquake [Soil Dynam. Earthquake Engng. 17 (1998) 41], the observations summarized in this paper show the fallacy of simplistic and popular interpretations, such as those that hold that in the near field the damage to buildings is caused by ‘soft’ or ‘bad ground’ conditions. In fact, significant reduction in the potential damage to buildings may be expected in the areas where the soil experiences ‘moderate to large’ strains.     

Estimating severity of liquefaction-induced damage near foundation

An empirical procedure for estimating the severity of liquefaction-induced ground damage at or near foundations of existing buildings is established. The procedure is based on an examination of 30 case histories from recent earthquakes. The data for these case histories consist of observations of the damage that resulted from liquefaction, and the subsurface soil conditions as revealed by cone penetration tests. These field observations are used to classify these cases into one of three damaging effect categories, ‘no damage’, ‘minor to moderate damage’, and ‘major damage’. The potential for liquefaction-induced ground failure at each site is calculated and expressed as the probability of ground failure. The relationship between the probability of ground failure and the damage class is established, which allows for the evaluation of the severity of liquefaction-induced ground damage at or near foundations. The procedure presented herein represents a significant attempt to address the issue of liquefaction effect. Caution must be exercised, however, when using the proposed model and procedure for estimating liquefaction damage severity, because they are developed based on limited number of case histories.     

Regional and local seismic hazard assessment

Recent earthquakes such as the M_JMA 7.2 Hyogo-ken Nambu earthquake and the M 7.4 Kocaeli earthquake demonstrate once again the need to include detailed soil investigation into hazard evaluation, that is the need of microzonation. Seismic hazard assessment evaluated at a regional scale generally does not consider soil effects but only in a limited way using an attenuation law that can be ‘soft soil’ or ‘rock’. However, the relevant role of seismic hazard in the assessment of seismic coefficients for the definition of the actions in seismic codes must be properly considered. That is to say, the level of protection of buildings is proportional to a definite level of hazard (generally considered to be the ground motion with 10% probability of exceedence in 50 years). When a microzonation is performed, this criterion cannot be ignored, therefore, a clear linkage must be established between hazard (regional scale) and microzonation. The crucial point is represented by the reference motion (or input motion) to be used for site effects analysis, that must be compatible with the regional seismic hazard. In this paper, three different approaches for reference motion evaluation are analysed: probabilistic; stochastic; and deterministic. Through the case history of Fabriano microzonation the three approaches are compared. It is shown that each approach presents advantages and disadvantages with respect to the others. For example, the probabilistic approach (the reference motion is directly derived from the expected response spectra for a given return period) is linked with hazard, but produces an overestimation in short periods range, while the deterministic approach correctly simulates the wave propagation, but it ends with a kind of conditional probability. Until now, clear criteria to choose the right approach do not appear to exist and the expert experience is of fundamental importance.     

An alternative approach for the Istanbul earthquake early warning system

Two recent catastrophic earthquakes that struck the Marmara Region on 17 August 1999 (M_w=7.4) and 12 November 1999 (M_w=7.2) caused major concern about future earthquake occurrences in Istanbul and the Marmara Region. As a result of the preparations for an expected earthquake may occur around Istanbul region, an earthquake early warning system has been established in 2002 with a simple and robust algorithm, based on the exceedance of specified thresholds of time domain amplitudes and the cumulative absolute velocity (CAV) levels (Erdik et al., 2003 [1]). In order to improve the capability of Istanbul earthquake early warning system (IEEWS) for giving early warning of a damaging earthquake in the Marmara Region, we explored an alternative approach with the use of a period parameter (τ_c) and a high-pass filtered vertical displacement amplitude parameter (Pd) from the initial 3 s of the P waveforms as proposed by Kanamori (2005) [2] and Wu and Kanamori (2005) 3 and 4. The empirical relationships both between τ_c and moment magnitude (M_w), and between Pd and peak ground velocity (PGV) for the Marmara Region are presented. These relationships can be used to detect a damaging earthquake within seconds after the arrival of P waves, and can provide on-site warning in the Marmara Region.     

Earthquake performance assessment and rehabilitation of two historical unreinforced masonry buildings

The paper describes the earthquake performance assessment of two historical buildings located in Istanbul exposed to a M_w = 7+ earthquake expected to hit the city and proposes solutions for their structural rehabilitation and/or strengthening. Both buildings are unreinforced clay brick masonry (URM) structures built in 1869 and 1885, respectively. The first building is a rectangular-shaped structure rising on four floors. The second one is L-shaped with one basement and three normal floors above ground. They survived the 1894, M_s = 7.0 Istanbul Earthquake, during which widespread damage to URM buildings took place in the city. Earthquake ground motion to be used in performance assessment and retrofit design is determined through probabilistic and deterministic seismic hazard assessment. Strength characteristics of the brick walls are assessed on the basis of Schmidt hammer test results and information reported in the literature. Dynamic properties of the buildings (fundamental vibration periods) are measured via ambient vibration tests. The buildings are modelled and analyzed as three-dimensional assembly of finite elements. Following the preliminary assessment based on the equivalent earthquake loads method, the dynamic analysis procedure of FEMA 356 (Pre-standard and commentary for the seismic rehabilitation of buildings, American Society of Civil Engineers, Reston, 2000) and ASCE/SEI 41-06 (Seismic rehabilitation of existing buildings, American Society of Civil Engineers, Reston, 2007) is followed to obtain dynamic structural response of the buildings and to evaluate their earthquake performance. In order to improve earthquake resistance of the buildings, reinforced cement jacketing of the main load carrying walls and application of fiber reinforced polymer bands to the secondary walls are proposed.     

Earthquake Damage Scenarios of the Building Stock of Potenza (Southern Italy) Including Site Effects

Damage scenarios relevant to the building stock of the town of Potenza, Southern Italy, are presented. A procedure for the preparation of scenarios has been purposely set up. In the first step, the inventory of the building stock has been made. Location and characteristics of buildings have been obtained from a survey carried out after the 1990 Potenza earthquake and further updated in 1999. In the second step, the absolute vulnerability of the buildings has been evaluated. A hybrid technique has been used, where typological analyses and expert judgement are combined together. Beyond the classes of vulnerability A, B and C of the MSK scale, the class D of EMS98 scale, for the less vulnerable buildings, has been considered. The third step has been the selection of the reference earthquakes by including also local amplification effects. Two events with 50 and 475 years return periods have been chosen as representative, respectively, of a damaging and of a destructive seismic event expected in Potenza. The sites that may exhibit important amplification effects have been identified using the first level method of the TC4 Manual. Damage scenarios of dwelling buildings have been prepared in the fourth step and reported in a GIS. They are relevant to the selected reference earthquakes, taking into account or not site effects. The generally low vulnerability of buildings results in a limited number of damaged buildings for the lower intensity earthquake, and of collapsed buildings, for the higher intensity earthquake. The influence of site effects on the damage distribution is significant. This revised version was published online in July 2006 with corrections to the Cover Date.     

Critical evaluation of strong motion in Kocaeli and Düzce (Turkey) earthquakes

Turkey was struck by two major events on August 17th and November 12th, 1999. Named Kocaeli (M_w=7.4) and Düzce (M_w=7.2) earthquakes, respectively, the two earthquakes provided the most extensive strong ground motion data set ever recorded in Turkey. The strong motion stations operated by the General Directorate of Disaster Affairs, the Kandilli Observatory and Earthquake Research Institute of Bogazici University and Istanbul Technical University have produced at least 27 strong motion records for the Kocaeli earthquake within 200 km of the fault. Kocaeli earthquake has generated six motions within 20 km of the fault adding significantly to the near-field database of ground motions for M_w>=7.0 strike–slip earthquakes. The paper discusses available strong motion data, studies their attenuation characteristics, analyses time domain, as well as spectral properties such as spectral accelerations with special emphasis on fault normal and fault parallel components and the elastic attenuation parameter, kappa. A simulation of the Kocaeli earthquake using code FINSIM is also presented.     

Nonlinear soil response as a natural passive isolation mechanism-the 1994 Northridge, California, earthquake

The spatial relationship between areas with severely damaged (red-tagged) buildings and areas with large strains in the soil (indicated by reported breaks in the water distribution system), observed during the 1994 Northridge earthquake, is analysed. It is shown that these areas can be separated almost everywhere. Minimal overlapping is observed only in the regions with very large amplitudes of shaking (peak ground velocity exceeding about 150 cm s⁻¹). One explanation for this remarkable separation is that the buildings on ‘soft’ soils, which experienced nonlinear strain levels, were damaged to a lesser degree, possibly because the soil absorbed a significant portion of the incident seismic wave energy. As a result, the total number of severely damaged (red-tagged) buildings in San Fernando Valley, Los Angeles and Santa Monica may have been reduced by a factor of two or more. This interpretation is consistent with the recorded peak accelerations of strong motion in the same area. It is concluded that significant reduction in the potential damage to wood frame single family dwellings may be expected in areas where the soil experiences ‘large’ strains (beyond the linear range) during strong earthquake shaking, but not significant differential motions, settlement or lateral spreading, near the surface.     

A rapid seismic safety assessment method for mid-rise reinforced concrete buildings

The majority of existing buildings are not safe against earthquakes in most of the developing countries. Existing building stocks should be assessed with a seismic safety assessment method before a devastating earthquake. Cheaper and quicker rapid seismic safety assessment methods can be used instead of code-based assessment methods to determine the seismic performance of existing buildings. In this study, an approach was introduced to determine the seismic performance of existing mid-rise reinforced concrete buildings with fewer parameters and process steps than code-based detailed assessment procedures. Calibration and regulation of the introduced method were conducted on the 39 collapsed buildings’ projects in 1999 Kocaeli, Turkey, earthquake. Finally, 55 existing buildings located in Eskisehir, Turkey, assessed with this calibrated method and the results were compared with the results of a code-based detailed assessment method; the results showed a very good agreement of about 83%. This study shows that the proposed method can be applied for the determination of the seismic performance of existing mid-rise reinforced concrete buildings quickly and without compromising reliability.     

Joint deconvolution of building and downhole seismic recordings: an application to three test cases

In this study, the joint deconvolution is applied to recordings of three test cases located in the cities of Bishkek, Kyrgyzstan, Istanbul, Turkey, and Mexico City, Mexico. Each test case consists of a building equipped with sensors and a nearby borehole installation in order to investigate different cases of coupling (impedance contrasts) between the building and the soil by analyzing the wave propagation through the building-soil-layers, and hence resolving the soil–structure-interactions. The three installations considering different dynamic characteristics of buildings and soil, and thus, different building-soil couplings, are investigated. The seismic input (i.e., the part of the wave field containing only the up-going waves after removing all down-going waves) and the part of the wave field that is associated with the waves radiated back from the building are separated by using the constrained deconvolution. The energy being radiated back from the building to the soil has been estimated for the three test cases. The values obtained show that even at great depths (and therefore distances), the amount of wave field radiated back by the building to the soil is significant (e.g., for the Bishkek case, at 145 m depth, 10% of the estimated real input energy is expected to be emitted back from the building; for Istanbul at 50 m depth, the value is also 10–15% of the estimated real input energy while for Mexico City at 45 m depth, it is 25–65% of the estimated real input energy). Such results confirm the active role of buildings in shaping the wave field.     

Seismic Vulnerability Assessment of Gravity Load Designed R/C Frames

The seismic vulnerability of some frame structures, typical of existing Reinforced Concrete buildings designed only to vertical loads, has been evaluated. They are representative of building types widely present in the Italian building stock of the last 30 years. A simulated design of the structures has been made with reference to the codes in force, the available handbooks and the current practice at the time of construction. The seismic response is calculated through non linear dynamic analyses with artificial and natural accelerograms. Three main types have been examined: bare frames, regularly infilled frames and pilotis frames. The results show a high vulnerability for the pilotis buildings: they can be assigned to the class B of the European Macroseismic Scale of 1998 (EMS98). On the contrary, a low vulnerability (class D of EMS98) can be attributed to the regularly infilled buildings: in this case collapse can be considered unlikely also with strong earthquakes. An intermediate seismic behavior is shown by buildings without infills, whose vulnerability can be placed between the classes B and C of EMS98. This revised version was published online in July 2006 with corrections to the Cover Date.     

Damages on reinforced concrete buildings due to consecutive earthquakes in Van

Consecutive earthquakes occurred on October 23rd, 2011 in Ercis and on November 9th, 2011 in Edremit that are townships located 90 km and 18 km far from Van city in Turkey, respectively. A total of 28,000 buildings were damaged or collapsed in the city center and the surrounding villages after the Ercis earthquake. This number reached 35,000 after the Edremit earthquake. In the area where the earthquakes occurred, almost all the reinforced concrete buildings were affected.This study presents field observations of damages on reinforced concrete buildings due to the consecutive earthquakes that occurred in Van, Turkey. Damages appearing in the buildings may occur due to several reasons such as site effect, poor construction quality, poor concrete strength, poor detailing in beam-column joints, detailing of stronger beam than column, soft stories, weak stories, inadequate reinforcement, short lap splices, incorrect end hook angle, and short columns. Aftershocks also caused progressive damages on the buildings within 17 days after the earthquakes. According to the results of this study, most of the damaged buildings were not designed and constructed according to the Turkish earthquake code, the so-called Specification for Buildings to be built in Seismic Zones.     

Loss estimation: a powerful tool for risk assessment and mitigation

Earlier loss estimation studies were limited to investigating particular scenarios and were carried out by highly specialized experts. Today, loss estimation techniques are translated into efficient software applications that are accessible by a large constituency of end-users. These techniques offer a high level of analysis sophistication and enable users to perform various ‘if–then’ scenarios to study the sensitivity of the results, to develop a better understanding of the outcomes and to gain insight on the consequences of the findings and decisions. Functionality of loss estimation models has improved significantly due to advances in information technology such as the introduction of Geographical Information Systems (GIS). GIS allows for easy display of input and output (in standard reports and maps) providing a critical functionality for communication of outcomes to emergency planners and decision-makers.     

Fuzzy neural network models for liquefaction prediction

Integrated fuzzy neural network models are developed for the assessment of liquefaction potential of a site. The models are trained with large databases of liquefaction case histories. A two-stage training algorithm is used to develop a fuzzy neural network model. In the preliminary training stage, the training case histories are used to determine initial network parameters. In the final training stage, the training case histories are processed one by one to develop membership functions for the network parameters. During the testing phase, input variables are described in linguistic terms such as ‘high’ and ‘low’. The prediction is made in terms of a liquefaction index representing the degree of liquefaction described in fuzzy terms such as ‘highly likely’, ‘likely’, or ‘unlikely’. The results from the model are compared with actual field observations and misclassified cases are identified. The models are found to have good predictive ability and are expected to be very useful for a preliminary evaluation of liquefaction potential of a site for which the input parameters are not well defined.