Identification of Elements at Risk for a Credible Tsunami Event for Istanbul

Physical and social elements at risk are identified for a credible tsunami event for Istanbul. For this purpose , inundation maps resulting from probabilistic tsunami hazard analysis for a 10 % probability of exceedance in 50 yr are utilised in combination with the geo-coded inventories of building stock, lifeline systems and demographic data. The built environment on Istanbul's shorelines that is exposed to tsunami inundation comprises residential, commercial, industrial , public (governmental/municipal, schools, hospitals, sports and religious), infrastructure (car parks, garages, fuel stations, electricity transformer buildings) and military buildings , as well as piers and ports, gas tanks and stations and other urban elements (e.g., recreational facilities). Along the Marmara Sea shore, Tuzla shipyards and important port and petrochemical facilities at Ambarlı are expected to be exposed to tsunami hazard. Significant lifeline systems of the city of Istanbul such as natural gas, electricity, telecommu-nication and sanitary and waste-water transmission, are also under the threat of tsunamis. In terms of social risk, it is estimated that there are about 32 000 inhabitants exposed to tsunami hazard.


Introduction
The city of Istanbul is under the threat of earthquakes expected to originate from the Main Marmara branch of the North Anatolian Fault System.In the Marmara region the earthquake hazard reached very high levels with 2 % annual probability of occurrence of a magnitude 7+ earthquake on the Main Marmara Fault (Erdik et al., 2004).Istanbul is the biggest city of the Marmara region as well as of Turkey with its almost 12 million inhabitants.It is home to 40 % of the industrial facilities in Turkey and operates as the financial and trade hub of the country.With the evidence of past earthquakes, the structural reliability of residential and industrial buildings, as well as that of lifelines including port and harbour structures in the country, is questionable (Erdik and Durukal, 2008;Durukal et al., 2008).These facts make the management of earthquake risks imperative for the reduction of physical and socio-economic losses.Yet the assets at risk along the shores of the city make a thorough assessment of tsunami risk essential.Important residential and industrial centres exist along the shores of the Marmara Sea.Particularly along the northern and eastern shores an uninterrupted settlement pattern with industries, businesses, commercial centres and ports and harbours in between is seen (Hancilar et al., 2008).
The population, structures, utilities, systems and socioeconomic activities constitute the "elements at risk" in urban areas.The physical elements are the built environment such as buildings and lifelines.Demographic data represent the social elements at risk.The objective of the present study is to identify the elements at risk based on a probabilistic tsunami hazard assessment for Istanbul carried out by OYO Co. (2007).The paper encompasses three parts.In the first part, tsunamigenic seismic sources for the surrounding seas of Turkey and particularly for the Marmara Sea region are identified and a database for tsunamigenic seismic zonation is provided.In the second part, inundation maps resulting from probabilistic as well as deterministic tsunami hazard assessments are represented.The methodology for the identification of elements at risk and the results are presented in the last part.

Tsunamigenic seismic sources in the Marmara Sea region
There are two potential sources for tsunami generation in the Marmara Sea region: earthquakes and sub-marine landslides.
As future earthquakes are expected to break segments of the North Anatolian Fault System, the possibility that tsunamis could be generated by the co-seismic displacement of the seafloor or by triggered sub-marine landslides should be considered (Hebert et al., 2005).It has been reported that the coasts of the Marmara Sea have been frequently struck by tsunamis; over 40 tsunamis could have occurred in the Marmara Sea between 120 and 1999 AD (Altinok et al., 2001a).
The most recent event identified as a tsunami in the Marmara Sea was triggered by the 1999 Kocaeli Earthquake (Altinok et al., 2001b).
In the present study, only seismic sources are considered.A seismic source zone can be defined as a seismically homogenous area, in which every point within the source zone is assumed to have the same probability of being the epicentre of a future earthquake (Erdik et al., 2000).Seismic source zones can be determined by the help of a seismicity profile and the tectonic regime of the region under consideration.A seismic zonation map of Turkey and neighbouring regions is provided by Erdik et al. (1999Erdik et al. ( , 2000)).This map was adopted for the compilation of an inventory database for tsunamigenic seismic sources in and around Turkey.In the compilation process, the tectonics, seismicity and topography of the study region were checked with related literature (i.e., McKenzie, 1970;Barka and Kadinsky-Cade, 1988;Saroǧlu et al., 1992;Barka and Reilinger, 1997;Armijo et al., 1999).Dominant faulting mechanisms, the dip angle, direction strike and rake angles were obtained through the searches of the CMT catalogue (http://www.globalcmt.org/CMTsearch.html).The Harvard CMT catalogue provides fault plane solutions for earthquakes with magnitudes greater than 5.0 in Turkey starting from 1976.Historical data provided in Ambraseys (2002), Altinok and Ersoy (2000), Altinok et al. (2001aAltinok et al. ( , b, 2003)), Yalciner et al. (2002) and Yolsal et al. (2007) on the past tsunami occurrences were utilised in the identification of seismic zones that have potential for tsunami generation.Only seismic zones that have produced earthquakes followed by a tsunami were included in the database.Hence, 26 zones were identified in total.Table 1 provides the minimum and maximum depths, expected seismogenic thicknesses and the associated faulting mechanisms for each zone.The zones are represented by polygons with up to eight vertexes with their corresponding geographical coordinates, i.e., latitudes and longitudes.The identified tsunamigenic seismic zones in the Marmara Sea region and partly in the Euro-Mediterranean and Black Sea regions are depicted in Fig. 1.For verification purposes, locations of the tsunamigenic sources, i.e., epicentres of historical tsunamigenic earthquakes, as provided by Y. Altinok (personal communication, 2010), are also illustrated in Fig. 1. Figure 2 represents a closer view on satellite imagery for the tsunamigenic seismic zones, namely Zone 16 and 17, in the Marmara Sea region.In this figure, an idealised fault segmentation model (Erdik et al., 2004) of the region as well as the epicentres of historical tsunamigenic earthquakes are also illustrated.

Inundation maps for Istanbul
Tsunami inundation maps show coastal land areas that become submerged during a tsunami.The area of land subject to inundation is a factor of (EERI, 2005): -distance of shoreline from the tsunami-generating source; -earthquake magnitude (primarily related to the earthquake source); -duration and periods of the waves; -run-up elevations (height above sea level likely to be flooded); -tide level at time of occurrence; -direction of shore with respect to propagated waves; -topography of the seabed in the vicinity (bathymetry).
This section provides a brief summary of the probabilistic and deterministic tsunami hazard studies where the tsunamigenic seismic sources described in the previous section were utilised.Probabilistic inundation maps for a credible tsunami event on the coastal lines of Istanbul are also represented.Probabilistic tsunami hazard analyses and inundation mapping for Istanbul were performed by OYO Co. (2007) for Istanbul Metropolitan Municipality in the framework of the project entitled "Simulation and Vulnerability Analysis of Tsunamis Affecting the Istanbul Coasts".For the assessment of tsunami hazard for a 10 % probability of exceedance in 50 yr, OYO Co. conducted 42 simulations by utilising 150-m-resolution bathymetry-topography data for the Marmara Sea region and 50-m-resolution was adopted for the better characterisation of the area covering the fault segments which were considered to produce the expected magnitude 7+ earthquake.In each simulation, not only different rupturing cases of individual fault segments and/or simultaneous rupturing of several segments were considered, but also cascade and no cascade models were taken into account.The reader is referred to OYO Co. ( 2007) and http://www.ibb.gov.tr/en-US/SubSites/IstanbulEarthquake/Pages/TsunamiHazardAnalysis.aspx for further information on the probabilistic hazard study.Maximum run-up heights and maximum inundation depths as the outcome among 42 simulations are mapped in Figs. 3 and  4, respectively.The results show that the eastern coasts of Istanbul are more hazardous than its western coastline.The highest run-up height exceeding 9 m is expected in the Prince Islands.Kartal and Kadıköy are the next hazardous areas on the Asian side.On the European side, run-up heights up to 3 to 4 m are expected in Bakırköy and Zeytinburnu districts.The inundation at the south of the Küc ¸ükc ¸ekmece Lake is remarkable.The maximum inundation distance from the coast reaches about 600 m.The coastline of Kadıköy and the coast from Kartal to Tuzla are also expected to suffer run-ups for 100 to 300 m.
Piatenesi and Romano (2009) and Roger et al. ( 2009) conducted deterministic tsunami hazard assessments for Istanbul within the activities of the EC FP6 project entitled "Tsunami Risk and the Strategies for the European Region-TRANSFER (http://www.transferproject.eu/)".Piatenesi and Romano ( 2009) studied a worst-case scenario event to take place in zone Z16 (see Table 1) on fault segment S6 (see Fig. 2).They reported that the eastern shores of Istanbul will likely be more affected than its western shores.The wave heights on the eastern shorelines reach  2009) studied five earthquake scenario cases for the modelling of tsunami hazard in Istanbul.They selected fault lines within zone Z16 corresponding to the fault segments S5, S6, S7, S8, S9 and S10 (see Fig. 2).They found that the whole shoreline of Istanbul is exposed to important tsunami heights (often exceeding 2 m) and the eastern shores are more impacted than the western shores.They also identified the most hazardous areas as the area between the Marmara Sea and the Küc ¸ükc ¸ekmece Lake, and the coasts along the Kartal-Pendik-Tuzla line exposed to tsunami heights possibly greater than 3 m and to important local inundations.It is seen that both deterministic tsunami hazard assessments for Istanbul produce inundation results highly comparable to those of the probabilistic analyses.

Exposure assessment
Methods for tsunami risk assessment generally include hazard identification and characterisation, assessment of exposure data and risk characterisation.Risk is defined as the probability of harmful consequences, or expected losses (deaths, injuries, property, livelihoods, economic activity disrupted or environment damaged) resulting from interactions between hazards and vulnerable conditions determined by physical, social, economic and environmental factors or processes, which increase the susceptibility of a community to the impact of hazards (UN/ISDR, 2004).Tobin and Montz (1997) give a definition of exposure as the measure of a population at risk.This definition typically refers to the spatial coincidence of a resource (e.g., a structure) and a hazard (e.g., tsunami).It is a spatial attribute and does not include the quality of the resource in question (e.g., building code level of a structure) or efforts already in place to minimize future losses (e.g., flood insurance, evacuation routes) (Wood and Stein, 2001).Elements at risk are the social and physical elements exposed to a hazard and usually include population, buildings, lifeline systems, transportation systems and infrastructures.For a general building stock the following parameters affect the damage and loss characteristics: structural (system, height and building practices), non-structural elements and occupancy (such as residential, commercial and governmental) (Hancilar et al., 2010).Comprehensive information on tsunami risk assessment methodologies can be found in Jelinek and Krausmann (2008); they denote that the literature on tsunami risk assessment is very limited compared to other natural hazard risk such as earthquakes, floods or landslides.In this section, the methodology for the identification of elements at risk is described and the results are presented.

Methodology
Tsunami vulnerability of the society as well as of the built environment has just been made evident again by the catastrophic tsunami which struck the Japanese coastlines after the Great East Japan Earthquake of moment magnitude 9.0 on 11 March 2011.As of 16 May 2011 the Japanese government communicated that the number of fully or partially collapsed buildings, essentially all due to the tsunami, is 126 800.According to Chock (2011), the cost of the damage and economic losses had reached about 309 billion US Dollars and the total number of fatalities and missing people was estimated at about 24 000.Chock (2011) also reported that the tsunami had imposed different types of loads on the buildings, including hydrostatic and hydrodynamic forces, debris damming and debris impact forces and scouring effects, which were sufficient to cause structural failures of low-to mid-rise buildings of any structural material.The need for credible fragility models and laboratory data to understand the interaction of tsunami with the built environment has been mentioned by different researchers and studies, e.g., Grundy et al. (2005); Bernard et al. (2007).Dall 'Osso et al. (2009) stated that no robust, well-constructed and validated building fragility model for assessing the vulnerability of buildings to tsunami had been developed yet.One of the recent studies on the derivation of structural fragility curves by the help of visual inspection of high-resolution satellite images taken before and after the 2004 Indian Ocean tsunami has been carried out by Suppasri et al. (2011).In order to make a preliminary investigation of the vulnerability of coastal areas to tsunami, a model, which is called "Papathoma tsunami vulnerability assessment model-PTVAM", was developed by Papathoma et al. (2003).The PTVAM was organised within a GIS framework to allow rapid data entry and visualisation of changing vulnerability by considering the production of new maps and was designed to be sensitive and capable of examining vulnerability at high resolution scales, i.e., at building to building scale (Dominey-Howesi and Papathoma, 2007).In this model a number of parameters/attributes which influence the vulnerability of buildings to tsunamis are identified, including but not limited to: number of stories, description of ground floor, building material, construction year and design code.Vulnerability of individual buildings is evaluated by considering the contributions made by those attributes.Concerning the social vulnerability, PTVAM estimates the number of people per building by taking into account population densities during the night and day times as well as in the summer and winter.The PTVAM requires that once data for physical and social elements at risk with pre-defined attributes are ready, they need inputting into the GIS environment and merged with inundation maps.
For the identification of socio-economic elements within the inundation zones on the shorelines of Istanbul, a similar methodology to PTVAM was followed.First, the building inventory data were classified and a unified database was compiled in GIS environment.Second, the unified inventory database was combined with the inundation maps and, buildings lying within the inundated areas were counted.The GISbased data for the building stock of Istanbul were obtained from the Istanbul Metropolitan Municipality (IMM-2009).Inundation maps resulting from the probabilistic tsunami hazard study by OYO Co. (2007) were used (Figs. 3 and 4).
The building inventory data were processed and classified in terms of structural system types, number of stories, building usage functions and existence of basement floors.The classification includes: -Structural System Types: reinforced concrete, masonry, steel and precast -Number of Stories: low-rise (1-4), mid-rise (5-8) and high-rise (>8) -Building Function and Usage: residential, commercial, public (schools, hospitals, governmental buildings, religious buildings, sports facilities), industrial, infrastructure (car parks, garages, fuel stations, electricity transformer buildings) -Existence of Basement Floor(s) According to this classification, low-rise and mid-rise reinforced concrete frame buildings constitute about 75 % of the building stock in Istanbul.The same building taxonomies utilised in the studies of seismic loss estimation for Istanbul (KOERI-2002, IMM-2009and Erdik et al., 2011) were adopted in the classification presented in this study.In this way, and considering that the credible tsunami will strike the city after an earthquake and damage to the buildings will primarily be associated with damage due to seismic action, it is aimed to provide a consistent database for future steps of tsunami loss assessment, i.e., computation of tsunami vulnerability functions.Although seismic fragility/vulnerability functions and conversion rates, in terms of repair-cost ratios, of physical damage to financial loss are available for the  buildings in Istanbul, there are currently no tsunami vulnerability and loss functions, neither empirically nor analytically derived.Since the present study only deals with exposure assessment and there is no estimation of physical tsunami damages to buildings, total monetary value of the exposed buildings was calculated as an indicative figure.Based on number of stories and structural system classification, a unit value in terms of construction costs by taking into account the construction practices in Istanbul, as provided in Durukal et al. (2006), was assumed for each class and then, it was multiplied by the number of buildings.
For the estimation of the social exposure, the number of people was calculated based on the number of exposed buildings (IMM-2009).The data from the 2009 Population Census were utilised in combination with the database of the Istanbul Metropolitan Municipality, IMM 5747 Mahalle-Koy-Nufus -Population Data for Districts, Sub-Districts and Villages.These data were spatially joined with the number of floors at each exposed building to estimate the number of people residing in each building.

Results
The built environment on the shorelines of Istanbul exposed to tsunami inundation, that results from a probabilistic hazard assessment for 10 % probability of exceedance in 50 yr, consists of residential, commercial, industrial, public (governmental/municipal, schools, hospitals, sports and religious), infrastructure (car parks, garages, fuel stations, electricity transformer buildings) and military buildings as well as piers and ports, gas tanks and stations, urban elements (recreational facilities).In the forthcoming sections, spatial distributions of the physical elements, i.e., buildings, lifeline systems, piers and ports, and other buildings and structures, as well as the population exposed to tsunami hazard are mapped.A preliminary estimation of the monetary values of the exposed buildings is provided.

Buildings and population
Total number of the exposed buildings within the inundation zone was estimated at 4922.The breakdown of the buildings, with respect to previously defined classification system, is given in Table 2.This table includes only the buildings for which at least two attributes of the classification system are known.The estimated total monetary value of the exposed buildings is 365 million Euros.The pie chart in Fig. 5 represents the distribution of this value for different building classes.The number of exposed people was estimated at about 32 000.The spatial distribution of the number of exposed population is shown in Fig. 6.The building stock along the inundation zone consists mostly of residential buildings.The spatial distribution of residential buildings within the inundation zone is shown in Fig. 7.The majority of the residential buildings are located on the Prince Islands, the east coast of the Marmara Sea as well as in Büyükc ¸ekmece district on the western coasts.Commercial buildings are the second largest group in the building stock and they are located mostly on the eastern coasts of the Marmara Sea.The spatial distribution of commercial buildings within the inundation zone is shown in Fig. 8.The eastern coasts of the Marmara Sea as well as Beyoǧlu, Fatih and Bes ¸iktas ¸districts on the European side are dense locations for public buildings.The distribution of public buildings within the inundation zone is shown in Fig. 9.The industrial buildings are mostly located in Tuzla Bay area and their spatial distribution within the inundation zone is shown in Fig. 10.The infrastructure buildings identified within the inundation zone spread throughout the city; their spatial distribution within the inundation zone is given in Fig. 11.

Piers and ports, other buildings and structures
It was identified that there are 44 small and large piers and ports within the inundation zone.Besides, two important commercial/industrial facilities are exposed to tsunami inundation: -The Ambarlı Port and Petrochemical Filling Facilities (Fig. 12): it should be noted that Ambarlı is the second largest port after Pire in the region of the Eastern Mediterranean and Black Sea.
-The Tuzla Shipyards (Fig. 12): there is a total of 36 companies actively working in Tuzla Bay area.
It was also identified that there are 17 fuel stations and tanks and 198 military buildings within the inundation zone.The military buildings are located next to the Tuzla Bay area.
Ambarlı and Tuzla are the two areas in Istanbul already recognised as particularly vulnerable to earthquake hazard (Durukal et al., 2008).

Lifeline systems
The inventory database for the lifeline systems of Istanbul, i.e., natural gas network, sanitary and waste-water transmission systems, electricity network, telecommunication stations and transportation network, was the one used in KOERI-2002 study.It was not possible to spatially join the inventory data with the inundation maps because of different formats and conversion errors.In order to present indicative maps of exposed elements, a uniform, 600 m-width buffer zone which corresponds to the maximum inundation distance from the coastline was shown with the spatial distribution of the components of each lifeline system.It can be seen that transmission lines and compressor stations of the natural gas network spread throughout both the western and eastern coastlines (Fig. 13).Most part of the transformer stations of electricity network exposed to tsunami hazard is located along the eastern coasts (Fig. 13).The reservoir stations of the sanitary and waste-water transmission systems are mostly located on the eastern shorelines as well as on the Prince Islands which are very close to the Main Marmara Fault System (Fig. 14).Some of the telecommunication buildings and secondary roads as well as some parts of the motorway lying between the Marmara Sea and Küc ¸ükc ¸ekmece Lake are also identified within the inundation zone (Fig. 15).

Concluding remarks
Tsunamigenic seismic sources for the surrounding seas of Turkey and particularly for the Marmara Sea region are provided and identification of physical and social elements at risk for a credible tsunami event for Istanbul is addressed.Based on the results of previous probabilistic and deterministic tsunami hazard assessments where tsunamigenic seismic sources provided herein were considered, it can be said that the eastern coasts of Istanbul are more hazardous than the western coastlines.At this point, it should be kept in mind that numerical modelling and hazard assessment considering sub-marine landslides, which are the non-seismic tsunamigenic sources in the study region, might result in different inundations on the shorelines of the city.
The built environment on the inundated coasts of Istanbul comprises residential, commercial, industrial, public (governmental/municipal, schools, hospitals, sports and religious), infrastructure (car parks, garages, fuel stations, electricity transformer buildings) and military buildings as well as piers and ports, gas tanks and stations, urban elements (recreational facilities).Total number of the exposed buildings and the number of inhabitants are estimated at 4922 and 32 000, respectively.The estimated total monetary value of the exposed buildings is 365 million Euros.Low-rise and mid-rise reinforced concrete frame buildings constitute about 75 % of the building stock in Istanbul.Regarding the relatively better structural resistance of those buildings comparing to wooden or adobe constructions, this can be considered as an advantage under the actions to be imposed by the credible tsunami event.On the other hand, a relatively high number of people reside/work in those multi-story buildings; this can increase the social vulnerability, especially for a daytime tsunami event.Significant lifeline systems for the city of Istanbul such as natural gas, electricity, telecommunication and sanitary and waste-water transmission as well as the important port and petrochemical facilities at Ambarlı and the Tuzla shipyards are also under the threat of tsunami inundation.It can be expected that economic losses due to business interruption and non-structural damage, i.e., damage to equipment and contents, in the industrial and infrastructure facilities will be higher than those resulting from structural damages in the buildings.
The aim of the project entitled "Simulation and Vulnerability Analysis of Tsunamis Affecting the Istanbul Coasts" by Istanbul Metropolitan Municipality was to produce a tsunami hazard map in order to assess tsunami risk and its impacts on Istanbul and to perform the necessary analyses for proper land use and development of strategies for mitigation of the possible effects of tsunami on Istanbul shorelines.The present study takes the efforts one step further by identifying the elements at risk in the framework of EC FP6 project entitled "Tsunami Risk and the Strategies for the European Region-TRANSFER".The study also underlines the significance of well-classified inventory databases in the risk assessment.The provided unified inventory data can be used for further steps in a tsunami loss assessment study for Istanbul.The exposed building data might be utilised for the derivation of tsunami fragility functions.The results of the study certainly help create more public awareness, might contribute to the risk mitigation efforts and may also provide useful information for the development of proper land use plans.

Figure 1 .Fig. 1 .
Figure 1.Tsunamigenic seismic zonation map for Turkey.Zones are shown by green 2 polygons and red dots represent the individual point sources as the epicentres of the 3 earthquakes that produced tsunamis in the past.4 Fig. 1.Tsunamigenic seismic zonation map for Turkey.Zones are shown by green polygons and red dots represent the individual point sources as the epicentres of the earthquakes that produced tsunamis in the past.

Figure 2 .Fig. 2 .
Figure 2. Tsunamigenic seismic zones (Zone 16 and 17) in the Marmara Sea region.Red dots 2 represent the epicentres of historical earthquakes and the numbers just below or above them 3 stand for the event's years.The fault segmentation model developed by Erdik et al. (2004) for 4 the North Western, South Western and Marmara Sea strands of the North Anatolian Fault 5 System is shown by dashed lines with the numbers of each segment (e.g.S1, S2 etc.).6 7 Fig. 2. Tsunamigenic seismic zones (Zone 16 and 17) in the Marmara Sea region.Red dots represent the epicentres of historical earthquakes and the numbers just below or above them stand for the event's years.The fault segmentation model developed by Erdik et al. (2004) for the North Western, South Western and Marmara Sea strands of the North Anatolian Fault System is shown by dashed lines with the numbers of each segment (e.g., S1, S2 etc.).

1Figure 3 .Fig. 3 .
Figure 3. Maximum run-up heights resulting from the probabilistic simulations for 10% probability of exceedance in 50 years.2 Fig. 3. Maximum run-up heights resulting from the probabilistic simulations for 10 % probability of exceedance in 50 years.

1Figure 4 .Fig. 4 .
Figure 4. Maximum inundation depths resulting from the probabilistic simulations for 10% probability of exceedance in 50 years.2 Fig. 4. Maximum inundation depths resulting from the probabilistic simulations for 10 % probability of exceedance in 50 years.

Figure 6 .Fig. 6 .
Figure 6.Spatial distribution of the number of people residing in the exposed buildings.2 3 Fig. 6.Spatial distribution of the number of people residing in the exposed buildings.

Figure 7 .Fig. 7 .
Figure 7. Spatial distribution of the residential buildings at risk: Buildings highlighted by red.Embedded frame: A closer view of the inundated 2 areas where the residential buildings are densely located.3 Fig. 7. Spatial distribution of the residential buildings at risk: buildings highlighted by red.Embedded frame: a closer view of the inundated areas where the residential buildings are densely located.

1Figure 8 .Fig. 8 .Figure 9 .
Figure 8. Spatial distribution of the commercial buildings at risk: Buildings highlighted by red.Embedded frame: A closer view of the inundated 2 areas where most of the commercial buildings are located.3 Fig. 8. Spatial distribution of the commercial buildings at risk: buildings highlighted in red.Embedded frame: a closer view of the inundated areas where most of the commercial buildings are located.

Fig. 9 .
Fig. 9. Spatial distribution of the public buildings at risk: buildings highlighted in red.Embedded frame: a closer view of the inundated areas where most of the public buildings are located.

Figure 10 .Fig. 10 .
Figure 10.Spatial distribution of the industrial buildings at risk: Buildings highlighted by red.Embedded frame: A closer view of the inundated 2 areas where most of the industrial buildings are located.3 Fig.10.Spatial distribution of the industrial buildings at risk: buildings highlighted in red.Embedded frame: a closer view of the inundated areas where most of the industrial buildings are located.

Figure 11 .
Figure 11.Spatial distribution of the infrastructure buildings at risk: Buildings highlighted by red.Embedded frame: A closer view of the 2 inundated areas where most of the infrastructure buildings are located.3

Fig. 11 .
Fig. 11.Spatial distribution of the infrastructure buildings at risk: buildings highlighted in red.Embedded frame: a closer view of the inundated areas where most of the infrastructure buildings are located.

1Figure 12 .Fig. 12 .Figure 13 .Fig. 13 .
Figure 12.Inundated areas within two important industrial/commercial zones of Istanbul.Left panel: Ambarlı Port area and its vicinity, Right 2 panel: Tuzla Bay area and the Tuzla shipyards.3 Fig.12. Inundated areas within two important industrial/commercial zones of Istanbul.Left panel: Ambarlı Port area and its vicinity, Right panel: Tuzla Bay area and the Tuzla shipyards.

Figure 14 .Fig. 14 .Figure 15 .Fig. 15 .
Figure14.Sanitary water transmission system (left) and waste-water transmission system (right) with a uniform, 600m-width buffer zone which 2 corresponds to the maximum inundation distance from the coastline (adopted fromKOERI, 2002).3 Fig.14.Sanitary water transmission system (left) and waste-water transmission system (right).A uniform, 600 m-width buffer zone which corresponds to the maximum inundation distance from the coastline is also shown (modified afterKOERI, 2002).

Table 1 .
Tsunamigenic seismic zones in and around Turkey: a summary table representing kinematic parameters and geographical coordinates for each zone.

Table 2 .
Number of buildings within the inundation zone with respect to structural system types, height-wise classification and usage functions.