**4.1 A brief information on the** *Mw* **5.8 Silivri earthquake**

On September 26, 2019, at 13:59 Turkish time (GMT +3), an *Mw* 5.8 offshore earthquake struck the Marmara Region, Turkey. This earthquake occurred at a depth of 6.99 kilometers from the ground and is located 21.99 kilometers from the nearest settlement, Silivri, in Istanbul. **Figure 6** depicts the epicenter of the earthquake. 150 aftershocks with magnitudes ranging from 1.0 to 4.1 were recorded between the mainshock and 07:40 on September 27, 2019 [62]. The Silivri earthquake is significant

**Figure 6.** *The epicenter of the Mw 5.8 Silivri earthquake.*

since it is the region's largest earthquake since the 1999 Kocaeli (*Mw* 7.6) and 1999 Düzce (*Mw* 7.1) earthquakes and their aftershocks. For further information, readers are advised to refer a recent paper by [63].

#### **4.2 Earthquake hazard maps during the Silivri earthquake**

The earthquake hazard maps are generated by grid-based distributions of ground motion parameters such as macroseismic intensity, PGA, and spectral acceleration for various vibration periods (SA02, SA10) based on the magnitude and location of the earthquake. This is achieved by making use of ground motion prediction equations (GMPEs) that are appropriate for Istanbul, as well as grid-based local site information such as Vs30 and a fault database. **Figure 7** illustrates the Vs30 map for metropolitan Istanbul. The choice of proper GMPEs can have a significant impact on design and safety evaluation as mentioned before (see Section 2.4). It is well known that attenuation relationships offer a satisfying estimation of ground motion parameters (PGA and SAs) depending on the magnitude and source-to-site distance.

The NGA2014's relationship with regional factors affecting Turkey in the aftermath of the mid-sized Marmara Sea earthquake was examined by [64]. According to their study, it appears that Chiou and Youngs 2014 (CY2014) provides the best fit to local PGA datasets recorded during the 26 September 2019 Silivri earthquake.

The CY2014 GMPE model is appropriate for predicting horizontal ground motion amplitudes associated with earthquakes in active tectonic zones that meet the following criteria [49]: (1) 3.5 ≤ M ≤ 8.5 for strike-slip earthquakes, (2) 3.5 ≤ M ≤ 8.0 for reverse and normal faulting earthquakes, (3) ZTOR ≤ 20 km, (4) 0 ≤ RRUP ≤ 300 km • 180 ≤ VS30 ≤ 1500 m/s. Therefore, in this work, CY2014 GMPE model is used for generating the earthquake hazard maps.

*Urban Damage Assessment after the* Mw *5.8 Silivri Earthquake: The Case of Istanbul City DOI: http://dx.doi.org/10.5772/intechopen.109758*

**Figure 7.** *Vs30 map for metropolitan Istanbul.*

The Modified Mercalli (MMI) conversion has been achieved through the regression relationships developed by [65]. The distribution maps for earthquake parameters (MMI, PGA, SA02, and SA10) are depicted in **Figures 8** and **9**. The European coastal part of Istanbul represents the highest intensity, MMI V (**Figure 7a**). PGA is found 0.1529 g, as seen in **Figure 8b**. Spectral accelerations descend from the coastline to the midland, as seen in **Figure 9**.

#### **4.3 Building damage assessment**

The loss estimation engine for the Level 1 (Loss assessment) module in ELER is based on macroseismic damage estimation tools and aims to analyze both building damage and casualties. In Level 1, the intensity-based empirical vulnerability relationships and casualty vulnerability models based on multiple approaches can be

*Earthquake hazard maps (ground motion distribution) using CY14 GMPE model: (a) MMI, (b) PGA.*

**Figure 9.** *Earthquake hazard maps (ground motion distribution) using CY14 GMPE model: (a) SA02 (b) SA10.*

utilized. The intensity grid should be a MATLAB (.mat) file containing a grid matrix and a reference vector created in the Hazard module. The building database file is essentially a Shapefile (.shp) containing the distribution of buildings within each cell. This file may also include the population of each cell for the purpose of computing casualties. In the absence of a population field in the building database, casualty estimates are approximated using the regional population (obtained from the Land Scan population distribution). The vulnerability-ductility as a MATLAB (.mat) file including a table describing the vulnerability, ductility, t parameter, and replacement cost of each building type. ELER scans the building database for each building type specified in the vulnerability-ductility table and, if found, calculates the damage.

For Level 1 analysis, the building inventory and population data consist of gridbased (geocell) building and population distribution. Based on the Risk UE building taxonomy, the building distribution for the Marmara region is used as sample data. In ELER, Turkey's data is supplied as a model for other regions and nations to develop/ incorporate their inventory data.

Damage estimation involves obtaining a cumulative damage probability using a normal distribution for each building type in Level 1. The damage probability distribution is dependent on the vulnerability and ductility factors of each building [66]. They derived the observed damage-based vulnerability approach known as the macroseismic method from the definition supplied by the European Macroseismic Scale [67] utilizing classical probability theory and fuzzy-set theory. As the purpose of a Macroseismic Scale is to acquire a measure of the earthquake's severity based on the damage sustained by the buildings, the scale itself can be used as a vulnerability model for predicting purposes to provide the likely damage distribution for a given intensity.

As discussed previously, the vulnerability table provides these parameters for each building type. The probability of cumulative damage is discretized to yield the five damage states as D1-slight damage, D2-moderate damage, D3-substantial to heavy damage, D4-very heavy damage, D5-destruction.

Since more significant damage is expected in coastal areas close to the epicenter, only Silivri, Çatalca, Büyükçekmece, Beylikdüzü, Esenyurt, and Avcılar are considered for damage distribution plots. **Figure 10** illustrates building damage distribution created employing the MMI results provided from Hazard Module in ELER. Slightly higher distribution levels of damage are seen in Esenyurt, Avcılar, and Beylikdüzü.

*Urban Damage Assessment after the* Mw *5.8 Silivri Earthquake: The Case of Istanbul City DOI: http://dx.doi.org/10.5772/intechopen.109758*

**Figure 10.**

*Building damage distribution obtained for the Mw 5.8 Silivri earthquake: (a) D1-slight damage, (b) D2-moderate damage.*
