**5. Near-real-time hazard and damage distribution maps generated using the data recorded by the strong motion network in Istanbul**

The Hazard module estimates spatially distributed intensity and ground motion parameters (PGA, PGV, Sa, and Sd) using region-specific ground motion prediction equations and gridded shear wave velocity data for a given earthquake magnitude and epicenter. In this section, near-real-time earthquake parameters (PGA, SA02, and SA10) and damage (slight and moderate) distribution maps are generated by utilizing data from Istanbul's strong motion network. The data collected from triggered stations during the Silivri earthquake were used as event data, and distribution maps were plotted based on the GMPE of CY2014 after bias correction of phantom stations by the data utilized from the strong motion network in İstanbul. The source type panel defines the source mechanism associated with the event. For small-magnitude events, the source can be given as a point; for large-magnitude events, the user can specify the source type as a finite fault. The source type was assigned as "point source" since we simulated a moderate earthquake in this case. ELER can use ground motion prediction equations with Vs30 [52, 68–70] as an input parameter to directly calculate the ground motion values at the surface. This feature was employed because the network recorded the event at the surface level. Custom site condition map is used as in the form of Vs30 grids.

**Figures 11**–**13** show the PGA, SA02, and SA10 distributions, respectively. Of 832 stations, 116 detected ground vibrations during the Silivri earthquake. Some data collected from acceleration stations is seen as a clear difference from the estimation of phantom stations. These abrupt and apparent variations in the distribution maps are shown as circles with a dashed line. As expected, the higher ground accelerations are seen in Silivri and Büyükçekmece's coastal region. This condition is also the same for SA02. However, Beylikdüzü also has a high SA10 spectral acceleration as other specified coastal counties. PGA is found to be 0.3496 g, as seen in **Figure 11**.

The building damage was estimated using the spectral acceleration-displacementbased vulnerability assessment methodology. ELER employs a variety of strategies for achieving this goal in the Level 2 (Loss assessment) module. There are currently two options for the seismic demand spectrum: (1) Euro Code 8 spectrum and (2) IBC 2006 spectrum. In this work, the construction of the 5%-damped elastic response spectra is selected as IBC (International Building Code) 2006. [71] provides a general

#### **Figure 11.**

*PGA distribution maps obtained for the Mw 5.8 Silivri earthquake based on the GMPE of CY2014 after bias correction of phantom stations by the strong motion network in İstanbul.*

#### **Figure 12.**

*SA02 spectral acceleration distribution maps obtained for the Mw 5.8 Silivri earthquake based on the GMPE of CY2014 after bias correction of phantom stations by the strong motion network in İstanbul.*

*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 13.**

*SA10 spectral acceleration distribution maps obtained for the Mw 5.8 Silivri earthquake based on the GMPE of CY2014 after bias correction of phantom stations by the strong motion network in İstanbul.*

horizontal elastic acceleration response spectrum. It is defined by (1) spectral accelerations at short period and 1-sec period, respectively, (2) short period and 1-sec period design response spectral accelerations adjusted for the specified site class and damping value (SDS, SD1), (3) corner periods of the constant spectral acceleration region given by T0 = 0.2TS and TS = SD1/SDS, and (4) long-period transition period. It is a regional-dependent parameter, and it is assumed that TL = 5 s herein. In IBC 2006 and NEHRP 2003 Provisions [72], the recommended values for site and damping adjustments are stated. Spectral acceleration values at 0.2 and 1 sec periods are required to generate the IBC-2006 demand spectrum for each geographical unit. The user provides two MATLAB (.mat) files including a grid matrix and a reference vector of 0.2 and 1.0 sec spectral accelerations.

These two spectral acceleration files can be obtained from the previous calculations in Hazard Module or it can be developed by the user in a proper format for Level 2. In this section the spectral accelerations recorded from the strong motion network in İstanbul during the Silivri earthquake were used.

For the estimation of building damage in a Level 2 module, analytical fragility relationships and spectral acceleration-displacement-based vulnerability assessment approaches are applied. In the current case study, Capacity Spectrum Method (CSM-ATC 40) [73] was employed. The CSM is an approximate heuristic method that essentially assumes that a complex non-linear multi-degree-of-freedom system, such as a multi-story building experiencing severe plastic deformations during an earthquake, can be modeled as an equivalent single-degree-of-freedom system with an appropriate level of inelasticity. The simplicity of the procedure is the advantage of the method since no time history analysis is required.

#### **Figure 14.**

*Slight damage distribution map of buildings obtained for the Mw 5.8 Silivri earthquake.*

#### **Figure 15.**

*Moderate damage distribution map of buildings obtained for the Mw 5.8 Silivri earthquake.*

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

The building and population data for Level 2 analysis are grid-based (geocell) urban building and demographic inventory. For building grouping, the European building taxonomy created as part of the EU-FP5 RISK-UE project [66, 74].

Since more significant damage is expected to occur in coastal areas close to the epicenter, only Silivri, Çatalca, Büyükçekmece, Beylikdüzü, Esenyurt, and Avcılar are considered for damage distribution. **Figures 14** and **15** show slight and moderate damage distribution maps of buildings obtained for the *Mw* 5.8 Silivri earthquake using the strong motion network data, respectively. The higher levels of damage are seen in Silivri, partially in the coastal regions of Büyükçekmece, Beylikdüzü, and Avcılar. In addition, a higher damage distribution is observed in Esenyurt than in other hinterland regions. Because extreme and complete damage are not observed, their maps were not created.
